From rendell@agbus.com Mon, 30 Apr 2001 11:56:14 -0700 Date: Mon, 30 Apr 2001 11:56:14 -0700 From: Rendell Requiro rendell@agbus.com Subject: AS400 V4R4M0 -- 6.07F FILE statement NEED option This is a multi-part message in MIME format. ------=_NextPart_000_006F_01C0D16C.8EB6ED40 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable Are there any limitations when using the NEED option on a DESIGNER file = statement? Sometimes the following code snippet works, other times no = DESIGNER file records are created. my file statement is =20 file TRANSFER-IN designer need 1 during a DESIGNER procedure I do a GET to the DESIGNER file TRANSFER-IN get TRANSFER-IN viaindex (...) via (...) using (...) optional=20 if not accessok then begin put TRANSFER-IN end Most of the time TRANSFER-IN records are created, maybe 20-30% of the = time, no records are created. I am baffled -- any ideas why this might = occur? Rendell Requiro AG Business Management, Inc. rendell@agbus.com 831-455-2355 x101 voice 831-455-2264 fax www.agbus.com ------=_NextPart_000_006F_01C0D16C.8EB6ED40 Content-Type: text/html; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable
Are there any limitations when using the NEED option = on a=20 DESIGNER file statement?  Sometimes the following code snippet = works, other=20 times no DESIGNER file records are created.
 
my file statement is 
 
    file TRANSFER-IN designer need=20 1
 
during a DESIGNER procedure I do a GET to the = DESIGNER file=20 TRANSFER-IN
 
    get TRANSFER-IN viaindex (...) = via (...)=20 using (...) optional
    if not accessok
    then begin
          &nbs= p;   =20 put TRANSFER-IN
          =20 end
 
Most of the time TRANSFER-IN records are created, = maybe 20-30%=20 of the time, no records are created.  I am baffled -- any ideas why = this=20 might occur?
 
Rendell Requiro
AG Business Management, = Inc.
rendell@agbus.com
831-455-2355&n= bsp; =20 x101 =20 voice
831-455-2264        &nbs= p;   =20 fax
www.agbus.com
------=_NextPart_000_006F_01C0D16C.8EB6ED40-- From oran@intertechsystems.com Mon, 30 Apr 2001 13:34:45 -0600 Date: Mon, 30 Apr 2001 13:34:45 -0600 From: Oran Shapitka oran@intertechsystems.com Subject: AS400 V4R4M0 -- 6.07F FILE statement NEED option This is a multi-part message in MIME format. ------=_NextPart_000_00A9_01C0D17A.51FE00B0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: 7bit Without looking at the code and the data, it would be difficult to know for sure. Knowing that the use of a designer file stipulates that all read\write activity is the responsibility of the programmer, I would guess that the NEED option would not doing anything for you (I know I should look this up in the manual, but after 17 years, you kind of know how the product behaves.) I would guess that your record complex has not had a changed from the default values and therefore your record is not being written out. I would take one of the fields on the file and change it to a bogus value, then set it back again. This will force the record complex to be in a state to be written. Also read in the manual about the reset command if you plan on writing multiple records out on the file. Hope this helps to point you in the right direction for solving the problem. Oran Warning: The advice and or opinions stated above may be worth exactly the compensation received for providing the same. -----Original Message----- From: powerh-l-admin@cube.swau.edu [mailto:powerh-l-admin@cube.swau.edu]On Behalf Of Rendell Requiro Sent: Monday, April 30, 2001 12:56 PM To: powerh-l@lists.swau.edu Subject: AS400 V4R4M0 -- 6.07F FILE statement NEED option Are there any limitations when using the NEED option on a DESIGNER file statement? Sometimes the following code snippet works, other times no DESIGNER file records are created. my file statement is file TRANSFER-IN designer need 1 during a DESIGNER procedure I do a GET to the DESIGNER file TRANSFER-IN get TRANSFER-IN viaindex (...) via (...) using (...) optional if not accessok then begin put TRANSFER-IN end Most of the time TRANSFER-IN records are created, maybe 20-30% of the time, no records are created. I am baffled -- any ideas why this might occur? Rendell Requiro AG Business Management, Inc. rendell@agbus.com 831-455-2355 x101 voice 831-455-2264 fax www.agbus.com ------=_NextPart_000_00A9_01C0D17A.51FE00B0 Content-Type: text/html; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable
Without looking at the code and the data, it = would be=20 difficult to know for sure.  Knowing that the use of a designer = file=20 stipulates that all read\write activity is the responsibility of the = programmer,=20 I would guess that the NEED option would not doing anything for you (I = know I=20 should look this up in the manual, but after 17 years, you kind of know = how the=20 product behaves.)  I would guess that your record complex has not = had a=20 changed from the default values and therefore your record is not being = written=20 out.  I would take one of the fields on the file and change it to a = bogus=20 value, then set it back again.  This will force the record complex = to be in=20 a state to be written.  Also read in the manual about the reset = command if=20 you plan on writing multiple records out on the = file.
 
Hope=20 this helps to point you in the right direction for solving the=20 problem.
 
Oran
 
Warning:  The advice and or = opinions stated=20 above may be worth exactly the compensation received for=20 providing the same.
 
-----Original=20 Message-----
From: powerh-l-admin@cube.swau.edu=20 [mailto:powerh-l-admin@cube.swau.edu]On Behalf Of Rendell=20 Requiro
Sent: Monday, April 30, 2001 12:56 PM
To:=20 powerh-l@lists.swau.edu
Subject: AS400 V4R4M0 -- 6.07F FILE = statement=20 NEED option

Are there any limitations when using the NEED option = on a=20 DESIGNER file statement?  Sometimes the following code snippet = works, other=20 times no DESIGNER file records are created.
 
my file statement is 
 
    file TRANSFER-IN designer need=20 1
 
during a DESIGNER procedure I do a GET to the = DESIGNER file=20 TRANSFER-IN
 
    get TRANSFER-IN viaindex (...) = via (...)=20 using (...) optional
    if not accessok
    then begin
          &nbs= p;   =20 put TRANSFER-IN
          =20 end
 
Most of the time TRANSFER-IN records are created, = maybe 20-30%=20 of the time, no records are created.  I am baffled -- any ideas why = this=20 might occur?
 
Rendell Requiro
AG Business Management, = Inc.
rendell@agbus.com
831-455-2355&n= bsp; =20 x101 =20 voice
831-455-2264        &nbs= p;   =20 fax
www.agbus.com
------=_NextPart_000_00A9_01C0D17A.51FE00B0-- From Roger.Kesterson@mtdsw.com Mon, 30 Apr 2001 13:50:29 -0700 Date: Mon, 30 Apr 2001 13:50:29 -0700 From: Kesterson, Roger Roger.Kesterson@mtdsw.com Subject: Abnormal Termination of a Quick Screen in Batch PH7.10.G2 Alpha OpenVMS 7.1 ----------------- Fellow PH Listers (no reference to Red Dwarf intended): We have some Quick programs that execute in batch, using a WHILE loop in the INITIALIZE procedure. QKOUT is sent to a file, but QKIN is not redefined in any way. One of these programs just terminated, and the only message in the QKOUT file is: "The limit for backup movements (\\) has been reached." Does any of you have any idea what might cause this? Thanks, Roger Kesterson Senior Programming Analyst MTD Southwest, Inc. 550 N. 54th Street Chandler, AZ 85226 Phone: 480-961-1002 x4357 Fax: 480-940-5035 From GGrimm@pcc-structurals.com Mon, 30 Apr 2001 16:40:25 -0700 Date: Mon, 30 Apr 2001 16:40:25 -0700 From: GGrimm@pcc-structurals.com GGrimm@pcc-structurals.com Subject: Delete Procedure On some situation (yet to be ascertained), a partial update occurs about 5% of the time when updating cached records if record(s), have been deleted with a range of numbers, for example: D-9/10. Has anyone an idea what could cause this from the below snippet without having to see the entire code? The data base is Oracle 8.06. Could changing a field in a deleted record after the delete cause a problem? Following is a snippet of the code, primary and three secondary files, no DELETE files, secondary retrieval done in FIND procedure with no records is possible on any of the three secondaries. The absence of secondary(s) is not the cause of the partial Update that happens. --------- FILE TIME_ATTEND PRIMARY OCCURS 13 NOITEMS CACHE 255 ACCESS VIA EMPLOYEENO, SHIFT_DATE, SHIFT_NO, ENTITY_CODE & USING EMPLOYEENO OF EMPLOYEES, T_SHIFT_DATE, & T_SHIFT, TIME_ATTEND_ENTITY & ORDERBY EMPLOYEENO, SHIFT_DATE, SEQ_NUM ACCESS VIA SHIFT_DATE, DEPT_CODE, SHIFT_NO, ENTITY_CODE & USING T_SHIFT_DATE, T_REQ_DEPT, T_SHIFT, TIME_ATTEND_ENTITY & ORDERBY EMPLOYEENO, SHIFT_DATE, SEQ_NUM FILE TIME_ATTEND_2 SECONDARY OCCURS WITH TIME_ATTEND NOITEMS FILE TIME_ATTEND_2 SECONDARY OCCURS WITH TIME_ATTEND NOITEMS & ALIAS TIME_ATTEND_2IN FILE TIME_ATTEND SECONDARY OCCURS WITH TIME_ATTEND NOITEMS & ALIAS TIME_ATTEND_IN . . . . . PROCEDURE DELETE BEGIN DELETE TIME_ATTEND_2IN DELETE TIME_ATTEND_IN DELETE TIME_ATTEND_2 DELETE TIME_ATTEND END From david.followes@ceo.woll.catholic.edu.au Tue, 1 May 2001 15:52:51 +1000 Date: Tue, 1 May 2001 15:52:51 +1000 From: David Followes david.followes@ceo.woll.catholic.edu.au Subject: Cognos NETD Does anyone know if there is a way to debug the NETD run as a service? Or what information it uses/needs from the registry or Cognos.ini file or ......??? We are trying to run PH 8.21D1 on Win 2000 Advance server SP1 and running Axiant thin client. When we start NETD as a service it runs for about 30 - 40 second before stopping again. The event log contains an error message stating "The Cognos Netd for NT service as reported an invalid current state 0." If we try to test the connection with Netinfo we get successful responses up until the service stops. Th interactive "Netd Console" stays working and tests with Netinfo perform correctly forever. As this isn't an ideal way to run a production server can anyone shed any light on this problem?? Regards David From t.e.burke@dundee.ac.uk Tue, 1 May 2001 13:03:24 +0100 Date: Tue, 1 May 2001 13:03:24 +0100 From: Terrence Burke t.e.burke@dundee.ac.uk Subject: Delete Procedure On 30 Apr 2001, at 16:40, GGrimm@pcc-structurals.com wrote: > On some situation (yet to be ascertained), a partial update occurs > about 5% of the time when updating cached records if record(s), have > been deleted with a range of numbers, for example: D-9/10. You are using a single file to provide BOTH the Primary and Secondary records for the screen. You have said nothing about the indexes, apart from listing [some of] the segments of two of them. Can you be certain that any of the records selected as Primary by the FIND procedure will NOT also be identified as a Secondary record for some other Primary record ? Could any record be selected as Secondary for two or more of the Primary records ? In either case, you will inevitably have update errors on occasions. Do you ever have any errors if the Primary file occurs just ONCE ? Regards Terry Burke > Has anyone an idea what could cause this from the below snippet > without having to see the entire code? The data base is Oracle 8.06. > Could changing a field in a deleted record after the delete cause a > problem? > > Following is a snippet of the code, primary and three secondary files, > no DELETE files, secondary retrieval done in FIND procedure with no > records is possible on any of the three secondaries. The absence of > secondary(s) is not the cause of the partial Update that happens. > > --------- > FILE TIME_ATTEND PRIMARY OCCURS 13 NOITEMS CACHE 255 > ACCESS VIA EMPLOYEENO, SHIFT_DATE, SHIFT_NO, ENTITY_CODE & > USING EMPLOYEENO OF EMPLOYEES, T_SHIFT_DATE, & > T_SHIFT, TIME_ATTEND_ENTITY & > ORDERBY EMPLOYEENO, SHIFT_DATE, SEQ_NUM > ACCESS VIA SHIFT_DATE, DEPT_CODE, SHIFT_NO, ENTITY_CODE & > USING T_SHIFT_DATE, T_REQ_DEPT, T_SHIFT, TIME_ATTEND_ENTITY & > ORDERBY EMPLOYEENO, SHIFT_DATE, SEQ_NUM > FILE TIME_ATTEND_2 SECONDARY OCCURS WITH TIME_ATTEND NOITEMS > FILE TIME_ATTEND_2 SECONDARY OCCURS WITH TIME_ATTEND NOITEMS & > ALIAS TIME_ATTEND_2IN > FILE TIME_ATTEND SECONDARY OCCURS WITH TIME_ATTEND NOITEMS & > ALIAS TIME_ATTEND_IN > . . . . . > PROCEDURE DELETE > BEGIN > DELETE TIME_ATTEND_2IN > DELETE TIME_ATTEND_IN > DELETE TIME_ATTEND_2 > DELETE TIME_ATTEND > END > > > > > > > > = = = = = = = = = = = = = = = = = = = = = = = = = = = = > Mailing list: powerh-l@lists.swau.edu > Subscribe: "subscribe" in message body to powerh-l-request@lists.swau.edu > Unsubscribe: "unsubscribe" in message body to powerh-l-request@lists.swau.edu > http://lists.swau.edu/mailman/listinfo/powerh-l > This list is closed, thus to post to the list you must be a subscriber. Dr T E Burke Admin Information Systems University of Dundee Dundee DD1 4HN Phone 01382 344022 From jthunter@nbnet.nb.ca Tue, 1 May 2001 10:14:25 -0300 Date: Tue, 1 May 2001 10:14:25 -0300 From: John Hunter jthunter@nbnet.nb.ca Subject: Cognos Scenario and 4Thought Hi Again, Thanks to all that responded to my question reguarding Powerplay vs. Oracle vs. Hyperion. I'll summarize and let you know the route we take. Has anyone tried Scenario and/or 4Thought to build 'what ifs' against Powerplay cubes? How did you make out? Thanks for suggestions (again), -John From bob.edis@fleetpride.com Tue, 1 May 2001 09:23:49 -0500 Date: Tue, 1 May 2001 09:23:49 -0500 From: Edis, Bob bob.edis@fleetpride.com Subject: FW: Delete Procedure What platform? what version? I suggest that you try using the debug feature of Quick. This has worked nicely for me in the past to find why the screen behaves differently for different records. Blue -----Original Message----- From: GGrimm@pcc-structurals.com [mailto:GGrimm@pcc-structurals.com] Sent: Monday, April 30, 2001 6:40 PM To: powerh-l@lists.swau.edu Subject: Delete Procedure On some situation (yet to be ascertained), a partial update occurs about 5% of the time when updating cached records if record(s), have been deleted with a range of numbers, for example: D-9/10. Has anyone an idea what could cause this from the below snippet without having to see the entire code? The data base is Oracle 8.06. Could changing a field in a deleted record after the delete cause a problem? Following is a snippet of the code, primary and three secondary files, no DELETE files, secondary retrieval done in FIND procedure with no records is possible on any of the three secondaries. The absence of secondary(s) is not the cause of the partial Update that happens. --------- FILE TIME_ATTEND PRIMARY OCCURS 13 NOITEMS CACHE 255 ACCESS VIA EMPLOYEENO, SHIFT_DATE, SHIFT_NO, ENTITY_CODE & USING EMPLOYEENO OF EMPLOYEES, T_SHIFT_DATE, & T_SHIFT, TIME_ATTEND_ENTITY & ORDERBY EMPLOYEENO, SHIFT_DATE, SEQ_NUM ACCESS VIA SHIFT_DATE, DEPT_CODE, SHIFT_NO, ENTITY_CODE & USING T_SHIFT_DATE, T_REQ_DEPT, T_SHIFT, TIME_ATTEND_ENTITY & ORDERBY EMPLOYEENO, SHIFT_DATE, SEQ_NUM FILE TIME_ATTEND_2 SECONDARY OCCURS WITH TIME_ATTEND NOITEMS FILE TIME_ATTEND_2 SECONDARY OCCURS WITH TIME_ATTEND NOITEMS & ALIAS TIME_ATTEND_2IN FILE TIME_ATTEND SECONDARY OCCURS WITH TIME_ATTEND NOITEMS & ALIAS TIME_ATTEND_IN . . . . . PROCEDURE DELETE BEGIN DELETE TIME_ATTEND_2IN DELETE TIME_ATTEND_IN DELETE TIME_ATTEND_2 DELETE TIME_ATTEND END = = = = = = = = = = = = = = = = = = = = = = = = = = = = Mailing list: powerh-l@lists.swau.edu Subscribe: "subscribe" in message body to powerh-l-request@lists.swau.edu Unsubscribe: "unsubscribe" in message body to powerh-l-request@lists.swau.edu http://lists.swau.edu/mailman/listinfo/powerh-l This list is closed, thus to post to the list you must be a subscriber. From bob.edis@fleetpride.com Tue, 1 May 2001 10:30:33 -0500 Date: Tue, 1 May 2001 10:30:33 -0500 From: Edis, Bob bob.edis@fleetpride.com Subject: Cognos NETD G'day David We are running PH 8.21D4 on MS Windows NT 4.0 (sp6). The NetD service has been running for ages without issue and starts automatically on system reboot. The machine is a Compaq Proliant 5500 quad CPU 550Mhz with 4GB RAM. We are not using Windows 2000 server. The Axiant client is running on a MS Windows 2000 workstation though. We only use it for development and user interface is via PH Web. Regards, Blue -----Original Message----- From: David Followes [mailto:david.followes@ceo.woll.catholic.edu.au] Sent: Tuesday, May 01, 2001 12:53 AM To: powerh-l@lists.swau.edu Subject: Cognos NETD Does anyone know if there is a way to debug the NETD run as a service? Or what information it uses/needs from the registry or Cognos.ini file or ......??? We are trying to run PH 8.21D1 on Win 2000 Advance server SP1 and running Axiant thin client. When we start NETD as a service it runs for about 30 - 40 second before stopping again. The event log contains an error message stating "The Cognos Netd for NT service as reported an invalid current state 0." If we try to test the connection with Netinfo we get successful responses up until the service stops. Th interactive "Netd Console" stays working and tests with Netinfo perform correctly forever. As this isn't an ideal way to run a production server can anyone shed any light on this problem?? Regards David = = = = = = = = = = = = = = = = = = = = = = = = = = = = Mailing list: powerh-l@lists.swau.edu Subscribe: "subscribe" in message body to powerh-l-request@lists.swau.edu Unsubscribe: "unsubscribe" in message body to powerh-l-request@lists.swau.edu http://lists.swau.edu/mailman/listinfo/powerh-l This list is closed, thus to post to the list you must be a subscriber. From JMBackus@TechGroupMD.com Tue, 1 May 2001 12:33:22 -0400 Date: Tue, 1 May 2001 12:33:22 -0400 From: Jonathan M. Backus JMBackus@TechGroupMD.com Subject: <> The new 2001 Summer Semester Catalog for MPE classes at Tech Group University is now available for downloading from our web site (www.TechGroupMD.com). It includes new classes for PowerHouse, Quiz, DBGeneral, Customer Relationship Management (CRM), QEdit, and Query. We now have well over 80 classes in the catalog that can be offered. It also contains updated information on the MPE Certification program that is being developed through www.MPECert.org. Thanx, Jonathan (Jon) M. Backus, MPE-CSM ~ President Tech Group ~ 15 Catawba Place ~ Hagerstown, MD ~ 21742-6515 Email: JMBackus@TechGroupMD.com ~ AIM: JMBackus Vmail: 301.988.0614 ~ Fmail: 301.714.1854 Web: www.TechGroupMD.com From JackN@datasoft.co.za Wed, 2 May 2001 07:06:49 +0200 Date: Wed, 2 May 2001 07:06:49 +0200 From: Jack Noppe JackN@datasoft.co.za Subject: Answer RE: Cognos NETD Hi David, This sounds like a problem we had when running the same software environment on an IBM Netfinity server. The problem turned out to be an incompatibility with the IBM Server Management Service. If this is your case, try shutting down the IBM Server Management Service and try Cognos NETD again. If you are running a different config, then try shutting down other non-essential services one-by-one until the Cognos NETD stays up. That should give you enough info to report the problem to Customer Support. Good Luck, Jack Noppé -----Original Message----- From: David Followes [mailto:david.followes@ceo.woll.catholic.edu.au] Sent: Tuesday, May 01, 2001 12:53 AM To: powerh-l@lists.swau.edu Subject: Cognos NETD Does anyone know if there is a way to debug the NETD run as a service? Or what information it uses/needs from the registry or Cognos.ini file or ......??? We are trying to run PH 8.21D1 on Win 2000 Advance server SP1 and running Axiant thin client. When we start NETD as a service it runs for about 30 - 40 second before stopping again. The event log contains an error message stating "The Cognos Netd for NT service as reported an invalid current state 0." If we try to test the connection with Netinfo we get successful responses up until the service stops. Th interactive "Netd Console" stays working and tests with Netinfo perform correctly forever. As this isn't an ideal way to run a production server can anyone shed any light on this problem?? Regards David = = = = = = = = = = = = = = = = = = = = = = = = = = = = Mailing list: powerh-l@lists.swau.edu Subscribe: "subscribe" in message body to powerh-l-request@lists.swau.edu Unsubscribe: "unsubscribe" in message body to powerh-l-request@lists.swau.edu http://lists.swau.edu/mailman/listinfo/powerh-l This list is closed, thus to post to the list you must be a subscriber. = = = = = = = = = = = = = = = = = = = = = = = = = = = = Mailing list: powerh-l@lists.swau.edu Subscribe: "subscribe" in message body to powerh-l-request@lists.swau.edu Unsubscribe: "unsubscribe" in message body to powerh-l-request@lists.swau.edu http://lists.swau.edu/mailman/listinfo/powerh-l This list is closed, thus to post to the list you must be a subscriber. From andreasa@scicom.gr Wed, 2 May 2001 12:04:53 +0100 Date: Wed, 2 May 2001 12:04:53 +0100 From: =?iso-8859-7?Q?=C1=D1=C3=D5=D1=C9=CF=D5_=C1=CD=C4=D1=C5=C1=D3?= andreasa@scicom.gr Subject: Cognos NETD Hello David, It is possible to direct output from the Netd service to a log file. I did this with Powerhouse 8.11C running on NT 4.0 and Axiant 2.03 thin client. You stop the Netd service and then start it specifying -dgfxC:\\netd.log in the startup parameters box, where netd.log is the name of the log file. Of course you need to be able to understand the contents of the log file. What I did was send the file to UK Telesupport (they were the ones who told me about the log file). I don't know if things are any different with PH 8.21D1 and Win 2000. Good luck. Andreas Argyriou Scicom S.A. Athens, Greece From Mark.Fry@COGNOS.com Wed, 2 May 2001 05:23:03 -0400 Date: Wed, 2 May 2001 05:23:03 -0400 From: Fry, Mark Mark.Fry@COGNOS.com Subject: Delete Procedure Hmmm, something else is going on here, I'd guess. Just a hunch, do you have a hard-coded UPDATE procedure? If so, what does it look like? (ie does it contain code that says something like... FOR TIME_ATTEND BEGIN PUT NOTDELETED TIME_ATTEND PUT TIME_ATTEND_2 PUT TIME_ATTEND_2IN PUT TIME_ATTEND_IN ; PUT TIME_ATTEND END Mark Fry Cognos Limited email: Mark.Fry@Cognos.com Cognos Limited Registered in England, number 1635598. Registered office: Westerly Point, Market Street, Bracknell, Berkshire, RG12 1QB This e-mail may contain confidential information and is confidential to the user of the e-mail address to which it is addressed. No-one else may copy or forward all or any of this e-mail in any form. If you receive this e-mail in error please destroy this message and notify the sender by return e-mail." -----Original Message----- From: GGrimm@pcc-structurals.com [mailto:GGrimm@pcc-structurals.com] Sent: 01 May 2001 00:40 To: powerh-l@lists.swau.edu Subject: Delete Procedure On some situation (yet to be ascertained), a partial update occurs about 5% of the time when updating cached records if record(s), have been deleted with a range of numbers, for example: D-9/10. Has anyone an idea what could cause this from the below snippet without having to see the entire code? The data base is Oracle 8.06. Could changing a field in a deleted record after the delete cause a problem? Following is a snippet of the code, primary and three secondary files, no DELETE files, secondary retrieval done in FIND procedure with no records is possible on any of the three secondaries. The absence of secondary(s) is not the cause of the partial Update that happens. --------- FILE TIME_ATTEND PRIMARY OCCURS 13 NOITEMS CACHE 255 ACCESS VIA EMPLOYEENO, SHIFT_DATE, SHIFT_NO, ENTITY_CODE & USING EMPLOYEENO OF EMPLOYEES, T_SHIFT_DATE, & T_SHIFT, TIME_ATTEND_ENTITY & ORDERBY EMPLOYEENO, SHIFT_DATE, SEQ_NUM ACCESS VIA SHIFT_DATE, DEPT_CODE, SHIFT_NO, ENTITY_CODE & USING T_SHIFT_DATE, T_REQ_DEPT, T_SHIFT, TIME_ATTEND_ENTITY & ORDERBY EMPLOYEENO, SHIFT_DATE, SEQ_NUM FILE TIME_ATTEND_2 SECONDARY OCCURS WITH TIME_ATTEND NOITEMS FILE TIME_ATTEND_2 SECONDARY OCCURS WITH TIME_ATTEND NOITEMS & ALIAS TIME_ATTEND_2IN FILE TIME_ATTEND SECONDARY OCCURS WITH TIME_ATTEND NOITEMS & ALIAS TIME_ATTEND_IN . . . . . PROCEDURE DELETE BEGIN DELETE TIME_ATTEND_2IN DELETE TIME_ATTEND_IN DELETE TIME_ATTEND_2 DELETE TIME_ATTEND END = = = = = = = = = = = = = = = = = = = = = = = = = = = = Mailing list: powerh-l@lists.swau.edu Subscribe: "subscribe" in message body to powerh-l-request@lists.swau.edu Unsubscribe: "unsubscribe" in message body to powerh-l-request@lists.swau.edu http://lists.swau.edu/mailman/listinfo/powerh-l This list is closed, thus to post to the list you must be a subscriber. From merol.newman@ramesys.com Wed, 2 May 2001 13:22:08 +0100 Date: Wed, 2 May 2001 13:22:08 +0100 From: Merol Newman merol.newman@ramesys.com Subject: Abnormal Termination of a Quick Screen in Batch Hello Roger I'm working on the same platform as you, and have found that this amazingly unhelpful error message in an INTERACTIVE session usually means the program has failed to perform a VMS command, such as copying a file, or if it runs a qtp, any error in that. But I think I have also had it happen while running Quick in batch, on a while retrieving loop in the initialize procedure. This was during development, so I can't find any record of it. It was also on a different platform - Unix running on an Alpha. I think it might have occurred on a non-optional GET if it didn't find a record. I don't think I found it till I added an audit file so I could find the data it failed on. The cure would be to make the GET optional and test accessok, but I suspect other errors in the Quick processing would come up with the same wonderful message! On second thoughts, I might have just displayed the record code on the screen to find this bug - the audit file came later. They get displayed too quick to read, but you can read the one it stops at! This doesn't make sense - I was obviously debugging the screen interactively, but it was designed to be run in batch. I don't think the interactive/batch thing is relevant. You could probably get the info line to be written to your output file. I've found the bit of code... refresh line 24 let t_student_count = t_student_count + 1 info = "Students processed = " + ascii(t_student_count) now Not ideal if the error is near the end of a big data file! Hope this helps. regards Merol merol.newman@ramesys.com Ramesys, Eldon Way, Crick, Northamptonshire, UK. NN6 7SL phone 01788 822133/823831, fax 01788 823601 > -----Original Message----- > From: powerh-l-admin@cube.swau.edu > [mailto:powerh-l-admin@cube.swau.edu]On Behalf Of Kesterson, Roger > Sent: 30 April 2001 21:50 > To: PowerHouse List Server (E-mail) > Subject: Abnormal Termination of a Quick Screen in Batch > > > PH7.10.G2 > Alpha OpenVMS 7.1 > ----------------- > > Fellow PH Listers (no reference to Red Dwarf intended): > > We have some Quick programs that execute in batch, using a WHILE loop in the > INITIALIZE procedure. QKOUT is sent to a file, but QKIN is not redefined in > any way. One of these programs just terminated, and the only message in the > QKOUT file is: > > "The limit for backup movements (\\) has been reached." > > Does any of you have any idea what might cause this? > > Thanks, > > Roger Kesterson > Senior Programming Analyst > MTD Southwest, Inc. > 550 N. 54th Street > Chandler, AZ 85226 > > Phone: 480-961-1002 x4357 > Fax: 480-940-5035 > > = = = = = = = = = = = = = = = = = = = = = = = = = = = = > Mailing list: powerh-l@lists.swau.edu > Subscribe: "subscribe" in message body to powerh-l-request@lists.swau.edu > Unsubscribe: "unsubscribe" in message body to powerh-l-request@lists.swau.edu > http://lists.swau.edu/mailman/listinfo/powerh-l > This list is closed, thus to post to the list you must be a subscriber. > From paul.macdermott@eds.com Wed, 2 May 2001 17:33:12 +0100 Date: Wed, 2 May 2001 17:33:12 +0100 From: MacDermott, Paul paul.macdermott@eds.com Subject: Scrolling popup screen Hello all, Can anybody help me please. I want to amend an existing popup screen in QUICK to be able to display scrolling data. I have attempted but unfortunately scrolling forward does not work. Have any of you done this before? The development environment is Axiant 4Gl version 2.03 for a VMS environment. Paul Mac Dermott EDS Ireland Treasury Building Lower Grand Canal Street Dublin 2, Ireland * paul.macdermott@eds.com * + 353 1 703 9217 / 8 996 9217 * + 353 1 607 9300 From GGrimm@pcc-structurals.com Wed, 02 May 2001 11:43:55 -0700 Date: Wed, 02 May 2001 11:43:55 -0700 From: GGrimm@pcc-structurals.com GGrimm@pcc-structurals.com Subject: Re[2]: Delete Procedure To all who offered food for thought, my sincere thanks (Bob Edis, Mark, Terrence Burk, Charlie Maloney). It will be a trial and error to fix this problem. To re-create the situation in my test environment is very difficult and attempts have been made with no reoccurrence of the partial updating problem. I shall ponder each of your ideas and try various approaches if the problem happens again. So far, the rarity of the problem allows me the liberty to not be too excited about a solution. For those of you who wondered about my environment, it is HP-UX B.11.00 A, Oracle Release 8.0.6.2.0, PowerHouse 8.13.D3. Regards, Gerry ______________________________ Reply Separator _________________________________ Subject: RE: Delete Procedure Author: "Fry Mark" at RDM-PO1 Date: 05/02/2001 5:23 AM Hmmm, something else is going on here, I'd guess. Just a hunch, do you have a hard-coded UPDATE procedure? If so, what does it look like? (ie does it contain code that says something like... FOR TIME_ATTEND BEGIN PUT NOTDELETED TIME_ATTEND PUT TIME_ATTEND_2 PUT TIME_ATTEND_2IN PUT TIME_ATTEND_IN ; PUT TIME_ATTEND END Mark Fry Cognos Limited email: Mark.Fry@Cognos.com Cognos Limited Registered in England, number 1635598. Registered office: Westerly Point, Market Street, Bracknell, Berkshire, RG12 1QB This e-mail may contain confidential information and is confidential to the user of the e-mail address to which it is addressed. No-one else may copy or forward all or any of this e-mail in any form. If you receive this e-mail in error please destroy this message and notify the sender by return e-mail." -----Original Message----- From: GGrimm@pcc-structurals.com [mailto:GGrimm@pcc-structurals.com] Sent: 01 May 2001 00:40 To: powerh-l@lists.swau.edu Subject: Delete Procedure On some situation (yet to be ascertained), a partial update occurs about 5% of the time when updating cached records if record(s), have been deleted with a range of numbers, for example: D-9/10. Has anyone an idea what could cause this from the below snippet without having to see the entire code? The data base is Oracle 8.06. Could changing a field in a deleted record after the delete cause a problem? Following is a snippet of the code, primary and three secondary files, no DELETE files, secondary retrieval done in FIND procedure with no records is possible on any of the three secondaries. The absence of secondary(s) is not the cause of the partial Update that happens. --------- FILE TIME_ATTEND PRIMARY OCCURS 13 NOITEMS CACHE 255 ACCESS VIA EMPLOYEENO, SHIFT_DATE, SHIFT_NO, ENTITY_CODE & USING EMPLOYEENO OF EMPLOYEES, T_SHIFT_DATE, & T_SHIFT, TIME_ATTEND_ENTITY & ORDERBY EMPLOYEENO, SHIFT_DATE, SEQ_NUM ACCESS VIA SHIFT_DATE, DEPT_CODE, SHIFT_NO, ENTITY_CODE & USING T_SHIFT_DATE, T_REQ_DEPT, T_SHIFT, TIME_ATTEND_ENTITY & ORDERBY EMPLOYEENO, SHIFT_DATE, SEQ_NUM FILE TIME_ATTEND_2 SECONDARY OCCURS WITH TIME_ATTEND NOITEMS FILE TIME_ATTEND_2 SECONDARY OCCURS WITH TIME_ATTEND NOITEMS & ALIAS TIME_ATTEND_2IN FILE TIME_ATTEND SECONDARY OCCURS WITH TIME_ATTEND NOITEMS & ALIAS TIME_ATTEND_IN . . . . . PROCEDURE DELETE BEGIN DELETE TIME_ATTEND_2IN DELETE TIME_ATTEND_IN DELETE TIME_ATTEND_2 DELETE TIME_ATTEND END = = = = = = = = = = = = = = = = = = = = = = = = = = = = Mailing list: powerh-l@lists.swau.edu Subscribe: "subscribe" in message body to powerh-l-request@lists.swau.edu Unsubscribe: "unsubscribe" in message body to powerh-l-request@lists.swau.edu http://lists.swau.edu/mailman/listinfo/powerh-l This list is closed, thus to post to the list you must be a subscriber. From nick_mickey@yahoo.com Wed, 2 May 2001 13:25:42 -0700 (PDT) Date: Wed, 2 May 2001 13:25:42 -0700 (PDT) From: Nicholas Mickey nick_mickey@yahoo.com Subject: commit transaction one at a time --0-1649760492-988835142=:22368 Content-Type: text/plain; charset=us-ascii I am working on HP3000 image system: I would like to know how to commit one transaction at time in QTP (OUTPUT statement doing update to tables) /nick Nicholas R. Mickey Sr. Programmer/Analyst Ajilon Consulting nick_mickey@yahoo.com 1-888-449-9595 --------------------------------- Do You Yahoo!? Yahoo! Auctions - buy the things you want at great prices --0-1649760492-988835142=:22368 Content-Type: text/html; charset=us-ascii

I am working on HP3000 image system:

 

I would like to know how to commit one transaction at time in QTP (OUTPUT statement doing update to tables)

 

/nick



Nicholas R. Mickey
Sr. Programmer/Analyst
Ajilon Consulting
nick_mickey@yahoo.com
1-888-449-9595



Do You Yahoo!?
Yahoo! Auctions - buy the things you want at great prices --0-1649760492-988835142=:22368-- From nick_mickey@yahoo.com Wed, 2 May 2001 13:25:58 -0700 (PDT) Date: Wed, 2 May 2001 13:25:58 -0700 (PDT) From: Nicholas Mickey nick_mickey@yahoo.com Subject: commit transaction one at a time --0-1804289383-988835158=:24205 Content-Type: text/plain; charset=us-ascii I am working on HP3000 image system: I would like to know how to commit one transaction at time in QTP (OUTPUT statement doing update to tables) /nick Nicholas R. Mickey Sr. Programmer/Analyst Ajilon Consulting nick_mickey@yahoo.com 1-888-449-9595 --------------------------------- Do You Yahoo!? Yahoo! Auctions - buy the things you want at great prices --0-1804289383-988835158=:24205 Content-Type: text/html; charset=us-ascii

I am working on HP3000 image system:


 


I would like to know how to commit one transaction at time in QTP (OUTPUT statement doing update to tables)


 


/nick



Nicholas R. Mickey
Sr. Programmer/Analyst
Ajilon Consulting
nick_mickey@yahoo.com
1-888-449-9595



Do You Yahoo!?
Yahoo! Auctions - buy the things you want at great prices --0-1804289383-988835158=:24205-- From nick_mickey@yahoo.com Wed, 2 May 2001 13:26:03 -0700 (PDT) Date: Wed, 2 May 2001 13:26:03 -0700 (PDT) From: Nicholas Mickey nick_mickey@yahoo.com Subject: commit transaction one at a time --0-1102520059-988835163=:44880 Content-Type: text/plain; charset=us-ascii I am working on HP3000 image system: I would like to know how to commit one transaction at time in QTP (OUTPUT statement doing update to tables) /nick Nicholas R. Mickey Sr. Programmer/Analyst Ajilon Consulting nick_mickey@yahoo.com 1-888-449-9595 --------------------------------- Do You Yahoo!? Yahoo! Auctions - buy the things you want at great prices --0-1102520059-988835163=:44880 Content-Type: text/html; charset=us-ascii

I am working on HP3000 image system:



 



I would like to know how to commit one transaction at time in QTP (OUTPUT statement doing update to tables)



 



/nick



Nicholas R. Mickey
Sr. Programmer/Analyst
Ajilon Consulting
nick_mickey@yahoo.com
1-888-449-9595



Do You Yahoo!?
Yahoo! Auctions - buy the things you want at great prices --0-1102520059-988835163=:44880-- From AKarman@USFreightways.com Wed, 2 May 2001 15:36:34 -0500 Date: Wed, 2 May 2001 15:36:34 -0500 From: Karman, Al AKarman@USFreightways.com Subject: commit transaction one at a time How about starting with commit to 1 email query at a time....-:) Al Karman IT Consultant US Freightways akarman@usfreightways.com 773.824.2284 -----Original Message----- From: Nicholas Mickey [mailto:nick_mickey@yahoo.com] Sent: Wednesday, May 02, 2001 3:26 PM To: powerh-l email Subject: commit transaction one at a time I am working on HP3000 image system: I would like to know how to commit one transaction at time in QTP (OUTPUT statement doing update to tables) /nick Nicholas R. Mickey Sr. Programmer/Analyst Ajilon Consulting nick_mickey@yahoo.com 1-888-449-9595 _____ Do You Yahoo!? Yahoo! Auctions - buy the things you want at great prices From david.followes@ceo.woll.catholic.edu.au Thu, 3 May 2001 10:10:54 +1000 Date: Thu, 3 May 2001 10:10:54 +1000 From: David Followes david.followes@ceo.woll.catholic.edu.au Subject: Answer RE: Cognos NETD Thanks to everyone who responded. It turns out that we had the 120 day trial of Exchange Server 2000 running on this box. After stopping the services for it NETD stays running. Thanks Jack Maybe Cognos can explain this one! Why NETD won't run with Exchange server. Regards David Followes -----Original Message----- From: powerh-l-admin@cube.swau.edu [mailto:powerh-l-admin@cube.swau.edu]On Behalf Of Jack Noppe Sent: Wednesday, 2 May 2001 3:07 PM To: 'powerh-l@list.swau.edu' Subject: Answer RE: Cognos NETD Hi David, This sounds like a problem we had when running the same software environment on an IBM Netfinity server. The problem turned out to be an incompatibility with the IBM Server Management Service. If this is your case, try shutting down the IBM Server Management Service and try Cognos NETD again. If you are running a different config, then try shutting down other non-essential services one-by-one until the Cognos NETD stays up. That should give you enough info to report the problem to Customer Support. Good Luck, Jack Noppé -----Original Message----- From: David Followes [mailto:david.followes@ceo.woll.catholic.edu.au] Sent: Tuesday, May 01, 2001 12:53 AM To: powerh-l@lists.swau.edu Subject: Cognos NETD Does anyone know if there is a way to debug the NETD run as a service? Or what information it uses/needs from the registry or Cognos.ini file or .......??? We are trying to run PH 8.21D1 on Win 2000 Advance server SP1 and running Axiant thin client. When we start NETD as a service it runs for about 30 - 40 second before stopping again. The event log contains an error message stating "The Cognos Netd for NT service as reported an invalid current state 0." If we try to test the connection with Netinfo we get successful responses up until the service stops. Th interactive "Netd Console" stays working and tests with Netinfo perform correctly forever. As this isn't an ideal way to run a production server can anyone shed any light on this problem?? Regards David = = = = = = = = = = = = = = = = = = = = = = = = = = = = Mailing list: powerh-l@lists.swau.edu Subscribe: "subscribe" in message body to powerh-l-request@lists.swau.edu Unsubscribe: "unsubscribe" in message body to powerh-l-request@lists.swau.edu http://lists.swau.edu/mailman/listinfo/powerh-l This list is closed, thus to post to the list you must be a subscriber. = = = = = = = = = = = = = = = = = = = = = = = = = = = = Mailing list: powerh-l@lists.swau.edu Subscribe: "subscribe" in message body to powerh-l-request@lists.swau.edu Unsubscribe: "unsubscribe" in message body to powerh-l-request@lists.swau.edu http://lists.swau.edu/mailman/listinfo/powerh-l This list is closed, thus to post to the list you must be a subscriber. = = = = = = = = = = = = = = = = = = = = = = = = = = = = Mailing list: powerh-l@lists.swau.edu Subscribe: "subscribe" in message body to powerh-l-request@lists.swau.edu Unsubscribe: "unsubscribe" in message body to powerh-l-request@lists.swau.edu http://lists.swau.edu/mailman/listinfo/powerh-l This list is closed, thus to post to the list you must be a subscriber. From neil@nha.co.za Thu, 3 May 2001 07:48:21 +0200 Date: Thu, 3 May 2001 07:48:21 +0200 From: Neil Harvey neil@nha.co.za Subject: Answer RE: Cognos NETD Maybe Microsoft can explain this one! ;) Neil -----Original Message----- From: David Followes [mailto:david.followes@ceo.woll.catholic.edu.au] Sent: 03 May 2001 02:11 To: 'powerh-l@list.swau.edu' Subject: RE: Answer RE: Cognos NETD Thanks to everyone who responded. It turns out that we had the 120 day trial of Exchange Server 2000 running on this box. After stopping the services for it NETD stays running. Thanks Jack Maybe Cognos can explain this one! Why NETD won't run with Exchange server. Regards David Followes -----Original Message----- From: powerh-l-admin@cube.swau.edu [mailto:powerh-l-admin@cube.swau.edu]On Behalf Of Jack Noppe Sent: Wednesday, 2 May 2001 3:07 PM To: 'powerh-l@list.swau.edu' Subject: Answer RE: Cognos NETD Hi David, This sounds like a problem we had when running the same software environment on an IBM Netfinity server. The problem turned out to be an incompatibility with the IBM Server Management Service. If this is your case, try shutting down the IBM Server Management Service and try Cognos NETD again. If you are running a different config, then try shutting down other non-essential services one-by-one until the Cognos NETD stays up. That should give you enough info to report the problem to Customer Support. Good Luck, Jack Noppé -----Original Message----- From: David Followes [mailto:david.followes@ceo.woll.catholic.edu.au] Sent: Tuesday, May 01, 2001 12:53 AM To: powerh-l@lists.swau.edu Subject: Cognos NETD Does anyone know if there is a way to debug the NETD run as a service? Or what information it uses/needs from the registry or Cognos.ini file or .......??? We are trying to run PH 8.21D1 on Win 2000 Advance server SP1 and running Axiant thin client. When we start NETD as a service it runs for about 30 - 40 second before stopping again. The event log contains an error message stating "The Cognos Netd for NT service as reported an invalid current state 0." If we try to test the connection with Netinfo we get successful responses up until the service stops. Th interactive "Netd Console" stays working and tests with Netinfo perform correctly forever. As this isn't an ideal way to run a production server can anyone shed any light on this problem?? Regards David = = = = = = = = = = = = = = = = = = = = = = = = = = = = Mailing list: powerh-l@lists.swau.edu Subscribe: "subscribe" in message body to powerh-l-request@lists.swau.edu Unsubscribe: "unsubscribe" in message body to powerh-l-request@lists.swau.edu http://lists.swau.edu/mailman/listinfo/powerh-l This list is closed, thus to post to the list you must be a subscriber. = = = = = = = = = = = = = = = = = = = = = = = = = = = = Mailing list: powerh-l@lists.swau.edu Subscribe: "subscribe" in message body to powerh-l-request@lists.swau.edu Unsubscribe: "unsubscribe" in message body to powerh-l-request@lists.swau.edu http://lists.swau.edu/mailman/listinfo/powerh-l This list is closed, thus to post to the list you must be a subscriber. = = = = = = = = = = = = = = = = = = = = = = = = = = = = Mailing list: powerh-l@lists.swau.edu Subscribe: "subscribe" in message body to powerh-l-request@lists.swau.edu Unsubscribe: "unsubscribe" in message body to powerh-l-request@lists.swau.edu http://lists.swau.edu/mailman/listinfo/powerh-l This list is closed, thus to post to the list you must be a subscriber. = = = = = = = = = = = = = = = = = = = = = = = = = = = = Mailing list: powerh-l@lists.swau.edu Subscribe: "subscribe" in message body to powerh-l-request@lists.swau.edu Unsubscribe: "unsubscribe" in message body to powerh-l-request@lists.swau.edu http://lists.swau.edu/mailman/listinfo/powerh-l This list is closed, thus to post to the list you must be a subscriber. From RessingM@unive.nl Thu, 3 May 2001 11:56:30 +0200 Date: Thu, 3 May 2001 11:56:30 +0200 From: RessingM@unive.nl RessingM@unive.nl Subject: Problem with building a non-shareable QKDRIVER > Hello, > > we are using PH 7.10F2 on Open VMS for Alpha 7.1. > Because we are having some trouble with external routines (C-routines) we > tried to build the QKDRIVER as non-shareable instead of shareable. > We didn't manage to do that because after calling 'buildexternal' with the > '/noshare' flag we get the following error: > > %LINK-F-OPENIN, error opening > PATH$POWERHOUSE:[710F2.PH_DEVELOPMENT]ALPHA_DRIVER_MAINLINE.OBJ; ... > > The file mentioned above doesn't exist on our system. > > Does anybody know how to get this object-file (is it part of the > installation software or can anybody send us a copy of this file?). > Or is it an error in the buildexternal command? > > Greetings, > > Martie Ressing > Unive, Zwolle (The Netherlands) > > From Chris.Sharman@ccagroup.co.uk Thu, 3 May 2001 11:33:58 +0100 Date: Thu, 3 May 2001 11:33:58 +0100 From: Chris Sharman Chris.Sharman@ccagroup.co.uk Subject: Problem with building a non-shareable QKDRIVER > we are using PH 7.10F2 on Open VMS for Alpha 7.1. > Because we are having some trouble with external routines (C-routines) we > tried to build the QKDRIVER as non-shareable instead of shareable. > We didn't manage to do that because after calling 'buildexternal' with the > '/noshare' flag we get the following error: Non-shareable won't work. An image is either shareable or executable, and you can only run one executable at a time (eg QUICKA710F.EXE) - the images it links to must be shareable. Shareable images by definition don't (mustn't) have a main() program, just some functions of global scope, declared as universal by a symbol_vector in the linker options file. I've never used buildexternal myself: I just link my own shareable images, and I've never had any trouble. In the past I've used ph_location:dalock.aobj (which I don't recommend). I've run into conflict between VAXC & DECC. (I think DALOCK used to be VAXC). I used to use the following in my linker options files: sys$share:vaxcrtl/share psect_attr=userbreak,noshr ! from DALOCK/VAXC but no longer need it, apparently (710G3, Alpha VMS 7.1). We're mostly Pascal here, rather than C, but we do have one DECC module I can think of which we've successfully put into a QKDRIVER image & called from Quick. If you need more specific help, you'll have to more clearly describe your problems. Regards, Chris Sharman _______________________________________________________________________ Chris.Sharman@CCAgroup.co.uk http://www.ccagroup.co.uk/ CCA Stationery Ltd, Eastway, Fulwood, Preston, Lancashire, PR2 9WS. ----------------------------------------------------------------------- Any views expressed in this message are those of the sender and not necessarily those of CCA Group. The unauthorized use, disclosure, copying or alteration of this message is forbidden. The contents of this message may be confidential and/or privileged, copyright CCA Group and are intended solely for the use of the individual or entity to whom they are addressed. Whilst this message has been scanned, CCA Group cannot guarantee that it is virus free or compatible with your systems and accepts no responsibility for any loss or damage arising from its use. The recipient is advised to run their own anti-virus software. If you receive this message in error please contact postmaster@ccagroup.co.uk immediately, destroy any copies and delete it from your computer systems. From Erik.Hurworth@cognos.com Thu, 3 May 2001 06:42:42 -0400 Date: Thu, 3 May 2001 06:42:42 -0400 From: Hurworth, Erik Erik.Hurworth@cognos.com Subject: Problem with building a non-shareable QKDRIVER Hello all, If you are in testing mode, it is easier to use the disk image of Quick then you don't have to keep running INSTALL to install and deinstall the QKDRIVER every time you make a change to your 3GL routines. Use BuildExternal to build your QKDRIVER.EXE object library in the normal way. Set up a symbol to run the uninstalled disk version of Quick:- $QUICK == "PH_LOCATION:QUICKA710F.EXE;" Note the ; character at the end of the symbol definition. Now, when you run the command QUICK, it will run the disk version and will then be able to find the QKDRIVER.EXE in the same directory. I hope this is what you are after, Regards Erik A. Hurworth Technical Supervisor Customer Support __________________________________________________ Cognos Ltd. Phone: +44 / (0)1344 / 707799 Westerly Point Market Street e-mail: erik.hurworth@cognos.com Bracknell Berkshire RG12 1QB Cognos Support Page: http://support.cognos.com Cognos Ltd. SeeBusiness. UniteBusiness. ShareBusiness. Please send any e-mail response regarding this communication to support.uk@cognos.com and it will be directed to the specialist dealing with your call. Please make sure that you include the call reference number in the subject line of your e-mail. (Cognos Limited Registered in England, number 1635598. Registered office: Westerly Point, Market Street, Bracknell, Berkshire, RG12 1QB This e-mail may contain confidential information and is confidential to the user of the e-mail address to which it is addressed. No-one else may copy or forward all or any of this e-mail in any form. If you receive this e-mail in error please destroy this message and notify the sender by return e-mail) -----Original Message----- From: Chris Sharman [mailto:Chris.Sharman@ccagroup.co.uk] Sent: 03 May 2001 11:34 To: RessingM@unive.nl Cc: Chris.Sharman@ccagroup.co.uk; powerh-l@lists.swau.edu Subject: RE: Problem with building a non-shareable QKDRIVER > we are using PH 7.10F2 on Open VMS for Alpha 7.1. > Because we are having some trouble with external routines (C-routines) we > tried to build the QKDRIVER as non-shareable instead of shareable. > We didn't manage to do that because after calling 'buildexternal' with the > '/noshare' flag we get the following error: Non-shareable won't work. An image is either shareable or executable, and you can only run one executable at a time (eg QUICKA710F.EXE) - the images it links to must be shareable. Shareable images by definition don't (mustn't) have a main() program, just some functions of global scope, declared as universal by a symbol_vector in the linker options file. I've never used buildexternal myself: I just link my own shareable images, and I've never had any trouble. In the past I've used ph_location:dalock.aobj (which I don't recommend). I've run into conflict between VAXC & DECC. (I think DALOCK used to be VAXC). I used to use the following in my linker options files: sys$share:vaxcrtl/share psect_attr=userbreak,noshr ! from DALOCK/VAXC but no longer need it, apparently (710G3, Alpha VMS 7.1). We're mostly Pascal here, rather than C, but we do have one DECC module I can think of which we've successfully put into a QKDRIVER image & called from Quick. If you need more specific help, you'll have to more clearly describe your problems. Regards, Chris Sharman _______________________________________________________________________ Chris.Sharman@CCAgroup.co.uk http://www.ccagroup.co.uk/ CCA Stationery Ltd, Eastway, Fulwood, Preston, Lancashire, PR2 9WS. ----------------------------------------------------------------------- Any views expressed in this message are those of the sender and not necessarily those of CCA Group. The unauthorized use, disclosure, copying or alteration of this message is forbidden. The contents of this message may be confidential and/or privileged, copyright CCA Group and are intended solely for the use of the individual or entity to whom they are addressed. Whilst this message has been scanned, CCA Group cannot guarantee that it is virus free or compatible with your systems and accepts no responsibility for any loss or damage arising from its use. The recipient is advised to run their own anti-virus software. If you receive this message in error please contact postmaster@ccagroup.co.uk immediately, destroy any copies and delete it from your computer systems. = = = = = = = = = = = = = = = = = = = = = = = = = = = = Mailing list: powerh-l@lists.swau.edu Subscribe: "subscribe" in message body to powerh-l-request@lists.swau.edu Unsubscribe: "unsubscribe" in message body to powerh-l-request@lists.swau.edu http://lists.swau.edu/mailman/listinfo/powerh-l This list is closed, thus to post to the list you must be a subscriber. From Bob.Deskin@Cognos.COM Thu, 3 May 2001 08:05:32 -0400 Date: Thu, 3 May 2001 08:05:32 -0400 From: Deskin, Bob Bob.Deskin@Cognos.COM Subject: commit transaction one at a time This message is in MIME format. Since your mail reader does not understand this format, some or all of this message may not be legible. ------_=_NextPart_001_01C0D3C9.5A2252E0 Content-Type: text/plain Check out the COMMIT statement in the QTP manual. You may want COMMIT AT UPDATE or COMMIT AT 1 TRANSACTION depending on your circumstances. Bob Deskin PowerHouse Web Product Manager, Application Development Tools, Cognos Inc. bob.deskin@cognos.com (613) 738-1338 ext 7268 FAX: (613) 727-1178 3755 Riverside Drive P.O. Box 9707 Stn. T, Ottawa ON K1G 4K9 CANADA -----Original Message----- From: Nicholas Mickey [mailto:nick_mickey@yahoo.com] Sent: Wednesday, May 02, 2001 4:26 PM To: powerh-l email Subject: commit transaction one at a time I am working on HP3000 image system: I would like to know how to commit one transaction at time in QTP (OUTPUT statement doing update to tables) /nick Nicholas R. Mickey Sr. Programmer/Analyst Ajilon Consulting nick_mickey@yahoo.com 1-888-449-9595 _____ Do You Yahoo!? Yahoo! Auctions - buy the things you want at great prices ------_=_NextPart_001_01C0D3C9.5A2252E0 Content-Type: text/html
Check out the COMMIT statement in the QTP manual. You may want COMMIT AT UPDATE or COMMIT AT 1 TRANSACTION depending on your circumstances.

Bob Deskin             
PowerHouse Web Product Manager, Application Development Tools, Cognos Inc.
bob.deskin@cognos.com (613) 738-1338 ext 7268 FAX: (613) 727-1178
3755 Riverside Drive P.O. Box 9707 Stn. T, Ottawa ON K1G 4K9 CANADA

-----Original Message-----
From: Nicholas Mickey [mailto:nick_mickey@yahoo.com]
Sent: Wednesday, May 02, 2001 4:26 PM
To: powerh-l email
Subject: commit transaction one at a time

I am working on HP3000 image system:



 



I would like to know how to commit one transaction at time in QTP (OUTPUT statement doing update to tables)



 



/nick



Nicholas R. Mickey
Sr. Programmer/Analyst
Ajilon Consulting
nick_mickey@yahoo.com
1-888-449-9595



Do You Yahoo!?
Yahoo! Auctions - buy the things you want at great prices
------_=_NextPart_001_01C0D3C9.5A2252E0-- From Allison.Hamilton@Cognos.COM Thu, 3 May 2001 10:51:26 -0400 Date: Thu, 3 May 2001 10:51:26 -0400 From: Hamilton, Allison Allison.Hamilton@Cognos.COM Subject: Problem with building a non-shareable QKDRIVER The dalock objects should be built with DECC since at least 710Gn on both platforms. We no longer use VAXC. -----Original Message----- From: Chris Sharman [mailto:Chris.Sharman@ccagroup.co.uk] Sent: Thursday, May 03, 2001 6:34 AM To: RessingM@unive.nl Cc: Chris.Sharman@ccagroup.co.uk; powerh-l@lists.swau.edu Subject: RE: Problem with building a non-shareable QKDRIVER > we are using PH 7.10F2 on Open VMS for Alpha 7.1. > Because we are having some trouble with external routines (C-routines) we > tried to build the QKDRIVER as non-shareable instead of shareable. > We didn't manage to do that because after calling 'buildexternal' with the > '/noshare' flag we get the following error: Non-shareable won't work. An image is either shareable or executable, and you can only run one executable at a time (eg QUICKA710F.EXE) - the images it links to must be shareable. Shareable images by definition don't (mustn't) have a main() program, just some functions of global scope, declared as universal by a symbol_vector in the linker options file. I've never used buildexternal myself: I just link my own shareable images, and I've never had any trouble. In the past I've used ph_location:dalock.aobj (which I don't recommend). I've run into conflict between VAXC & DECC. (I think DALOCK used to be VAXC). I used to use the following in my linker options files: sys$share:vaxcrtl/share psect_attr=userbreak,noshr ! from DALOCK/VAXC but no longer need it, apparently (710G3, Alpha VMS 7.1). We're mostly Pascal here, rather than C, but we do have one DECC module I can think of which we've successfully put into a QKDRIVER image & called from Quick. If you need more specific help, you'll have to more clearly describe your problems. Regards, Chris Sharman _______________________________________________________________________ Chris.Sharman@CCAgroup.co.uk http://www.ccagroup.co.uk/ CCA Stationery Ltd, Eastway, Fulwood, Preston, Lancashire, PR2 9WS. ----------------------------------------------------------------------- Any views expressed in this message are those of the sender and not necessarily those of CCA Group. The unauthorized use, disclosure, copying or alteration of this message is forbidden. The contents of this message may be confidential and/or privileged, copyright CCA Group and are intended solely for the use of the individual or entity to whom they are addressed. Whilst this message has been scanned, CCA Group cannot guarantee that it is virus free or compatible with your systems and accepts no responsibility for any loss or damage arising from its use. The recipient is advised to run their own anti-virus software. If you receive this message in error please contact postmaster@ccagroup.co.uk immediately, destroy any copies and delete it from your computer systems. = = = = = = = = = = = = = = = = = = = = = = = = = = = = Mailing list: powerh-l@lists.swau.edu Subscribe: "subscribe" in message body to powerh-l-request@lists.swau.edu Unsubscribe: "unsubscribe" in message body to powerh-l-request@lists.swau.edu http://lists.swau.edu/mailman/listinfo/powerh-l This list is closed, thus to post to the list you must be a subscriber. From GGrimm@pcc-structurals.com Thu, 03 May 2001 09:01:33 -0700 Date: Thu, 03 May 2001 09:01:33 -0700 From: GGrimm@pcc-structurals.com GGrimm@pcc-structurals.com Subject: Re[2]: Delete Procedure Hi Mark, Thanks for your private email suggesting that I "relax" my PUT constraints on both 'IN' tables in the UPDATE with the OR (ALTEREDRECORD OF .... AND DELETEDRECORD OF ...) due to potential errors in programatically controlling the existance and non-existance of CLOS records in the different tables. Now for a late breaking bit of NEWS!! We now have put on our thinking caps and 'recall' that several times a day a BATCH job runs that also maintains the same data a screen user may be concurrently updating! Please don't ask why we didn't remember that small fact before yesterday:-) Brain dead is my only response! I will be testing a method using GETSYSTEMVAL looking for a value which will be set by the batch job when it is running in my PREUPDATE Proc and dissallow updating when that value exists. I will issue an error message and let them redo their changes later. So thank you each and all for your kind suggestions which I shall selectively include my next maintanence to this Quick program. As always this language is robust and has so many nuances, I don't know how I will be challenged when I retire in 8 months:-) Regards, Gerry ______________________________ Reply Separator _________________________________ Subject: RE: Delete Procedure Author: "Fry Mark" at RDM-PO1 Date: 05/02/2001 5:23 AM Hmmm, something else is going on here, I'd guess. Just a hunch, do you have a hard-coded UPDATE procedure? If so, what does it look like? (ie does it contain code that says something like... FOR TIME_ATTEND BEGIN PUT NOTDELETED TIME_ATTEND PUT TIME_ATTEND_2 PUT TIME_ATTEND_2IN PUT TIME_ATTEND_IN ; PUT TIME_ATTEND END Mark Fry Cognos Limited email: Mark.Fry@Cognos.com Cognos Limited Registered in England, number 1635598. Registered office: Westerly Point, Market Street, Bracknell, Berkshire, RG12 1QB This e-mail may contain confidential information and is confidential to the user of the e-mail address to which it is addressed. No-one else may copy or forward all or any of this e-mail in any form. If you receive this e-mail in error please destroy this message and notify the sender by return e-mail." -----Original Message----- From: GGrimm@pcc-structurals.com [mailto:GGrimm@pcc-structurals.com] Sent: 01 May 2001 00:40 To: powerh-l@lists.swau.edu Subject: Delete Procedure On some situation (yet to be ascertained), a partial update occurs about 5% of the time when updating cached records if record(s), have been deleted with a range of numbers, for example: D-9/10. Has anyone an idea what could cause this from the below snippet without having to see the entire code? The data base is Oracle 8.06. Could changing a field in a deleted record after the delete cause a problem? Following is a snippet of the code, primary and three secondary files, no DELETE files, secondary retrieval done in FIND procedure with no records is possible on any of the three secondaries. The absence of secondary(s) is not the cause of the partial Update that happens. --------- FILE TIME_ATTEND PRIMARY OCCURS 13 NOITEMS CACHE 255 ACCESS VIA EMPLOYEENO, SHIFT_DATE, SHIFT_NO, ENTITY_CODE & USING EMPLOYEENO OF EMPLOYEES, T_SHIFT_DATE, & T_SHIFT, TIME_ATTEND_ENTITY & ORDERBY EMPLOYEENO, SHIFT_DATE, SEQ_NUM ACCESS VIA SHIFT_DATE, DEPT_CODE, SHIFT_NO, ENTITY_CODE & USING T_SHIFT_DATE, T_REQ_DEPT, T_SHIFT, TIME_ATTEND_ENTITY & ORDERBY EMPLOYEENO, SHIFT_DATE, SEQ_NUM FILE TIME_ATTEND_2 SECONDARY OCCURS WITH TIME_ATTEND NOITEMS FILE TIME_ATTEND_2 SECONDARY OCCURS WITH TIME_ATTEND NOITEMS & ALIAS TIME_ATTEND_2IN FILE TIME_ATTEND SECONDARY OCCURS WITH TIME_ATTEND NOITEMS & ALIAS TIME_ATTEND_IN . . . . . PROCEDURE DELETE BEGIN DELETE TIME_ATTEND_2IN DELETE TIME_ATTEND_IN DELETE TIME_ATTEND_2 DELETE TIME_ATTEND END = = = = = = = = = = = = = = = = = = = = = = = = = = = = Mailing list: powerh-l@lists.swau.edu Subscribe: "subscribe" in message body to powerh-l-request@lists.swau.edu Unsubscribe: "unsubscribe" in message body to powerh-l-request@lists.swau.edu http://lists.swau.edu/mailman/listinfo/powerh-l This list is closed, thus to post to the list you must be a subscriber. From bob.edis@fleetpride.com Thu, 3 May 2001 14:01:04 -0500 Date: Thu, 3 May 2001 14:01:04 -0500 From: Edis, Bob bob.edis@fleetpride.com Subject: Strange results from DB2 in Quiz G'day all Environment: PH 8.21D4, WinNT 4.0, HitODBC 400, DB/400 We are using HitODBC to pull data from a DB2 database on an AS/400 machine and are getting erroneous results in PowerHouse Quiz. E.g. Case record: DB2 shows the cost amount column as 3.670 DecisionStream using the above ODBC driver shows 3.670 Quiz using the above ODBC driver shows 3.610. In another case Quiz reports 4.950 whereas DB2 and DS show 5.00. The values reported by DB2 and DecisionStream are the correct ones. The column is stored to 3 decimal places (float) in the DB2 table. The discrepancy between Quiz and DecisionStream is NOT consistent. I.e. some records in the table match exactly. Any ideas why Quiz is reporting incorrect values? Regards, Robert Edis Manager - Business Intelligence FleetPride 520 Lake Cook Rd., Suite 100 Deerfield IL 60015 Tel: 847.572.8039 Fax: 847.444.1096 www.fleetpride.com From PICKERIJ@norbord.com Thu, 3 May 2001 15:40:30 -0400 Date: Thu, 3 May 2001 15:40:30 -0400 From: Pickering, John (NORBORD) PICKERIJ@norbord.com Subject: Supportlink: who edits the ADT stuff? Mr. Deskin Please take a look at page 67 of the just arrived (on my desk anyway) Supportlink volume 11 number 2, April - June 2001. This is the sort of extravagant code that looks important but half of it is unnecessary! The kind of stuff one inherits from others and spends heaps of time wading through the muck trying to find the pony. Not a really brilliant example of PowerHouse code for such an august publication :-( Why test ACCESSOK in the statement after the WHILE RETRIEVING? This verb will execute the statement after it ONLY if it retrieves a record. A simple BEGIN...END would be much clearer! And since ACCESSOK will always be false when you read to the end of file or read all the values for a single key, it seems kind of pointless to test it. Lose this one too! And since t_bill_no will always have the last value retrieved (or the initial value if the file is empty), why not simply pass it? There seems little point in moving it to another temporary item! Or use the t_final_xxx temporary items inside the WHILE RETRIEVING loop. And for those who don't like unnecessary compile time warnings such as "*W* the PRIMARY file is missing." you should probably use the SCREEN statement MENU option. And I won't even start on what the user will be presented with upon return from the lower level screen! Sheesh! Regards, JWP From byron.welch@creatcomp.com Thu, 3 May 2001 15:46:26 -0400 Date: Thu, 3 May 2001 15:46:26 -0400 From: Byron Welch byron.welch@creatcomp.com Subject: Strange results from DB2 in Quiz Bob, Two comments but unfortunately I don't have a solution:( I think you are getting different results between PowerHouse and DecisionStream because they are using different database layers. My understanding is that they will be merged in the next major release but for now you get slightly different database layers between PowerHouse and the newer products. I think you are running into a precession error. In a couple of cases when I was using floating point in the past I came across similar results. If I remember correctly the PH manuals used to have a caveat about floating point numbers. It was something like for fractions or very large numbers a loss of precision can occur. i.e. .2 might actually be stored as .1999. This used to prevent me from comparing a Floating point to an integer in a select statement. I usually solved this by multiplying by 100 and then dividing by 100 in a define. Byron Welch Manager Consulting Services Creative Computing, Inc. 100 Middle Street Lincoln, RI 02865 (401) 727-0183 -----Original Message----- From: Edis, Bob [mailto:bob.edis@fleetpride.com] Sent: Thursday, May 03, 2001 3:01 PM To: 'powerh-l@list.swau.edu' Subject: Strange results from DB2 in Quiz G'day all Environment: PH 8.21D4, WinNT 4.0, HitODBC 400, DB/400 We are using HitODBC to pull data from a DB2 database on an AS/400 machine and are getting erroneous results in PowerHouse Quiz. E.g. Case record: DB2 shows the cost amount column as 3.670 DecisionStream using the above ODBC driver shows 3.670 Quiz using the above ODBC driver shows 3.610. In another case Quiz reports 4.950 whereas DB2 and DS show 5.00. The values reported by DB2 and DecisionStream are the correct ones. The column is stored to 3 decimal places (float) in the DB2 table. The discrepancy between Quiz and DecisionStream is NOT consistent. I.e. some records in the table match exactly. Any ideas why Quiz is reporting incorrect values? Regards, Robert Edis Manager - Business Intelligence FleetPride 520 Lake Cook Rd., Suite 100 Deerfield IL 60015 Tel: 847.572.8039 Fax: 847.444.1096 www.fleetpride.com = = = = = = = = = = = = = = = = = = = = = = = = = = = = Mailing list: powerh-l@lists.swau.edu Subscribe: "subscribe" in message body to powerh-l-request@lists.swau.edu Unsubscribe: "unsubscribe" in message body to powerh-l-request@lists.swau.edu http://lists.swau.edu/mailman/listinfo/powerh-l This list is closed, thus to post to the list you must be a subscriber. From Bob.Deskin@Cognos.COM Thu, 3 May 2001 15:55:50 -0400 Date: Thu, 3 May 2001 15:55:50 -0400 From: Deskin, Bob Bob.Deskin@Cognos.COM Subject: Supportlink: who edits the ADT stuff? I will forward your comments to the SupportLink editor. Bob -----Original Message----- From: Pickering, John (NORBORD) [mailto:PICKERIJ@norbord.com] Sent: Thursday, May 03, 2001 3:41 PM To: 'Deskin, Bob' Cc: 'Powerhouse List' Subject: Supportlink: who edits the ADT stuff? Mr. Deskin Please take a look at page 67 of the just arrived (on my desk anyway) Supportlink volume 11 number 2, April - June 2001. This is the sort of extravagant code that looks important but half of it is unnecessary! The kind of stuff one inherits from others and spends heaps of time wading through the muck trying to find the pony. Not a really brilliant example of PowerHouse code for such an august publication :-( Why test ACCESSOK in the statement after the WHILE RETRIEVING? This verb will execute the statement after it ONLY if it retrieves a record. A simple BEGIN...END would be much clearer! And since ACCESSOK will always be false when you read to the end of file or read all the values for a single key, it seems kind of pointless to test it. Lose this one too! And since t_bill_no will always have the last value retrieved (or the initial value if the file is empty), why not simply pass it? There seems little point in moving it to another temporary item! Or use the t_final_xxx temporary items inside the WHILE RETRIEVING loop. And for those who don't like unnecessary compile time warnings such as "*W* the PRIMARY file is missing." you should probably use the SCREEN statement MENU option. And I won't even start on what the user will be presented with upon return from the lower level screen! Sheesh! Regards, JWP = = = = = = = = = = = = = = = = = = = = = = = = = = = = Mailing list: powerh-l@lists.swau.edu Subscribe: "subscribe" in message body to powerh-l-request@lists.swau.edu Unsubscribe: "unsubscribe" in message body to powerh-l-request@lists.swau.edu http://lists.swau.edu/mailman/listinfo/powerh-l This list is closed, thus to post to the list you must be a subscriber. From PICKERIJ@norbord.com Thu, 3 May 2001 17:26:40 -0400 Date: Thu, 3 May 2001 17:26:40 -0400 From: Pickering, John (NORBORD) PICKERIJ@norbord.com Subject: Strange results from DB2 in Quiz Byron I can't comment on the "different layers" theory in your first paragraph. But if both DecisionStream and Quiz are using the same ODBC driver then how can they affect DB2 on the remote server? Are they making different sorts of requests of the ODBC driver? I would heartily concur with your comments about the inappropriateness of floating point numbers in typical business calculations. But what Bob is experiencing (as per his example, not below) seems much greater than what can be explained by simple floating point precision problems -- e.g. >DecisionStream using the above ODBC driver shows 3.670 >Quiz using the above ODBC driver shows 3.610. Were it just float problems I would expect 3.70 versus 3.6699999999999999999 or some such value. And this could be cured by judicious rounding when converting to something sensible like integer datatypes. Seems like something else is going on here. Regards, JWP > -----Original Message----- > From: Byron Welch [SMTP:byron.welch@creatcomp.com] > Sent: Thursday, May 03, 2001 3:46 PM > To: 'Edis, Bob'; 'powerh-l@list.swau.edu' > Subject: RE: Strange results from DB2 in Quiz > > Bob, > Two comments but unfortunately I don't have a solution:( > > I think you are getting different results between PowerHouse and > DecisionStream because they are using different database layers. My > understanding is that they will be merged in the next major release but > for > now you get slightly different database layers between PowerHouse and the > newer products. > > I think you are running into a precession error. In a couple of cases > when I was using floating point in the past I came across similar results. > If I remember correctly the PH manuals used to have a caveat about > floating > point numbers. It was something like for fractions or very large numbers > a > loss of precision can occur. i.e. .2 might actually be stored as .1999. > This used to prevent me from comparing a Floating point to an integer in a > select statement. I usually solved this by multiplying by 100 and then > dividing by 100 in a define. > > Byron Welch > Manager Consulting Services > Creative Computing, Inc. > 100 Middle Street > Lincoln, RI 02865 > (401) 727-0183 > > > -----Original Message----- > From: Edis, Bob [mailto:bob.edis@fleetpride.com] > Sent: Thursday, May 03, 2001 3:01 PM > To: 'powerh-l@list.swau.edu' > Subject: Strange results from DB2 in Quiz > > > G'day all > > Environment: PH 8.21D4, WinNT 4.0, HitODBC 400, DB/400 > > We are using HitODBC to pull data from a DB2 database on an AS/400 machine > and are getting erroneous results in PowerHouse Quiz. E.g. > > Case record: > > DB2 shows the cost amount column as 3.670 > DecisionStream using the above ODBC driver shows 3.670 > Quiz using the above ODBC driver shows 3.610. > > In another case Quiz reports 4.950 whereas DB2 and DS show 5.00. > > The values reported by DB2 and DecisionStream are the correct ones. The > column is stored to 3 decimal places (float) in the DB2 table. > > The discrepancy between Quiz and DecisionStream is NOT consistent. I.e. > some records in the table match exactly. > > Any ideas why Quiz is reporting incorrect values? > > Regards, > > Robert Edis > Manager - Business Intelligence > FleetPride > 520 Lake Cook Rd., Suite 100 > Deerfield IL 60015 > Tel: 847.572.8039 > Fax: 847.444.1096 > www.fleetpride.com > > From bob.edis@fleetpride.com Thu, 3 May 2001 16:37:17 -0500 Date: Thu, 3 May 2001 16:37:17 -0500 From: Edis, Bob bob.edis@fleetpride.com Subject: Strange results from DB2 in Quiz G'day all This is a case of not examining the code closely enough. The Quiz code was NOT looking at the same column as the DecisionStream code! (a very close alias though) In my experience most PowreHouse issues stem from either bad data or bad facts. This is a case of the latter. Blue -----Original Message----- From: Edis, Bob [mailto:bob.edis@fleetpride.com] Sent: Thursday, May 03, 2001 2:01 PM To: 'powerh-l@list.swau.edu' Subject: Strange results from DB2 in Quiz G'day all Environment: PH 8.21D4, WinNT 4.0, HitODBC 400, DB/400 We are using HitODBC to pull data from a DB2 database on an AS/400 machine and are getting erroneous results in PowerHouse Quiz. E.g. Case record: DB2 shows the cost amount column as 3.670 DecisionStream using the above ODBC driver shows 3.670 Quiz using the above ODBC driver shows 3.610. In another case Quiz reports 4.950 whereas DB2 and DS show 5.00. The values reported by DB2 and DecisionStream are the correct ones. The column is stored to 3 decimal places (float) in the DB2 table. The discrepancy between Quiz and DecisionStream is NOT consistent. I.e. some records in the table match exactly. Any ideas why Quiz is reporting incorrect values? Regards, Robert Edis Manager - Business Intelligence FleetPride 520 Lake Cook Rd., Suite 100 Deerfield IL 60015 Tel: 847.572.8039 Fax: 847.444.1096 www.fleetpride.com = = = = = = = = = = = = = = = = = = = = = = = = = = = = Mailing list: powerh-l@lists.swau.edu Subscribe: "subscribe" in message body to powerh-l-request@lists.swau.edu Unsubscribe: "unsubscribe" in message body to powerh-l-request@lists.swau.edu http://lists.swau.edu/mailman/listinfo/powerh-l This list is closed, thus to post to the list you must be a subscriber. From bourchih@unive.nl Fri, 4 May 2001 11:22:46 +0200 Date: Fri, 4 May 2001 11:22:46 +0200 From: bourchih@unive.nl bourchih@unive.nl Subject: Don't open this mail. It's just a test. Sorry! I am just testing my powerhouse list email adres. sorry. From Leonard_Berkowitz@harvardpilgrim.org Fri, 4 May 2001 09:34:04 -0400 Date: Fri, 4 May 2001 09:34:04 -0400 From: Leonard_Berkowitz@harvardpilgrim.org Leonard_Berkowitz@harvardpilgrim.org Subject: Quick: superfluous update records HP3000; PowerHouse v. 8.19 Our major application uses Quick screens. Recently we have installed the Bridge product (a subset or superset of Quest's Netbase) that creates transaction logging files. Now that we have these files, we have spent some time analyzing transactions. One phenomenon that we have noticed is that there is a huge number of update transactions. The before-image of the record is mostly spaces, with only the key values present. The after-image, of course, is all of the data that the user entered. Quite a lot of I/O overhead, I would say. The question is: Is this S.O.P. for Quick or is there something in the application that if forcing incomplete records to be written to the data base and then updated? =================== Leonard S. Berkowitz Perot Health Care Systems (Harvard Pilgrim Health Care account) voice: 617-509-1212 fax: 617-509-3737 pager: 781-226-2431 From Bob.Deskin@Cognos.COM Fri, 4 May 2001 09:51:42 -0400 Date: Fri, 4 May 2001 09:51:42 -0400 From: Deskin, Bob Bob.Deskin@Cognos.COM Subject: Quick: superfluous update records I'm assuming you're using IMAGE. It depends on your screen structure. If you look at the standard UPDATE procedure, you'll see that it does a PUT for new records and then later does a PUT to update. That's because IMAGE requires master dataset records to be added before corresponding detail dataset records (for those of you who are not MPE/iX people, we're talking IMAGE master-detail datasets and not QUICK MASTER DETAIL files). If you change the content of the master record while updating the details, for example from an ITEM FINAL, then you may well do an update on the master giving you two PUTs. By the way, if you're adding to an IMAGE master in QTP, it automatically does an ADD and an UPDATE when you use an OUTPUT ADD for the same reason. A typical sequence is OUTPUT image-detail ADD OUTPUT image-master ADD AT key-item Since you need the IMAGE master before you can add the details, QTP adds the master with key values only AT START OF key-item, and then updates AT "end of" key-item. Bob Deskin PowerHouse Web Product Manager, Application Development Tools, Cognos Inc. bob.deskin@cognos.com (613) 738-1338 ext 7268 FAX: (613) 727-1178 3755 Riverside Drive P.O. Box 9707 Stn. T, Ottawa ON K1G 4K9 CANADA -----Original Message----- From: Leonard_Berkowitz@harvardpilgrim.org [mailto:Leonard_Berkowitz@harvardpilgrim.org] Sent: Friday, May 04, 2001 9:34 AM To: powerh-l@lists.swau.edu Subject: Quick: superfluous update records HP3000; PowerHouse v. 8.19 Our major application uses Quick screens. Recently we have installed the Bridge product (a subset or superset of Quest's Netbase) that creates transaction logging files. Now that we have these files, we have spent some time analyzing transactions. One phenomenon that we have noticed is that there is a huge number of update transactions. The before-image of the record is mostly spaces, with only the key values present. The after-image, of course, is all of the data that the user entered. Quite a lot of I/O overhead, I would say. The question is: Is this S.O.P. for Quick or is there something in the application that if forcing incomplete records to be written to the data base and then updated? =================== Leonard S. Berkowitz Perot Health Care Systems (Harvard Pilgrim Health Care account) voice: 617-509-1212 fax: 617-509-3737 pager: 781-226-2431 = = = = = = = = = = = = = = = = = = = = = = = = = = = = Mailing list: powerh-l@lists.swau.edu Subscribe: "subscribe" in message body to powerh-l-request@lists.swau.edu Unsubscribe: "unsubscribe" in message body to powerh-l-request@lists.swau.edu http://lists.swau.edu/mailman/listinfo/powerh-l This list is closed, thus to post to the list you must be a subscriber. From bob.edis@fleetpride.com Fri, 4 May 2001 12:38:31 -0500 Date: Fri, 4 May 2001 12:38:31 -0500 From: Edis, Bob bob.edis@fleetpride.com Subject: using env vars in PH on NT4 G'day all Has anyone been using DOS environment variables with their Quiz programs? I am trying to use something like 'USE %phuse%myusefile.use' but the variable is not translating. Instead I get an error message that says: *W* No such file or directory The variable phuse is set in the .BAT file that executes the Quiz program as SET phuse=d:\dirpath\. Regards, Robert Edis Manager - Business Intelligence FleetPride 520 Lake Cook Rd., Suite 100 Deerfield IL 60015 Tel: 847.572.8039 Fax: 847.444.1096 www.fleetpride.com From bob.edis@fleetpride.com Fri, 4 May 2001 12:49:26 -0500 Date: Fri, 4 May 2001 12:49:26 -0500 From: Edis, Bob bob.edis@fleetpride.com Subject: FW: using env vars in PH on NT4 Thanks Christina We have the environment variables set as system variables but even the batch (.BAT) files wouldn't translate them, let alone Quiz. So instead we placed all the relevant variables, phd, phuse, phtemp, phexe, phlog, etc., in a common .BAT file and call that from the jobs that run the PowerHouse programs. The system variables DO work when in interactive mode however. E.g. phedit is working correctly when issuing a REVISE command in Qdesign. Regards, Blue -----Original Message----- From: Hasse, Christina [mailto:Christina.Hasse@COGNOS.com] Sent: Friday, May 04, 2001 12:46 PM To: 'Edis, Bob' Subject: RE: using env vars in PH on NT4 Hi Bob, You should try setting phuse as a system variable and see if it works that way. Regards, Christina Hasse ADT North American Services Mgr. COGNOS CORPORATION christina.hasse@cognos.com Office: 847 - 285 - 2905 Cell: 847 - 612 - 6854 -----Original Message----- From: Edis, Bob [mailto:bob.edis@fleetpride.com] Sent: Friday, May 04, 2001 12:39 PM To: 'powerh-l@list.swau.edu' Subject: using env vars in PH on NT4 G'day all Has anyone been using DOS environment variables with their Quiz programs? I am trying to use something like 'USE %phuse%myusefile.use' but the variable is not translating. Instead I get an error message that says: *W* No such file or directory The variable phuse is set in the .BAT file that executes the Quiz program as SET phuse=d:\dirpath\. Regards, Robert Edis Manager - Business Intelligence FleetPride 520 Lake Cook Rd., Suite 100 Deerfield IL 60015 Tel: 847.572.8039 Fax: 847.444.1096 www.fleetpride.com = = = = = = = = = = = = = = = = = = = = = = = = = = = = Mailing list: powerh-l@lists.swau.edu Subscribe: "subscribe" in message body to powerh-l-request@lists.swau.edu Unsubscribe: "unsubscribe" in message body to powerh-l-request@lists.swau.edu http://lists.swau.edu/mailman/listinfo/powerh-l This list is closed, thus to post to the list you must be a subscriber. From bob.edis@fleetpride.com Mon, 7 May 2001 08:23:40 -0500 Date: Mon, 7 May 2001 08:23:40 -0500 From: Edis, Bob bob.edis@fleetpride.com Subject: Cognos PowerHouse programmer wanted in Dallas G'day all I heard about this PowerHouse programming gig and thought someone on the list might be interested, so here it is. Regards, Blue > >Here is the job description!! > >This is a 6 to 9 month contract. >Must have at least 5 years experience (current) with Powerhouse in a DEC/VAX >environment. Prefer to know DEC/VAX Basic . Prefer to know DCL. > >This person will work on a small team developing and enhancing existing >mortgage/financial applications. > >Thank you, >Stephanie >Stephanie Bilotta >Partner >SearchNetwork, L.L.C. >5220 Spring Valley Rd #250 >Dallas, Tx 75006 >wk (972) 503-8181 ext.102 >fax (972) 503-8192 >stephanieb@SearchNetwork.net From bob.edis@fleetpride.com Mon, 7 May 2001 08:47:00 -0500 Date: Mon, 7 May 2001 08:47:00 -0500 From: Edis, Bob bob.edis@fleetpride.com Subject: local user group meeting near Chicago G'day all There is some interest in the area where I work (Deerfield, Illinois, USA - 45 minutes north of Chicago) to re-establish a local Cognos user group focusing on application development. To this end I have received permission to utilize my company's conference room and plan on sponsoring regular meetings (once a month) here. My idea for the meeting agenda is: 1. presentation by participant on new tool or method 2. workshop where people can present tool/method issues and the group can brainstorm on ideas. The meeting should be no more than 1 to 1.5 hours long. The area of discussion should be restricted to Cognos ADT tools except where those tools are being used in concert with other tools, e.g. Impromptu or MS FrontPage or Apache IS. Any interested party can attend but I will be specifically directing invitations to people working/living within 1/2 hour of Deerfield. The first meeting, an informal get together to establish how future meetings should work, will be held on May 10th, 2001 at the address below. If anyone would like to attend this and future meetings, please let me know and I can provide directions. Any suggestions on how to make these meetings productive and enjoyable are also appreciated. Regards, Robert Edis Manager - Business Intelligence FleetPride 520 Lake Cook Rd., Suite 100 Deerfield IL 60015 Tel: 847.572.8039 Fax: 847.444.1096 www.fleetpride.com From james.stanton@execrelo.com Mon, 7 May 2001 11:19:29 -0400 Date: Mon, 7 May 2001 11:19:29 -0400 From: james.stanton@execrelo.com james.stanton@execrelo.com Subject: Powerhouse terminal code to Axiant thin client In working with Axiant 3.0 on a HPe3000, I am trying to get this designer to work. The code first builds and then loads a case number to print in the parameter file, then it run a quiz report to access the case number and print a temp file to be processed by fantasia which prints logos, etc.... I am running netd to stream a new job for each user. The Cognos support person stated the following: "The reason the Quiz report is not working when you call it from the Axiant screen is that the various commands are not being run in the same session, as they are when you run the screen online. You can run all of the commands together by putting them into a single command file, and calling the command file from the Axiant screen procedure with a single RUN COMMAND." Will this type of Powerhouse terminal code work in Axiant thin client? PROCEDURE DESIGNER p HELP "Print Consultant Chronology." BEGIN IF ALTEREDRECORD OF CounslChronology THEN BEGIN INFORMATION "You haven't updated (use US then try again)." END ELSE BEGIN RUN COMMAND "reset parm" CLEAR ALL NOWARN RUN COMMAND "purge parm,temp" NOWARN RUN COMMAND "build parm;rec=-150,1,f,ascii;disc=10;temp" NOWARN GET Parameter opt LET ParameterNumber0 OF Parameter = CaseNumber OF Case PUT Parameter CLOSE Parameter RUN COMMAND "PURGE CHRONLST,TEMP" RUN COMMAND & "BUILD CHRONLST;REC=-132,,F,ASCII;CCTL;DISC=10000;TEMP" RUN COMMAND "file OUTPUT=CHRONLST,OLDTEMP;CCTL;DEV=DISC" RUN COMMAND "file quizuse=BT0002QZ" RUN COMMAND "QUIZ" RUN COMMAND "RUN LASER.LASER.LARC;INFO='CHRONO.FENV QDEV=!LJPRINT'" & NOWARN RUN COMMAND "reset quizuse" REFRESH ALL INFORMATION " Consultant Chronology printed." END END From joann.l.olsen@us.abb.com Mon, 7 May 2001 12:59:37 -0500 Date: Mon, 7 May 2001 12:59:37 -0500 From: joann.l.olsen@us.abb.com joann.l.olsen@us.abb.com Subject: PH Access to Data Areas We are on an AS/400 platform, but using some PowerHouse for creating reports. When a user runs a Dream Writer, the version that they have selected is held in the Logical Data Area (LDA). When the job ends the LDA ceases to exist. I would like to be able to use the version in my PowerHouse Quiz routine to obtain the responses to the various prompts in the Data Selection Values. The version is part of the index needed to read the F98301 & F9831 files. How can I read this data area or move the information contained in the data area to a subfile? Thanks for your help. From frenkel@worldnet.att.net Mon, 7 May 2001 22:16:54 -0700 Date: Mon, 7 May 2001 22:16:54 -0700 From: david frenkel frenkel@worldnet.att.net Subject: Employment I am an experienced HP3000/Powerhouse P/A looking for contract or direct work in WA, WI, MN or IL. Thanks. Dave Frenkel 425-260-5030 From Conrad.Whittall@Cognos.COM Tue, 8 May 2001 15:06:19 -0400 Date: Tue, 8 May 2001 15:06:19 -0400 From: Whittall, Conrad Conrad.Whittall@Cognos.COM Subject: PH Access to Data Areas OK, please bear with me here as I haven't touched an AS/400 (unfortunately) for a few years. However, I do remember that one of the enhancements introduced with PowerHouse 6.07 was the ability to treat data areas just like any other PowerHouse data FILE. You need to declare the FILE in your PowerHouse data dictionary as normal, except that the file TYPE will be DATA AREA...which may be a local, process or group data area (which you specify as *LDA, *PDA or *GDA with no library name as the OPEN name for the file). PowerHouse treats these data areas as single record files with a PowerHouse ORGANIZATION of DIRECT. You can find a discussion of this in the "PowerHouse for AS/400 Dictionary and Utilities Reference" manual for version 6.07 of PowerHouse, under the FILE statement on pages 215-235. My guess is that once you have declared your data areas as FILEs in the PowerHouse data dictionary then you can use either QUICK or QTP to read and write to these data areas, as well as using QUIZ to read them. I hope that this points you in the right direction. Best regards, Conrad Conrad Whittall Marketing Manager, Application Development Tools Cognos Incorporated, 3755 Riverside Drive, Ottawa, Ontario, K1G 4K9, Canada Developer productivity never goes out of style! For details of the Cognos PowerHouse family of high-productivity development tools for Web, Windows and terminal-based business applications, and to request your own free copy of our new product demo CD, simply click here http://www.cognos.com/powerhouse. -----Original Message----- From: joann.l.olsen@us.abb.com [mailto:joann.l.olsen@us.abb.com] Sent: Monday 7 May 2001 14:00 To: 'powerh-l@list.swau.edu' Subject: PH Access to Data Areas We are on an AS/400 platform, but using some PowerHouse for creating reports. When a user runs a Dream Writer, the version that they have selected is held in the Logical Data Area (LDA). When the job ends the LDA ceases to exist. I would like to be able to use the version in my PowerHouse Quiz routine to obtain the responses to the various prompts in the Data Selection Values. The version is part of the index needed to read the F98301 & F9831 files. How can I read this data area or move the information contained in the data area to a subfile? Thanks for your help. From Joe.Boyle@cognos.com Wed, 9 May 2001 05:20:08 -0400 Date: Wed, 9 May 2001 05:20:08 -0400 From: Boyle, Joe Joe.Boyle@cognos.com Subject: using env vars in PH on NT4 Hi Bob, try the following, suppose you have variables 'set myd=c' and 'set mydir=liz' as below, U:\>set my myd=c mydir=liz In your bat file add a command such as 'set myloc=%myd%:\%mydir%' then when you issue the use command, quiz will action the file c:\liz\quse.txt, see below, U:\>quiz *E* No dictionary file with the specified name Q U I Z (PowerHouse 8.21.D4) Copyright 2000 COGNOS INCORPORATED Licensed PH8-TEMPORARY for customer: 071257000 > use $myloc\quse.txt list > set dict D:\PW2201EB\Demos\phd.pdc > U:\>set my myd=c mydir=liz myloc=c:\liz regards, Joe Boyle. -----Original Message----- From: Edis, Bob [mailto:bob.edis@fleetpride.com] Sent: 04 May 2001 18:49 To: 'powerh-l@list.swau.edu' Subject: FW: using env vars in PH on NT4 Thanks Christina We have the environment variables set as system variables but even the batch (.BAT) files wouldn't translate them, let alone Quiz. So instead we placed all the relevant variables, phd, phuse, phtemp, phexe, phlog, etc., in a common .BAT file and call that from the jobs that run the PowerHouse programs. The system variables DO work when in interactive mode however. E.g. phedit is working correctly when issuing a REVISE command in Qdesign. Regards, Blue -----Original Message----- From: Hasse, Christina [mailto:Christina.Hasse@COGNOS.com] Sent: Friday, May 04, 2001 12:46 PM To: 'Edis, Bob' Subject: RE: using env vars in PH on NT4 Hi Bob, You should try setting phuse as a system variable and see if it works that way. Regards, Christina Hasse ADT North American Services Mgr. COGNOS CORPORATION christina.hasse@cognos.com Office: 847 - 285 - 2905 Cell: 847 - 612 - 6854 -----Original Message----- From: Edis, Bob [mailto:bob.edis@fleetpride.com] Sent: Friday, May 04, 2001 12:39 PM To: 'powerh-l@list.swau.edu' Subject: using env vars in PH on NT4 G'day all Has anyone been using DOS environment variables with their Quiz programs? I am trying to use something like 'USE %phuse%myusefile.use' but the variable is not translating. Instead I get an error message that says: *W* No such file or directory The variable phuse is set in the .BAT file that executes the Quiz program as SET phuse=d:\dirpath\. Regards, Robert Edis Manager - Business Intelligence FleetPride 520 Lake Cook Rd., Suite 100 Deerfield IL 60015 Tel: 847.572.8039 Fax: 847.444.1096 www.fleetpride.com = = = = = = = = = = = = = = = = = = = = = = = = = = = = Mailing list: powerh-l@lists.swau.edu Subscribe: "subscribe" in message body to powerh-l-request@lists.swau.edu Unsubscribe: "unsubscribe" in message body to powerh-l-request@lists.swau.edu http://lists.swau.edu/mailman/listinfo/powerh-l This list is closed, thus to post to the list you must be a subscriber. = = = = = = = = = = = = = = = = = = = = = = = = = = = = Mailing list: powerh-l@lists.swau.edu Subscribe: "subscribe" in message body to powerh-l-request@lists.swau.edu Unsubscribe: "unsubscribe" in message body to powerh-l-request@lists.swau.edu http://lists.swau.edu/mailman/listinfo/powerh-l This list is closed, thus to post to the list you must be a subscriber. From Bjorn.Selfors@kby.BERGEN-KOMMUNE.telemax.no 09 May 2001 13:16:07 Z Date: 09 May 2001 13:16:07 Z From: Bjorn.Selfors@kby.BERGEN-KOMMUNE.telemax.no Bjorn.Selfors@kby.BERGEN-KOMMUNE.telemax.no Subject: Axiant 2.03 - 3.0 We are planning to upgrade from Axiant 2.03 to 3.0 in a NT / Oracle environment. Anybody have any experience about this upgrade? What should we be aware of? Bjřrn Selfors Delfi Data a.s. Tlf +47 5556 8887 http://www.delfidata.no From MartinM@Rosebys.com Wed, 9 May 2001 14:52:04 +0100 Date: Wed, 9 May 2001 14:52:04 +0100 From: Martin McDonough MartinM@Rosebys.com Subject: Axiant 2.03 - 3.0 You will find in axiant 3 hot areas are not needed. In version 2, to enable a user to change data in a field you needed to add a hot area over the field. The advantage of this was, if you didn't want the user to change a piece of data, don't put a hot area over it. Now, in version 3, any fields in the screen (unless from a reference file) will be available for change, all the user has to do is click in the field. You may have to go through your screens and change these fields to display only. That was the main one, I'll let you know if any others come to mind Regards Martin McDonough. > -----Original Message----- > From: Bjorn.Selfors@kby.BERGEN-KOMMUNE.telemax.no > [SMTP:Bjorn.Selfors@kby.BERGEN-KOMMUNE.telemax.no] > Sent: 09 May 2001 14:16 > To: powerh-l@lists.swau.edu > Subject: Axiant 2.03 - 3.0 > > > We are planning to upgrade from Axiant 2.03 to 3.0 in a NT / Oracle > environment. > Anybody have any experience about this upgrade? What should we be aware > of? > > Bjřrn Selfors > Delfi Data a.s. > Tlf +47 5556 8887 > http://www.delfidata.no > > = = = = = = = = = = = = = = = = = = = = = = = = = = = = > Mailing list: powerh-l@lists.swau.edu > Subscribe: "subscribe" in message body to powerh-l-request@lists.swau.edu > Unsubscribe: "unsubscribe" in message body to > powerh-l-request@lists.swau.edu > http://lists.swau.edu/mailman/listinfo/powerh-l > This list is closed, thus to post to the list you must be a subscriber. From steve_franklin@yahoo.com Wed, 9 May 2001 09:33:13 -0500 Date: Wed, 9 May 2001 09:33:13 -0500 From: Steve Franklin steve_franklin@yahoo.com Subject: Axiant question This is a multi-part message in MIME format. ------=_NextPart_000_0001_01C0D86B.1802A260 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: 7bit I am doing a pilot conversion from Powerhouse on AIX to Axiant 3.0. In the Powerhouse screens on AIX, we use function keys extensively to perform operations such as F9 = skip all, designer proc. Is there a way in Axiant to assign two different commands to either a button or a function key. I have looked through commands and events and could not determine how to do this. Thank you for your assistance, Steve Franklin Stw Inc. Email: steve_franklin@yahoo.com ------=_NextPart_000_0001_01C0D86B.1802A260 Content-Type: text/html; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable

I am doing a pilot conversion from Powerhouse on AIX = to Axiant=20 3.0.  In the Powerhouse screens on AIX, we use function keys = extensively to=20 perform operations such as F9 =3D skip all, designer proc.  Is = there a way in=20 Axiant to assign two different commands to either a button or a function = key.  I have looked through commands and events and could not = determine how=20 to do this.

Thank you for your assistance,

Steve = Franklin
Stw=20 Inc.

Email: = steve_franklin@yahoo.com

------=_NextPart_000_0001_01C0D86B.1802A260-- _________________________________________________________ Do You Yahoo!? Get your free @yahoo.com address at http://mail.yahoo.com From Conrad.Whittall@Cognos.COM Wed, 9 May 2001 11:08:16 -0400 Date: Wed, 9 May 2001 11:08:16 -0400 From: Whittall, Conrad Conrad.Whittall@Cognos.COM Subject: Axiant 2.03 - 3.0 Hello Bjorn, I'm sure that others will contribute their experiences and recommendations for upgrading from Axiant 2.03 to 3.0, but there is one particular feature of the upgrade process that I strongly recommend that you use. That feature is having the Axiant 3.0 repository upgrade process automatically add Event and Toolbar sections to all of your existing screen programs. See the section entitled "Upgrading Axiant 4GL 2.02 or 2.03 Repositories" in the Axiant 3.0 Windows Help, under "Axiant 4GL Release Notes | Upgrade Information for Existing Axiant 4GL Users". Doing this will mean that you can immediately benefit from the new default point-and-click interface for fields in Axiant screen programs, and from the new user toolbars. Even if your application currently uses hot area buttons over data fields to simulate a point-and-click interface, I would recommend that you have Axiant 3.0 automatically add the Event section that controls this as you upgrade your existing repository. The processing triggered by the new Event section will not take effect until you remove the hot area that is positioned over the field, since the hot area is effectively masking the field from the user and so when a user clicks over the field they are clicking the hot area and not the field itself (where the Event section would be triggered). You may then be able to use the find and change capabilities within Axiant 3.0 to find and remove the hot areas (and simply delete the corresponding procedures from the Procedure Section if they are no longer needed). Even if you don't plan on using the capabilities of the new Event and Toolbar sections immediately it is far easier to have the Axiant 3.0 upgrade process add the inheritance of these sections to your existing screen programs than for you to go back and add them manually after converting your repository to the 3.0 format. If you don't want to use the capabilities of these sections straightaway you could simply delete the objects that appear in the default sections that are being inherited (in the library in which they exist) so that these sections are empty when inherited by the programs at build time. Best regards, Conrad Conrad Whittall Marketing Manager, Application Development Tools Cognos Incorporated, 3755 Riverside Drive, Ottawa, Ontario, K1G 4K9, Canada Developer productivity never goes out of style! For details of the Cognos PowerHouse family of high-productivity development tools for Web, Windows and terminal-based business applications, and to request your own free copy of our new product demo CD, simply click here http://www.cognos.com/powerhouse. -----Original Message----- From: Bjorn.Selfors@kby.BERGEN-KOMMUNE.telemax.no [mailto:Bjorn.Selfors@kby.BERGEN-KOMMUNE.telemax.no] Sent: Wednesday 9 May 2001 09:16 To: powerh-l@lists.swau.edu Subject: Axiant 2.03 - 3.0 We are planning to upgrade from Axiant 2.03 to 3.0 in a NT / Oracle environment. Anybody have any experience about this upgrade? What should we be aware of? Bjorn Selfors Delfi Data a.s. Tlf +47 5556 8887 http://www.delfidata.no From robert.mills@windsong-services.co.uk Thu, 10 May 2001 16:19:30 +0100 Date: Thu, 10 May 2001 16:19:30 +0100 From: Robert Mills robert.mills@windsong-services.co.uk Subject: Clear All not working in an Internal Quick procedure Hi fellow listers, I have an internal procedure that is coded as follows: > Clear All > While Retrieving > Begin > Run Command Of > End ; While Retrieving > Refresh All It looks as if the 'Clear All' is not working. I've even tried using 'Clear Screen' but it still don't work. Anybody know why? Environment is 819c5 running under MPE/iX 6.5 regards, +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ |Robert W.Mills (Systems Development Manager) | |Windsong Services, St. Mary Cray, ENGLAND | |Tel : +44 (0)1689 870622 x3005 | |Fax : +44 (0)1689 899026 | |Email: Robert.Mills@windsong-services.co.uk | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ From Bob.Deskin@Cognos.COM Thu, 10 May 2001 11:30:03 -0400 Date: Thu, 10 May 2001 11:30:03 -0400 From: Deskin, Bob Bob.Deskin@Cognos.COM Subject: Clear All not working in an Internal Quick procedure Typically the results of doing anything to the display are not obvious until there's some output to the display, whether for display or prompting. Do you expect the display to clear while you're doing the WHILE RETRIEVING? Try adding INFO "" NOW after the CLEAR. Bob Deskin PowerHouse Web Product Manager, Application Development Tools, Cognos Inc. bob.deskin@cognos.com (613) 738-1338 ext 7268 FAX: (613) 727-1178 3755 Riverside Drive P.O. Box 9707 Stn. T, Ottawa ON K1G 4K9 CANADA -----Original Message----- From: Robert Mills [mailto:robert.mills@windsong-services.co.uk] Sent: Thursday, May 10, 2001 11:20 AM To: 'PowerHouse List' Subject: Clear All not working in an Internal Quick procedure Hi fellow listers, I have an internal procedure that is coded as follows: > Clear All > While Retrieving > Begin > Run Command Of > End ; While Retrieving > Refresh All It looks as if the 'Clear All' is not working. I've even tried using 'Clear Screen' but it still don't work. Anybody know why? Environment is 819c5 running under MPE/iX 6.5 regards, +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ |Robert W.Mills (Systems Development Manager) | |Windsong Services, St. Mary Cray, ENGLAND | |Tel : +44 (0)1689 870622 x3005 | |Fax : +44 (0)1689 899026 | |Email: Robert.Mills@windsong-services.co.uk | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ = = = = = = = = = = = = = = = = = = = = = = = = = = = = Mailing list: powerh-l@lists.swau.edu Subscribe: "subscribe" in message body to powerh-l-request@lists.swau.edu Unsubscribe: "unsubscribe" in message body to powerh-l-request@lists.swau.edu http://lists.swau.edu/mailman/listinfo/powerh-l This list is closed, thus to post to the list you must be a subscriber. From Chris.Sharman@ccagroup.co.uk Thu, 10 May 2001 16:48:51 +0100 Date: Thu, 10 May 2001 16:48:51 +0100 From: Chris Sharman Chris.Sharman@ccagroup.co.uk Subject: Clear All not working in an Internal Quick procedure >> Clear All >> While Retrieving >> Begin >> Run Command Of >> End ; While Retrieving >> Refresh All > >It looks as if the 'Clear All' is not working. >I've even tried using 'Clear Screen' but it still don't work. >Anybody know why? >Environment is 819c5 running under MPE/iX 6.5 According to my book (VMS, which may be different to yours): Clear & Refresh both need the "restore=lines" program parameter. Also, they're only done when Quick is ready to prompt, which won't be until some time after the refresh all. I'd recommend using just one command (which could invoke a script if necessary), and the necessary clear/refresh clauses on the one run command. Keeping shell scripts in a database seems like unnecessary hard work. An alternative would be to write a script on the fly from the database, then invoke it. Chris _______________________________________________________________________ Chris.Sharman@CCAgroup.co.uk http://www.ccagroup.co.uk/ CCA Stationery Ltd, Eastway, Fulwood, Preston, Lancashire, PR2 9WS. ----------------------------------------------------------------------- Any views expressed in this message are those of the sender and not necessarily those of CCA Group. The unauthorized use, disclosure, copying or alteration of this message is forbidden. The contents of this message may be confidential and/or privileged, copyright CCA Group and are intended solely for the use of the individual or entity to whom they are addressed. Whilst this message has been scanned, CCA Group cannot guarantee that it is virus free or compatible with your systems and accepts no responsibility for any loss or damage arising from its use. The recipient is advised to run their own anti-virus software. If you receive this message in error please contact postmaster@ccagroup.co.uk immediately, destroy any copies and delete it from your computer systems. From bob.edis@fleetpride.com Thu, 10 May 2001 12:37:28 -0500 Date: Thu, 10 May 2001 12:37:28 -0500 From: Edis, Bob bob.edis@fleetpride.com Subject: Cognos ADT user group meeting G'day all For those interested in attending the inaugural user group meeting in Deerfield, IL, USA tonight (Thursday, May 10) at 6PM to 7:30PM CST, the directions are: >From north of Deerfield, come South on I94 and exit at Lake Cook Rd., turn left (East) and continue under railway bridge then take first turn on left. Follow this road and 520 is the second building on the left. Enter through the revolving doors and check in with the security desk. FleetPride is the first suite on the right as you enter the building. >From south of Deerfield come north on either I94 or I294 and exit at Lake Cook Rd. then turn right and the rest is the same as above. >From the West, follow Lake Cook Rd. east until you pass over I94 then same as above. I hope we can get something interesting, exciting and edifying happening here. Regards, Robert Edis Manager - Business Intelligence FleetPride 520 Lake Cook Rd., Suite 100 Deerfield IL 60015 Tel: 847.572.8039 Fax: 847.444.1096 www.fleetpride.com From Aci.Polajnar@rcc-irc.si Fri, 11 May 2001 10:06:55 +0200 Date: Fri, 11 May 2001 10:06:55 +0200 From: Aci Polajnar Aci.Polajnar@rcc-irc.si Subject: QUICK Fatal Error Hi there! I have long running QUICK screen, working with RMS files on ALPHA OPEN/VMS. My PowerHouse is 7.10G1. The processing is quite complicated and is started on 4th screen level. Some time QUICK finish O.K. but if I run it on a huge amount of data I get *Fatal Error *635* Notify .......... and OS prompt. Any ideas ???????? Best regards -- _\\///_ (' o-o ') _______________ooOo_( )_oOoo_____________________________________ Aci Polajnar Tel : +386(0)3 4274174 IS Consultant Fax : +386(0)3 4274198 Razvojni Center Celje e-mail: acip@rcc-irc.si 3000 Celje, Slovenia URL : http://www.rcc-irc.si _______________oooO____Oooo______________________________________ ( ) ( ) \ ( ) / \_) (_/ From steve_franklin@yahoo.com Fri, 11 May 2001 08:47:56 -0500 Date: Fri, 11 May 2001 08:47:56 -0500 From: Steve Franklin steve_franklin@yahoo.com Subject: Axiant Question I am trying to determine the best way to structure my application. Is it better to use the various default sections or is it better to create new ones? If I create new sections, should I also create a new library to put them or should they go in the Default Section Library? Thanks for the help, Steve Franklin Stw Inc. (952) 432-2203 Email: steve_franklin@yahoo.com _________________________________________________________ Do You Yahoo!? Get your free @yahoo.com address at http://mail.yahoo.com From bob.edis@fleetpride.com Fri, 11 May 2001 08:49:07 -0500 Date: Fri, 11 May 2001 08:49:07 -0500 From: Edis, Bob bob.edis@fleetpride.com Subject: QUICK Fatal Error Aci I can't remember the various OpenVMS issues precisely but I think the problem has to do with quotas for the job/account Quick is running in. At least it didn't crash the system like some other OS environments! Regards, Blue -----Original Message----- From: Aci Polajnar [mailto:Aci.Polajnar@rcc-irc.si] Sent: Friday, May 11, 2001 3:07 AM To: 'PowerHouse List' Subject: QUICK Fatal Error Hi there! I have long running QUICK screen, working with RMS files on ALPHA OPEN/VMS. My PowerHouse is 7.10G1. The processing is quite complicated and is started on 4th screen level. Some time QUICK finish O.K. but if I run it on a huge amount of data I get *Fatal Error *635* Notify .......... and OS prompt. Any ideas ???????? Best regards -- _\\///_ (' o-o ') _______________ooOo_( )_oOoo_____________________________________ Aci Polajnar Tel : +386(0)3 4274174 IS Consultant Fax : +386(0)3 4274198 Razvojni Center Celje e-mail: acip@rcc-irc.si 3000 Celje, Slovenia URL : http://www.rcc-irc.si _______________oooO____Oooo______________________________________ ( ) ( ) \ ( ) / \_) (_/ = = = = = = = = = = = = = = = = = = = = = = = = = = = = Mailing list: powerh-l@lists.swau.edu Subscribe: "subscribe" in message body to powerh-l-request@lists.swau.edu Unsubscribe: "unsubscribe" in message body to powerh-l-request@lists.swau.edu http://lists.swau.edu/mailman/listinfo/powerh-l This list is closed, thus to post to the list you must be a subscriber. From darren_reely@latticesemi.com Fri, 11 May 2001 10:22:06 -0700 Date: Fri, 11 May 2001 10:22:06 -0700 From: Darren Reely darren_reely@latticesemi.com Subject: QUICK Fatal Error Aci Polajnar wrote: > > Hi there! > > I have long running QUICK screen, working with RMS files on > ALPHA OPEN/VMS. My PowerHouse is 7.10G1. The processing is quite complicated > and is started on 4th screen level. > Some time QUICK finish O.K. but if I run it on a huge amount of data I get > > *Fatal Error *635* Notify .......... > > and OS prompt. > > Any ideas ???????? I would first look at the 'quotas for the job/account' suggestion by Bob. It's been a few years, but I too remember having that problem. But... If this is a screen called repetively for some kind of processing, you may be hitting an artificial call limit of 65527. I hit this problem on a Sun Solaris box when testing a 26 base addition function screen. I saw this within a While loop, but it is NOT a while loop limit. Why Quick has this limit is beyond me. Darren. From Chris.Sharman@ccagroup.co.uk Mon, 14 May 2001 09:12:42 +0100 Date: Mon, 14 May 2001 09:12:42 +0100 From: Chris Sharman Chris.Sharman@ccagroup.co.uk Subject: QUICK Fatal Error >I have long running QUICK screen, working with RMS files on >ALPHA OPEN/VMS. My PowerHouse is 7.10G1. The processing is quite complicated >and is started on 4th screen level. >Some time QUICK finish O.K. but if I run it on a huge amount of data I get > >*Fatal Error *635* Notify .......... > > and OS prompt. Definitely look at quotas. I've seen this, and it's virtually impossible to finally nail down: telephone support (UK) spent a lot of time (& several upgrades) trying - incidence is now very low, but it still happens occasionally. Raise UAF quotas: WSdef/quo/extent & PGflquota - I've got 2k/4k/16k & 100k, but my screen depth is less than yours. The WS quotas do seem to matter, although they really shouldn't. Note also that the VIRTUALPAGECNT system parameter is effectively an upper bound for PGflquota, so be prepared to raise that too. The WS system parameters should be out of sight already on a well configured system (see recommendations for backup). I'd also recommend upgrading to 710G3/4 - I'm on G3 which significantly reduced the incidence of the problem. Also talk to Telephone Support, if you haven't already: I don't recall fatal error 635 - I got %system-f-accvio. Chris _______________________________________________________________________ Chris.Sharman@CCAgroup.co.uk http://www.ccagroup.co.uk/ CCA Stationery Ltd, Eastway, Fulwood, Preston, Lancashire, PR2 9WS. ----------------------------------------------------------------------- Any views expressed in this message are those of the sender and not necessarily those of CCA Group. The unauthorized use, disclosure, copying or alteration of this message is forbidden. The contents of this message may be confidential and/or privileged, copyright CCA Group and are intended solely for the use of the individual or entity to whom they are addressed. Whilst this message has been scanned, CCA Group cannot guarantee that it is virus free or compatible with your systems and accepts no responsibility for any loss or damage arising from its use. The recipient is advised to run their own anti-virus software. If you receive this message in error please contact postmaster@ccagroup.co.uk immediately, destroy any copies and delete it from your computer systems. From Joe.Boyle@cognos.com Mon, 14 May 2001 05:25:23 -0400 Date: Mon, 14 May 2001 05:25:23 -0400 From: Boyle, Joe Joe.Boyle@cognos.com Subject: Abnormal Termination of a Quick Screen in Batch Hi Roger, you could try debugging the cause of the problem by adding records to a sequential log file at key stages within the screen processing ( or at every stage ) in order to identify the point at which the failure occurs. You will get an idea of why the process stopped, hopefully, by observing the last 'run command' or procedure processed in conjunction with the data values in use at the point of failure. see very crude example below, includes quick source code and log file output as viewed from Quiz. *____________________________________________________________________* can cle set dict sys$scratch:bulllock set defa screen sys$scratch:LOCKTEST.QKc temp t-key char size 5 reset at startup define d'getsystemval char * 2 = & getsystemval ( "batch-flag", logical ,"lnm$job") file bullp designer file bulls designer file rep-stats designer temp t-p integer *10 reset at startup initial 0 temp t-s integer *10 reset at startup initial 0 PROCEDURE internal getdets begin let rep-stats-det of rep-stats & = "starting loop in getdets proc " + " > " + ascii (systime ) put rep-stats reset while ret bulls VIA SCHED-COMMENT-CD using SCHED-COMMENT-CD of bullp begin let t-s = t-s + 1 end end PROCEDURE initial BEGIN let rep-stats-det of rep-stats & = "@ing dev$coms:setup com file in initial proc " + " > " + ascii (systime ) put rep-stats reset run command "@dev$coms:setup.com" on error terminate let rep-stats-det of rep-stats & = "starting loop in initial proc " + " > " + ascii (systime ) put rep-stats reset while ret bullp sequential begin let rep-stats-det of rep-stats & = "rec " + SCHED-COMMENT-CD of bullp + & " in loop in initial proc " + " > " + ascii (systime ) put rep-stats reset do internal getdets let t-p = t-p + 1 end let rep-stats-det of rep-stats & = ascii ( t-p ) + " > " + ascii ( t-s ) & + " > " + ascii (systime ) + d'getsystemval put rep-stats reset let rep-stats-det of rep-stats & = "closing files " + " > " + ascii (systime ) put rep-stats reset close bulls close bullp close rep-stats RETURN end build list detail *____________________________________________________________________* Viewed from a Quiz such as, can cle set default set pag len 0 set rep nolim acc rep-stats rep all set pag wid 132 go you would see log detail like that below, @ing dev$coms:setup com file in initial proc > 5134007 starting loop in initial proc > 5134150 rec A in loop in initial proc > 5134222 starting loop in getdets proc > 5134225 rec AA in loop in initial proc > 5134227 starting loop in getdets proc > 5134230 rec B in loop in initial proc > 5134235 starting loop in getdets proc > 5134237 ... more details ... rec C in loop in initial proc > 5134249 starting loop in getdets proc > 5134251 rec D in loop in initial proc > 5134263 starting loop in getdets proc > 5134266 26 > 21538 > 5134551 closing files > 5134561 regards, Joe Boyle. -----Original Message----- From: Kesterson, Roger [mailto:Roger.Kesterson@mtdsw.com] Sent: 30 April 2001 21:50 To: PowerHouse List Server (E-mail) Subject: Abnormal Termination of a Quick Screen in Batch PH7.10.G2 Alpha OpenVMS 7.1 ----------------- Fellow PH Listers (no reference to Red Dwarf intended): We have some Quick programs that execute in batch, using a WHILE loop in the INITIALIZE procedure. QKOUT is sent to a file, but QKIN is not redefined in any way. One of these programs just terminated, and the only message in the QKOUT file is: "The limit for backup movements (\\) has been reached." Does any of you have any idea what might cause this? Thanks, Roger Kesterson Senior Programming Analyst MTD Southwest, Inc. 550 N. 54th Street Chandler, AZ 85226 Phone: 480-961-1002 x4357 Fax: 480-940-5035 = = = = = = = = = = = = = = = = = = = = = = = = = = = = Mailing list: powerh-l@lists.swau.edu Subscribe: "subscribe" in message body to powerh-l-request@lists.swau.edu Unsubscribe: "unsubscribe" in message body to powerh-l-request@lists.swau.edu http://lists.swau.edu/mailman/listinfo/powerh-l This list is closed, thus to post to the list you must be a subscriber. From abahrs@denkor.com Mon, 14 May 2001 12:56:50 -0700 Date: Mon, 14 May 2001 12:56:50 -0700 From: Art Bahrs abahrs@denkor.com Subject: PowerHouse Web Setup???? Hi All :) Ok... Dumb Question time.... I am trying to bring up PowerHouse Web Demo on a HP e3000 s928 using IIS on a NT server and have the Dispatcher and PH WebServer running on the HP 3k 928.... IIS is running on the NT box.... I have created a test HTML Quick Screen via QDesign 8.29 D3... The page is not being found (the infamous error 404 ...) so... I checked the NT server has proper entries in DNS for the 928 and can ping the 928.... the 928 has proper entries in the hosts.net.sys file for the NT server and can ping the NT server by IP address... I am new to most of this ... so be nice to me :) but hey... if you got the answer a little friendly abuse would, of course, not be out of line :) hehe Art "Waiting for return call from Cognos... but fishing for all info! hehe" Bahrs From Bob.Deskin@Cognos.COM Mon, 14 May 2001 16:10:00 -0400 Date: Mon, 14 May 2001 16:10:00 -0400 From: Deskin, Bob Bob.Deskin@Cognos.COM Subject: PowerHouse Web Setup???? I have found that Most of the time, working through the configuration step-by-step resolves the problem. If you're getting a 404, then I don't know that you've even gotten to PH Web. The first thing to check is that you can bring up a basic page on your web server. Next, bring up a basic page with a link to the PH Web page. Check the configuration information in the Administrator to ensure that the locations for the dictionary, the screen files, and the HTML Templates files are correct. They should be MPE/iX paths. Note that the PH_QKC and PH_HTML names should be MPE/iX file names and should not have extensions. As you work through this, watch the messages that you get. They will often indicate exactly how far along in the process you've gotten and what the problem is likely (but not always) to be. Turn logging on for the PHCGI, Dispatcher, and PH Web Server using the Administrator. That will help trace the process as well. Bob Deskin PowerHouse Web Product Manager, Application Development Tools, Cognos Inc. bob.deskin@cognos.com (613) 738-1338 ext 7268 FAX: (613) 727-1178 3755 Riverside Drive P.O. Box 9707 Stn. T, Ottawa ON K1G 4K9 CANADA -----Original Message----- From: Art Bahrs [mailto:abahrs@denkor.com] Sent: Monday, May 14, 2001 3:57 PM To: Power House ListServ Subject: PowerHouse Web Setup???? Hi All :) Ok... Dumb Question time.... I am trying to bring up PowerHouse Web Demo on a HP e3000 s928 using IIS on a NT server and have the Dispatcher and PH WebServer running on the HP 3k 928.... IIS is running on the NT box.... I have created a test HTML Quick Screen via QDesign 8.29 D3... The page is not being found (the infamous error 404 ...) so... I checked the NT server has proper entries in DNS for the 928 and can ping the 928.... the 928 has proper entries in the hosts.net.sys file for the NT server and can ping the NT server by IP address... I am new to most of this ... so be nice to me :) but hey... if you got the answer a little friendly abuse would, of course, not be out of line :) hehe Art "Waiting for return call from Cognos... but fishing for all info! hehe" Bahrs = = = = = = = = = = = = = = = = = = = = = = = = = = = = Mailing list: powerh-l@lists.swau.edu Subscribe: "subscribe" in message body to powerh-l-request@lists.swau.edu Unsubscribe: "unsubscribe" in message body to powerh-l-request@lists.swau.edu http://lists.swau.edu/mailman/listinfo/powerh-l This list is closed, thus to post to the list you must be a subscriber. From bob.edis@fleetpride.com Mon, 14 May 2001 15:42:17 -0500 Date: Mon, 14 May 2001 15:42:17 -0500 From: Edis, Bob bob.edis@fleetpride.com Subject: PowerHouse Web Setup???? G'day Art Are you declaring a specific path for the QKC and HTML files in the screen source? What do you have in the default HTML and QKC directory sections of PH Web Administrator? Have the virtual directories been declared/exist in the IIS Console? Are you trying to access the screen page directly or through a basic HTML menu page? We found we have to put the initial page in the wwwroot directory on the web server machine and have this page call the web screen which is located on a different box. We also use the default QKC and HTML directory sections of the PH Web Admin tool for the PH Web server rather than explicit path names in the source code. Regards, Blue -----Original Message----- From: Art Bahrs [mailto:abahrs@denkor.com] Sent: Monday, May 14, 2001 2:57 PM To: Power House ListServ Subject: PowerHouse Web Setup???? Hi All :) Ok... Dumb Question time.... I am trying to bring up PowerHouse Web Demo on a HP e3000 s928 using IIS on a NT server and have the Dispatcher and PH WebServer running on the HP 3k 928.... IIS is running on the NT box.... I have created a test HTML Quick Screen via QDesign 8.29 D3... The page is not being found (the infamous error 404 ...) so... I checked the NT server has proper entries in DNS for the 928 and can ping the 928.... the 928 has proper entries in the hosts.net.sys file for the NT server and can ping the NT server by IP address... I am new to most of this ... so be nice to me :) but hey... if you got the answer a little friendly abuse would, of course, not be out of line :) hehe Art "Waiting for return call from Cognos... but fishing for all info! hehe" Bahrs = = = = = = = = = = = = = = = = = = = = = = = = = = = = Mailing list: powerh-l@lists.swau.edu Subscribe: "subscribe" in message body to powerh-l-request@lists.swau.edu Unsubscribe: "unsubscribe" in message body to powerh-l-request@lists.swau.edu http://lists.swau.edu/mailman/listinfo/powerh-l This list is closed, thus to post to the list you must be a subscriber. From abahrs@denkor.com Mon, 14 May 2001 16:13:01 -0700 Date: Mon, 14 May 2001 16:13:01 -0700 From: Art Bahrs abahrs@denkor.com Subject: PowerHouse Web Setup???? Hi Blue :) Ok... You mentioned something I haven't checked (mainly 'cause I really don't understand it) .... ie "Have the virtual directories been declared/exist in the IIS Console?" Um... which directories are your referring to? I have set up the administrator and the PH web server stuff on our 928 as per all the books and guidelines I have.... btw.... the books and docs I got with the demo kit are very good! Art "losing what *little* hair I have left ! hehe" Bahrs ----- Original Message ----- From: "Edis, Bob" To: "'powerh-l@list.swau.edu'" Sent: Monday, May 14, 2001 1:42 PM Subject: RE: PowerHouse Web Setup???? > G'day Art > > Are you declaring a specific path for the QKC and HTML files in the screen > source? > What do you have in the default HTML and QKC directory sections of PH Web > Administrator? > Have the virtual directories been declared/exist in the IIS Console? > Are you trying to access the screen page directly or through a basic HTML > menu page? > > We found we have to put the initial page in the wwwroot directory on the web > server machine and have this page call the web screen which is located on a > different box. > We also use the default QKC and HTML directory sections of the PH Web Admin > tool for the PH Web server rather than explicit path names in the source > code. > > Regards, > Blue > > > -----Original Message----- > From: Art Bahrs [mailto:abahrs@denkor.com] > Sent: Monday, May 14, 2001 2:57 PM > To: Power House ListServ > Subject: PowerHouse Web Setup???? > > > Hi All :) > Ok... Dumb Question time.... I am trying to bring up PowerHouse Web Demo > on a HP e3000 s928 using IIS on a NT server and have the Dispatcher and PH > WebServer running on the HP 3k 928.... IIS is running on the NT box.... I > have created a test HTML Quick Screen via QDesign 8.29 D3... > > The page is not being found (the infamous error 404 ...) so... I checked > the NT server has proper entries in DNS for the 928 and can ping the 928.... > the 928 has proper entries in the hosts.net.sys file for the NT server and > can ping the NT server by IP address... > > I am new to most of this ... so be nice to me :) but hey... if you got the > answer a little friendly abuse would, of course, not be out of line :) hehe > > Art "Waiting for return call from Cognos... but fishing for all info! hehe" > Bahrs > > > > > = = = = = = = = = = = = = = = = = = = = = = = = = = = = > Mailing list: powerh-l@lists.swau.edu > Subscribe: "subscribe" in message body to powerh-l-request@lists.swau.edu > Unsubscribe: "unsubscribe" in message body to > powerh-l-request@lists.swau.edu > http://lists.swau.edu/mailman/listinfo/powerh-l > This list is closed, thus to post to the list you must be a subscriber. > > = = = = = = = = = = = = = = = = = = = = = = = = = = = = > Mailing list: powerh-l@lists.swau.edu > Subscribe: "subscribe" in message body to powerh-l-request@lists.swau.edu > Unsubscribe: "unsubscribe" in message body to powerh-l-request@lists.swau.edu > http://lists.swau.edu/mailman/listinfo/powerh-l > This list is closed, thus to post to the list you must be a subscriber. From bob.edis@fleetpride.com Mon, 14 May 2001 18:18:17 -0500 Date: Mon, 14 May 2001 18:18:17 -0500 From: Edis, Bob bob.edis@fleetpride.com Subject: PowerHouse Web Setup???? Art I am referring to the IIS administrator. To get there on the NT web server box: 1. Start->Programs->Windows NT 4.0 Option Pack->Microsoft Internet Information Server->Internet service Manager 2. Expand your web site folder; you should see a folder called 'cgi-bin'. This is a virtual directory. 3. Select this folder, right click and select properties. 4. Under 'Local Path:' you should see something like 'C:\InetPub\wwwroot\cgi-bin'. C:\InetPub\wwwroot is where we put a HTML menu pages that refer to the PH Web screens. E.g. dw_menu.htm is a simple HTML page containing references like . You *could* build this menu page using Qdesign but it must be copied to the wwwroot directory. We reference the menu in the web browser as //cogdev-02/dw_menu.htm. The '//cogdev-02' is the name of the IIS web server machine. Regards, Blue -----Original Message----- From: Art Bahrs [mailto:abahrs@denkor.com] Sent: Monday, May 14, 2001 6:13 PM To: Edis, Bob; 'powerh-l@list.swau.edu' Subject: Re: PowerHouse Web Setup???? Hi Blue :) Ok... You mentioned something I haven't checked (mainly 'cause I really don't understand it) .... ie "Have the virtual directories been declared/exist in the IIS Console?" Um... which directories are your referring to? I have set up the administrator and the PH web server stuff on our 928 as per all the books and guidelines I have.... btw.... the books and docs I got with the demo kit are very good! Art "losing what *little* hair I have left ! hehe" Bahrs ----- Original Message ----- From: "Edis, Bob" To: "'powerh-l@list.swau.edu'" Sent: Monday, May 14, 2001 1:42 PM Subject: RE: PowerHouse Web Setup???? > G'day Art > > Are you declaring a specific path for the QKC and HTML files in the screen > source? > What do you have in the default HTML and QKC directory sections of PH Web > Administrator? > Have the virtual directories been declared/exist in the IIS Console? > Are you trying to access the screen page directly or through a basic HTML > menu page? > > We found we have to put the initial page in the wwwroot directory on the web > server machine and have this page call the web screen which is located on a > different box. > We also use the default QKC and HTML directory sections of the PH Web Admin > tool for the PH Web server rather than explicit path names in the source > code. > > Regards, > Blue > > > -----Original Message----- > From: Art Bahrs [mailto:abahrs@denkor.com] > Sent: Monday, May 14, 2001 2:57 PM > To: Power House ListServ > Subject: PowerHouse Web Setup???? > > > Hi All :) > Ok... Dumb Question time.... I am trying to bring up PowerHouse Web Demo > on a HP e3000 s928 using IIS on a NT server and have the Dispatcher and PH > WebServer running on the HP 3k 928.... IIS is running on the NT box.... I > have created a test HTML Quick Screen via QDesign 8.29 D3... > > The page is not being found (the infamous error 404 ...) so... I checked > the NT server has proper entries in DNS for the 928 and can ping the 928.... > the 928 has proper entries in the hosts.net.sys file for the NT server and > can ping the NT server by IP address... > > I am new to most of this ... so be nice to me :) but hey... if you got the > answer a little friendly abuse would, of course, not be out of line :) hehe > > Art "Waiting for return call from Cognos... but fishing for all info! hehe" > Bahrs > > > > > = = = = = = = = = = = = = = = = = = = = = = = = = = = = > Mailing list: powerh-l@lists.swau.edu > Subscribe: "subscribe" in message body to powerh-l-request@lists.swau.edu > Unsubscribe: "unsubscribe" in message body to > powerh-l-request@lists.swau.edu > http://lists.swau.edu/mailman/listinfo/powerh-l > This list is closed, thus to post to the list you must be a subscriber. > > = = = = = = = = = = = = = = = = = = = = = = = = = = = = > Mailing list: powerh-l@lists.swau.edu > Subscribe: "subscribe" in message body to powerh-l-request@lists.swau.edu > Unsubscribe: "unsubscribe" in message body to powerh-l-request@lists.swau.edu > http://lists.swau.edu/mailman/listinfo/powerh-l > This list is closed, thus to post to the list you must be a subscriber. From Joe.Boyle@cognos.com Tue, 15 May 2001 04:36:44 -0400 Date: Tue, 15 May 2001 04:36:44 -0400 From: Boyle, Joe Joe.Boyle@cognos.com Subject: PowerHouse Web Setup???? Hi Art, Bob, everyone, I'm sorry to state the obvious but just in case - you have actually copied the phcgi.exe from the install directory on the HP server to the directory 'C:\InetPub\wwwroot\cgi-bin' on the NT box ? Also, you will very likely have to change the security access of the phcgi.dat file to 'everyone', as I seem to recall that the 404 error cannot find the phcgi.dat script, see below for details. If the PH Web Administrator or the IIS web server is on Windows 2000, the phcgi.dat file will be saved with limited permissions when a Save Changes is done in the PH Web Administrator. To resolve, change the permissions on the phcgi.dat file in the cgi-bin directory, opening it to Everyone with Full Control, after changes have been saved in the PH Web Administrator. You may wish to download some of the PHweb docs for MPE from the Cognos web site at the URL below, http://support.cognos.com/supported/support/support_library/phweb/phweb22doc .html Some PHweb docs available are listed below, 2.29D2 Readme (MPEiX) Release notes and how to install PowerHouse Web for MPEiX updated for 2.29D2. PH Web Getting Started (MPEiX) How to install PowerHouse Web 2.29 for MPEiX. PowerHouse Web Guide Revised manual intended for PowerHouse programmers who wish to create web applications. regards, Joe Boyle. -----Original Message----- From: Edis, Bob [mailto:bob.edis@fleetpride.com] Sent: 15 May 2001 00:18 To: 'powerh-l@list.swau.edu' Subject: RE: PowerHouse Web Setup???? Art I am referring to the IIS administrator. To get there on the NT web server box: 1. Start->Programs->Windows NT 4.0 Option Pack->Microsoft Internet Information Server->Internet service Manager 2. Expand your web site folder; you should see a folder called 'cgi-bin'. This is a virtual directory. 3. Select this folder, right click and select properties. 4. Under 'Local Path:' you should see something like 'C:\InetPub\wwwroot\cgi-bin'. C:\InetPub\wwwroot is where we put a HTML menu pages that refer to the PH Web screens. E.g. dw_menu.htm is a simple HTML page containing references like . You *could* build this menu page using Qdesign but it must be copied to the wwwroot directory. We reference the menu in the web browser as //cogdev-02/dw_menu.htm. The '//cogdev-02' is the name of the IIS web server machine. Regards, Blue -----Original Message----- From: Art Bahrs [mailto:abahrs@denkor.com] Sent: Monday, May 14, 2001 6:13 PM To: Edis, Bob; 'powerh-l@list.swau.edu' Subject: Re: PowerHouse Web Setup???? Hi Blue :) Ok... You mentioned something I haven't checked (mainly 'cause I really don't understand it) .... ie "Have the virtual directories been declared/exist in the IIS Console?" Um... which directories are your referring to? I have set up the administrator and the PH web server stuff on our 928 as per all the books and guidelines I have.... btw.... the books and docs I got with the demo kit are very good! Art "losing what *little* hair I have left ! hehe" Bahrs ----- Original Message ----- From: "Edis, Bob" To: "'powerh-l@list.swau.edu'" Sent: Monday, May 14, 2001 1:42 PM Subject: RE: PowerHouse Web Setup???? > G'day Art > > Are you declaring a specific path for the QKC and HTML files in the screen > source? > What do you have in the default HTML and QKC directory sections of PH Web > Administrator? > Have the virtual directories been declared/exist in the IIS Console? > Are you trying to access the screen page directly or through a basic HTML > menu page? > > We found we have to put the initial page in the wwwroot directory on the web > server machine and have this page call the web screen which is located on a > different box. > We also use the default QKC and HTML directory sections of the PH Web Admin > tool for the PH Web server rather than explicit path names in the source > code. > > Regards, > Blue > > > -----Original Message----- > From: Art Bahrs [mailto:abahrs@denkor.com] > Sent: Monday, May 14, 2001 2:57 PM > To: Power House ListServ > Subject: PowerHouse Web Setup???? > > > Hi All :) > Ok... Dumb Question time.... I am trying to bring up PowerHouse Web Demo > on a HP e3000 s928 using IIS on a NT server and have the Dispatcher and PH > WebServer running on the HP 3k 928.... IIS is running on the NT box.... I > have created a test HTML Quick Screen via QDesign 8.29 D3... > > The page is not being found (the infamous error 404 ...) so... I checked > the NT server has proper entries in DNS for the 928 and can ping the 928.... > the 928 has proper entries in the hosts.net.sys file for the NT server and > can ping the NT server by IP address... > > I am new to most of this ... so be nice to me :) but hey... if you got the > answer a little friendly abuse would, of course, not be out of line :) hehe > > Art "Waiting for return call from Cognos... but fishing for all info! hehe" > Bahrs > > > > > = = = = = = = = = = = = = = = = = = = = = = = = = = = = > Mailing list: powerh-l@lists.swau.edu > Subscribe: "subscribe" in message body to powerh-l-request@lists.swau.edu > Unsubscribe: "unsubscribe" in message body to > powerh-l-request@lists.swau.edu > http://lists.swau.edu/mailman/listinfo/powerh-l > This list is closed, thus to post to the list you must be a subscriber. > > = = = = = = = = = = = = = = = = = = = = = = = = = = = = > Mailing list: powerh-l@lists.swau.edu > Subscribe: "subscribe" in message body to powerh-l-request@lists.swau.edu > Unsubscribe: "unsubscribe" in message body to powerh-l-request@lists.swau.edu > http://lists.swau.edu/mailman/listinfo/powerh-l > This list is closed, thus to post to the list you must be a subscriber. = = = = = = = = = = = = = = = = = = = = = = = = = = = = Mailing list: powerh-l@lists.swau.edu Subscribe: "subscribe" in message body to powerh-l-request@lists.swau.edu Unsubscribe: "unsubscribe" in message body to powerh-l-request@lists.swau.edu http://lists.swau.edu/mailman/listinfo/powerh-l This list is closed, thus to post to the list you must be a subscriber. From Bob.Deskin@Cognos.COM Tue, 15 May 2001 09:07:18 -0400 Date: Tue, 15 May 2001 09:07:18 -0400 From: Deskin, Bob Bob.Deskin@Cognos.COM Subject: PowerHouse Web Setup???? Joe, Art, and everyone... Things are only obvious if they're okay or if you already know them. File permissions seem to be an issue, but not all the time. Just one more thing to check. I've found that most configuration issues in PH Web are "obvious". I have a PowerPoint presentation I did for our annual kickoff meetings that I'd be more than happy to e-mail to anyone who's interested. I'm sure it doesn't cover everything, but it's better than nothing (he said modestly). I will be adding more configuration information into the documentation for the next release. If you're interested in the presentation, send me an e-mail - don't reply all. The presentation's about 150K zipped. Bob Deskin PowerHouse Web Product Manager, Application Development Tools, Cognos Inc. bob.deskin@cognos.com (613) 738-1338 ext 7268 FAX: (613) 727-1178 3755 Riverside Drive P.O. Box 9707 Stn. T, Ottawa ON K1G 4K9 CANADA -----Original Message----- From: Boyle, Joe [mailto:Joe.Boyle@cognos.com] Sent: Tuesday, May 15, 2001 4:37 AM To: 'abahrs@denkor.com'; 'powerh-l@list.swau.edu' Subject: RE: PowerHouse Web Setup???? Hi Art, Bob, everyone, I'm sorry to state the obvious but just in case - you have actually copied the phcgi.exe from the install directory on the HP server to the directory 'C:\InetPub\wwwroot\cgi-bin' on the NT box ? Also, you will very likely have to change the security access of the phcgi.dat file to 'everyone', as I seem to recall that the 404 error cannot find the phcgi.dat script, see below for details. If the PH Web Administrator or the IIS web server is on Windows 2000, the phcgi.dat file will be saved with limited permissions when a Save Changes is done in the PH Web Administrator. To resolve, change the permissions on the phcgi.dat file in the cgi-bin directory, opening it to Everyone with Full Control, after changes have been saved in the PH Web Administrator. You may wish to download some of the PHweb docs for MPE from the Cognos web site at the URL below, http://support.cognos.com/supported/support/support_library/phweb/phweb22doc .html Some PHweb docs available are listed below, 2.29D2 Readme (MPEiX) Release notes and how to install PowerHouse Web for MPEiX updated for 2.29D2. PH Web Getting Started (MPEiX) How to install PowerHouse Web 2.29 for MPEiX. PowerHouse Web Guide Revised manual intended for PowerHouse programmers who wish to create web applications. regards, Joe Boyle. -----Original Message----- From: Edis, Bob [mailto:bob.edis@fleetpride.com] Sent: 15 May 2001 00:18 To: 'powerh-l@list.swau.edu' Subject: RE: PowerHouse Web Setup???? Art I am referring to the IIS administrator. To get there on the NT web server box: 1. Start->Programs->Windows NT 4.0 Option Pack->Microsoft Internet Information Server->Internet service Manager 2. Expand your web site folder; you should see a folder called 'cgi-bin'. This is a virtual directory. 3. Select this folder, right click and select properties. 4. Under 'Local Path:' you should see something like 'C:\InetPub\wwwroot\cgi-bin'. C:\InetPub\wwwroot is where we put a HTML menu pages that refer to the PH Web screens. E.g. dw_menu.htm is a simple HTML page containing references like . You *could* build this menu page using Qdesign but it must be copied to the wwwroot directory. We reference the menu in the web browser as //cogdev-02/dw_menu.htm. The '//cogdev-02' is the name of the IIS web server machine. Regards, Blue -----Original Message----- From: Art Bahrs [mailto:abahrs@denkor.com] Sent: Monday, May 14, 2001 6:13 PM To: Edis, Bob; 'powerh-l@list.swau.edu' Subject: Re: PowerHouse Web Setup???? Hi Blue :) Ok... You mentioned something I haven't checked (mainly 'cause I really don't understand it) .... ie "Have the virtual directories been declared/exist in the IIS Console?" Um... which directories are your referring to? I have set up the administrator and the PH web server stuff on our 928 as per all the books and guidelines I have.... btw.... the books and docs I got with the demo kit are very good! Art "losing what *little* hair I have left ! hehe" Bahrs ----- Original Message ----- From: "Edis, Bob" To: "'powerh-l@list.swau.edu'" Sent: Monday, May 14, 2001 1:42 PM Subject: RE: PowerHouse Web Setup???? > G'day Art > > Are you declaring a specific path for the QKC and HTML files in the screen > source? > What do you have in the default HTML and QKC directory sections of PH Web > Administrator? > Have the virtual directories been declared/exist in the IIS Console? > Are you trying to access the screen page directly or through a basic HTML > menu page? > > We found we have to put the initial page in the wwwroot directory on the web > server machine and have this page call the web screen which is located on a > different box. > We also use the default QKC and HTML directory sections of the PH Web Admin > tool for the PH Web server rather than explicit path names in the source > code. > > Regards, > Blue > > > -----Original Message----- > From: Art Bahrs [mailto:abahrs@denkor.com] > Sent: Monday, May 14, 2001 2:57 PM > To: Power House ListServ > Subject: PowerHouse Web Setup???? > > > Hi All :) > Ok... Dumb Question time.... I am trying to bring up PowerHouse Web Demo > on a HP e3000 s928 using IIS on a NT server and have the Dispatcher and PH > WebServer running on the HP 3k 928.... IIS is running on the NT box.... I > have created a test HTML Quick Screen via QDesign 8.29 D3... > > The page is not being found (the infamous error 404 ...) so... I checked > the NT server has proper entries in DNS for the 928 and can ping the 928.... > the 928 has proper entries in the hosts.net.sys file for the NT server and > can ping the NT server by IP address... > > I am new to most of this ... so be nice to me :) but hey... if you got the > answer a little friendly abuse would, of course, not be out of line :) hehe > > Art "Waiting for return call from Cognos... but fishing for all info! hehe" > Bahrs > > > > > = = = = = = = = = = = = = = = = = = = = = = = = = = = = > Mailing list: powerh-l@lists.swau.edu > Subscribe: "subscribe" in message body to powerh-l-request@lists.swau.edu > Unsubscribe: "unsubscribe" in message body to > powerh-l-request@lists.swau.edu > http://lists.swau.edu/mailman/listinfo/powerh-l > This list is closed, thus to post to the list you must be a subscriber. > > = = = = = = = = = = = = = = = = = = = = = = = = = = = = > Mailing list: powerh-l@lists.swau.edu > Subscribe: "subscribe" in message body to powerh-l-request@lists.swau.edu > Unsubscribe: "unsubscribe" in message body to powerh-l-request@lists.swau.edu > http://lists.swau.edu/mailman/listinfo/powerh-l > This list is closed, thus to post to the list you must be a subscriber. = = = = = = = = = = = = = = = = = = = = = = = = = = = = Mailing list: powerh-l@lists.swau.edu Subscribe: "subscribe" in message body to powerh-l-request@lists.swau.edu Unsubscribe: "unsubscribe" in message body to powerh-l-request@lists.swau.edu http://lists.swau.edu/mailman/listinfo/powerh-l This list is closed, thus to post to the list you must be a subscriber. = = = = = = = = = = = = = = = = = = = = = = = = = = = = Mailing list: powerh-l@lists.swau.edu Subscribe: "subscribe" in message body to powerh-l-request@lists.swau.edu Unsubscribe: "unsubscribe" in message body to powerh-l-request@lists.swau.edu http://lists.swau.edu/mailman/listinfo/powerh-l This list is closed, thus to post to the list you must be a subscriber. From mjohns@accubanc.com Tue, 15 May 2001 08:31:57 -0500 Date: Tue, 15 May 2001 08:31:57 -0500 From: mjohns@accubanc.com mjohns@accubanc.com Subject: PowerHouse Web Setup???? is anyone else on the list receiving the same messages more than once. "Deskin, Bob" on 05/15/2001 08:07:18 AM To: "'powerh-l@list.swau.edu'" cc: (bcc: Margaret Johns/Contractor/Dallas/AccuBanc) Subject: RE: PowerHouse Web Setup???? Joe, Art, and everyone... Things are only obvious if they're okay or if you already know them. File permissions seem to be an issue, but not all the time. Just one more thing to check. I've found that most configuration issues in PH Web are "obvious". I have a PowerPoint presentation I did for our annual kickoff meetings that I'd be more than happy to e-mail to anyone who's interested. I'm sure it doesn't cover everything, but it's better than nothing (he said modestly). I will be adding more configuration information into the documentation for the next release. If you're interested in the presentation, send me an e-mail - don't reply all. The presentation's about 150K zipped. Bob Deskin PowerHouse Web Product Manager, Application Development Tools, Cognos Inc. bob.deskin@cognos.com (613) 738-1338 ext 7268 FAX: (613) 727-1178 3755 Riverside Drive P.O. Box 9707 Stn. T, Ottawa ON K1G 4K9 CANADA -----Original Message----- From: Boyle, Joe [mailto:Joe.Boyle@cognos.com] Sent: Tuesday, May 15, 2001 4:37 AM To: 'abahrs@denkor.com'; 'powerh-l@list.swau.edu' Subject: RE: PowerHouse Web Setup???? Hi Art, Bob, everyone, I'm sorry to state the obvious but just in case - you have actually copied the phcgi.exe from the install directory on the HP server to the directory 'C:\InetPub\wwwroot\cgi-bin' on the NT box ? Also, you will very likely have to change the security access of the phcgi.dat file to 'everyone', as I seem to recall that the 404 error cannot find the phcgi.dat script, see below for details. If the PH Web Administrator or the IIS web server is on Windows 2000, the phcgi.dat file will be saved with limited permissions when a Save Changes is done in the PH Web Administrator. To resolve, change the permissions on the phcgi.dat file in the cgi-bin directory, opening it to Everyone with Full Control, after changes have been saved in the PH Web Administrator. You may wish to download some of the PHweb docs for MPE from the Cognos web site at the URL below, http://support.cognos.com/supported/support/support_library/phweb/phweb22doc .html Some PHweb docs available are listed below, 2.29D2 Readme (MPEiX) Release notes and how to install PowerHouse Web for MPEiX updated for 2.29D2. PH Web Getting Started (MPEiX) How to install PowerHouse Web 2.29 for MPEiX. PowerHouse Web Guide Revised manual intended for PowerHouse programmers who wish to create web applications. regards, Joe Boyle. -----Original Message----- From: Edis, Bob [mailto:bob.edis@fleetpride.com] Sent: 15 May 2001 00:18 To: 'powerh-l@list.swau.edu' Subject: RE: PowerHouse Web Setup???? Art I am referring to the IIS administrator. To get there on the NT web server box: 1. Start->Programs->Windows NT 4.0 Option Pack->Microsoft Internet Information Server->Internet service Manager 2. Expand your web site folder; you should see a folder called 'cgi-bin'. This is a virtual directory. 3. Select this folder, right click and select properties. 4. Under 'Local Path:' you should see something like 'C:\InetPub\wwwroot\cgi-bin'. C:\InetPub\wwwroot is where we put a HTML menu pages that refer to the PH Web screens. E.g. dw_menu.htm is a simple HTML page containing references like . You *could* build this menu page using Qdesign but it must be copied to the wwwroot directory. We reference the menu in the web browser as //cogdev-02/dw_menu.htm. The '//cogdev-02' is the name of the IIS web server machine. Regards, Blue -----Original Message----- From: Art Bahrs [mailto:abahrs@denkor.com] Sent: Monday, May 14, 2001 6:13 PM To: Edis, Bob; 'powerh-l@list.swau.edu' Subject: Re: PowerHouse Web Setup???? Hi Blue :) Ok... You mentioned something I haven't checked (mainly 'cause I really don't understand it) .... ie "Have the virtual directories been declared/exist in the IIS Console?" Um... which directories are your referring to? I have set up the administrator and the PH web server stuff on our 928 as per all the books and guidelines I have.... btw.... the books and docs I got with the demo kit are very good! Art "losing what *little* hair I have left ! hehe" Bahrs ----- Original Message ----- From: "Edis, Bob" To: "'powerh-l@list.swau.edu'" Sent: Monday, May 14, 2001 1:42 PM Subject: RE: PowerHouse Web Setup???? > G'day Art > > Are you declaring a specific path for the QKC and HTML files in the screen > source? > What do you have in the default HTML and QKC directory sections of PH Web > Administrator? > Have the virtual directories been declared/exist in the IIS Console? > Are you trying to access the screen page directly or through a basic HTML > menu page? > > We found we have to put the initial page in the wwwroot directory on the web > server machine and have this page call the web screen which is located on a > different box. > We also use the default QKC and HTML directory sections of the PH Web Admin > tool for the PH Web server rather than explicit path names in the source > code. > > Regards, > Blue > > > -----Original Message----- > From: Art Bahrs [mailto:abahrs@denkor.com] > Sent: Monday, May 14, 2001 2:57 PM > To: Power House ListServ > Subject: PowerHouse Web Setup???? > > > Hi All :) > Ok... Dumb Question time.... I am trying to bring up PowerHouse Web Demo > on a HP e3000 s928 using IIS on a NT server and have the Dispatcher and PH > WebServer running on the HP 3k 928.... IIS is running on the NT box.... I > have created a test HTML Quick Screen via QDesign 8.29 D3... > > The page is not being found (the infamous error 404 ...) so... I checked > the NT server has proper entries in DNS for the 928 and can ping the 928.... > the 928 has proper entries in the hosts.net.sys file for the NT server and > can ping the NT server by IP address... > > I am new to most of this ... so be nice to me :) but hey... if you got the > answer a little friendly abuse would, of course, not be out of line :) hehe > > Art "Waiting for return call from Cognos... but fishing for all info! hehe" > Bahrs > > > > > = = = = = = = = = = = = = = = = = = = = = = = = = = = = > Mailing list: powerh-l@lists.swau.edu > Subscribe: "subscribe" in message body to powerh-l-request@lists.swau.edu > Unsubscribe: "unsubscribe" in message body to > powerh-l-request@lists.swau.edu > http://lists.swau.edu/mailman/listinfo/powerh-l > This list is closed, thus to post to the list you must be a subscriber. > > = = = = = = = = = = = = = = = = = = = = = = = = = = = = > Mailing list: powerh-l@lists.swau.edu > Subscribe: "subscribe" in message body to powerh-l-request@lists.swau.edu > Unsubscribe: "unsubscribe" in message body to powerh-l-request@lists.swau.edu > http://lists.swau.edu/mailman/listinfo/powerh-l > This list is closed, thus to post to the list you must be a subscriber. = = = = = = = = = = = = = = = = = = = = = = = = = = = = Mailing list: powerh-l@lists.swau.edu Subscribe: "subscribe" in message body to powerh-l-request@lists.swau.edu Unsubscribe: "unsubscribe" in message body to powerh-l-request@lists.swau.edu http://lists.swau.edu/mailman/listinfo/powerh-l This list is closed, thus to post to the list you must be a subscriber. = = = = = = = = = = = = = = = = = = = = = = = = = = = = Mailing list: powerh-l@lists.swau.edu Subscribe: "subscribe" in message body to powerh-l-request@lists.swau.edu Unsubscribe: "unsubscribe" in message body to powerh-l-request@lists.swau.edu http://lists.swau.edu/mailman/listinfo/powerh-l This list is closed, thus to post to the list you must be a subscriber. = = = = = = = = = = = = = = = = = = = = = = = = = = = = Mailing list: powerh-l@lists.swau.edu Subscribe: "subscribe" in message body to powerh-l-request@lists.swau.edu Unsubscribe: "unsubscribe" in message body to powerh-l-request@lists.swau.edu http://lists.swau.edu/mailman/listinfo/powerh-l This list is closed, thus to post to the list you must be a subscriber. From ryappert@cmotor.com Tue, 15 May 2001 06:44:13 -0700 Date: Tue, 15 May 2001 06:44:13 -0700 From: Roxanna Yappert ryappert@cmotor.com Subject: PowerHouse Web Setup???? Yes At 08:31 AM 5/15/01 -0500, mjohns@accubanc.com wrote: >is anyone else on the list receiving the same messages more than once. > > > > > >"Deskin, Bob" on 05/15/2001 08:07:18 AM > >To: "'powerh-l@list.swau.edu'" >cc: (bcc: Margaret Johns/Contractor/Dallas/AccuBanc) > >Subject: RE: PowerHouse Web Setup???? > > > > >Joe, Art, and everyone... > >Things are only obvious if they're okay or if you already know them. > >File permissions seem to be an issue, but not all the time. Just one more >thing to check. > >I've found that most configuration issues in PH Web are "obvious". I have a >PowerPoint presentation I did for our annual kickoff meetings that I'd be >more than happy to e-mail to anyone who's interested. I'm sure it doesn't >cover everything, but it's better than nothing (he said modestly). I will be >adding more configuration information into the documentation for the next >release. > >If you're interested in the presentation, send me an e-mail - don't reply >all. The presentation's about 150K zipped. > >Bob Deskin >PowerHouse Web Product Manager, Application Development Tools, Cognos Inc. >bob.deskin@cognos.com (613) 738-1338 ext 7268 FAX: (613) 727-1178 >3755 Riverside Drive P.O. Box 9707 Stn. T, Ottawa ON K1G 4K9 CANADA > >-----Original Message----- >From: Boyle, Joe [mailto:Joe.Boyle@cognos.com] >Sent: Tuesday, May 15, 2001 4:37 AM >To: 'abahrs@denkor.com'; 'powerh-l@list.swau.edu' >Subject: RE: PowerHouse Web Setup???? > > >Hi Art, Bob, everyone, > >I'm sorry to state the obvious but just in case - you have actually copied >the phcgi.exe from the install directory on the HP server to the directory >'C:\InetPub\wwwroot\cgi-bin' on the NT box ? > >Also, you will very likely have to change the security access of the >phcgi.dat file to 'everyone', as I seem to recall that the 404 error cannot >find the phcgi.dat script, see below for details. > >If the PH Web Administrator or the IIS web server is on Windows 2000, the >phcgi.dat file will be saved with limited permissions when a Save Changes is >done in the PH Web Administrator. To resolve, change the permissions on the >phcgi.dat file in the cgi-bin directory, opening it to Everyone with Full >Control, after changes have been saved in the PH Web Administrator. > > >You may wish to download some of the PHweb docs for MPE from the Cognos web >site at the URL below, > >http://support.cognos.com/supported/support/support_library/phweb/phweb22doc >.html > > >Some PHweb docs available are listed below, > >2.29D2 Readme (MPEiX) >Release notes and how to install PowerHouse Web for MPEiX updated for >2.29D2. > >PH Web Getting Started (MPEiX) >How to install PowerHouse Web 2.29 for MPEiX. > >PowerHouse Web Guide >Revised manual intended for PowerHouse programmers who wish to create web >applications. > > > >regards, >Joe Boyle. > > >-----Original Message----- >From: Edis, Bob [mailto:bob.edis@fleetpride.com] >Sent: 15 May 2001 00:18 >To: 'powerh-l@list.swau.edu' >Subject: RE: PowerHouse Web Setup???? > > >Art > >I am referring to the IIS administrator. To get there on the NT web server >box: > >1. Start->Programs->Windows NT 4.0 Option Pack->Microsoft Internet >Information Server->Internet service Manager >2. Expand your web site folder; you should see a folder called 'cgi-bin'. >This is a virtual directory. >3. Select this folder, right click and select properties. >4. Under 'Local Path:' you should see something like >'C:\InetPub\wwwroot\cgi-bin'. C:\InetPub\wwwroot is where we put a HTML >menu pages that refer to the PH Web screens. E.g. dw_menu.htm is a simple >HTML page containing references like href="/cgi-bin/phcgi.exe?PH_QKC=dw_lu_part_list&PH_HTML=dw_lu_part_list.htm& >PH_APP=DataWare_App">. You *could* build this menu page using Qdesign but >it must be copied to the wwwroot directory. We reference the menu in the >web browser as //cogdev-02/dw_menu.htm. The '//cogdev-02' is the name of >the IIS web server machine. > >Regards, >Blue > >-----Original Message----- >From: Art Bahrs [mailto:abahrs@denkor.com] >Sent: Monday, May 14, 2001 6:13 PM >To: Edis, Bob; 'powerh-l@list.swau.edu' >Subject: Re: PowerHouse Web Setup???? > > >Hi Blue :) > Ok... You mentioned something I haven't checked (mainly 'cause I really >don't understand it) .... ie "Have the virtual directories been >declared/exist in the IIS Console?" > > Um... which directories are your referring to? I have set up the >administrator and the PH web server stuff on our 928 as per all the books >and guidelines I have.... btw.... the books and docs I got with the demo kit >are very good! > >Art "losing what *little* hair I have left ! hehe" Bahrs > >----- Original Message ----- >From: "Edis, Bob" >To: "'powerh-l@list.swau.edu'" >Sent: Monday, May 14, 2001 1:42 PM >Subject: RE: PowerHouse Web Setup???? > > > > G'day Art > > > > Are you declaring a specific path for the QKC and HTML files in the screen > > source? > > What do you have in the default HTML and QKC directory sections of PH Web > > Administrator? > > Have the virtual directories been declared/exist in the IIS Console? > > Are you trying to access the screen page directly or through a basic HTML > > menu page? > > > > We found we have to put the initial page in the wwwroot directory on the >web > > server machine and have this page call the web screen which is located on >a > > different box. > > We also use the default QKC and HTML directory sections of the PH Web >Admin > > tool for the PH Web server rather than explicit path names in the source > > code. > > > > Regards, > > Blue > > > > > > -----Original Message----- > > From: Art Bahrs [mailto:abahrs@denkor.com] > > Sent: Monday, May 14, 2001 2:57 PM > > To: Power House ListServ > > Subject: PowerHouse Web Setup???? > > > > > > Hi All :) > > Ok... Dumb Question time.... I am trying to bring up PowerHouse Web >Demo > > on a HP e3000 s928 using IIS on a NT server and have the Dispatcher and PH > > WebServer running on the HP 3k 928.... IIS is running on the NT box.... I > > have created a test HTML Quick Screen via QDesign 8.29 D3... > > > > The page is not being found (the infamous error 404 ...) so... I >checked > > the NT server has proper entries in DNS for the 928 and can ping the >928.... > > the 928 has proper entries in the hosts.net.sys file for the NT server and > > can ping the NT server by IP address... > > > > I am new to most of this ... so be nice to me :) but hey... if you got >the > > answer a little friendly abuse would, of course, not be out of line :) >hehe > > > > Art "Waiting for return call from Cognos... but fishing for all info! >hehe" > > Bahrs > > > > > > > > > > = = = = = = = = = = = = = = = = = = = = = = = = = = = = > > Mailing list: powerh-l@lists.swau.edu > > Subscribe: "subscribe" in message body to powerh-l-request@lists.swau.edu > > Unsubscribe: "unsubscribe" in message body to > > powerh-l-request@lists.swau.edu > > http://lists.swau.edu/mailman/listinfo/powerh-l > > This list is closed, thus to post to the list you must be a subscriber. > > > > = = = = = = = = = = = = = = = = = = = = = = = = = = = = > > Mailing list: powerh-l@lists.swau.edu > > Subscribe: "subscribe" in message body to powerh-l-request@lists.swau.edu > > Unsubscribe: "unsubscribe" in message body to >powerh-l-request@lists.swau.edu > > http://lists.swau.edu/mailman/listinfo/powerh-l > > This list is closed, thus to post to the list you must be a subscriber. > >= = = = = = = = = = = = = = = = = = = = = = = = = = = = >Mailing list: powerh-l@lists.swau.edu >Subscribe: "subscribe" in message body to powerh-l-request@lists.swau.edu >Unsubscribe: "unsubscribe" in message body to >powerh-l-request@lists.swau.edu >http://lists.swau.edu/mailman/listinfo/powerh-l >This list is closed, thus to post to the list you must be a subscriber. > >= = = = = = = = = = = = = = = = = = = = = = = = = = = = >Mailing list: powerh-l@lists.swau.edu >Subscribe: "subscribe" in message body to powerh-l-request@lists.swau.edu >Unsubscribe: "unsubscribe" in message body to >powerh-l-request@lists.swau.edu >http://lists.swau.edu/mailman/listinfo/powerh-l >This list is closed, thus to post to the list you must be a subscriber. > >= = = = = = = = = = = = = = = = = = = = = = = = = = = = >Mailing list: powerh-l@lists.swau.edu >Subscribe: "subscribe" in message body to powerh-l-request@lists.swau.edu >Unsubscribe: "unsubscribe" in message body to powerh-l-request@lists.swau.edu >http://lists.swau.edu/mailman/listinfo/powerh-l >This list is closed, thus to post to the list you must be a subscriber. > > > > > >= = = = = = = = = = = = = = = = = = = = = = = = = = = = >Mailing list: powerh-l@lists.swau.edu >Subscribe: "subscribe" in message body to powerh-l-request@lists.swau.edu >Unsubscribe: "unsubscribe" in message body to powerh-l-request@lists.swau.edu >http://lists.swau.edu/mailman/listinfo/powerh-l >This list is closed, thus to post to the list you must be a subscriber. From SPhillips@CONTDISC.com Tue, 15 May 2001 08:43:45 -0500 Date: Tue, 15 May 2001 08:43:45 -0500 From: Shirley Phillips SPhillips@CONTDISC.com Subject: PowerHouse Web Setup???? YES! Shirley Phillips Information Systems Manager Continental Disc Corporation 3160 W. Heartland Drive Liberty, MO 64068 (816) 407-5562 -----Original Message----- From: mjohns@accubanc.com [mailto:mjohns@accubanc.com] Sent: Tuesday, May 15, 2001 8:32 AM To: Deskin, Bob Cc: 'powerh-l@list.swau.edu' Subject: RE: PowerHouse Web Setup???? is anyone else on the list receiving the same messages more than once. From ntietz@MCARE.MED.UMICH.EDU Tue, 15 May 2001 09:43:37 -0400 Date: Tue, 15 May 2001 09:43:37 -0400 From: Nancy Tietz ntietz@MCARE.MED.UMICH.EDU Subject: PowerHouse Web Setup???? No. I'm not. -----Original Message----- From: Shirley Phillips [mailto:SPhillips@CONTDISC.com] Sent: Tuesday, May 15, 2001 9:44 AM To: 'powerh-l@list.swau.edu' Subject: RE: PowerHouse Web Setup???? YES! Shirley Phillips Information Systems Manager Continental Disc Corporation 3160 W. Heartland Drive Liberty, MO 64068 (816) 407-5562 -----Original Message----- From: mjohns@accubanc.com [mailto:mjohns@accubanc.com] Sent: Tuesday, May 15, 2001 8:32 AM To: Deskin, Bob Cc: 'powerh-l@list.swau.edu' Subject: RE: PowerHouse Web Setup???? is anyone else on the list receiving the same messages more than once. = = = = = = = = = = = = = = = = = = = = = = = = = = = = Mailing list: powerh-l@lists.swau.edu Subscribe: "subscribe" in message body to powerh-l-request@lists.swau.edu Unsubscribe: "unsubscribe" in message body to powerh-l-request@lists.swau.edu http://lists.swau.edu/mailman/listinfo/powerh-l This list is closed, thus to post to the list you must be a subscriber. From tropiano@MCARE.MED.UMICH.EDU Tue, 15 May 2001 09:47:38 -0400 Date: Tue, 15 May 2001 09:47:38 -0400 From: Jim Tropiano tropiano@MCARE.MED.UMICH.EDU Subject: PowerHouse Web Setup???? no -----Original Message----- From: Roxanna Yappert [mailto:ryappert@cmotor.com] Sent: Tuesday, May 15, 2001 9:44 AM To: 'powerh-l@list.swau.edu' Subject: RE: PowerHouse Web Setup???? Yes At 08:31 AM 5/15/01 -0500, mjohns@accubanc.com wrote: >is anyone else on the list receiving the same messages more than once. > > > > > >"Deskin, Bob" on 05/15/2001 08:07:18 AM > >To: "'powerh-l@list.swau.edu'" >cc: (bcc: Margaret Johns/Contractor/Dallas/AccuBanc) > >Subject: RE: PowerHouse Web Setup???? > > > > >Joe, Art, and everyone... > >Things are only obvious if they're okay or if you already know them. > >File permissions seem to be an issue, but not all the time. Just one more >thing to check. > >I've found that most configuration issues in PH Web are "obvious". I have a >PowerPoint presentation I did for our annual kickoff meetings that I'd be >more than happy to e-mail to anyone who's interested. I'm sure it doesn't >cover everything, but it's better than nothing (he said modestly). I will be >adding more configuration information into the documentation for the next >release. > >If you're interested in the presentation, send me an e-mail - don't reply >all. The presentation's about 150K zipped. > >Bob Deskin >PowerHouse Web Product Manager, Application Development Tools, Cognos Inc. >bob.deskin@cognos.com (613) 738-1338 ext 7268 FAX: (613) 727-1178 >3755 Riverside Drive P.O. Box 9707 Stn. T, Ottawa ON K1G 4K9 CANADA > >-----Original Message----- >From: Boyle, Joe [mailto:Joe.Boyle@cognos.com] >Sent: Tuesday, May 15, 2001 4:37 AM >To: 'abahrs@denkor.com'; 'powerh-l@list.swau.edu' >Subject: RE: PowerHouse Web Setup???? > > >Hi Art, Bob, everyone, > >I'm sorry to state the obvious but just in case - you have actually copied >the phcgi.exe from the install directory on the HP server to the directory >'C:\InetPub\wwwroot\cgi-bin' on the NT box ? > >Also, you will very likely have to change the security access of the >phcgi.dat file to 'everyone', as I seem to recall that the 404 error cannot >find the phcgi.dat script, see below for details. > >If the PH Web Administrator or the IIS web server is on Windows 2000, the >phcgi.dat file will be saved with limited permissions when a Save Changes is >done in the PH Web Administrator. To resolve, change the permissions on the >phcgi.dat file in the cgi-bin directory, opening it to Everyone with Full >Control, after changes have been saved in the PH Web Administrator. > > >You may wish to download some of the PHweb docs for MPE from the Cognos web >site at the URL below, > >http://support.cognos.com/supported/support/support_library/phweb/phweb22do c >.html > > >Some PHweb docs available are listed below, > >2.29D2 Readme (MPEiX) >Release notes and how to install PowerHouse Web for MPEiX updated for >2.29D2. > >PH Web Getting Started (MPEiX) >How to install PowerHouse Web 2.29 for MPEiX. > >PowerHouse Web Guide >Revised manual intended for PowerHouse programmers who wish to create web >applications. > > > >regards, >Joe Boyle. > > >-----Original Message----- >From: Edis, Bob [mailto:bob.edis@fleetpride.com] >Sent: 15 May 2001 00:18 >To: 'powerh-l@list.swau.edu' >Subject: RE: PowerHouse Web Setup???? > > >Art > >I am referring to the IIS administrator. To get there on the NT web server >box: > >1. Start->Programs->Windows NT 4.0 Option Pack->Microsoft Internet >Information Server->Internet service Manager >2. Expand your web site folder; you should see a folder called 'cgi-bin'. >This is a virtual directory. >3. Select this folder, right click and select properties. >4. Under 'Local Path:' you should see something like >'C:\InetPub\wwwroot\cgi-bin'. C:\InetPub\wwwroot is where we put a HTML >menu pages that refer to the PH Web screens. E.g. dw_menu.htm is a simple >HTML page containing references like href="/cgi-bin/phcgi.exe?PH_QKC=dw_lu_part_list&PH_HTML=dw_lu_part_list.htm & >PH_APP=DataWare_App">. You *could* build this menu page using Qdesign but >it must be copied to the wwwroot directory. We reference the menu in the >web browser as //cogdev-02/dw_menu.htm. The '//cogdev-02' is the name of >the IIS web server machine. > >Regards, >Blue > >-----Original Message----- >From: Art Bahrs [mailto:abahrs@denkor.com] >Sent: Monday, May 14, 2001 6:13 PM >To: Edis, Bob; 'powerh-l@list.swau.edu' >Subject: Re: PowerHouse Web Setup???? > > >Hi Blue :) > Ok... You mentioned something I haven't checked (mainly 'cause I really >don't understand it) .... ie "Have the virtual directories been >declared/exist in the IIS Console?" > > Um... which directories are your referring to? I have set up the >administrator and the PH web server stuff on our 928 as per all the books >and guidelines I have.... btw.... the books and docs I got with the demo kit >are very good! > >Art "losing what *little* hair I have left ! hehe" Bahrs > >----- Original Message ----- >From: "Edis, Bob" >To: "'powerh-l@list.swau.edu'" >Sent: Monday, May 14, 2001 1:42 PM >Subject: RE: PowerHouse Web Setup???? > > > > G'day Art > > > > Are you declaring a specific path for the QKC and HTML files in the screen > > source? > > What do you have in the default HTML and QKC directory sections of PH Web > > Administrator? > > Have the virtual directories been declared/exist in the IIS Console? > > Are you trying to access the screen page directly or through a basic HTML > > menu page? > > > > We found we have to put the initial page in the wwwroot directory on the >web > > server machine and have this page call the web screen which is located on >a > > different box. > > We also use the default QKC and HTML directory sections of the PH Web >Admin > > tool for the PH Web server rather than explicit path names in the source > > code. > > > > Regards, > > Blue > > > > > > -----Original Message----- > > From: Art Bahrs [mailto:abahrs@denkor.com] > > Sent: Monday, May 14, 2001 2:57 PM > > To: Power House ListServ > > Subject: PowerHouse Web Setup???? > > > > > > Hi All :) > > Ok... Dumb Question time.... I am trying to bring up PowerHouse Web >Demo > > on a HP e3000 s928 using IIS on a NT server and have the Dispatcher and PH > > WebServer running on the HP 3k 928.... IIS is running on the NT box.... I > > have created a test HTML Quick Screen via QDesign 8.29 D3... > > > > The page is not being found (the infamous error 404 ...) so... I >checked > > the NT server has proper entries in DNS for the 928 and can ping the >928.... > > the 928 has proper entries in the hosts.net.sys file for the NT server and > > can ping the NT server by IP address... > > > > I am new to most of this ... so be nice to me :) but hey... if you got >the > > answer a little friendly abuse would, of course, not be out of line :) >hehe > > > > Art "Waiting for return call from Cognos... but fishing for all info! >hehe" > > Bahrs > > > > > > > > > > = = = = = = = = = = = = = = = = = = = = = = = = = = = = > > Mailing list: powerh-l@lists.swau.edu > > Subscribe: "subscribe" in message body to powerh-l-request@lists.swau.edu > > Unsubscribe: "unsubscribe" in message body to > > powerh-l-request@lists.swau.edu > > http://lists.swau.edu/mailman/listinfo/powerh-l > > This list is closed, thus to post to the list you must be a subscriber. > > > > = = = = = = = = = = = = = = = = = = = = = = = = = = = = > > Mailing list: powerh-l@lists.swau.edu > > Subscribe: "subscribe" in message body to powerh-l-request@lists.swau.edu > > Unsubscribe: "unsubscribe" in message body to >powerh-l-request@lists.swau.edu > > http://lists.swau.edu/mailman/listinfo/powerh-l > > This list is closed, thus to post to the list you must be a subscriber. > >= = = = = = = = = = = = = = = = = = = = = = = = = = = = >Mailing list: powerh-l@lists.swau.edu >Subscribe: "subscribe" in message body to powerh-l-request@lists.swau.edu >Unsubscribe: "unsubscribe" in message body to >powerh-l-request@lists.swau.edu >http://lists.swau.edu/mailman/listinfo/powerh-l >This list is closed, thus to post to the list you must be a subscriber. > >= = = = = = = = = = = = = = = = = = = = = = = = = = = = >Mailing list: powerh-l@lists.swau.edu >Subscribe: "subscribe" in message body to powerh-l-request@lists.swau.edu >Unsubscribe: "unsubscribe" in message body to >powerh-l-request@lists.swau.edu >http://lists.swau.edu/mailman/listinfo/powerh-l >This list is closed, thus to post to the list you must be a subscriber. > >= = = = = = = = = = = = = = = = = = = = = = = = = = = = >Mailing list: powerh-l@lists.swau.edu >Subscribe: "subscribe" in message body to powerh-l-request@lists.swau.edu >Unsubscribe: "unsubscribe" in message body to powerh-l-request@lists.swau.edu >http://lists.swau.edu/mailman/listinfo/powerh-l >This list is closed, thus to post to the list you must be a subscriber. > > > > > >= = = = = = = = = = = = = = = = = = = = = = = = = = = = >Mailing list: powerh-l@lists.swau.edu >Subscribe: "subscribe" in message body to powerh-l-request@lists.swau.edu >Unsubscribe: "unsubscribe" in message body to powerh-l-request@lists.swau.edu >http://lists.swau.edu/mailman/listinfo/powerh-l >This list is closed, thus to post to the list you must be a subscriber. = = = = = = = = = = = = = = = = = = = = = = = = = = = = Mailing list: powerh-l@lists.swau.edu Subscribe: "subscribe" in message body to powerh-l-request@lists.swau.edu Unsubscribe: "unsubscribe" in message body to powerh-l-request@lists.swau.edu http://lists.swau.edu/mailman/listinfo/powerh-l This list is closed, thus to post to the list you must be a subscriber. From DMielke@aicfunds.com Tue, 15 May 2001 16:59:58 -0400 Date: Tue, 15 May 2001 16:59:58 -0400 From: Daniel Mielke DMielke@aicfunds.com Subject: AS/400 - QUICK Control Function keys F7 & F8 Hello all, I'm looking to control the block transfer of records in a cluster for a display screen. The following screen displays the races for a specified Rat's Ratlets. RATRACES Rat Races Screen Rat number: 123456 Rat Name: Slick Rat Ratlet-no Ratlet Race number Earnings Final position 1 Mini rat 456 $3,450.00 1 7 Super rat 466 $ 70.00 5 13 Slow rat 542 $ 0.00 16 COMMAND ===>________________________________ If there were additional races for Rat 123456, F8 or [ENTER] would list the next cluster of RAT-RACES. However, since there are no more races, the F8 or [ENTER] clears the list and returns to the command in find mode. The same occurs for [F7] when the user is at the initial list of races. How do I control these function keys such that the user may scroll through the existing records stopping at both the end and the beginning without clearing the screen? Special thanks to Chuck Reinke for the use of the Rat/Ratlets analogy > BUY. HOLD. AND PROSPER. > > Daniel Mielke > Senior User Analyst > AIC Limited > 1-888-710-4242 (4614) > dmielke@aicfunds.com > www.aicfunds.com > From Conrad.Whittall@Cognos.COM Tue, 15 May 2001 17:29:11 -0400 Date: Tue, 15 May 2001 17:29:11 -0400 From: Whittall, Conrad Conrad.Whittall@Cognos.COM Subject: AS/400 - QUICK Control Function keys F7 & F8 Hello Daniel, I thought I detected a touch of the "Reinke" influence in that example :-) Your in luck because you are on an AS/400...where the HOLD FIND option of the SCREEN statement is available. As the manual says, "If you specify HOLD, QUICK doesn't automatically start a new find sequence after the last record from the primary file is displayed. When the last record of a detail file (or the first record when PAGEBACK is used) is displayed, the user must use the Next (N) command to display the next primary record." I remember using this quite extensively with an AS/400 customer in London (England that is, not Ontario). In their applications we also declared an expression within the screen to make use of the ENDOFDATA function, like this... DEFINE d_more CHARACTER*7 & = "End" IF ENDOFDATA = 1 & ELSE "Start" IF ENDOFDATA = -1 & ELSE "More..." ...which we displayed in the bottom right-hand corner of the screen in true "CUA" fashion :-) Check the PowerHouse for AS/400 QDESIGN Statement Reference (version 6.07) manual for details. See page 452 for a discussion of the HOLD FIND option of the SCREEN statement, and page 178 for the ENDOFDATA function. Best regards, Conrad Conrad Whittall Marketing Manager, Application Development Tools Cognos Incorporated, 3755 Riverside Drive, Ottawa, Ontario, K1G 4K9, Canada Developer productivity never goes out of style! For details of the Cognos PowerHouse family of high-productivity development tools for Web, Windows and terminal-based business applications, and to request your own free copy of our new product demo CD, simply click here http://www.cognos.com/powerhouse. -----Original Message----- From: Daniel Mielke [mailto:DMielke@aicfunds.com] Sent: Tuesday 15 May 2001 17:00 To: Powerhouse List (E-mail) Subject: AS/400 - QUICK Control Function keys F7 & F8 Hello all, I'm looking to control the block transfer of records in a cluster for a display screen. The following screen displays the races for a specified Rat's Ratlets. RATRACES Rat Races Screen Rat number: 123456 Rat Name: Slick Rat Ratlet-no Ratlet Race number Earnings Final position 1 Mini rat 456 $3,450.00 1 7 Super rat 466 $ 70.00 5 13 Slow rat 542 $ 0.00 16 COMMAND ===>________________________________ If there were additional races for Rat 123456, F8 or [ENTER] would list the next cluster of RAT-RACES. However, since there are no more races, the F8 or [ENTER] clears the list and returns to the command in find mode. The same occurs for [F7] when the user is at the initial list of races. How do I control these function keys such that the user may scroll through the existing records stopping at both the end and the beginning without clearing the screen? Special thanks to Chuck Reinke for the use of the Rat/Ratlets analogy > BUY. HOLD. AND PROSPER. > > Daniel Mielke > Senior User Analyst > AIC Limited > 1-888-710-4242 (4614) > dmielke@aicfunds.com > www.aicfunds.com > = = = = = = = = = = = = = = = = = = = = = = = = = = = = Mailing list: powerh-l@lists.swau.edu Subscribe: "subscribe" in message body to powerh-l-request@lists.swau.edu Unsubscribe: "unsubscribe" in message body to powerh-l-request@lists.swau.edu http://lists.swau.edu/mailman/listinfo/powerh-l This list is closed, thus to post to the list you must be a subscriber. From wonicon@bigpond.net.au Wed, 16 May 2001 07:41:39 +1000 Date: Wed, 16 May 2001 07:41:39 +1000 From: Jeff Hoffman wonicon@bigpond.net.au Subject: PowerHouse Web Setup???? No At 08:31 AM 15-05-01 -0500, you wrote: > > >is anyone else on the list receiving the same messages more than once. > > > > > >"Deskin, Bob" on 05/15/2001 08:07:18 AM > >To: "'powerh-l@list.swau.edu'" >cc: (bcc: Margaret Johns/Contractor/Dallas/AccuBanc) > >Subject: RE: PowerHouse Web Setup???? > > > > >Joe, Art, and everyone... > >Things are only obvious if they're okay or if you already know them. > >File permissions seem to be an issue, but not all the time. Just one more >thing to check. > >I've found that most configuration issues in PH Web are "obvious". I have a >PowerPoint presentation I did for our annual kickoff meetings that I'd be >more than happy to e-mail to anyone who's interested. I'm sure it doesn't >cover everything, but it's better than nothing (he said modestly). I will be >adding more configuration information into the documentation for the next >release. > >If you're interested in the presentation, send me an e-mail - don't reply >all. The presentation's about 150K zipped. > >Bob Deskin >PowerHouse Web Product Manager, Application Development Tools, Cognos Inc. >bob.deskin@cognos.com (613) 738-1338 ext 7268 FAX: (613) 727-1178 >3755 Riverside Drive P.O. Box 9707 Stn. T, Ottawa ON K1G 4K9 CANADA > >-----Original Message----- >From: Boyle, Joe [mailto:Joe.Boyle@cognos.com] >Sent: Tuesday, May 15, 2001 4:37 AM >To: 'abahrs@denkor.com'; 'powerh-l@list.swau.edu' >Subject: RE: PowerHouse Web Setup???? > > >Hi Art, Bob, everyone, > >I'm sorry to state the obvious but just in case - you have actually copied >the phcgi.exe from the install directory on the HP server to the directory >'C:\InetPub\wwwroot\cgi-bin' on the NT box ? > >Also, you will very likely have to change the security access of the >phcgi.dat file to 'everyone', as I seem to recall that the 404 error cannot >find the phcgi.dat script, see below for details. > >If the PH Web Administrator or the IIS web server is on Windows 2000, the >phcgi.dat file will be saved with limited permissions when a Save Changes is >done in the PH Web Administrator. To resolve, change the permissions on the >phcgi.dat file in the cgi-bin directory, opening it to Everyone with Full >Control, after changes have been saved in the PH Web Administrator. > > >You may wish to download some of the PHweb docs for MPE from the Cognos web >site at the URL below, > >http://support.cognos.com/supported/support/support_library/phweb/phweb22doc >.html > > >Some PHweb docs available are listed below, > >2.29D2 Readme (MPEiX) >Release notes and how to install PowerHouse Web for MPEiX updated for >2.29D2. > >PH Web Getting Started (MPEiX) >How to install PowerHouse Web 2.29 for MPEiX. > >PowerHouse Web Guide >Revised manual intended for PowerHouse programmers who wish to create web >applications. > > > >regards, >Joe Boyle. > > >-----Original Message----- >From: Edis, Bob [mailto:bob.edis@fleetpride.com] >Sent: 15 May 2001 00:18 >To: 'powerh-l@list.swau.edu' >Subject: RE: PowerHouse Web Setup???? > > >Art > >I am referring to the IIS administrator. To get there on the NT web server >box: > >1. Start->Programs->Windows NT 4.0 Option Pack->Microsoft Internet >Information Server->Internet service Manager >2. Expand your web site folder; you should see a folder called 'cgi-bin'. >This is a virtual directory. >3. Select this folder, right click and select properties. >4. Under 'Local Path:' you should see something like >'C:\InetPub\wwwroot\cgi-bin'. C:\InetPub\wwwroot is where we put a HTML >menu pages that refer to the PH Web screens. E.g. dw_menu.htm is a simple >HTML page containing references like href="/cgi-bin/phcgi.exe?PH_QKC=dw_lu_part_list&PH_HTML=dw_lu_part_list.htm& >PH_APP=DataWare_App">. You *could* build this menu page using Qdesign but >it must be copied to the wwwroot directory. We reference the menu in the >web browser as //cogdev-02/dw_menu.htm. The '//cogdev-02' is the name of >the IIS web server machine. > >Regards, >Blue > >-----Original Message----- >From: Art Bahrs [mailto:abahrs@denkor.com] >Sent: Monday, May 14, 2001 6:13 PM >To: Edis, Bob; 'powerh-l@list.swau.edu' >Subject: Re: PowerHouse Web Setup???? > > >Hi Blue :) > Ok... You mentioned something I haven't checked (mainly 'cause I really >don't understand it) .... ie "Have the virtual directories been >declared/exist in the IIS Console?" > > Um... which directories are your referring to? I have set up the >administrator and the PH web server stuff on our 928 as per all the books >and guidelines I have.... btw.... the books and docs I got with the demo kit >are very good! > >Art "losing what *little* hair I have left ! hehe" Bahrs > >----- Original Message ----- >From: "Edis, Bob" >To: "'powerh-l@list.swau.edu'" >Sent: Monday, May 14, 2001 1:42 PM >Subject: RE: PowerHouse Web Setup???? > > >> G'day Art >> >> Are you declaring a specific path for the QKC and HTML files in the screen >> source? >> What do you have in the default HTML and QKC directory sections of PH Web >> Administrator? >> Have the virtual directories been declared/exist in the IIS Console? >> Are you trying to access the screen page directly or through a basic HTML >> menu page? >> >> We found we have to put the initial page in the wwwroot directory on the >web >> server machine and have this page call the web screen which is located on >a >> different box. >> We also use the default QKC and HTML directory sections of the PH Web >Admin >> tool for the PH Web server rather than explicit path names in the source >> code. >> >> Regards, >> Blue >> >> >> -----Original Message----- >> From: Art Bahrs [mailto:abahrs@denkor.com] >> Sent: Monday, May 14, 2001 2:57 PM >> To: Power House ListServ >> Subject: PowerHouse Web Setup???? >> >> >> Hi All :) >> Ok... Dumb Question time.... I am trying to bring up PowerHouse Web >Demo >> on a HP e3000 s928 using IIS on a NT server and have the Dispatcher and PH >> WebServer running on the HP 3k 928.... IIS is running on the NT box.... I >> have created a test HTML Quick Screen via QDesign 8.29 D3... >> >> The page is not being found (the infamous error 404 ...) so... I >checked >> the NT server has proper entries in DNS for the 928 and can ping the >928.... >> the 928 has proper entries in the hosts.net.sys file for the NT server and >> can ping the NT server by IP address... >> >> I am new to most of this ... so be nice to me :) but hey... if you got >the >> answer a little friendly abuse would, of course, not be out of line :) >hehe >> >> Art "Waiting for return call from Cognos... but fishing for all info! >hehe" >> Bahrs >> >> >> >> >> = = = = = = = = = = = = = = = = = = = = = = = = = = = = >> Mailing list: powerh-l@lists.swau.edu >> Subscribe: "subscribe" in message body to powerh-l-request@lists.swau.edu >> Unsubscribe: "unsubscribe" in message body to >> powerh-l-request@lists.swau.edu >> http://lists.swau.edu/mailman/listinfo/powerh-l >> This list is closed, thus to post to the list you must be a subscriber. >> >> = = = = = = = = = = = = = = = = = = = = = = = = = = = = >> Mailing list: powerh-l@lists.swau.edu >> Subscribe: "subscribe" in message body to powerh-l-request@lists.swau.edu >> Unsubscribe: "unsubscribe" in message body to >powerh-l-request@lists.swau.edu >> http://lists.swau.edu/mailman/listinfo/powerh-l >> This list is closed, thus to post to the list you must be a subscriber. > >= = = = = = = = = = = = = = = = = = = = = = = = = = = = >Mailing list: powerh-l@lists.swau.edu >Subscribe: "subscribe" in message body to powerh-l-request@lists.swau.edu >Unsubscribe: "unsubscribe" in message body to >powerh-l-request@lists.swau.edu >http://lists.swau.edu/mailman/listinfo/powerh-l >This list is closed, thus to post to the list you must be a subscriber. > >= = = = = = = = = = = = = = = = = = = = = = = = = = = = >Mailing list: powerh-l@lists.swau.edu >Subscribe: "subscribe" in message body to powerh-l-request@lists.swau.edu >Unsubscribe: "unsubscribe" in message body to >powerh-l-request@lists.swau.edu >http://lists.swau.edu/mailman/listinfo/powerh-l >This list is closed, thus to post to the list you must be a subscriber. > >= = = = = = = = = = = = = = = = = = = = = = = = = = = = >Mailing list: powerh-l@lists.swau.edu >Subscribe: "subscribe" in message body to powerh-l-request@lists.swau.edu >Unsubscribe: "unsubscribe" in message body to powerh-l-request@lists.swau.edu >http://lists.swau.edu/mailman/listinfo/powerh-l >This list is closed, thus to post to the list you must be a subscriber. > > > > > >= = = = = = = = = = = = = = = = = = = = = = = = = = = = >Mailing list: powerh-l@lists.swau.edu >Subscribe: "subscribe" in message body to powerh-l-request@lists.swau.edu >Unsubscribe: "unsubscribe" in message body to powerh-l-request@lists.swau.edu >http://lists.swau.edu/mailman/listinfo/powerh-l >This list is closed, thus to post to the list you must be a subscriber. > From Chris.Sharman@ccagroup.co.uk Wed, 16 May 2001 09:01:42 +0100 Date: Wed, 16 May 2001 09:01:42 +0100 From: Chris Sharman Chris.Sharman@ccagroup.co.uk Subject: PowerHouse Web Setup???? >is anyone else on the list receiving the same messages more than once. Yes, me too. Chris _______________________________________________________________________ Chris.Sharman@CCAgroup.co.uk http://www.ccagroup.co.uk/ CCA Stationery Ltd, Eastway, Fulwood, Preston, Lancashire, PR2 9WS. ----------------------------------------------------------------------- Any views expressed in this message are those of the sender and not necessarily those of CCA Group. The unauthorized use, disclosure, copying or alteration of this message is forbidden. The contents of this message may be confidential and/or privileged, copyright CCA Group and are intended solely for the use of the individual or entity to whom they are addressed. Whilst this message has been scanned, CCA Group cannot guarantee that it is virus free or compatible with your systems and accepts no responsibility for any loss or damage arising from its use. The recipient is advised to run their own anti-virus software. If you receive this message in error please contact postmaster@ccagroup.co.uk immediately, destroy any copies and delete it from your computer systems. From markus.grossrieder@alba-systems.com Wed, 16 May 2001 11:24:38 +0200 Date: Wed, 16 May 2001 11:24:38 +0200 From: Markus Grossrieder markus.grossrieder@alba-systems.com Subject: Action command S (select) on dates/numbers Hi all, I just wonder if it is possible to use the S action for dates (or numbers) in combination with wildcards, i.e. - all invoices with '1999' in the date - all clients with '000' in the clients ID number - etc. I never find out if it's possible and (if yes) how to do it, or (if not) why not (Conrad, Bob) ? TIA, Markus ps: PH 8.13d (HP Unix), 7.10g (VMS) From merol.newman@ramesys.com Wed, 16 May 2001 11:07:29 +0100 Date: Wed, 16 May 2001 11:07:29 +0100 From: Merol Newman merol.newman@ramesys.com Subject: duplicate messages I got the one from Bob Deskin on Mon 14/05/01 21:10 twice, but not any others. merol.newman@ramesys.com Ramesys, Eldon Way, Crick, Northamptonshire, UK. NN6 7SL phone 01788 822133/823831, fax 01788 823601 > -----Original Message----- > From: powerh-l-admin@cube.swau.edu > [mailto:powerh-l-admin@cube.swau.edu]On Behalf Of Chris Sharman > Sent: 16 May 2001 09:02 > To: mjohns@accubanc.com > Cc: Chris.Sharman@ccagroup.co.uk; powerh-l@lists.swau.edu > Subject: RE: PowerHouse Web Setup???? > > > > >is anyone else on the list receiving the same messages more than once. > > Yes, me too. > Chris > _______________________________________________________________________ > Chris.Sharman@CCAgroup.co.uk http://www.ccagroup.co.uk/ > CCA Stationery Ltd, Eastway, Fulwood, Preston, Lancashire, PR2 9WS. > > > ----------------------------------------------------------------------- > > Any views expressed in this message are those of the sender and not > necessarily those of CCA Group. The unauthorized use, disclosure, > copying or alteration of this message is forbidden. The contents of > this message may be confidential and/or privileged, copyright CCA Group > and are intended solely for the use of the individual or entity to whom > they are addressed. Whilst this message has been scanned, CCA Group > cannot guarantee that it is virus free or compatible with your systems > and accepts no responsibility for any loss or damage arising from its > use. The recipient is advised to run their own anti-virus software. If > you receive this message in error please contact > postmaster@ccagroup.co.uk immediately, destroy any copies and delete it > from your computer systems. > > = = = = = = = = = = = = = = = = = = = = = = = = = = = = > Mailing list: powerh-l@lists.swau.edu > Subscribe: "subscribe" in message body to powerh-l-request@lists.swau.edu > Unsubscribe: "unsubscribe" in message body to powerh-l-request@lists.swau.edu > http://lists.swau.edu/mailman/listinfo/powerh-l > This list is closed, thus to post to the list you must be a subscriber. > From Chris.Sharman@ccagroup.co.uk Wed, 16 May 2001 11:15:55 +0100 Date: Wed, 16 May 2001 11:15:55 +0100 From: Chris Sharman Chris.Sharman@ccagroup.co.uk Subject: Action command S (select) on dates/numbers >I just wonder if it is possible to use the S action for dates (or numbers) in >combination with wildcards, i.e. >- all invoices with '1999' in the date >- all clients with '000' in the clients ID number >- etc. >I never find out if it's possible and If they're really numbers, then NO, you can't (in general). Come to that, you can't find from a date or number (the "@@" find syntax). If your numbers are zoned, you can relatively easily redefine them as characters, and then everything works. If they're binary, redefinition is still possible, but setting up a suitable pattern would be difficult or impossible, as non-printing characters are disallowed in patterns. It does work well with "@@" though (RMS). Eg, for a field MYDATE (I typically use VMSDATE, but it would work the same for ordinary PH dates): add a redefinition MYDATEC and an index MYDATEC to the dictionary. DON'T rebuild the file - Powerhouse will use the MYDATE index without complaining. Add an access lineto your screen: ACCESS VIA MYDATEC USING MYDATEC+"@@" REQUEST MYDATE >(if yes) how to do it, or >(if not) why not (Conrad, Bob) ? Well, it would be possible to implement. OTOH, there is a workaround (use character or zoned datatypes, even for date/numeric fields). So it's a question of priorities, like always. Chris _______________________________________________________________________ Chris.Sharman@CCAgroup.co.uk http://www.ccagroup.co.uk/ CCA Stationery Ltd, Eastway, Fulwood, Preston, Lancashire, PR2 9WS. ----------------------------------------------------------------------- Any views expressed in this message are those of the sender and not necessarily those of CCA Group. The unauthorized use, disclosure, copying or alteration of this message is forbidden. The contents of this message may be confidential and/or privileged, copyright CCA Group and are intended solely for the use of the individual or entity to whom they are addressed. Whilst this message has been scanned, CCA Group cannot guarantee that it is virus free or compatible with your systems and accepts no responsibility for any loss or damage arising from its use. The recipient is advised to run their own anti-virus software. If you receive this message in error please contact postmaster@ccagroup.co.uk immediately, destroy any copies and delete it from your computer systems. From robert.mills@windsong-services.co.uk Wed, 16 May 2001 11:21:39 +0100 Date: Wed, 16 May 2001 11:21:39 +0100 From: Robert Mills robert.mills@windsong-services.co.uk Subject: PowerHouse Features (was AS/400 - QUICK Control Function keys F7 & F8) Conrad, I see that Release 8, which I understand uses a 'common code base', is now available on all platforms except the AS/400. Is the AS/400 joining this group and if so will features like HOLD FIND be available to the rest of us? regards, +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ |Robert W.Mills (Systems Development Manager) | |Windsong Services, St. Mary Cray, ENGLAND | |Tel : +44 (0)1689 870622 x3005 | |Fax : +44 (0)1689 899026 | |Email: Robert.Mills@windsong-services.co.uk | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ -----Original Message----- From: Whittall, Conrad [mailto:Conrad.Whittall@Cognos.COM] Sent: Tuesday, May 15, 2001 10:29 PM To: 'Daniel Mielke'; Powerhouse List (E-mail) Subject: RE: AS/400 - QUICK Control Function keys F7 & F8 Hello Daniel, I thought I detected a touch of the "Reinke" influence in that example :-) Your in luck because you are on an AS/400...where the HOLD FIND option of the SCREEN statement is available. As the manual says, "If you specify HOLD, QUICK doesn't automatically start a new find sequence after the last record from the primary file is displayed. When the last record of a detail file (or the first record when PAGEBACK is used) is displayed, the user must use the Next (N) command to display the next primary record." I remember using this quite extensively with an AS/400 customer in London (England that is, not Ontario). In their applications we also declared an expression within the screen to make use of the ENDOFDATA function, like this... DEFINE d_more CHARACTER*7 & = "End" IF ENDOFDATA = 1 & ELSE "Start" IF ENDOFDATA = -1 & ELSE "More..." ...which we displayed in the bottom right-hand corner of the screen in true "CUA" fashion :-) Check the PowerHouse for AS/400 QDESIGN Statement Reference (version 6.07) manual for details. See page 452 for a discussion of the HOLD FIND option of the SCREEN statement, and page 178 for the ENDOFDATA function. Best regards, Conrad Conrad Whittall Marketing Manager, Application Development Tools Cognos Incorporated, 3755 Riverside Drive, Ottawa, Ontario, K1G 4K9, Canada Developer productivity never goes out of style! For details of the Cognos PowerHouse family of high-productivity development tools for Web, Windows and terminal-based business applications, and to request your own free copy of our new product demo CD, simply click here http://www.cognos.com/powerhouse. -----Original Message----- From: Daniel Mielke [mailto:DMielke@aicfunds.com] Sent: Tuesday 15 May 2001 17:00 To: Powerhouse List (E-mail) Subject: AS/400 - QUICK Control Function keys F7 & F8 Hello all, I'm looking to control the block transfer of records in a cluster for a display screen. The following screen displays the races for a specified Rat's Ratlets. RATRACES Rat Races Screen Rat number: 123456 Rat Name: Slick Rat Ratlet-no Ratlet Race number Earnings Final position 1 Mini rat 456 $3,450.00 1 7 Super rat 466 $ 70.00 5 13 Slow rat 542 $ 0.00 16 COMMAND ===>________________________________ If there were additional races for Rat 123456, F8 or [ENTER] would list the next cluster of RAT-RACES. However, since there are no more races, the F8 or [ENTER] clears the list and returns to the command in find mode. The same occurs for [F7] when the user is at the initial list of races. How do I control these function keys such that the user may scroll through the existing records stopping at both the end and the beginning without clearing the screen? Special thanks to Chuck Reinke for the use of the Rat/Ratlets analogy > BUY. HOLD. AND PROSPER. > > Daniel Mielke > Senior User Analyst > AIC Limited > 1-888-710-4242 (4614) > dmielke@aicfunds.com > www.aicfunds.com > = = = = = = = = = = = = = = = = = = = = = = = = = = = = Mailing list: powerh-l@lists.swau.edu Subscribe: "subscribe" in message body to powerh-l-request@lists.swau.edu Unsubscribe: "unsubscribe" in message body to powerh-l-request@lists.swau.edu http://lists.swau.edu/mailman/listinfo/powerh-l This list is closed, thus to post to the list you must be a subscriber. = = = = = = = = = = = = = = = = = = = = = = = = = = = = Mailing list: powerh-l@lists.swau.edu Subscribe: "subscribe" in message body to powerh-l-request@lists.swau.edu Unsubscribe: "unsubscribe" in message body to powerh-l-request@lists.swau.edu http://lists.swau.edu/mailman/listinfo/powerh-l This list is closed, thus to post to the list you must be a subscriber. From Joe.Boyle@cognos.com Wed, 16 May 2001 06:55:36 -0400 Date: Wed, 16 May 2001 06:55:36 -0400 From: Boyle, Joe Joe.Boyle@cognos.com Subject: Action command S (select) on dates/numbers Hi Markus, I am assuming that by your comment 'S action for dates (or numbers)' you are referring to select processing. I've found that select patterns do work e.g. entering %@6@ will find all numeric values containing the digit 6. The field structure as seen from Quick is EMPLOYEE_NO OF EMPLOYEES NUMERIC -9(6), this is a c-isam file on Unix. For a list of other patterns which may work, look in the 'Pattern Matching in Powerhouse' section of your Qdesign statement manual. I have also found that syntax like that below will achieve the equivalent of the '@@' function when used with indexes containing numeric segments, but only with btree type file systems such as RMS, Xbase etc... I seem to recall that Oracle and SQL did'nt like this. proc input TEST1_DATO begin ; this input procedure is required only if you have to rework a date format entered by the user. Not required if ; not a date ; if 5 lt size(truncate(fieldtext)) then begin if 7 gt size(truncate(fieldtext)) then begin if 20 gt nconvert ( fieldtext[5:2] ) then let t1 = nconvert ( & ( "20" + fieldtext[5:2] & + fieldtext [3:2] + fieldtext [1:2] )) else let t1 = nconvert ( & ( "19" + fieldtext[5:2] & + fieldtext [3:2] + fieldtext [1:2] )) end else let t1 = nconvert ( & ( fieldtext[5:4] & + fieldtext [3:2] + fieldtext [1:2] )) end END PROCEDURE PATH BEGIN let t-flag = "P" REQUEST TEST1_DATO OF TEST1 IF PROMPTOK THEN LET PATH = 1 IF PATH = 0 THEN BEGIN LET PATH = 2 END END PROCEDURE FIND BEGIN for display test1 begin IF PATH = 1 THEN begin if t-flag = "P" then begin let t-flag = "F" ; next, set the pointer in the file GET TEST1 unique & VIA TEST1_DATO using T1 optional end else GET TEST1 SEQUENTIAL end IF PATH = 2 THEN GET TEST1 SEQUENTIAL end END regards, Joe Boyle. -----Original Message----- From: Markus Grossrieder [mailto:markus.grossrieder@alba-systems.com] Sent: 16 May 2001 10:25 To: Powerhouse List (E-mail) Subject: Action command S (select) on dates/numbers Hi all, I just wonder if it is possible to use the S action for dates (or numbers) in combination with wildcards, i.e. - all invoices with '1999' in the date - all clients with '000' in the clients ID number - etc. I never find out if it's possible and (if yes) how to do it, or (if not) why not (Conrad, Bob) ? TIA, Markus ps: PH 8.13d (HP Unix), 7.10g (VMS) = = = = = = = = = = = = = = = = = = = = = = = = = = = = Mailing list: powerh-l@lists.swau.edu Subscribe: "subscribe" in message body to powerh-l-request@lists.swau.edu Unsubscribe: "unsubscribe" in message body to powerh-l-request@lists.swau.edu http://lists.swau.edu/mailman/listinfo/powerh-l This list is closed, thus to post to the list you must be a subscriber. From Joe.Boyle@cognos.com Wed, 16 May 2001 07:41:46 -0400 Date: Wed, 16 May 2001 07:41:46 -0400 From: Boyle, Joe Joe.Boyle@cognos.com Subject: Action command S (select) on dates/numbers an afterthought, you may have tried the 'entering %@6@ ' approach and found that it failed. The problem could be that you are entering the pattern too early. Select processing first prompts for request responses in the path procedure and then places you back in the action field. Once back in the action field you should see a message like "Enter ID-numbers for select values or press NEXT DATA to start data retrieval.", it is at this point that you can enter the id of a field in which to enter a pattern. MODE:S ACTION:S 01 Employee No 02 Last Name Using the example above where both fields are index segments on separate indexes, the dialogue would go as below where Q is the quick prompt and R is your response, Q1. action field R1. S path procedure Q2. ID 1 R2. return Q3. ID 2 R3. return Q4. action field R4. 1 Q5. ID 1 R5. %@6@ Q6. action field R6. return MODE:S ACTION: 01 Employee No 2,560 02 Last Name ???jo4 03 Employee No 655,360 04 Last Name ??jo3 The full data set is below, MODE:F ACTION: 01 Employee No 0 02 Last Name ?????jo6 03 Employee No 1 04 Last Name jo 05 Employee No 10 06 Last Name ????jo5 07 Employee No 2,560 08 Last Name ???jo4 09 Employee No 655,360 10 Last Name ??jo3 11 Employee No ######## 12 Last Name ?jo2 Alas, for some strange reason the message 'This entry must be numeric.' appears at the bottom of the screen. regards, Joe Boyle. -----Original Message----- From: Boyle, Joe [mailto:Joe.Boyle@cognos.com] Sent: 16 May 2001 11:56 To: 'Markus Grossrieder'; Powerhouse List (E-mail) Subject: RE: Action command S (select) on dates/numbers Hi Markus, I am assuming that by your comment 'S action for dates (or numbers)' you are referring to select processing. I've found that select patterns do work e.g. entering %@6@ will find all numeric values containing the digit 6. The field structure as seen from Quick is EMPLOYEE_NO OF EMPLOYEES NUMERIC -9(6), this is a c-isam file on Unix. For a list of other patterns which may work, look in the 'Pattern Matching in Powerhouse' section of your Qdesign statement manual. I have also found that syntax like that below will achieve the equivalent of the '@@' function when used with indexes containing numeric segments, but only with btree type file systems such as RMS, Xbase etc... I seem to recall that Oracle and SQL did'nt like this. proc input TEST1_DATO begin ; this input procedure is required only if you have to rework a date format entered by the user. Not required if ; not a date ; if 5 lt size(truncate(fieldtext)) then begin if 7 gt size(truncate(fieldtext)) then begin if 20 gt nconvert ( fieldtext[5:2] ) then let t1 = nconvert ( & ( "20" + fieldtext[5:2] & + fieldtext [3:2] + fieldtext [1:2] )) else let t1 = nconvert ( & ( "19" + fieldtext[5:2] & + fieldtext [3:2] + fieldtext [1:2] )) end else let t1 = nconvert ( & ( fieldtext[5:4] & + fieldtext [3:2] + fieldtext [1:2] )) end END PROCEDURE PATH BEGIN let t-flag = "P" REQUEST TEST1_DATO OF TEST1 IF PROMPTOK THEN LET PATH = 1 IF PATH = 0 THEN BEGIN LET PATH = 2 END END PROCEDURE FIND BEGIN for display test1 begin IF PATH = 1 THEN begin if t-flag = "P" then begin let t-flag = "F" ; next, set the pointer in the file GET TEST1 unique & VIA TEST1_DATO using T1 optional end else GET TEST1 SEQUENTIAL end IF PATH = 2 THEN GET TEST1 SEQUENTIAL end END regards, Joe Boyle. -----Original Message----- From: Markus Grossrieder [mailto:markus.grossrieder@alba-systems.com] Sent: 16 May 2001 10:25 To: Powerhouse List (E-mail) Subject: Action command S (select) on dates/numbers Hi all, I just wonder if it is possible to use the S action for dates (or numbers) in combination with wildcards, i.e. - all invoices with '1999' in the date - all clients with '000' in the clients ID number - etc. I never find out if it's possible and (if yes) how to do it, or (if not) why not (Conrad, Bob) ? TIA, Markus ps: PH 8.13d (HP Unix), 7.10g (VMS) = = = = = = = = = = = = = = = = = = = = = = = = = = = = Mailing list: powerh-l@lists.swau.edu Subscribe: "subscribe" in message body to powerh-l-request@lists.swau.edu Unsubscribe: "unsubscribe" in message body to powerh-l-request@lists.swau.edu http://lists.swau.edu/mailman/listinfo/powerh-l This list is closed, thus to post to the list you must be a subscriber. = = = = = = = = = = = = = = = = = = = = = = = = = = = = Mailing list: powerh-l@lists.swau.edu Subscribe: "subscribe" in message body to powerh-l-request@lists.swau.edu Unsubscribe: "unsubscribe" in message body to powerh-l-request@lists.swau.edu http://lists.swau.edu/mailman/listinfo/powerh-l This list is closed, thus to post to the list you must be a subscriber. From johan@vandijkgroep.nl Wed, 16 May 2001 14:03:04 +0200 Date: Wed, 16 May 2001 14:03:04 +0200 From: Johan Koelewijn johan@vandijkgroep.nl Subject: "RE: Action command S (select) on dates/numbers" Only works in select mode when the selected field is long enough to contain the pattern. Johan Koelewijn > an afterthought, you may have tried the 'entering %@6@ ' approach and found > that it failed. > > The problem could be that you are entering the pattern too early. > > Select processing first prompts for request responses in the path procedure > and then places you back in the action field. Once back in the action field > you should see a message like "Enter ID-numbers for select values or press > NEXT DATA to start data retrieval.", it is at this point that you can enter > the id of a field in which to enter a pattern. > > MODE:S ACTION:S > 01 Employee No > 02 Last Name > > Using the example above where both fields are index segments on separate > indexes, the dialogue would go as below where Q is the quick prompt and R is > your response, > > Q1. action field > R1. S > path procedure > Q2. ID 1 > R2. return > Q3. ID 2 > R3. return > Q4. action field > R4. 1 > Q5. ID 1 > R5. %@6@ > Q6. action field > R6. return > > MODE:S ACTION: > 01 Employee No 2,560 > 02 Last Name ???jo4 > 03 Employee No 655,360 > 04 Last Name ??jo3 > > The full data set is below, > > MODE:F ACTION: > 01 Employee No 0 > 02 Last Name ?????jo6 > 03 Employee No 1 > 04 Last Name jo > 05 Employee No 10 > 06 Last Name ????jo5 > 07 Employee No 2,560 > 08 Last Name ???jo4 > 09 Employee No 655,360 > 10 Last Name ??jo3 > 11 Employee No ######## > 12 Last Name ?jo2 > > > Alas, for some strange reason the message 'This entry must be numeric.' > appears at the bottom of the screen. > > > regards, > Joe Boyle. > > -----Original Message----- > From: Boyle, Joe [mailto:Joe.Boyle@cognos.com] > Sent: 16 May 2001 11:56 > To: 'Markus Grossrieder'; Powerhouse List (E-mail) > Subject: RE: Action command S (select) on dates/numbers > > > Hi Markus, > > I am assuming that by your comment 'S action for dates (or numbers)' you are > referring to select processing. I've found that select patterns do work e.g. > entering %@6@ will find all numeric values containing the digit 6. > > The field structure as seen from Quick is EMPLOYEE_NO OF EMPLOYEES NUMERIC > -9(6), this is a c-isam file on Unix. > > For a list of other patterns which may work, look in the 'Pattern Matching > in Powerhouse' section of your Qdesign statement manual. > > I have also found that syntax like that below will achieve the equivalent of > the '@@' function when used with indexes containing numeric segments, but > only with btree type file systems such as RMS, Xbase etc... I seem to recall > that Oracle and SQL did'nt like this. > > > proc input TEST1_DATO > begin > > ; this input procedure is required only if you have to rework a date format > entered by the user. Not required if ; not a date > ; > > > if 5 lt size(truncate(fieldtext)) > then begin > > if 7 gt size(truncate(fieldtext)) > then begin > if 20 gt nconvert ( fieldtext[5:2] ) > then let t1 = nconvert ( & > ( "20" + fieldtext[5:2] & > + fieldtext [3:2] + fieldtext [1:2] )) > else let t1 = nconvert ( & > ( "19" + fieldtext[5:2] & > + fieldtext [3:2] + fieldtext [1:2] )) > end > else let t1 = nconvert ( & > ( fieldtext[5:4] & > + fieldtext [3:2] + fieldtext [1:2] )) > > > end > > END > > > PROCEDURE PATH > BEGIN > let t-flag = "P" > > REQUEST TEST1_DATO OF TEST1 > IF PROMPTOK > THEN LET PATH = 1 > IF PATH = 0 > THEN BEGIN > LET PATH = 2 > END > END > > > PROCEDURE FIND > BEGIN > > > for display test1 > begin > > IF PATH = 1 > THEN begin > > if t-flag = "P" > then begin > let t-flag = "F" > ; next, set the pointer in the file > GET TEST1 unique & > VIA TEST1_DATO using T1 optional > end > > else GET TEST1 SEQUENTIAL > > end > > > IF PATH = 2 > THEN GET TEST1 SEQUENTIAL > > end > END > > > regards, > Joe Boyle. > > > -----Original Message----- > From: Markus Grossrieder [mailto:markus.grossrieder@alba-systems.com] > Sent: 16 May 2001 10:25 > To: Powerhouse List (E-mail) > Subject: Action command S (select) on dates/numbers > > > Hi all, > > I just wonder if it is possible to use the S action for dates (or numbers) > in > combination with wildcards, i.e. > - all invoices with '1999' in the date > - all clients with '000' in the clients ID number > - etc. > I never find out if it's possible and > (if yes) how to do it, or > (if not) why not (Conrad, Bob) ? > > TIA, Markus > > ps: PH 8.13d (HP Unix), 7.10g (VMS) > > > = = = = = = = = = = = = = = = = = = = = = = = = = = = = > Mailing list: powerh-l@lists.swau.edu > Subscribe: "subscribe" in message body to powerh-l-request@lists.swau.edu > Unsubscribe: "unsubscribe" in message body to > powerh-l-request@lists.swau.edu > http://lists.swau.edu/mailman/listinfo/powerh-l > This list is closed, thus to post to the list you must be a subscriber. > > = = = = = = = = = = = = = = = = = = = = = = = = = = = = > Mailing list: powerh-l@lists.swau.edu > Subscribe: "subscribe" in message body to powerh-l-request@lists.swau.edu > Unsubscribe: "unsubscribe" in message body to > powerh-l-request@lists.swau.edu > http://lists.swau.edu/mailman/listinfo/powerh-l > This list is closed, thus to post to the list you must be a subscriber. > > = = = = = = = = = = = = = = = = = = = = = = = = = = = = > Mailing list: powerh-l@lists.swau.edu > Subscribe: "subscribe" in message body to powerh-l-request@lists.swau.edu > Unsubscribe: "unsubscribe" in message body to powerh-l-request@lists.swau.edu > http://lists.swau.edu/mailman/listinfo/powerh-l > This list is closed, thus to post to the list you must be a subscriber. > From Chris.Sharman@ccagroup.co.uk Wed, 16 May 2001 13:15:30 +0100 Date: Wed, 16 May 2001 13:15:30 +0100 From: Chris Sharman Chris.Sharman@ccagroup.co.uk Subject: Action command S (select) on dates/numbers >I am assuming that by your comment 'S action for dates (or numbers)' you are >referring to select processing. I've found that select patterns do work e.g. >entering %@6@ will find all numeric values containing the digit 6. So it does ! I stand corrected. I'll admit I've never really wanted to select on numbers, just dates. It's very easy for a pattern to get too big for its field too. >For a list of other patterns which may work, look in the 'Pattern Matching >in Powerhouse' section of your Qdesign statement manual. Indeed. Sorry for posting a wrong answer. Chris _______________________________________________________________________ Chris.Sharman@CCAgroup.co.uk http://www.ccagroup.co.uk/ CCA Stationery Ltd, Eastway, Fulwood, Preston, Lancashire, PR2 9WS. ----------------------------------------------------------------------- Any views expressed in this message are those of the sender and not necessarily those of CCA Group. The unauthorized use, disclosure, copying or alteration of this message is forbidden. The contents of this message may be confidential and/or privileged, copyright CCA Group and are intended solely for the use of the individual or entity to whom they are addressed. Whilst this message has been scanned, CCA Group cannot guarantee that it is virus free or compatible with your systems and accepts no responsibility for any loss or damage arising from its use. The recipient is advised to run their own anti-virus software. If you receive this message in error please contact postmaster@ccagroup.co.uk immediately, destroy any copies and delete it from your computer systems. From Joe.Boyle@cognos.com Wed, 16 May 2001 12:15:40 -0400 Date: Wed, 16 May 2001 12:15:40 -0400 From: Boyle, Joe Joe.Boyle@cognos.com Subject: PowerHouse Web Setup???? Hi Art, Bob, Bob and everyone, Oh dear, I'm afraid its humble pie time. So much for stating the obvious, the suggestion to copy the phcgi.exe from the install directory on the HP server to the directory 'C:\InetPub\wwwroot\cgi-bin' on the NT box, was a complete misinterpretation. The phcgi.exe file for the NT box was present all of the time, but not on the NT server, it was still in the box in which it arrived. Once the install kit for the NT box was installed, the phcgi.exe file appeared. Apologies all. regards, Joe Boyle. -----Original Message----- From: Boyle, Joe Sent: 15 May 2001 09:37 To: 'abahrs@denkor.com'; 'powerh-l@list.swau.edu' Subject: RE: PowerHouse Web Setup???? Hi Art, Bob, everyone, I'm sorry to state the obvious but just in case - you have actually copied the phcgi.exe from the install directory on the HP server to the directory 'C:\InetPub\wwwroot\cgi-bin' on the NT box ? Also, you will very likely have to change the security access of the phcgi.dat file to 'everyone', as I seem to recall that the 404 error cannot find the phcgi.dat script, see below for details. If the PH Web Administrator or the IIS web server is on Windows 2000, the phcgi.dat file will be saved with limited permissions when a Save Changes is done in the PH Web Administrator. To resolve, change the permissions on the phcgi.dat file in the cgi-bin directory, opening it to Everyone with Full Control, after changes have been saved in the PH Web Administrator. You may wish to download some of the PHweb docs for MPE from the Cognos web site at the URL below, http://support.cognos.com/supported/support/support_library/phweb/phweb22doc .html Some PHweb docs available are listed below, 2.29D2 Readme (MPEiX) Release notes and how to install PowerHouse Web for MPEiX updated for 2.29D2. PH Web Getting Started (MPEiX) How to install PowerHouse Web 2.29 for MPEiX. PowerHouse Web Guide Revised manual intended for PowerHouse programmers who wish to create web applications. regards, Joe Boyle. -----Original Message----- From: Edis, Bob [mailto:bob.edis@fleetpride.com] Sent: 15 May 2001 00:18 To: 'powerh-l@list.swau.edu' Subject: RE: PowerHouse Web Setup???? Art I am referring to the IIS administrator. To get there on the NT web server box: 1. Start->Programs->Windows NT 4.0 Option Pack->Microsoft Internet Information Server->Internet service Manager 2. Expand your web site folder; you should see a folder called 'cgi-bin'. This is a virtual directory. 3. Select this folder, right click and select properties. 4. Under 'Local Path:' you should see something like 'C:\InetPub\wwwroot\cgi-bin'. C:\InetPub\wwwroot is where we put a HTML menu pages that refer to the PH Web screens. E.g. dw_menu.htm is a simple HTML page containing references like . You *could* build this menu page using Qdesign but it must be copied to the wwwroot directory. We reference the menu in the web browser as //cogdev-02/dw_menu.htm. The '//cogdev-02' is the name of the IIS web server machine. Regards, Blue -----Original Message----- From: Art Bahrs [mailto:abahrs@denkor.com] Sent: Monday, May 14, 2001 6:13 PM To: Edis, Bob; 'powerh-l@list.swau.edu' Subject: Re: PowerHouse Web Setup???? Hi Blue :) Ok... You mentioned something I haven't checked (mainly 'cause I really don't understand it) .... ie "Have the virtual directories been declared/exist in the IIS Console?" Um... which directories are your referring to? I have set up the administrator and the PH web server stuff on our 928 as per all the books and guidelines I have.... btw.... the books and docs I got with the demo kit are very good! Art "losing what *little* hair I have left ! hehe" Bahrs ----- Original Message ----- From: "Edis, Bob" To: "'powerh-l@list.swau.edu'" Sent: Monday, May 14, 2001 1:42 PM Subject: RE: PowerHouse Web Setup???? > G'day Art > > Are you declaring a specific path for the QKC and HTML files in the screen > source? > What do you have in the default HTML and QKC directory sections of PH Web > Administrator? > Have the virtual directories been declared/exist in the IIS Console? > Are you trying to access the screen page directly or through a basic HTML > menu page? > > We found we have to put the initial page in the wwwroot directory on the web > server machine and have this page call the web screen which is located on a > different box. > We also use the default QKC and HTML directory sections of the PH Web Admin > tool for the PH Web server rather than explicit path names in the source > code. > > Regards, > Blue > > > -----Original Message----- > From: Art Bahrs [mailto:abahrs@denkor.com] > Sent: Monday, May 14, 2001 2:57 PM > To: Power House ListServ > Subject: PowerHouse Web Setup???? > > > Hi All :) > Ok... Dumb Question time.... I am trying to bring up PowerHouse Web Demo > on a HP e3000 s928 using IIS on a NT server and have the Dispatcher and PH > WebServer running on the HP 3k 928.... IIS is running on the NT box.... I > have created a test HTML Quick Screen via QDesign 8.29 D3... > > The page is not being found (the infamous error 404 ...) so... I checked > the NT server has proper entries in DNS for the 928 and can ping the 928.... > the 928 has proper entries in the hosts.net.sys file for the NT server and > can ping the NT server by IP address... > > I am new to most of this ... so be nice to me :) but hey... if you got the > answer a little friendly abuse would, of course, not be out of line :) hehe > > Art "Waiting for return call from Cognos... but fishing for all info! hehe" > Bahrs > > > > > = = = = = = = = = = = = = = = = = = = = = = = = = = = = > Mailing list: powerh-l@lists.swau.edu > Subscribe: "subscribe" in message body to powerh-l-request@lists.swau.edu > Unsubscribe: "unsubscribe" in message body to > powerh-l-request@lists.swau.edu > http://lists.swau.edu/mailman/listinfo/powerh-l > This list is closed, thus to post to the list you must be a subscriber. > > = = = = = = = = = = = = = = = = = = = = = = = = = = = = > Mailing list: powerh-l@lists.swau.edu > Subscribe: "subscribe" in message body to powerh-l-request@lists.swau.edu > Unsubscribe: "unsubscribe" in message body to powerh-l-request@lists.swau.edu > http://lists.swau.edu/mailman/listinfo/powerh-l > This list is closed, thus to post to the list you must be a subscriber. = = = = = = = = = = = = = = = = = = = = = = = = = = = = Mailing list: powerh-l@lists.swau.edu Subscribe: "subscribe" in message body to powerh-l-request@lists.swau.edu Unsubscribe: "unsubscribe" in message body to powerh-l-request@lists.swau.edu http://lists.swau.edu/mailman/listinfo/powerh-l This list is closed, thus to post to the list you must be a subscriber. From bengt.johansson@thoreb.se Wed, 16 May 2001 19:05:22 +0200 Date: Wed, 16 May 2001 19:05:22 +0200 From: Bengt Johansson bengt.johansson@thoreb.se Subject: Delays in connection to HP3000 via VT-MGR vs TELNET Hi everybody ! This question is not pure PowerHouse but it shows up when using PowerHouse. We have an HP3000 at our office in Gothenburg and our customers are spread out all over Sweden. Our customers mostly connects to the HP3000 using the Internet, but when there are problems with Internet or their local network they all have a lifeline using the old technique with direct telephone modem ( 19200 bps ), and that is always working. When using Internet our customers use either the VT-MGR or the TELNET protocoll in 'Connection Setup' in Reflection. Some customers never have problems ( and that is good ). But some customers almost always have problems ( and that is no good ), and so have I myself sometimes when connecting from my home i Halmstad, where I also have an office. When using TELNET there are sometimes delays for every character sent. I myself have seen delays up to 15 seconds but one customer have seen delays up to 40 seconds. This delay is from having pressed one button till the character pops up at the screen. Very annoying. When using TELNET every character ( plus 40 extra characters ), when pressing the button, are sent as a package from the user, to the HP3000 and then back to the user before the character pops up at the screen. This TELNET delay became a bit shorter = better ( but not good enough ) after having changed 'Retransmission Interval Lower Bound' from default 4 seconds to 1 second in NETXPORT.GPROT.TCP in NMMGR. Now to the question: When using VT-MGR there is ALWAYS a delay in some PowerHouse Quick screens, especially one heavily used screen with a primary file and a detail file with many occurrancies and much data on the screen in Find Mode after having read one set of data from having pressed Return to get the next set of data till the new data pops up at the screen. >From home with my relatively slow Internet connection this delay is always 13 seconds with VT-MGR in this screen filled with data. Using TELNET this delay is only between 1 and 2 seconds, but as I said TELNET sometimes is too slow in another way. Inside our office in Gothenburg with a fast LAN ( 100 Mbit/sec ) the same screen gives a delay between 4 and 5 seconds with VT-MGR ( when the TELNET delay is under 1 second ), and that is too much isn't it ? And the 13 seconds mentioned above I can't stand. Would you ? When using common MPE commands like LISTF or other, there are no delays at all inside the fast LAN neither with VT-MGR nor TELNET, so why when using PowerHouse ? When using VT-MGR every pressed button is echoed to the screen locally and there is a package sent to the HP3000 only when Return is pressed, so this protocoll should be more efficient, but at least in PowerHouse it seems not to be so efficiant. Does anyone know a prescription to solve this problem ? Some of my customers hate these delays so much that they use the telephone modem instead of Internet, and have to pay the telephone cost for that, when Internet is free when you have a fixed connection. I have asked HP Support, but they have not given me any solution yet. Another queer question is why some other customers NEVER have problems using TELNET. We use MPE 6.0 and PowerHouse 8.29.D3. Best regards Bengt Johansson AB Thoreb From PICKERIJ@norbord.com Wed, 16 May 2001 13:38:48 -0400 Date: Wed, 16 May 2001 13:38:48 -0400 From: Pickering, John (NORBORD) PICKERIJ@norbord.com Subject: Delays in connection to HP3000 via VT-MGR vs TELNET Do you run Quick the same way in every case? With the Quick run time setting of "READ=CHAR" even vt-mgr sessions will send each keystroke as a packet. If you don't explicitly set the READ or LINEREAD parameter then the Quick udc will attempt to alter it in some cases. As for other sources of difference I think this question might be better addressed to the 3000-L listserver (hp3000-l@raven.utc.edu). There's lots of experts on that list who could answer telnet vs. vt-mgr vs. modem questions. They'll be happy to tell you about an advanced version of telnet that should improve things. And some of them even use PowerHouse! Regards, JWP > -----Original Message----- > From: Bengt Johansson [SMTP:bengt.johansson@thoreb.se] > Sent: Wednesday, May 16, 2001 1:05 PM > To: 'Powerhouse List Group' > Subject: Delays in connection to HP3000 via VT-MGR vs TELNET > > Hi everybody ! > > This question is not pure PowerHouse but it shows up when using > PowerHouse. > > We have an HP3000 at our office in Gothenburg and our customers are > spread out all over Sweden. Our customers mostly connects to the HP3000 > using the Internet, but when there are problems with Internet or their > local network they all have a lifeline using the old technique with > direct telephone modem ( 19200 bps ), and that is always working. > > When using Internet our customers use either the VT-MGR or the TELNET > protocoll in 'Connection Setup' in Reflection. > Some customers never have problems ( and that is good ). > But some customers almost always have problems ( and that is no good ), > and so have I myself sometimes when connecting from my home i Halmstad, > where I also have an office. > > When using TELNET there are sometimes delays for every character sent. I > myself have seen delays up to 15 seconds but one customer have seen > delays up to 40 seconds. > This delay is from having pressed one button till the character pops up > at the screen. > Very annoying. > When using TELNET every character ( plus 40 extra characters ), when > pressing the button, are sent as a package from the user, to the HP3000 > and then back to the user before the character pops up at the screen. > This TELNET delay became a bit shorter = better ( but not good enough ) > after having changed 'Retransmission Interval Lower Bound' from default > 4 seconds to 1 second in NETXPORT.GPROT.TCP in NMMGR. > > Now to the question: > When using VT-MGR there is ALWAYS a delay in some PowerHouse Quick > screens, especially one heavily used screen with a primary file and a > detail file with many occurrancies and much data on the screen in Find > Mode after having read one set of data from having pressed Return to get > the next set of data till the new data pops up at the screen. > From home with my relatively slow Internet connection this delay is > always 13 seconds with VT-MGR in this screen filled with data. > Using TELNET this delay is only between 1 and 2 seconds, but as I said > TELNET sometimes is too slow in another way. > Inside our office in Gothenburg with a fast LAN ( 100 Mbit/sec ) the > same screen gives a delay between 4 and 5 seconds with VT-MGR ( when the > TELNET delay is under 1 second ), and that is too much isn't it ? And > the 13 seconds mentioned above I can't stand. Would you ? > When using common MPE commands like LISTF or other, there are no delays > at all inside the fast LAN neither with VT-MGR nor TELNET, so why when > using PowerHouse ? > > When using VT-MGR every pressed button is echoed to the screen locally > and there is a package sent to the HP3000 only when Return is pressed, > so this protocoll should be more efficient, but at least in PowerHouse > it seems not to be so efficiant. > > Does anyone know a prescription to solve this problem ? > Some of my customers hate these delays so much that they use the > telephone modem instead of Internet, and have to pay the telephone cost > for that, when Internet is free when you have a fixed connection. > I have asked HP Support, but they have not given me any solution yet. > > Another queer question is why some other customers NEVER have problems > using TELNET. > > We use MPE 6.0 and PowerHouse 8.29.D3. > > Best regards > Bengt Johansson > AB Thoreb > From Glenn_Waters@ncsu.edu Wed, 16 May 2001 13:41:26 -0400 Date: Wed, 16 May 2001 13:41:26 -0400 From: Glenn Waters Glenn_Waters@ncsu.edu Subject: Delays in connection to HP3000 via VT-MGR vs TELNET Just a couple of comments that hopefully might help. Based on one of your comments early on "When using Internet our customers use either the VT-MGR or the TELNET protocoll in 'Connection Setup' in Reflection." If you are using Reflections why are you using Telenet? Wouldn't it be easier to just emulate one of the regular HP terminal types(using Reflection). It's been a while since I was on the 3000, I seem to remember you have to configure the modem port slightly less "generically". But that is a dim memory, so I may be wrong on that one. If you are able to use the HP term types though, instead of telnet. That should increase your dial up responsivness. If thats not possible, have you tried other telenet clients. There are vast differences in the assorted clients out there, you definitely do NOT want to use the generic one that comes with Windows. We currently use Putty (because its free), but there are plenty of others. Maybe you'd have better luck with another telenet package? At 07:05 PM 05/16/2001 +0200, you wrote: >Hi everybody ! > >This question is not pure PowerHouse but it shows up when using >PowerHouse. > >We have an HP3000 at our office in Gothenburg and our customers are >spread out all over Sweden. Our customers mostly connects to the HP3000 >using the Internet, but when there are problems with Internet or their >local network they all have a lifeline using the old technique with >direct telephone modem ( 19200 bps ), and that is always working. > >When using Internet our customers use either the VT-MGR or the TELNET >protocoll in 'Connection Setup' in Reflection. >Some customers never have problems ( and that is good ). >But some customers almost always have problems ( and that is no good ), >and so have I myself sometimes when connecting from my home i Halmstad, >where I also have an office. > >When using TELNET there are sometimes delays for every character sent. I >myself have seen delays up to 15 seconds but one customer have seen >delays up to 40 seconds. >This delay is from having pressed one button till the character pops up >at the screen. >Very annoying. >When using TELNET every character ( plus 40 extra characters ), when >pressing the button, are sent as a package from the user, to the HP3000 >and then back to the user before the character pops up at the screen. >This TELNET delay became a bit shorter = better ( but not good enough ) >after having changed 'Retransmission Interval Lower Bound' from default >4 seconds to 1 second in NETXPORT.GPROT.TCP in NMMGR. > >Now to the question: >When using VT-MGR there is ALWAYS a delay in some PowerHouse Quick >screens, especially one heavily used screen with a primary file and a >detail file with many occurrancies and much data on the screen in Find >Mode after having read one set of data from having pressed Return to get >the next set of data till the new data pops up at the screen. > From home with my relatively slow Internet connection this delay is >always 13 seconds with VT-MGR in this screen filled with data. >Using TELNET this delay is only between 1 and 2 seconds, but as I said >TELNET sometimes is too slow in another way. >Inside our office in Gothenburg with a fast LAN ( 100 Mbit/sec ) the >same screen gives a delay between 4 and 5 seconds with VT-MGR ( when the >TELNET delay is under 1 second ), and that is too much isn't it ? And >the 13 seconds mentioned above I can't stand. Would you ? >When using common MPE commands like LISTF or other, there are no delays >at all inside the fast LAN neither with VT-MGR nor TELNET, so why when >using PowerHouse ? > >When using VT-MGR every pressed button is echoed to the screen locally >and there is a package sent to the HP3000 only when Return is pressed, >so this protocoll should be more efficient, but at least in PowerHouse >it seems not to be so efficiant. > >Does anyone know a prescription to solve this problem ? >Some of my customers hate these delays so much that they use the >telephone modem instead of Internet, and have to pay the telephone cost >for that, when Internet is free when you have a fixed connection. >I have asked HP Support, but they have not given me any solution yet. > >Another queer question is why some other customers NEVER have problems >using TELNET. > >We use MPE 6.0 and PowerHouse 8.29.D3. > >Best regards >Bengt Johansson >AB Thoreb > > >= = = = = = = = = = = = = = = = = = = = = = = = = = = = >Mailing list: powerh-l@lists.swau.edu >Subscribe: "subscribe" in message body to powerh-l-request@lists.swau.edu >Unsubscribe: "unsubscribe" in message body to powerh-l-request@lists.swau.edu >http://lists.swau.edu/mailman/listinfo/powerh-l >This list is closed, thus to post to the list you must be a subscriber. Glenn Waters North Carolina State University College of Veterinary Medicine 4700 Hillsborough St., Box 8401 Raleigh, NC 27695-8401 Phone: (919) 513-6652 Fax: (919) 513-6452 Email: Glenn_Waters@ncsu.edu From abahrs@denkor.com Wed, 16 May 2001 11:06:06 -0700 Date: Wed, 16 May 2001 11:06:06 -0700 From: Art Bahrs abahrs@denkor.com Subject: Delays in connection to HP3000 via VT-MGR vs TELNET Hi Glenn and All :) Ok... Glenn has a great point.... if the 3k in question is on the network then you can use NS/VT in Reflection to contact your HP e3k... and get network style connections from home.... especially if you have DSL! (we do this here all the time... the firewall authenticates us then lets us use NS/VT to the 3k... Art "much better connections! hehe " Bahrs We do this here... this way we don't even use our support modem... just come in via the firewall.... ----- Original Message ----- From: "Glenn Waters" To: Sent: Wednesday, May 16, 2001 10:41 AM Subject: Re: Delays in connection to HP3000 via VT-MGR vs TELNET > Just a couple of comments that hopefully might help. > > Based on one of your comments early on > > "When using Internet our customers use either the VT-MGR or the TELNET > protocoll in 'Connection Setup' in Reflection." > > If you are using Reflections why are you using Telenet? Wouldn't it be easier to just emulate one of the regular HP terminal types(using Reflection). It's been a while since I was on the 3000, I seem to remember you have to configure the modem port slightly less "generically". But that is a dim memory, so I may be wrong on that one. If you are able to use the HP term types though, instead of telnet. That should increase your dial up responsivness. > > If thats not possible, have you tried other telenet clients. There are vast differences in the assorted clients out there, you definitely do NOT want to use the generic one that comes with Windows. We currently use Putty (because its free), but there are plenty of others. Maybe you'd have better luck with another telenet package? > > > > > > At 07:05 PM 05/16/2001 +0200, you wrote: > >Hi everybody ! > > > >This question is not pure PowerHouse but it shows up when using > >PowerHouse. > > > >We have an HP3000 at our office in Gothenburg and our customers are > >spread out all over Sweden. Our customers mostly connects to the HP3000 > >using the Internet, but when there are problems with Internet or their > >local network they all have a lifeline using the old technique with > >direct telephone modem ( 19200 bps ), and that is always working. > > > >When using Internet our customers use either the VT-MGR or the TELNET > >protocoll in 'Connection Setup' in Reflection. > >Some customers never have problems ( and that is good ). > >But some customers almost always have problems ( and that is no good ), > >and so have I myself sometimes when connecting from my home i Halmstad, > >where I also have an office. > > > >When using TELNET there are sometimes delays for every character sent. I > >myself have seen delays up to 15 seconds but one customer have seen > >delays up to 40 seconds. > >This delay is from having pressed one button till the character pops up > >at the screen. > >Very annoying. > >When using TELNET every character ( plus 40 extra characters ), when > >pressing the button, are sent as a package from the user, to the HP3000 > >and then back to the user before the character pops up at the screen. > >This TELNET delay became a bit shorter = better ( but not good enough ) > >after having changed 'Retransmission Interval Lower Bound' from default > >4 seconds to 1 second in NETXPORT.GPROT.TCP in NMMGR. > > > >Now to the question: > >When using VT-MGR there is ALWAYS a delay in some PowerHouse Quick > >screens, especially one heavily used screen with a primary file and a > >detail file with many occurrancies and much data on the screen in Find > >Mode after having read one set of data from having pressed Return to get > >the next set of data till the new data pops up at the screen. > > From home with my relatively slow Internet connection this delay is > >always 13 seconds with VT-MGR in this screen filled with data. > >Using TELNET this delay is only between 1 and 2 seconds, but as I said > >TELNET sometimes is too slow in another way. > >Inside our office in Gothenburg with a fast LAN ( 100 Mbit/sec ) the > >same screen gives a delay between 4 and 5 seconds with VT-MGR ( when the > >TELNET delay is under 1 second ), and that is too much isn't it ? And > >the 13 seconds mentioned above I can't stand. Would you ? > >When using common MPE commands like LISTF or other, there are no delays > >at all inside the fast LAN neither with VT-MGR nor TELNET, so why when > >using PowerHouse ? > > > >When using VT-MGR every pressed button is echoed to the screen locally > >and there is a package sent to the HP3000 only when Return is pressed, > >so this protocoll should be more efficient, but at least in PowerHouse > >it seems not to be so efficiant. > > > >Does anyone know a prescription to solve this problem ? > >Some of my customers hate these delays so much that they use the > >telephone modem instead of Internet, and have to pay the telephone cost > >for that, when Internet is free when you have a fixed connection. > >I have asked HP Support, but they have not given me any solution yet. > > > >Another queer question is why some other customers NEVER have problems > >using TELNET. > > > >We use MPE 6.0 and PowerHouse 8.29.D3. > > > >Best regards > >Bengt Johansson > >AB Thoreb > > > > > >= = = = = = = = = = = = = = = = = = = = = = = = = = = = > >Mailing list: powerh-l@lists.swau.edu > >Subscribe: "subscribe" in message body to powerh-l-request@lists.swau.edu > >Unsubscribe: "unsubscribe" in message body to powerh-l-request@lists.swau.edu > >http://lists.swau.edu/mailman/listinfo/powerh-l > >This list is closed, thus to post to the list you must be a subscriber. > > > Glenn Waters > North Carolina State University > College of Veterinary Medicine > 4700 Hillsborough St., Box 8401 > Raleigh, NC 27695-8401 > Phone: (919) 513-6652 > Fax: (919) 513-6452 > Email: Glenn_Waters@ncsu.edu > > > > = = = = = = = = = = = = = = = = = = = = = = = = = = = = > Mailing list: powerh-l@lists.swau.edu > Subscribe: "subscribe" in message body to powerh-l-request@lists.swau.edu > Unsubscribe: "unsubscribe" in message body to powerh-l-request@lists.swau.edu > http://lists.swau.edu/mailman/listinfo/powerh-l > This list is closed, thus to post to the list you must be a subscriber. From jerryh@psc-usa.com 16 May 2001 13:44:37 -0500 Date: 16 May 2001 13:44:37 -0500 From: Jerry Hirschel jerryh@psc-usa.com Subject: Help Hi all I have a client in a suburb of Chicago looking for a VAX Powerhouse person for contract. Anybody interested? -- Jerry S. Hirschel, CPC Professional Search Centre, Ltd. 847-330-3250 847-330-3255 fax jerryh@psc-usa.com http://www.psc-usa.com From markus.grossrieder@alba-systems.com Wed, 16 May 2001 20:46:07 +0200 Date: Wed, 16 May 2001 20:46:07 +0200 From: Markus Grossrieder markus.grossrieder@alba-systems.com Subject: Action command S (select) on dates/numbers Thanks to Chris, Joe & Johan for the fast answers ! I'll give the propositions a try when I'll have some time. I was hoping, against all experience, that there is some way, pattern, or holy sign to use the 'select processing' for numeric data just as it works for strings ... Woudn't it be nice if you could say 'Show me everything from 2001/01/01', or 'bigger than 10000' ? Bob, Conrad, something planned ? cu, Markus From oh@scanconsult.dk Wed, 16 May 2001 22:46:42 +0200 Date: Wed, 16 May 2001 22:46:42 +0200 From: Ole Hansen, ScanConsult oh@scanconsult.dk Subject: Sv: Delays in connection to HP3000 via VT-MGR vs TELNET Hi Bengt, In addition to John Pickering's explanation of READ=CHAR vs. LINEREAD, I recall having similar problems, when we first connected remote sites over X.25-connections. At the advice of Cognos we changed our startup-command to: QUICK;LINEREAD=YES;TERM=HP2392-X25; etc. Now that we use Internet-borne connections instead, we still have TERM=HP2392-X25 (not because we did some clever thinking, but we didn't change anything, and had no problems when we switched). Maybe somebody at Cognos may want to comment on the significance of TERM in this context. Slightly off topic, we have noticed another peculiarity when running PowerHouse screens over 'long' connections (internet or satelite links). We have a few screens that run WHILE RETRIEVING statements and does a lot of spin-off processing, a.o. some RUN COMMAND statements, but with no display activity (i.e. no DISPLAY or INFORM verbs). These screens run 5-10 times slower over 'long' connections than when run over a local network. We have observed, that even if the screen is not changing its appearance, there is a lot of network traffic during the spin-off processing, i.e. a lot of packets are exchanged between QUICK on the HP3000 and Reflection on the PC - thus causing the slower response we think. Best regards Ole Hansen ==================================================================== E-mail: oh@scanconsult.dk Cell Phone: +45 40 40 11 50 ==================================================================== ScanConsult IT-Partners ApS Voice : +45 87 38 74 74 Jegstrupvej 96A Fax : +45 87 38 74 75 DK-8361 Hasselager Support : support@scanconsult.dk Denmark Sales : sales@scanconsult.dk http://www.scanconsult.dk === ParaSuite : A Suite of Air Cargo Handling solutions === === ParaSafe : A Suite of IT-Security solutions === === We partner with Hewlett-Packard, Cognos, Lotus and others === -----Oprindelig meddelelse----- Fra: Bengt Johansson Til: 'Powerhouse List Group' Dato: 16. maj 2001 19:10 Emne: Delays in connection to HP3000 via VT-MGR vs TELNET >Hi everybody ! > >This question is not pure PowerHouse but it shows up when using >PowerHouse. > >We have an HP3000 at our office in Gothenburg and our customers are >spread out all over Sweden. Our customers mostly connects to the HP3000 >using the Internet, but when there are problems with Internet or their >local network they all have a lifeline using the old technique with >direct telephone modem ( 19200 bps ), and that is always working. > >When using Internet our customers use either the VT-MGR or the TELNET >protocoll in 'Connection Setup' in Reflection. >Some customers never have problems ( and that is good ). >But some customers almost always have problems ( and that is no good ), >and so have I myself sometimes when connecting from my home i Halmstad, >where I also have an office. > >When using TELNET there are sometimes delays for every character sent. I >myself have seen delays up to 15 seconds but one customer have seen >delays up to 40 seconds. >This delay is from having pressed one button till the character pops up >at the screen. >Very annoying. >When using TELNET every character ( plus 40 extra characters ), when >pressing the button, are sent as a package from the user, to the HP3000 >and then back to the user before the character pops up at the screen. >This TELNET delay became a bit shorter = better ( but not good enough ) >after having changed 'Retransmission Interval Lower Bound' from default >4 seconds to 1 second in NETXPORT.GPROT.TCP in NMMGR. > >Now to the question: >When using VT-MGR there is ALWAYS a delay in some PowerHouse Quick >screens, especially one heavily used screen with a primary file and a >detail file with many occurrancies and much data on the screen in Find >Mode after having read one set of data from having pressed Return to get >the next set of data till the new data pops up at the screen. >>From home with my relatively slow Internet connection this delay is >always 13 seconds with VT-MGR in this screen filled with data. >Using TELNET this delay is only between 1 and 2 seconds, but as I said >TELNET sometimes is too slow in another way. >Inside our office in Gothenburg with a fast LAN ( 100 Mbit/sec ) the >same screen gives a delay between 4 and 5 seconds with VT-MGR ( when the >TELNET delay is under 1 second ), and that is too much isn't it ? And >the 13 seconds mentioned above I can't stand. Would you ? >When using common MPE commands like LISTF or other, there are no delays >at all inside the fast LAN neither with VT-MGR nor TELNET, so why when >using PowerHouse ? > >When using VT-MGR every pressed button is echoed to the screen locally >and there is a package sent to the HP3000 only when Return is pressed, >so this protocoll should be more efficient, but at least in PowerHouse >it seems not to be so efficiant. > >Does anyone know a prescription to solve this problem ? >Some of my customers hate these delays so much that they use the >telephone modem instead of Internet, and have to pay the telephone cost >for that, when Internet is free when you have a fixed connection. >I have asked HP Support, but they have not given me any solution yet. > >Another queer question is why some other customers NEVER have problems >using TELNET. > >We use MPE 6.0 and PowerHouse 8.29.D3. > >Best regards >Bengt Johansson >AB Thoreb > > >= = = = = = = = = = = = = = = = = = = = = = = = = = = = >Mailing list: powerh-l@lists.swau.edu >Subscribe: "subscribe" in message body to powerh-l-request@lists.swau.edu >Unsubscribe: "unsubscribe" in message body to powerh-l-request@lists.swau.edu >http://lists.swau.edu/mailman/listinfo/powerh-l >This list is closed, thus to post to the list you must be a subscriber. > From Esko.Lahdensivu@soon.fi Thu, 17 May 2001 10:02:49 +0300 Date: Thu, 17 May 2001 10:02:49 +0300 From: Esko Lahdensivu Esko.Lahdensivu@soon.fi Subject: Action command S (select) on dates/numbers Markus Grossrieder wrote: > Thanks to Chris, Joe & Johan for the fast answers ! I'll give the propositions a > try when I'll have some time. > > I was hoping, against all experience, that there is some way, pattern, or holy > sign to use the 'select processing' for numeric data just as it works for > strings ... > MPE/iX, PH 8.29. If I recall correctly, the original question was about searching dates. well, in Quick, dates can be searched ! We use this feature extensively. Say you have a field "TTASPV OF STTYOTIE DATE DD.MM.YYYY" : to review all entries from a certain year and month, you would enter (in Select mode) %##.04.2001 % = enter pattern matching mode ## = any two digits (ie., any day of the month) the rest is obvious. > > Woudn't it be nice if you could say 'Show me everything from 2001/01/01', or This can be done (but only for year 2001) with "%##.##.2001" depending, of course, from your date format. > 'bigger than 10000' ? In Quick pattern matching, this would be a nice feature. > > Bob, Conrad, something planned ? > > cu, Markus > regards, Esko From bengt.johansson@thoreb.se Thu, 17 May 2001 14:34:02 +0200 Date: Thu, 17 May 2001 14:34:02 +0200 From: Bengt Johansson bengt.johansson@thoreb.se Subject: Sv: Delays in connection to HP3000 via VT-MGR vs TELNET Hi again everybody ! I have tested the advices. ( Thank you for them ! ) First only ( in the UDC: ) QUICK;LINEREAD=YES, then both QUICK;LINEREAD=YES; and TERM=HP2392-X25; and also READ="LINE", ( from the beginning there was 'QUICK;LINEREAD=NO; and TERM=HP2392-192-B-ANY' ), but I am sorry to say that nothing helped. There are still the same delays. I have now addressed the suggested HP 3000-L listserver (hp3000-l@raven.utc.edu) with the same question. ( Thank you, John Pickering, for that tip. ) I will report back to you if they have a solution. Best regards Bengt Johansson "Ole Hansen, ScanConsult" skrev: > Hi Bengt, > > In addition to John Pickering's explanation of READ=CHAR vs. LINEREAD, I recall having similar problems, when we first connected remote sites over X.25-connections. At the advice of Cognos we changed our startup-command to: > > QUICK;LINEREAD=YES;TERM=HP2392-X25; etc. > > Now that we use Internet-borne connections instead, we still have TERM=HP2392-X25 (not because we did some clever thinking, but we didn't change anything, and had no problems when we switched). Maybe somebody at Cognos may want to comment on the significance of TERM in this context. > > Slightly off topic, we have noticed another peculiarity when running PowerHouse screens over 'long' connections (internet or satelite links). We have a few screens that run WHILE RETRIEVING statements and does a lot of spin-off processing, a.o. some RUN COMMAND statements, but with no display activity (i.e. no DISPLAY or INFORM verbs). These screens run 5-10 times slower over 'long' connections than when run over a local network. We have observed, that even if the screen is not changing its appearance, there is a lot of network traffic during the spin-off processing, i.e. a lot of packets are exchanged between QUICK on the HP3000 and Reflection on the PC - thus causing the slower response we think. > > Best regards > Ole Hansen > > ==================================================================== > E-mail: oh@scanconsult.dk Cell Phone: +45 40 40 11 50 > ==================================================================== > ScanConsult IT-Partners ApS Voice : +45 87 38 74 74 > Jegstrupvej 96A Fax : +45 87 38 74 75 > DK-8361 Hasselager Support : support@scanconsult.dk > Denmark Sales : sales@scanconsult.dk > http://www.scanconsult.dk > === ParaSuite : A Suite of Air Cargo Handling solutions === > === ParaSafe : A Suite of IT-Security solutions === > === We partner with Hewlett-Packard, Cognos, Lotus and others === > > -----Oprindelig meddelelse----- > Fra: Bengt Johansson > Til: 'Powerhouse List Group' > Dato: 16. maj 2001 19:10 > Emne: Delays in connection to HP3000 via VT-MGR vs TELNET > > >Hi everybody ! > > > >This question is not pure PowerHouse but it shows up when using > >PowerHouse. > > > >We have an HP3000 at our office in Gothenburg and our customers are > >spread out all over Sweden. Our customers mostly connects to the HP3000 > >using the Internet, but when there are problems with Internet or their > >local network they all have a lifeline using the old technique with > >direct telephone modem ( 19200 bps ), and that is always working. > > > >When using Internet our customers use either the VT-MGR or the TELNET > >protocoll in 'Connection Setup' in Reflection. > >Some customers never have problems ( and that is good ). > >But some customers almost always have problems ( and that is no good ), > >and so have I myself sometimes when connecting from my home i Halmstad, > >where I also have an office. > > > >When using TELNET there are sometimes delays for every character sent. I > >myself have seen delays up to 15 seconds but one customer have seen > >delays up to 40 seconds. > >This delay is from having pressed one button till the character pops up > >at the screen. > >Very annoying. > >When using TELNET every character ( plus 40 extra characters ), when > >pressing the button, are sent as a package from the user, to the HP3000 > >and then back to the user before the character pops up at the screen. > >This TELNET delay became a bit shorter = better ( but not good enough ) > >after having changed 'Retransmission Interval Lower Bound' from default > >4 seconds to 1 second in NETXPORT.GPROT.TCP in NMMGR. > > > >Now to the question: > >When using VT-MGR there is ALWAYS a delay in some PowerHouse Quick > >screens, especially one heavily used screen with a primary file and a > >detail file with many occurrancies and much data on the screen in Find > >Mode after having read one set of data from having pressed Return to get > >the next set of data till the new data pops up at the screen. > >>From home with my relatively slow Internet connection this delay is > >always 13 seconds with VT-MGR in this screen filled with data. > >Using TELNET this delay is only between 1 and 2 seconds, but as I said > >TELNET sometimes is too slow in another way. > >Inside our office in Gothenburg with a fast LAN ( 100 Mbit/sec ) the > >same screen gives a delay between 4 and 5 seconds with VT-MGR ( when the > >TELNET delay is under 1 second ), and that is too much isn't it ? And > >the 13 seconds mentioned above I can't stand. Would you ? > >When using common MPE commands like LISTF or other, there are no delays > >at all inside the fast LAN neither with VT-MGR nor TELNET, so why when > >using PowerHouse ? > > > >When using VT-MGR every pressed button is echoed to the screen locally > >and there is a package sent to the HP3000 only when Return is pressed, > >so this protocoll should be more efficient, but at least in PowerHouse > >it seems not to be so efficiant. > > > >Does anyone know a prescription to solve this problem ? > >Some of my customers hate these delays so much that they use the > >telephone modem instead of Internet, and have to pay the telephone cost > >for that, when Internet is free when you have a fixed connection. > >I have asked HP Support, but they have not given me any solution yet. > > > >Another queer question is why some other customers NEVER have problems > >using TELNET. > > > >We use MPE 6.0 and PowerHouse 8.29.D3. > > > >Best regards > >Bengt Johansson > >AB Thoreb > > From Roger.Kesterson@mtdsw.com Thu, 17 May 2001 09:30:41 -0700 Date: Thu, 17 May 2001 09:30:41 -0700 From: Kesterson, Roger Roger.Kesterson@mtdsw.com Subject: QTP file/record locking PH 7.10.G2, OpenVMS 7.1 I need to run QTP to update what could be thousands of records in a particular file during the day while other users are in the system in QUICK and QUIZ, and I don't want to prevent them from accessing the same file for read and write. They will not be going after any of the records that QTP will be updating. If I am understanding the manual, I should be able to accomplish this by using "SET LOCK RECORD UPDATE", and using an ALIAS on the OUTPUT. Is this correct? Thanks in advance for your advice. Roger Kesterson Senior Programming Analyst MTD Southwest, Inc. 550 N. 54th Street Chandler, AZ 85226 Phone: 480-961-1002 x4357 Fax: 480-940-5035 From Bob.Deskin@Cognos.COM Thu, 17 May 2001 12:48:55 -0400 Date: Thu, 17 May 2001 12:48:55 -0400 From: Deskin, Bob Bob.Deskin@Cognos.COM Subject: QTP file/record locking SET LOCK RECORD UPDATE will lock around the specific update. It will not lock the files during read. Because of the multiple locks, processing will be slower. If you're accessing and updating the same file, you do not need an ALIAS unless you're using the ADD option of OUTPUT. In fact, if you want to read and update the same record, you're best not to use an ALIAS. Bob Deskin PowerHouse Web Product Manager, Application Development Tools, Cognos Inc. bob.deskin@cognos.com (613) 738-1338 ext 7268 FAX: (613) 727-1178 3755 Riverside Drive P.O. Box 9707 Stn. T, Ottawa ON K1G 4K9 CANADA -----Original Message----- From: Kesterson, Roger [mailto:Roger.Kesterson@mtdsw.com] Sent: Thursday, May 17, 2001 12:31 PM To: PowerHouse List Server (E-mail) Subject: QTP file/record locking PH 7.10.G2, OpenVMS 7.1 I need to run QTP to update what could be thousands of records in a particular file during the day while other users are in the system in QUICK and QUIZ, and I don't want to prevent them from accessing the same file for read and write. They will not be going after any of the records that QTP will be updating. If I am understanding the manual, I should be able to accomplish this by using "SET LOCK RECORD UPDATE", and using an ALIAS on the OUTPUT. Is this correct? Thanks in advance for your advice. Roger Kesterson Senior Programming Analyst MTD Southwest, Inc. 550 N. 54th Street Chandler, AZ 85226 Phone: 480-961-1002 x4357 Fax: 480-940-5035 = = = = = = = = = = = = = = = = = = = = = = = = = = = = Mailing list: powerh-l@lists.swau.edu Subscribe: "subscribe" in message body to powerh-l-request@lists.swau.edu Unsubscribe: "unsubscribe" in message body to powerh-l-request@lists.swau.edu http://lists.swau.edu/mailman/listinfo/powerh-l This list is closed, thus to post to the list you must be a subscriber. From abahrs@denkor.com Thu, 17 May 2001 09:58:32 -0700 Date: Thu, 17 May 2001 09:58:32 -0700 From: Art Bahrs abahrs@denkor.com Subject: The Meaning of PowerHouse Web Error MEssages?? Hi All :) Any clue which file is not being opened? or what Message id 23 means??? I put the following into my browser and PowerHouse Web responds with: http://opus.denkor.com/cgi-bin/PHCGI?PH_QKC=TESTOPUS&PH_HTML=TESTOPUW&PH_APP =DEFAULT Message file open error, message id is 23. Art "Being Clueless yet again! hehe" Bahrs From Bob.Deskin@Cognos.COM Thu, 17 May 2001 13:31:14 -0400 Date: Thu, 17 May 2001 13:31:14 -0400 From: Deskin, Bob Bob.Deskin@Cognos.COM Subject: The Meaning of PowerHouse Web Error MEssages?? If I recall correctly, you have your PHCGI on NT. You should have the following files: phcgi.exe phcgi.dat phwebcgimsg.dll std-2.0.1-vc5.0-mt-noalloc.dll The third file is the message catalog. The 23rd message is a time-out message. Bob Deskin PowerHouse Web Product Manager, Application Development Tools, Cognos Inc. bob.deskin@cognos.com (613) 738-1338 ext 7268 FAX: (613) 727-1178 3755 Riverside Drive P.O. Box 9707 Stn. T, Ottawa ON K1G 4K9 CANADA -----Original Message----- From: Art Bahrs [mailto:abahrs@denkor.com] Sent: Thursday, May 17, 2001 12:59 PM To: Power House ListServ Cc: John Shields Subject: The Meaning of PowerHouse Web Error MEssages?? Hi All :) Any clue which file is not being opened? or what Message id 23 means??? I put the following into my browser and PowerHouse Web responds with: http://opus.denkor.com/cgi-bin/PHCGI?PH_QKC=TESTOPUS&PH_HTML=TESTOPUW&PH_APP =DEFAULT Message file open error, message id is 23. Art "Being Clueless yet again! hehe" Bahrs = = = = = = = = = = = = = = = = = = = = = = = = = = = = Mailing list: powerh-l@lists.swau.edu Subscribe: "subscribe" in message body to powerh-l-request@lists.swau.edu Unsubscribe: "unsubscribe" in message body to powerh-l-request@lists.swau.edu http://lists.swau.edu/mailman/listinfo/powerh-l This list is closed, thus to post to the list you must be a subscriber. From PICKERIJ@norbord.com Thu, 17 May 2001 13:41:02 -0400 Date: Thu, 17 May 2001 13:41:02 -0400 From: Pickering, John (NORBORD) PICKERIJ@norbord.com Subject: The Meaning of PowerHouse Web Error MEssages?? Wild a** guess time ... ... can your Quick screen find the qkmsg and/or qkmsgdes message files? By the way, qkmsge records 22 and 23 (is the first record 0 or 1?) are: 22 *d* The screen linkage parameters don't agree with local definitions. 23 *d* QDESIGN ^ and this version of QUICK are incompatible releases. > -----Original Message----- > From: Art Bahrs [SMTP:abahrs@denkor.com] > Sent: Thursday, May 17, 2001 12:59 PM > To: Power House ListServ > Cc: John Shields > Subject: The Meaning of PowerHouse Web Error MEssages?? > > Hi All :) > Any clue which file is not being opened? or what Message id 23 > means??? > > I put the following into my browser and PowerHouse Web responds with: > http://opus.denkor.com/cgi-bin/PHCGI?PH_QKC=TESTOPUS&PH_HTML=TESTOPUW&PH_A > PP > =DEFAULT > > Message file open error, message id is 23. > > Art "Being Clueless yet again! hehe" Bahrs > From abahrs@denkor.com Thu, 17 May 2001 11:42:50 -0700 Date: Thu, 17 May 2001 11:42:50 -0700 From: Art Bahrs abahrs@denkor.com Subject: PowerHouse Web on 3k - Group Rights? Hi All :) Ok... PowerHouseWeb generates a screen (file code 641) and HTML code for the screen.... and I put the HTML code file in .HTML.COGACCT and the screen in OBJST.COGACCT .... I seem to be having a File Open issue .... what rights do these groups need? and what rights does the account need? Thanks, Art "starting to look for the SFA's! hehe" Bahrs From Bob.Deskin@Cognos.COM Thu, 17 May 2001 15:46:16 -0400 Date: Thu, 17 May 2001 15:46:16 -0400 From: Deskin, Bob Bob.Deskin@Cognos.COM Subject: PowerHouse Web on 3k - Group Rights? The HTML Template and compiled screen files are MPE/iX files. They must be readable by the PowerHouse Web Server. Typically they'll have to be readable across accounts. You'll probably need lock access across accounts as well. Bob -----Original Message----- From: Art Bahrs [mailto:abahrs@denkor.com] Sent: Thursday, May 17, 2001 2:43 PM To: Power House ListServ Cc: John Shields Subject: PowerHouse Web on 3k - Group Rights? Hi All :) Ok... PowerHouseWeb generates a screen (file code 641) and HTML code for the screen.... and I put the HTML code file in .HTML.COGACCT and the screen in OBJST.COGACCT .... I seem to be having a File Open issue .... what rights do these groups need? and what rights does the account need? Thanks, Art "starting to look for the SFA's! hehe" Bahrs = = = = = = = = = = = = = = = = = = = = = = = = = = = = Mailing list: powerh-l@lists.swau.edu Subscribe: "subscribe" in message body to powerh-l-request@lists.swau.edu Unsubscribe: "unsubscribe" in message body to powerh-l-request@lists.swau.edu http://lists.swau.edu/mailman/listinfo/powerh-l This list is closed, thus to post to the list you must be a subscriber. From GGrimm@pcc-structurals.com Thu, 17 May 2001 14:35:52 -0700 Date: Thu, 17 May 2001 14:35:52 -0700 From: GGrimm@pcc-structurals.com GGrimm@pcc-structurals.com Subject: QTP rollback & commit I'm looking for a way to easily do a commit or a rollback in a dynamically run QTP based on a visual review of the run statistics for "reasonableness". Anybody have an idea how? I have seen QTP do a rollback when "badness" happens so know there is a way:=) I've tried a COMMIT at RUN and then done a Ctrl Y which sometimes causes an abort but nothing much happened. (PowerHouse 8.13.D3)/Oracle 8.0.5.0.0/HP-UX B.11.00 See my simple code: > access *TA_one link to TIME_ATTEND link to TIME_ATTEND_2 Access TIME_ATTEND viaindex TIME_ATTEND_EMPLOYEENO_1. Access TIME_ATTEND_2 viaindex TIME_ATTEND_2_PK. > output TIME_ATTEND delete > output TIME_ATTEND_2 delete > go Executing request 1 ... Records read: TA_ONE 1 TIME_ATTEND 134 TIME_ATTEND_2 134 Transactions processed: 134 Records processed: Added Updated Unchanged Deleted TIME_ATTEND 0 0 0 134 TIME_ATTEND_2 0 0 0 134 Finished. From Bob.Deskin@Cognos.COM Fri, 18 May 2001 08:51:19 -0400 Date: Fri, 18 May 2001 08:51:19 -0400 From: Deskin, Bob Bob.Deskin@Cognos.COM Subject: QTP rollback & commit I don't know of any way to commit conditionally, but I'm curious as to the requirement. I would expect that if the data is correct and the update logic is correct that the processing would be correct and the commit should happen. If it's otherwise, I'd suggest doing something with subfiles as output just to see the process works and if so, do the real thing. Bob Deskin PowerHouse Web Product Manager, Application Development Tools, Cognos Inc. bob.deskin@cognos.com (613) 738-1338 ext 7268 FAX: (613) 727-1178 3755 Riverside Drive P.O. Box 9707 Stn. T, Ottawa ON K1G 4K9 CANADA -----Original Message----- From: GGrimm@pcc-structurals.com [mailto:GGrimm@pcc-structurals.com] Sent: Thursday, May 17, 2001 5:36 PM To: powerh-l@lists.swau.edu Subject: QTP rollback & commit I'm looking for a way to easily do a commit or a rollback in a dynamically run QTP based on a visual review of the run statistics for "reasonableness". Anybody have an idea how? I have seen QTP do a rollback when "badness" happens so know there is a way:=) I've tried a COMMIT at RUN and then done a Ctrl Y which sometimes causes an abort but nothing much happened. (PowerHouse 8.13.D3)/Oracle 8.0.5.0.0/HP-UX B.11.00 See my simple code: > access *TA_one link to TIME_ATTEND link to TIME_ATTEND_2 Access TIME_ATTEND viaindex TIME_ATTEND_EMPLOYEENO_1. Access TIME_ATTEND_2 viaindex TIME_ATTEND_2_PK. > output TIME_ATTEND delete > output TIME_ATTEND_2 delete > go Executing request 1 ... Records read: TA_ONE 1 TIME_ATTEND 134 TIME_ATTEND_2 134 Transactions processed: 134 Records processed: Added Updated Unchanged Deleted TIME_ATTEND 0 0 0 134 TIME_ATTEND_2 0 0 0 134 Finished. = = = = = = = = = = = = = = = = = = = = = = = = = = = = Mailing list: powerh-l@lists.swau.edu Subscribe: "subscribe" in message body to powerh-l-request@lists.swau.edu Unsubscribe: "unsubscribe" in message body to powerh-l-request@lists.swau.edu http://lists.swau.edu/mailman/listinfo/powerh-l This list is closed, thus to post to the list you must be a subscriber. From simopoulos@scoplife.gr Fri, 18 May 2001 16:22:10 +0300 Date: Fri, 18 May 2001 16:22:10 +0300 From: SD simopoulos@scoplife.gr Subject: QTP rollback & commit This is a multi-part message in MIME format. ------=_NextPart_000_0014_01C0DFB6.B09A0660 Content-Type: text/plain; charset="iso-8859-7" Content-Transfer-Encoding: quoted-printable In Mpe/Ix the comand is "SET SCRATCH" I don't know if that commans is available for Oracle. Simopoulos Dimitris Programmer Scoplife AEAEZ ------=_NextPart_000_0014_01C0DFB6.B09A0660 Content-Type: text/html; charset="iso-8859-7" Content-Transfer-Encoding: quoted-printable
In Mpe/Ix  the comand is "SET=20 SCRATCH"
I don't know if that commans is = available for=20 Oracle.
 
Simopoulos Dimitris
Programmer
Scoplife AEAEZ
 
------=_NextPart_000_0014_01C0DFB6.B09A0660-- From Chris.Sharman@ccagroup.co.uk Fri, 18 May 2001 14:55:48 +0100 Date: Fri, 18 May 2001 14:55:48 +0100 From: Chris Sharman Chris.Sharman@ccagroup.co.uk Subject: VMS 7.3 No Query Record Locking VMS, so those disinterested please skip now. VMS V7.3 (shipping) includes various RMS performance enhancements, which I believe could give us significant performance improvements. One interesting one is 'No query locking' which can "be enabled externally, potentially without any application change". To be specific: if the application (eg Quiz) uses both of the existing RMS flags RAB$V_NLK (no lock) and RAB$V_RRL (read regardless of lock), it will now be possible to avoid record locking altogether, on either a per-process or system-wide basis. Earlier versions of VMS probe the lock in order to appropriately return either normal success or 'OK_RRL' (or OK_RLK). This lock probe carries overhead comparable to locking. This overhead may be avoided under VMS 7.3 (and it won't know to return OK_RRL or OK_RLK). My question therefore is: Do Quiz (& Quick & QTP for readonly files) use either or both of the NLK & RRL flags ? As far as I can tell, 7.10G3 Quiz just uses read-ahead (& write behind, which is meaningless). Is 8.30 (due July??) the same ? If not, could a patch be released ? (ideally back to PH 7.10 or earlier, of course). Or perhaps select one or other behaviour via a logical, program parameter, or sitehook. Setting the new NQL bits would obviously be incompatible with VMS 7.2 & earlier, and is therefore a non-starter, but setting NLK+RRL which have been there forever, afaik, would create no compatibility issues. My Quiz reports presently spend a lot of time in kernel mode, which I can only put down to locking, and which would therefore be eliminated by this. Regards, Chris _______________________________________________________________________ Chris.Sharman@CCAgroup.co.uk http://www.ccagroup.co.uk/ CCA Stationery Ltd, Eastway, Fulwood, Preston, Lancashire, PR2 9WS. ----------------------------------------------------------------------- Any views expressed in this message are those of the sender and not necessarily those of CCA Group. The unauthorized use, disclosure, copying or alteration of this message is forbidden. The contents of this message may be confidential and/or privileged, copyright CCA Group and are intended solely for the use of the individual or entity to whom they are addressed. Whilst this message has been scanned, CCA Group cannot guarantee that it is virus free or compatible with your systems and accepts no responsibility for any loss or damage arising from its use. The recipient is advised to run their own anti-virus software. If you receive this message in error please contact postmaster@ccagroup.co.uk immediately, destroy any copies and delete it from your computer systems. From Matt.Ohmes@COGNOS.com Fri, 18 May 2001 10:07:10 -0400 Date: Fri, 18 May 2001 10:07:10 -0400 From: Ohmes, Matt Matt.Ohmes@COGNOS.com Subject: QTP rollback & commit QTP is essentially a batch process without much allowance for "visual review" before proceeding. Once it goes, it goes. To do what you want, I would go with Bob's suggestion on using Subfiles. In a pre-pass, create a subfile of all the transactions you want to process. You could do this with QTP (my personal favorite for creating subfiles) but it might make more sense to use Quiz. Regardless, I'd produce a report in Quiz for the subfile to let the users "visually review" it. Then, if it passes their review, have the users fire off the QTP pass that picks up the subfile and use that subfile to drive the process of deleting (updating, whatever) the processing. Matt Matt.Ohmes@Cognos.Com Cognos Corporation 909 E. Las Colinas Blvd. Suite 1900 Irving, TX 75039 214-259-6200 "Matt doesn't really know anything. He just likes to pontificate a lot. We refuse to acknowledge that he works for Cognos or that we have ever met him or anyone with whom he's ever been associated. Don't lend him money and don't let him talk to your sister!" ;-) -----Original Message----- From: Deskin, Bob [mailto:Bob.Deskin@cognos.com] Sent: Friday, May 18, 2001 7:51 AM To: powerh-l@lists.swau.edu Subject: RE: QTP rollback & commit I don't know of any way to commit conditionally, but I'm curious as to the requirement. I would expect that if the data is correct and the update logic is correct that the processing would be correct and the commit should happen. If it's otherwise, I'd suggest doing something with subfiles as output just to see the process works and if so, do the real thing. Bob Deskin PowerHouse Web Product Manager, Application Development Tools, Cognos Inc. bob.deskin@cognos.com (613) 738-1338 ext 7268 FAX: (613) 727-1178 3755 Riverside Drive P.O. Box 9707 Stn. T, Ottawa ON K1G 4K9 CANADA -----Original Message----- From: GGrimm@pcc-structurals.com [mailto:GGrimm@pcc-structurals.com] Sent: Thursday, May 17, 2001 5:36 PM To: powerh-l@lists.swau.edu Subject: QTP rollback & commit I'm looking for a way to easily do a commit or a rollback in a dynamically run QTP based on a visual review of the run statistics for "reasonableness". Anybody have an idea how? I have seen QTP do a rollback when "badness" happens so know there is a way:=) I've tried a COMMIT at RUN and then done a Ctrl Y which sometimes causes an abort but nothing much happened. (PowerHouse 8.13.D3)/Oracle 8.0.5.0.0/HP-UX B.11.00 See my simple code: > access *TA_one link to TIME_ATTEND link to TIME_ATTEND_2 Access TIME_ATTEND viaindex TIME_ATTEND_EMPLOYEENO_1. Access TIME_ATTEND_2 viaindex TIME_ATTEND_2_PK. > output TIME_ATTEND delete > output TIME_ATTEND_2 delete > go Executing request 1 ... Records read: TA_ONE 1 TIME_ATTEND 134 TIME_ATTEND_2 134 Transactions processed: 134 Records processed: Added Updated Unchanged Deleted TIME_ATTEND 0 0 0 134 TIME_ATTEND_2 0 0 0 134 Finished. From Bill.Michael@ipaper.com Fri, 18 May 2001 09:39:40 -0500 Date: Fri, 18 May 2001 09:39:40 -0500 From: Bill D Michael Bill.Michael@ipaper.com Subject: QTP rollback & commit This thread is of interest because we're hitting the same type of problems... QTP 8.20D6, Oracle 8, on VMS. We may have a batch process with two or three multi-request QTPs, followed by some Quizes; we'll get an Oracle "attach failure" in the middle of one of the QTPs that terminates the request or run. Anything running after this is getting garbage, or nothing; we can work around that part easily enough (though it will require code & DCL changes), but since the _first_ part of the QTP(s) were successful, our data is now in an "in between" state. With 7.10 and RMS files, failures "mid run" were exceedingly rare (a CPU crash or running out of disk space) - now we're getting failures several times a week, and we're spending all our time analyzing and fixing data. One thought we've had - subfile off the tables that are _about_ to be changed by the process. Check for overall success of the process at the end (preferably programmatically). If there was a failure, truncate the table and reload it from the subfile, then try the process again. Not ideal, by far, but it's got to be better than the current "randomly works or not" situation! Doing a "subfile first, then if it's good, the real data" won't work, because the subfile might work fine, then the real run still fail; we need to validate the 'real run'. At least we've gotten past the "a failure locks up all PH users until fixed" problem (subdict=search) with workarounds! Bill "Ohmes, Matt" @cube.swau.edu on 05/18/2001 09:07:10 AM Sent by: powerh-l-admin@cube.swau.edu To: powerh-l@lists.swau.edu cc: Subject: RE: QTP rollback & commit QTP is essentially a batch process without much allowance for "visual review" before proceeding. Once it goes, it goes. To do what you want, I would go with Bob's suggestion on using Subfiles. In a pre-pass, create a subfile of all the transactions you want to process. You could do this with QTP (my personal favorite for creating subfiles) but it might make more sense to use Quiz. Regardless, I'd produce a report in Quiz for the subfile to let the users "visually review" it. Then, if it passes their review, have the users fire off the QTP pass that picks up the subfile and use that subfile to drive the process of deleting (updating, whatever) the processing. Matt Matt.Ohmes@Cognos.Com Cognos Corporation 909 E. Las Colinas Blvd. Suite 1900 Irving, TX 75039 214-259-6200 "Matt doesn't really know anything. He just likes to pontificate a lot. We refuse to acknowledge that he works for Cognos or that we have ever met him or anyone with whom he's ever been associated. Don't lend him money and don't let him talk to your sister!" ;-) -----Original Message----- From: Deskin, Bob [mailto:Bob.Deskin@cognos.com] Sent: Friday, May 18, 2001 7:51 AM To: powerh-l@lists.swau.edu Subject: RE: QTP rollback & commit I don't know of any way to commit conditionally, but I'm curious as to the requirement. I would expect that if the data is correct and the update logic is correct that the processing would be correct and the commit should happen. If it's otherwise, I'd suggest doing something with subfiles as output just to see the process works and if so, do the real thing. Bob Deskin PowerHouse Web Product Manager, Application Development Tools, Cognos Inc. bob.deskin@cognos.com (613) 738-1338 ext 7268 FAX: (613) 727-1178 3755 Riverside Drive P.O. Box 9707 Stn. T, Ottawa ON K1G 4K9 CANADA -----Original Message----- From: GGrimm@pcc-structurals.com [mailto:GGrimm@pcc-structurals.com] Sent: Thursday, May 17, 2001 5:36 PM To: powerh-l@lists.swau.edu Subject: QTP rollback & commit I'm looking for a way to easily do a commit or a rollback in a dynamically run QTP based on a visual review of the run statistics for "reasonableness". Anybody have an idea how? I have seen QTP do a rollback when "badness" happens so know there is a way:=) I've tried a COMMIT at RUN and then done a Ctrl Y which sometimes causes an abort but nothing much happened. (PowerHouse 8.13.D3)/Oracle 8.0.5.0.0/HP-UX B.11.00 See my simple code: > access *TA_one link to TIME_ATTEND link to TIME_ATTEND_2 Access TIME_ATTEND viaindex TIME_ATTEND_EMPLOYEENO_1. Access TIME_ATTEND_2 viaindex TIME_ATTEND_2_PK. > output TIME_ATTEND delete > output TIME_ATTEND_2 delete > go Executing request 1 ... Records read: TA_ONE 1 TIME_ATTEND 134 TIME_ATTEND_2 134 Transactions processed: 134 Records processed: Added Updated Unchanged Deleted TIME_ATTEND 0 0 0 134 TIME_ATTEND_2 0 0 0 134 Finished. = = = = = = = = = = = = = = = = = = = = = = = = = = = = Mailing list: powerh-l@lists.swau.edu Subscribe: "subscribe" in message body to powerh-l-request@lists.swau.edu Unsubscribe: "unsubscribe" in message body to powerh-l-request@lists.swau.edu http://lists.swau.edu/mailman/listinfo/powerh-l This list is closed, thus to post to the list you must be a subscriber. From GGrimm@pcc-structurals.com Fri, 18 May 2001 07:45:16 -0700 Date: Fri, 18 May 2001 07:45:16 -0700 From: GGrimm@pcc-structurals.com GGrimm@pcc-structurals.com Subject: Re[2]: QTP rollback & commit Thanks for the reply, Bob, Charlie, and others. The requirement btw, is to do some "corrective" data administration on the fly. We have used the subfile technique before and I just got to wondering if there was some other method that was being overlooked. It seems not (knot?) ;=) ______________________________ Reply Separator _________________________________ Subject: RE: QTP rollback & commit Author: "Deskin Bob" at RDM-PO1 Date: 05/18/2001 8:51 AM I don't know of any way to commit conditionally, but I'm curious as to the requirement. I would expect that if the data is correct and the update logic is correct that the processing would be correct and the commit should happen. If it's otherwise, I'd suggest doing something with subfiles as output just to see the process works and if so, do the real thing. Bob Deskin PowerHouse Web Product Manager, Application Development Tools, Cognos Inc. bob.deskin@cognos.com (613) 738-1338 ext 7268 FAX: (613) 727-1178 3755 Riverside Drive P.O. Box 9707 Stn. T, Ottawa ON K1G 4K9 CANADA -----Original Message----- From: GGrimm@pcc-structurals.com [mailto:GGrimm@pcc-structurals.com] Sent: Thursday, May 17, 2001 5:36 PM To: powerh-l@lists.swau.edu Subject: QTP rollback & commit I'm looking for a way to easily do a commit or a rollback in a dynamically run QTP based on a visual review of the run statistics for "reasonableness". Anybody have an idea how? I have seen QTP do a rollback when "badness" happens so know there is a way:=) I've tried a COMMIT at RUN and then done a Ctrl Y which sometimes causes an abort but nothing much happened. (PowerHouse 8.13.D3)/Oracle 8.0.5.0.0/HP-UX B.11.00 See my simple code: > access *TA_one link to TIME_ATTEND link to TIME_ATTEND_2 Access TIME_ATTEND viaindex TIME_ATTEND_EMPLOYEENO_1. Access TIME_ATTEND_2 viaindex TIME_ATTEND_2_PK. > output TIME_ATTEND delete > output TIME_ATTEND_2 delete > go Executing request 1 ... Records read: TA_ONE 1 TIME_ATTEND 134 TIME_ATTEND_2 134 Transactions processed: 134 Records processed: Added Updated Unchanged Deleted TIME_ATTEND 0 0 0 134 TIME_ATTEND_2 0 0 0 134 Finished. From Matt.Ohmes@COGNOS.com Fri, 18 May 2001 10:46:24 -0400 Date: Fri, 18 May 2001 10:46:24 -0400 From: Ohmes, Matt Matt.Ohmes@COGNOS.com Subject: QTP rollback & commit Hi Bill, Your are right, of course, when you say that creating a "transaction subfile" first won't solve the problem of the "real" run failing. The real run can fail for a number of reasons (as you've discovered), which is why you should definitely have backups of your tables, either using DB utilities or subfiles. But that doesn't negate the value of a transaction subfile. I would very often create a transaction subfile even when I didn't need a pre-update review of the data. In addition to the previously stated reason, it also served as the basis for a "what DID I process" report. While QTP's statistics are fine, they obviously don't list any details of exactly WHICH rows were processed. I've always been a big fan of subfiles. :-) Matt -----Original Message----- From: Bill D Michael [mailto:Bill.Michael@ipaper.com] Sent: Friday, May 18, 2001 9:40 AM To: powerh-l@lists.swau.edu Subject: RE: QTP rollback & commit This thread is of interest because we're hitting the same type of problems... QTP 8.20D6, Oracle 8, on VMS. We may have a batch process with two or three multi-request QTPs, followed by some Quizes; we'll get an Oracle "attach failure" in the middle of one of the QTPs that terminates the request or run. Anything running after this is getting garbage, or nothing; we can work around that part easily enough (though it will require code & DCL changes), but since the _first_ part of the QTP(s) were successful, our data is now in an "in between" state. With 7.10 and RMS files, failures "mid run" were exceedingly rare (a CPU crash or running out of disk space) - now we're getting failures several times a week, and we're spending all our time analyzing and fixing data. One thought we've had - subfile off the tables that are _about_ to be changed by the process. Check for overall success of the process at the end (preferably programmatically). If there was a failure, truncate the table and reload it from the subfile, then try the process again. Not ideal, by far, but it's got to be better than the current "randomly works or not" situation! Doing a "subfile first, then if it's good, the real data" won't work, because the subfile might work fine, then the real run still fail; we need to validate the 'real run'. At least we've gotten past the "a failure locks up all PH users until fixed" problem (subdict=search) with workarounds! Bill From Matt.Ohmes@COGNOS.com Fri, 18 May 2001 10:47:54 -0400 Date: Fri, 18 May 2001 10:47:54 -0400 From: Ohmes, Matt Matt.Ohmes@COGNOS.com Subject: Re[2]: QTP rollback & commit There may be some other approach, but if you're performing corrective data administration, the subfile approach is probably the best approach by far. :-) Matt -----Original Message----- From: GGrimm@pcc-structurals.com [mailto:GGrimm@pcc-structurals.com] Sent: Friday, May 18, 2001 9:45 AM To: powerh-l@lists.swau.edu Subject: Re[2]: QTP rollback & commit Thanks for the reply, Bob, Charlie, and others. The requirement btw, is to do some "corrective" data administration on the fly. We have used the subfile technique before and I just got to wondering if there was some other method that was being overlooked. It seems not (knot?) ;=) From bob.edis@fleetpride.com Fri, 18 May 2001 09:47:06 -0500 Date: Fri, 18 May 2001 09:47:06 -0500 From: Edis, Bob bob.edis@fleetpride.com Subject: QTP rollback & commit If you are using the default COMMIT AT RUN you could test the QTP result in DCL then branch to a rollback routine on failure. The primary key of the records previously written would need to be kept in a subfile in order to do this or alternatively a temp Oracle table, e.g. like an audit file. (please excuse poor syntax - it's been 1.5 years since I worked on VMS) $QTP AUTO=run1 does some update to the Oracle table and writes keys to audit file or subfile EXIT $! $QTP AUTO=run2 EXIT $ test $status value and if failure GOTO run1_rb $ ... $GOTO end $run1_rb: $QTP AUTO=run1_rollback rollback logic using audit file/subfile as driver EXIT $! $end: $EXIT Blue -----Original Message----- From: Bill D Michael [mailto:Bill.Michael@ipaper.com] Sent: Friday, May 18, 2001 9:40 AM To: powerh-l@lists.swau.edu Subject: RE: QTP rollback & commit This thread is of interest because we're hitting the same type of problems... QTP 8.20D6, Oracle 8, on VMS. We may have a batch process with two or three multi-request QTPs, followed by some Quizes; we'll get an Oracle "attach failure" in the middle of one of the QTPs that terminates the request or run. Anything running after this is getting garbage, or nothing; we can work around that part easily enough (though it will require code & DCL changes), but since the _first_ part of the QTP(s) were successful, our data is now in an "in between" state. With 7.10 and RMS files, failures "mid run" were exceedingly rare (a CPU crash or running out of disk space) - now we're getting failures several times a week, and we're spending all our time analyzing and fixing data. One thought we've had - subfile off the tables that are _about_ to be changed by the process. Check for overall success of the process at the end (preferably programmatically). If there was a failure, truncate the table and reload it from the subfile, then try the process again. Not ideal, by far, but it's got to be better than the current "randomly works or not" situation! Doing a "subfile first, then if it's good, the real data" won't work, because the subfile might work fine, then the real run still fail; we need to validate the 'real run'. At least we've gotten past the "a failure locks up all PH users until fixed" problem (subdict=search) with workarounds! Bill "Ohmes, Matt" @cube.swau.edu on 05/18/2001 09:07:10 AM Sent by: powerh-l-admin@cube.swau.edu To: powerh-l@lists.swau.edu cc: Subject: RE: QTP rollback & commit QTP is essentially a batch process without much allowance for "visual review" before proceeding. Once it goes, it goes. To do what you want, I would go with Bob's suggestion on using Subfiles. In a pre-pass, create a subfile of all the transactions you want to process. You could do this with QTP (my personal favorite for creating subfiles) but it might make more sense to use Quiz. Regardless, I'd produce a report in Quiz for the subfile to let the users "visually review" it. Then, if it passes their review, have the users fire off the QTP pass that picks up the subfile and use that subfile to drive the process of deleting (updating, whatever) the processing. Matt Matt.Ohmes@Cognos.Com Cognos Corporation 909 E. Las Colinas Blvd. Suite 1900 Irving, TX 75039 214-259-6200 "Matt doesn't really know anything. He just likes to pontificate a lot. We refuse to acknowledge that he works for Cognos or that we have ever met him or anyone with whom he's ever been associated. Don't lend him money and don't let him talk to your sister!" ;-) -----Original Message----- From: Deskin, Bob [mailto:Bob.Deskin@cognos.com] Sent: Friday, May 18, 2001 7:51 AM To: powerh-l@lists.swau.edu Subject: RE: QTP rollback & commit I don't know of any way to commit conditionally, but I'm curious as to the requirement. I would expect that if the data is correct and the update logic is correct that the processing would be correct and the commit should happen. If it's otherwise, I'd suggest doing something with subfiles as output just to see the process works and if so, do the real thing. Bob Deskin PowerHouse Web Product Manager, Application Development Tools, Cognos Inc. bob.deskin@cognos.com (613) 738-1338 ext 7268 FAX: (613) 727-1178 3755 Riverside Drive P.O. Box 9707 Stn. T, Ottawa ON K1G 4K9 CANADA -----Original Message----- From: GGrimm@pcc-structurals.com [mailto:GGrimm@pcc-structurals.com] Sent: Thursday, May 17, 2001 5:36 PM To: powerh-l@lists.swau.edu Subject: QTP rollback & commit I'm looking for a way to easily do a commit or a rollback in a dynamically run QTP based on a visual review of the run statistics for "reasonableness". Anybody have an idea how? I have seen QTP do a rollback when "badness" happens so know there is a way:=) I've tried a COMMIT at RUN and then done a Ctrl Y which sometimes causes an abort but nothing much happened. (PowerHouse 8.13.D3)/Oracle 8.0.5.0.0/HP-UX B.11.00 See my simple code: > access *TA_one link to TIME_ATTEND link to TIME_ATTEND_2 Access TIME_ATTEND viaindex TIME_ATTEND_EMPLOYEENO_1. Access TIME_ATTEND_2 viaindex TIME_ATTEND_2_PK. > output TIME_ATTEND delete > output TIME_ATTEND_2 delete > go Executing request 1 ... Records read: TA_ONE 1 TIME_ATTEND 134 TIME_ATTEND_2 134 Transactions processed: 134 Records processed: Added Updated Unchanged Deleted TIME_ATTEND 0 0 0 134 TIME_ATTEND_2 0 0 0 134 Finished. = = = = = = = = = = = = = = = = = = = = = = = = = = = = Mailing list: powerh-l@lists.swau.edu Subscribe: "subscribe" in message body to powerh-l-request@lists.swau.edu Unsubscribe: "unsubscribe" in message body to powerh-l-request@lists.swau.edu http://lists.swau.edu/mailman/listinfo/powerh-l This list is closed, thus to post to the list you must be a subscriber. = = = = = = = = = = = = = = = = = = = = = = = = = = = = Mailing list: powerh-l@lists.swau.edu Subscribe: "subscribe" in message body to powerh-l-request@lists.swau.edu Unsubscribe: "unsubscribe" in message body to powerh-l-request@lists.swau.edu http://lists.swau.edu/mailman/listinfo/powerh-l This list is closed, thus to post to the list you must be a subscriber. From Jari.S.Kaljunen@telia.se Fri, 18 May 2001 17:08:00 +0200 Date: Fri, 18 May 2001 17:08:00 +0200 From: Jari.S.Kaljunen@telia.se Jari.S.Kaljunen@telia.se Subject: SV: QTP rollback & commit Try this : first make it krach at final source test.qts: acc frontline choose flkod "JAR" def a char*1 = parm prompt "shall we ?" def b int*1 = nconv(a) if a = "N" output frontline update item bkod final "AJK" sub e at final inc b go acc frontline choose flkod "JAR" def a char*1 = parm prompt "shall we ?" def b int*1 = nconv(a) if a = "N" output frontline update item bkod final "AJK" sub e at final inc b go results: Q T P (7.10.G) Copyright 1998 COGNOS INC. (ALPHA) Licensed PH-AXP-RUNTIME for customer: 8200290001 Sergel Kredttjanster AB axp QTP> use test QTP> acc frontline QTP> QTP> choose flkod "JAR" Choose viaindex FRONTLINE_FLKOD_IX. QTP> QTP> def a char*1 = parm prompt "shall we ?" QTP> QTP> def b int*1 = nconv(a) if a = "N" QTP> QTP> output frontline update QTP> item bkod final "AJK" QTP> QTP> sub e at final inc b QTP> QTP> go Executing request 1 ... shall we ?N ------------------------------------------------------------------------ ------- Data conversion error. [1] Item: B File: E Action Taken: Run terminated. ------------------------------------------------------------------------ ------- Records read: FRONTLINE 1 Transactions processed: 1 Records processed: Added Updated Unchanged Deleted FRONTLINE 0 1 0 0 E 0 0 0 0 Finished. ------------------------------------------------------------------------ ------- Changes made since the last commit have been rolled back to a stable state. Therefore, statistics reported may be incorrect. ------------------------------------------------------------------------ ------- QTP> use test QTP> acc frontline QTP> QTP> choose flkod "JAR" Choose viaindex FRONTLINE_FLKOD_IX. QTP> QTP> def a char*1 = parm prompt "shall we ?" QTP> QTP> def b int*1 = nconv(a) if a = "N" QTP> QTP> output frontline update QTP> item bkod final "AJK" QTP> QTP> sub e at final inc b A temporary subfile with the same name already exists. Ok to delete?y QTP> QTP> go Executing request 1 ... shall we ?J Records read: FRONTLINE 1 Transactions processed: 1 Records processed: Added Updated Unchanged Deleted FRONTLINE 0 1 0 0 E 1 0 0 0 Finished. QTP> -----Ursprungligt meddelande----- Frĺn: GGrimm@pcc-structurals.com [SMTP:GGrimm@pcc-structurals.com] Skickat: den 17 maj 2001 23:36 Till: powerh-l@lists.swau.edu Ämne: QTP rollback & commit I'm looking for a way to easily do a commit or a rollback in a dynamically run QTP based on a visual review of the run statistics for "reasonableness". Anybody have an idea how? I have seen QTP do a rollback when "badness" happens so know there is a way:=) I've tried a COMMIT at RUN and then done a Ctrl Y which sometimes causes an abort but nothing much happened. (PowerHouse 8.13.D3)/Oracle 8.0.5.0.0/HP-UX B.11.00 See my simple code: > access *TA_one link to TIME_ATTEND link to TIME_ATTEND_2 Access TIME_ATTEND viaindex TIME_ATTEND_EMPLOYEENO_1. Access TIME_ATTEND_2 viaindex TIME_ATTEND_2_PK. > output TIME_ATTEND delete > output TIME_ATTEND_2 delete > go Executing request 1 ... Records read: TA_ONE 1 TIME_ATTEND 134 TIME_ATTEND_2 134 Transactions processed: 134 Records processed: Added Updated Unchanged Deleted TIME_ATTEND 0 0 0 134 TIME_ATTEND_2 0 0 0 134 Finished. = = = = = = = = = = = = = = = = = = = = = = = = = = = = Mailing list: powerh-l@lists.swau.edu Subscribe: "subscribe" in message body to powerh-l-request@lists.swau.edu Unsubscribe: "unsubscribe" in message body to powerh-l-request@lists.swau.edu http://lists.swau.edu/mailman/listinfo/powerh-l This list is closed, thus to post to the list you must be a subscriber. <> From PICKERIJ@norbord.com Fri, 18 May 2001 11:39:19 -0400 Date: Fri, 18 May 2001 11:39:19 -0400 From: Pickering, John (NORBORD) PICKERIJ@norbord.com Subject: was QTP rollback & commit > From: Ohmes, Matt Very nice to see Matt Oh-not-zero posting his usually excellent suggestions to the list again. It's been quite a while and we've missed you. Come back to ADT and leave that BI stuff alone :-) Regards, JWP From Matt.Ohmes@COGNOS.com Fri, 18 May 2001 11:53:00 -0400 Date: Fri, 18 May 2001 11:53:00 -0400 From: Ohmes, Matt Matt.Ohmes@COGNOS.com Subject: was QTP rollback & commit Oh, I'll never really leave ADT. :-) But the BI guys need some help on building those Data Warehouses. ;-) -----Original Message----- From: Pickering, John (NORBORD) [mailto:PICKERIJ@norbord.com] Sent: Friday, May 18, 2001 10:39 AM To: 'Ohmes, Matt'; 'Powerhouse List' Subject: was QTP rollback & commit > From: Ohmes, Matt Very nice to see Matt Oh-not-zero posting his usually excellent suggestions to the list again. It's been quite a while and we've missed you. Come back to ADT and leave that BI stuff alone :-) Regards, JWP = = = = = = = = = = = = = = = = = = = = = = = = = = = = Mailing list: powerh-l@lists.swau.edu Subscribe: "subscribe" in message body to powerh-l-request@lists.swau.edu Unsubscribe: "unsubscribe" in message body to powerh-l-request@lists.swau.edu http://lists.swau.edu/mailman/listinfo/powerh-l This list is closed, thus to post to the list you must be a subscriber. From Bob.Deskin@Cognos.COM Fri, 18 May 2001 12:57:19 -0400 Date: Fri, 18 May 2001 12:57:19 -0400 From: Deskin, Bob Bob.Deskin@Cognos.COM Subject: Quick enhancement request Suggestion received. Bob Deskin PowerHouse Web Product Manager, Application Development Tools, Cognos Inc. bob.deskin@cognos.com (613) 738-1338 ext 7268 FAX: (613) 727-1178 3755 Riverside Drive P.O. Box 9707 Stn. T, Ottawa ON K1G 4K9 CANADA -----Original Message----- From: Pickering, John (NORBORD) [mailto:PICKERIJ@norbord.com] Sent: Friday, May 18, 2001 12:52 PM To: 'Powerhouse List' Cc: 'Deskin, Bob' Subject: Quick enhancement request How about adding a PREAPPEND procedure? I want to do some security checking before I allow additions to a table. When the screen has only an ENTRY procedure then the PREENTRY procedure seems the correct place to do this. But when the screen has a CLUSTER and therefore an APPEND procedure the user can bypass the PREENTRY procedure by simply using Append. And while you're in the code you might as well add PREDELETE for just the same kinds of reasons. I know I could modify the generated procedures but, since Matt has renewed his interest in this list, I must resist the temptation :-) Regards, JWP From PICKERIJ@norbord.com Fri, 18 May 2001 12:51:39 -0400 Date: Fri, 18 May 2001 12:51:39 -0400 From: Pickering, John (NORBORD) PICKERIJ@norbord.com Subject: Quick enhancement request How about adding a PREAPPEND procedure? I want to do some security checking before I allow additions to a table. When the screen has only an ENTRY procedure then the PREENTRY procedure seems the correct place to do this. But when the screen has a CLUSTER and therefore an APPEND procedure the user can bypass the PREENTRY procedure by simply using Append. And while you're in the code you might as well add PREDELETE for just the same kinds of reasons. I know I could modify the generated procedures but, since Matt has renewed his interest in this list, I must resist the temptation :-) Regards, JWP From steve_franklin@yahoo.com Fri, 18 May 2001 15:23:08 -0500 Date: Fri, 18 May 2001 15:23:08 -0500 From: Steve Franklin steve_franklin@yahoo.com Subject: Dumb Axiant Question I have a radio button that I would like the options to display horizontal rather than vertical. What option do I use to accomplish this? I could not find it anywhere within the help system. Steve Franklin Stw Inc. Email: steve_franklin@yahoo.com _________________________________________________________ Do You Yahoo!? Get your free @yahoo.com address at http://mail.yahoo.com From robert.mills@windsong-services.co.uk Mon, 21 May 2001 14:15:10 +0100 Date: Mon, 21 May 2001 14:15:10 +0100 From: Robert Mills robert.mills@windsong-services.co.uk Subject: MPE/Reflection script fails when run in PowerHouse Quick Hi fellow list members, I have a script file (RFT, appended below) that transfers file from an HPe3000/979-200 (MPE/iX 6.5) to a PC (Windows 98SE) using Reflection for HP (Version 7.0+). RFT does some validation and then sends (echo's) an RBS script to Reflection. It has been working fine, when invoked at the colon (:) prompt, without any problems. We then had the requirement for a PowerHouse Quick (8.19c5) screen to be able to copy a file from the HP to a PC. So instead of writing something new we added a 'run command' to the screen to invoke RFT with the needed parameters. At the time no problem was anticipated. All attempts to make it work have failed. The RBS code is invoking the error handler when it is executing the WRQReceiveFile method. It is displaying the error '10032 -- No response from host transfer program'. I did some checking and the only difference in the environments at the time WRQReceiveFile was invoked involved that HP type-ahead. At the colon prompt it was disabled, and in Quick it was enabled. I added code to into the script to handle this but it still made no change. Any ideas? regards, +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ |Robert W.Mills (Systems Development Manager) | |Windsong Services, St. Mary Cray, ENGLAND | |Tel : +44 (0)1689 870622 x3005 | |Fax : +44 (0)1689 899026 | |Email: Robert.Mills@windsong-services.co.uk | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ ++++ start of script. parm hpfile='?', pcfile='', folder='', drive='', ftm='ASCII', exists='ASK' if lft('!hpfile',1) = '?' then setvar s chr(14) setvar b chr(14)+':'+chr(15) echo ![s+'R;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;'] echo ![b+' RFT - HPe3000 to PC File Transfer via Reflection.'] echo ![s+'5;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;'] echo ![b+''] echo ![b+' Syntax:'] echo ![b+' RFT hpfile[, pcfile, folder, drive, ftm, exists]'] echo ![b+''] echo ![b+' Parameters:'] echo ![b+' hpfile Name of file on HPe3000 to be transfered.'] echo ![b+' pcfile Name of file on PC (default: hpfile)'] echo ![b+' folder Destination folder on the PC.'] echo ![b+' Note: Surround in double-quotes if it contains spaces.'] echo ![b+' drive Destination drive on the PC (default: H)'] echo ![b+' ftm File Transfer Method to be used.'] echo ![b+' Options are: ASCII (default), BINARY and LABELS'] echo ![b+' exists Specifies what to do if file already exists on the PC.'] echo ![b+' Options are: ASK (default) and DELETE'] echo ![b+''] echo ![b+' Restrictions:'] echo ![b+' WRQ Reflection (v6 and newer).'] echo ![b+''] echo ![s+'F;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;'] # # Author: Robert W.Mills # # Modification History can be found at the end of this file. # deletevar s, b return endif if not finfo('!hpfile','exists') then echo ERROR: HPe3000 file (![ups('!hpfile')]) does not exist -- TRANSFER ABORTED return endif setvar rft_hpfile ups('!hpfile') if '!pcfile' <> '' then setvar rft_pcfile ups('!pcfile') else setvar rft_pcfile '!rft_hpfile' endif setvar rft_folder '!folder' if rht('!folder',1) <> '\' setvar rft_folder '!folder\' endif if '!drive' <> '' then setvar rft_drive lft(ups('!drive'),1)+':\' else setvar rft_drive 'H:\' endif setvar rft_ftm ups('!ftm') if pos('|![lft("!rft_ftm",1)]|','|A|B|L|') = 0 then echo ERROR: Requested file transfer method (!rft_ftm) not recognised. return endif setvar rft_exists ups('!exists') if pos('|![lft("!rft_exists",1)]|','|A|D|') = 0 then echo ERROR: Requested PC file exists action (!rft_exists) not recognised. return endif if HPTYPEAHEAD then setvar rft_hptypeahead true setvar HPTYPEAHEAD false else setvar rft_hptypeahead false endif #---- Start of Reflection code. echo ![chr(27)]&o2G echo Const REMOTE_FILE = "!rft_hpfile" echo Const LOCAL_FILE = "!rft_pcfile" echo Const TRANSFER = " File Transferred : !rft_hpfile To !rft_pcfile" echo Const DIALOG_TITLE = "HPe3000 to PC File Transfer v01.05" echo Const INST_LINE_1 = " Select a folder and press the [OK] button" echo Const INST_LINE_2 = " OR -- Press the [Cancel] button to exit." echo Const CANCELLED = " Transfer has been cancelled by user." echo Const DESTINATION = " Destination Folder : " echo Const ERROR_L1 = " Please contact the Help Desk on x3014 and tell them:" echo Const ERROR_L2 = " An error occured in RTF.CMD.SYSTEMS during the download." echo Const ERROR_L3 = " Reflection reports " if lft('!rft_exists',1) = 'D' then echo Const IF_FILE_EXISTS = rcDelete else echo Const IF_FILE_EXISTS = rcAskUser endif if '!rft_ftm' = 'L' then echo Const TRANSFER_TYPE = rcLabels echo Const METHOD = " File Transfer Method : LABELS(WRQ)" elseif '!rft_ftm' = 'B' then echo Const TRANSFER_TYPE = rcBinary echo Const METHOD = " File Transfer Method : BINARY" else echo Const TRANSFER_TYPE = rcAscii echo Const METHOD = " File Transfer Method : ASCII" endif echo Function BrowseForFolder$ echo Dim InstLines$ echo InstLines$ = INST_LINE_1 & Chr$(rcCR) & INST_LINE_2 echo BrowseForFolder$ = Application.GetFolderName(DIALOG_TITLE, InstLines$, "!rft_drive") echo End Function echo Sub Main echo Dim Folder$ echo Dim ErrorMsg$ echo ErrorMsg$ = ERROR_L1 & Chr$(rcCR) & Chr$(rcCR) & ERROR_L2 & Chr$(rcCR) & ERROR_L3 & Chr$(rcCR) if '!folder' = '' then echo Folder$ = BrowseForFolder$ echo If Folder$ = "" Then echo MsgBox CANCELLED, 64, DIALOG_TITLE echo Exit Sub echo End If else echo Folder$ = "!rft_drive!rft_folder" endif echo With Application echo If Right(Folder$,1) <> "\" Then echo Folder$ = Folder$ + "\" echo End If echo On Error GoTo ErrorHandler echo .WRQReceiveFile Trim$(Folder$+LOCAL_FILE), REMOTE_FILE, TRANSFER_TYPE, IF_FILE_EXISTS echo On Error GoTo 0 echo MsgBox TRANSFER & Chr$(rcCR) & DESTINATION & Folder$ & Chr$(rcCR) & METHOD, 64, DIALOG_TITLE echo .Transmit Chr$(rcCR) echo End With echo Exit Sub echo ErrorHandler: echo MsgBox ErrorMsg$ & " [" & Str$(err) & " -- " & Error$(err) & "]", 16, DIALOG_TITLE echo End Sub echo ![chr(27)]&oH #---- End of Reflection code. if rft_hptypeahead then setvar HPTYPEAHEAD true endif deletevar rft_@ return #-- DO NOT ENTER EXECUTABLE CODE AFTER ABOVE 'RETURN' STATEMENT. Modification History: Versn | Date | Who | Reason for Change ------+------------+-----+------------------------------------------ 01.00 | 07/03/2001 | RWM | Initial version. | | | Based on TRANSFER command file written | | | by Eben Yong (Health Plan of San Mateo) 01.01 | 12/03/2001 | RWM | Allow user to select name to be used for | | | PC file (defaults to HP filename). 01.02 | 20/03/2001 | RWM | Added Drive parameter (defaults to H:\). 01.03 | 22/03/2001 | RWM | Added error file transfer handling. 01.04 | 19/04/2001 | RWM | Enhanced file transfer error nessage. 01.05 | 18/05/2001 | RWM | Disable type ahead during run if set. #=================================================================== # End of command file. #=================================================================== ++++ end of script. From Bjorn.Selfors@kby.BERGEN-KOMMUNE.telemax.no 21 May 2001 14:20:38 Z Date: 21 May 2001 14:20:38 Z From: Bjorn.Selfors@kby.BERGEN-KOMMUNE.telemax.no Bjorn.Selfors@kby.BERGEN-KOMMUNE.telemax.no Subject: MPE/Reflection script fails when run in PowerHouse Quick I have done it like this: run command "RUN CI.PUB.SYS;INFO='SC2PAR';PARM=2;STDIN=PCLNK" & clear 16 to 23 refresh 16 to 23 on error continue nowarn accept host PCLNK is a file 3 lines: RUN PCLINK2.PUB.SYS : EXIT SC2PAR is a file on the HP3000 : invoking my (RCL) script on the pc. ( See the Reflection Command Language Manual for details: Issuing Commands from the host) (You need some esc sequences in there) "host" is just a temporary item, receiving the Completion code from Reflection. Bjřrn Selfors Delfi Data a.s. Tlf +47 5556 8887 http://www.delfidata.no From bengt.johansson@thoreb.se Tue, 22 May 2001 15:55:10 +0200 Date: Tue, 22 May 2001 15:55:10 +0200 From: Bengt Johansson bengt.johansson@thoreb.se Subject: Solved: Delays in connection to HP3000 via VT-MGR vs TELNET Hi everybody ! I am now happy to say that my delay problem from last week is solved ! >From Chuck Ryan ( ) I got this information: 'I had a similar problem on our local lan where serially connected users were getting better performance than lan users using quick screens. To correct this I set a file equation of "FILE STDIN=$NULL" just before running the quick screen. Then, in the Initialize procedure of the first screen I issue a "RESET STDIN" command as some of my screens run quiz reports and the STDIN file equation prevents quiz from running. It's kind of a hack but we are talking about Powerhouse here. :>' I ( Bengt again ) have to confess that I firstly did not believe that this 'hack' would help, but it did ! And still I can't understand why it does. This is magic to me at least. Can somebody ( maybe somebody at Cognos ) explain to me why it helps, please. Best regards from a happy Bengt Johansson Bengt Johansson skrev: > Hi again everybody ! > > I have tested the advices. ( Thank you for them ! ) First only ( in the UDC: ) QUICK;LINEREAD=YES, then both QUICK;LINEREAD=YES; and TERM=HP2392-X25; and also READ="LINE", ( from the beginning there was 'QUICK;LINEREAD=NO; and TERM=HP2392-192-B-ANY' ), but I am sorry to say that nothing helped. There are still the same delays. > I have now addressed the suggested HP 3000-L listserver (hp3000-l@raven.utc.edu) with the same question. ( Thank you, John Pickering, for that tip. ) > I will report back to you if they have a solution. > > Best regards > Bengt Johansson > > "Ole Hansen, ScanConsult" skrev: > > > Hi Bengt, > > > > In addition to John Pickering's explanation of READ=CHAR vs. LINEREAD, I recall having similar problems, when we first connected remote sites over X.25-connections. At the advice of Cognos we changed our startup-command to: > > > > QUICK;LINEREAD=YES;TERM=HP2392-X25; etc. > > > > Now that we use Internet-borne connections instead, we still have TERM=HP2392-X25 (not because we did some clever thinking, but we didn't change anything, and had no problems when we switched). Maybe somebody at Cognos may want to comment on the significance of TERM in this context. > > > > Slightly off topic, we have noticed another peculiarity when running PowerHouse screens over 'long' connections (internet or satelite links). We have a few screens that run WHILE RETRIEVING statements and does a lot of spin-off processing, a.o. some RUN COMMAND statements, but with no display activity (i.e. no DISPLAY or INFORM verbs). These screens run 5-10 times slower over 'long' connections than when run over a local network. We have observed, that even if the screen is not changing its appearance, there is a lot of network traffic during the spin-off processing, i.e. a lot of packets are exchanged between QUICK on the HP3000 and Reflection on the PC - thus causing the slower response we think. > > > > Best regards > > Ole Hansen > > > > ==================================================================== > > E-mail: oh@scanconsult.dk Cell Phone: +45 40 40 11 50 > > ==================================================================== > > ScanConsult IT-Partners ApS Voice : +45 87 38 74 74 > > Jegstrupvej 96A Fax : +45 87 38 74 75 > > DK-8361 Hasselager Support : support@scanconsult.dk > > Denmark Sales : sales@scanconsult.dk > > http://www.scanconsult.dk > > === ParaSuite : A Suite of Air Cargo Handling solutions === > > === ParaSafe : A Suite of IT-Security solutions === > > === We partner with Hewlett-Packard, Cognos, Lotus and others === > > > > -----Oprindelig meddelelse----- > > Fra: Bengt Johansson > > Til: 'Powerhouse List Group' > > Dato: 16. maj 2001 19:10 > > Emne: Delays in connection to HP3000 via VT-MGR vs TELNET > > > > >Hi everybody ! > > > > > >This question is not pure PowerHouse but it shows up when using > > >PowerHouse. > > > > > >We have an HP3000 at our office in Gothenburg and our customers are > > >spread out all over Sweden. Our customers mostly connects to the HP3000 > > >using the Internet, but when there are problems with Internet or their > > >local network they all have a lifeline using the old technique with > > >direct telephone modem ( 19200 bps ), and that is always working. > > > > > >When using Internet our customers use either the VT-MGR or the TELNET > > >protocoll in 'Connection Setup' in Reflection. > > >Some customers never have problems ( and that is good ). > > >But some customers almost always have problems ( and that is no good ), > > >and so have I myself sometimes when connecting from my home i Halmstad, > > >where I also have an office. > > > > > >When using TELNET there are sometimes delays for every character sent. I > > >myself have seen delays up to 15 seconds but one customer have seen > > >delays up to 40 seconds. > > >This delay is from having pressed one button till the character pops up > > >at the screen. > > >Very annoying. > > >When using TELNET every character ( plus 40 extra characters ), when > > >pressing the button, are sent as a package from the user, to the HP3000 > > >and then back to the user before the character pops up at the screen. > > >This TELNET delay became a bit shorter = better ( but not good enough ) > > >after having changed 'Retransmission Interval Lower Bound' from default > > >4 seconds to 1 second in NETXPORT.GPROT.TCP in NMMGR. > > > > > >Now to the question: > > >When using VT-MGR there is ALWAYS a delay in some PowerHouse Quick > > >screens, especially one heavily used screen with a primary file and a > > >detail file with many occurrancies and much data on the screen in Find > > >Mode after having read one set of data from having pressed Return to get > > >the next set of data till the new data pops up at the screen. > > >>From home with my relatively slow Internet connection this delay is > > >always 13 seconds with VT-MGR in this screen filled with data. > > >Using TELNET this delay is only between 1 and 2 seconds, but as I said > > >TELNET sometimes is too slow in another way. > > >Inside our office in Gothenburg with a fast LAN ( 100 Mbit/sec ) the > > >same screen gives a delay between 4 and 5 seconds with VT-MGR ( when the > > >TELNET delay is under 1 second ), and that is too much isn't it ? And > > >the 13 seconds mentioned above I can't stand. Would you ? > > >When using common MPE commands like LISTF or other, there are no delays > > >at all inside the fast LAN neither with VT-MGR nor TELNET, so why when > > >using PowerHouse ? > > > > > >When using VT-MGR every pressed button is echoed to the screen locally > > >and there is a package sent to the HP3000 only when Return is pressed, > > >so this protocoll should be more efficient, but at least in PowerHouse > > >it seems not to be so efficiant. > > > > > >Does anyone know a prescription to solve this problem ? > > >Some of my customers hate these delays so much that they use the > > >telephone modem instead of Internet, and have to pay the telephone cost > > >for that, when Internet is free when you have a fixed connection. > > >I have asked HP Support, but they have not given me any solution yet. > > > > > >Another queer question is why some other customers NEVER have problems > > >using TELNET. > > > > > >We use MPE 6.0 and PowerHouse 8.29.D3. > > > > > >Best regards > > >Bengt Johansson > > >AB Thoreb > > > From steve_franklin@yahoo.com Tue, 22 May 2001 13:31:57 -0500 Date: Tue, 22 May 2001 13:31:57 -0500 From: Steve Franklin steve_franklin@yahoo.com Subject: Axiant Inheritance Question Does anyone have a good description of how inheritance is suppose to work? I set up a style sheet under the Default Section Library. This was done after the screens were migrated. Now I have to go into each screen to specify the inherited style section. I would expect to do this since I created the style sheet after migrating. What I did not expect to have to do was go into each field, label, button, etc. and specify the specific style from my Style Section I already specified in the form. What did I do wrong? Steve Franklin Stw Inc. (952) 432-2203 Email: steve_franklin@yahoo.com _________________________________________________________ Do You Yahoo!? Get your free @yahoo.com address at http://mail.yahoo.com From Roger.Kesterson@mtdsw.com Wed, 23 May 2001 07:53:39 -0700 Date: Wed, 23 May 2001 07:53:39 -0700 From: Kesterson, Roger Roger.Kesterson@mtdsw.com Subject: COMMIT This might seem like a silly question, but when might one use a COMMIT command in QUICK? From Bob.Deskin@Cognos.COM Wed, 23 May 2001 11:24:53 -0400 Date: Wed, 23 May 2001 11:24:53 -0400 From: Deskin, Bob Bob.Deskin@Cognos.COM Subject: COMMIT This is typical PowerHouse. We provide intelligent defaults plus an override capability. You might be doing relational updating in a DESIGNER procedure or you might want to force a commit outside of the normal flow of events. Or even conditionally commit. Note that there is also a ROLLBACK verb. Bob Deskin PowerHouse Web Product Manager, Application Development Tools, Cognos Inc. bob.deskin@cognos.com (613) 738-1338 ext 7268 FAX: (613) 727-1178 3755 Riverside Drive P.O. Box 9707 Stn. T, Ottawa ON K1G 4K9 CANADA -----Original Message----- From: Kesterson, Roger [mailto:Roger.Kesterson@mtdsw.com] Sent: Wednesday, May 23, 2001 10:54 AM To: PowerHouse List Server (E-mail) Subject: COMMIT This might seem like a silly question, but when might one use a COMMIT command in QUICK? = = = = = = = = = = = = = = = = = = = = = = = = = = = = Mailing list: powerh-l@lists.swau.edu Subscribe: "subscribe" in message body to powerh-l-request@lists.swau.edu Unsubscribe: "unsubscribe" in message body to powerh-l-request@lists.swau.edu http://lists.swau.edu/mailman/listinfo/powerh-l This list is closed, thus to post to the list you must be a subscriber. From Mark.Fry@COGNOS.com Wed, 23 May 2001 11:31:58 -0400 Date: Wed, 23 May 2001 11:31:58 -0400 From: Fry, Mark Mark.Fry@COGNOS.com Subject: COMMIT Hi Roger, It's purpose is to commit a transaction used with a relational database. For example, if you are writing a QUICK screen for use with ORACLE tables, you could procedurally commit the current UPDATE transaction your screen is using with the command COMMIT UPDATE. Also, if you have created your own custom transaction called (say) TRAN1, you can procedurally commit it with the command COMMIT TRAN1 [or roll it back with the command ROLLBACK TRAN1]. Hope that helps, Mark Fry Cognos Limited email: Mark.Fry@Cognos.com Cognos Limited Registered in England, number 1635598. Registered office: Westerly Point, Market Street, Bracknell, Berkshire, RG12 1QB This e-mail may contain confidential information and is confidential to the user of the e-mail address to which it is addressed. No-one else may copy or forward all or any of this e-mail in any form. If you receive this e-mail in error please destroy this message and notify the sender by return e-mail." -----Original Message----- From: Kesterson, Roger [mailto:Roger.Kesterson@mtdsw.com] Sent: 23 May 2001 15:54 To: PowerHouse List Server (E-mail) Subject: COMMIT This might seem like a silly question, but when might one use a COMMIT command in QUICK? = = = = = = = = = = = = = = = = = = = = = = = = = = = = Mailing list: powerh-l@lists.swau.edu Subscribe: "subscribe" in message body to powerh-l-request@lists.swau.edu Unsubscribe: "unsubscribe" in message body to powerh-l-request@lists.swau.edu http://lists.swau.edu/mailman/listinfo/powerh-l This list is closed, thus to post to the list you must be a subscriber. From JMBackus@TechGroupMD.com Wed, 23 May 2001 14:15:09 -0400 Date: Wed, 23 May 2001 14:15:09 -0400 From: Jonathan M. Backus JMBackus@TechGroupMD.com Subject: <<>> For the System Manager that wants to understand the queuing model the MPE uses to prioritize things and how it can be fine tuned using HP's Workload Manager or other 3rd party options, we have a 1 day MPE Process Queue Management class scheduled for Friday, July 27th. This class is being taught by Steve Cole of OuterBanks and follows his Advanced MPE System Management class the other 4 days of that week. It is highly recommended that you either attend the Advanced System Manager class or be a pretty seasoned veteran for this class. Details about this class can be found on the University web site - www.TechGroupMD.com Thanx, Jonathan (Jon) M. Backus, MPE-CSM ~ President Tech Group ~ 15 Catawba Place ~ Hagerstown, MD ~ 21742-6515 Email: JMBackus@TechGroupMD.com ~ AIM: JMBackus Vmail: 301.988.0614 ~ Fmail: 301.714.1854 Web: www.TechGroupMD.com From DMielke@aicfunds.com Wed, 23 May 2001 14:45:19 -0400 Date: Wed, 23 May 2001 14:45:19 -0400 From: Daniel Mielke DMielke@aicfunds.com Subject: AS/400 "Refresh" of a defined item as part of a cluster Hello all, I want to build a function key to toggle the display of a 35 character field in the available 20 character display. Error Error Field Name Nbr Description xxxxxxxxxxxxxxxxxxxx xxxx xxxxxxxxxxxxxxxxxxxx xxxxxxxxxxxxxxxxxxxx xxxx xxxxxxxxxxxxxxxxxxxx xxxxxxxxxxxxxxxxxxxx xxxx xxxxxxxxxxxxxxxxxxxx xxxxxxxxxxxxxxxxxxxx xxxx xxxxxxxxxxxxxxxxxxxx xxxxxxxxxxxxxxxxxxxx xxxx xxxxxxxxxxxxxxxxxxxx xxxxxxxxxxxxxxxxxxxx xxxx xxxxxxxxxxxxxxxxxxxx xxxxxxxxxxxxxxxxxxxx xxxx xxxxxxxxxxxxxxxxxxxx xxxxxxxxxxxxxxxxxxxx 1234 PROBLEM EXISTS BETWE to... xxxxxxxxxxxxxxxxxxxx 1234 BETWEEN CHAIR&KEYBRD The source below seems to work, but only for the first item in the cluster. How is it coded for multiple items? I'd like to see the entire block toggled. >key 2 local label "Desc.Toggle" ACTION AND DATA DESIGNER TOGGLE >temporary show char*1 initial "B" reset at mode >define description char*20 = substring(DESCR-35,1,20) & > if show = "B" & > else substring(DESCR-35,16,20) & > if show = "E" >PROCEDURE DESIGNER TOGGLE > BEGIN > if SHOW = "B" > then let SHOW = "E" > else let SHOW = "B" > display description > END > BUY. HOLD. AND PROSPER. > > Daniel Mielke > Senior User Analyst > AIC Limited > 1-888-710-4242 (4614) > dmielke@aicfunds.com > www.aicfunds.com > From Allen.Jones@vpgroup.com Wed, 23 May 2001 11:56:51 -0700 Date: Wed, 23 May 2001 11:56:51 -0700 From: Jones Allen (Van) Allen.Jones@vpgroup.com Subject: AS/400 "Refresh" of a defined item as part of a cluster You need a FOR statement to run the DISPLAY for each occurrence of your cluster: ... FOR DISPLAY description ... Use the file name in the FOR statement to specify the number of times the FOR loop should execute, in case you change the occurrence in later screen maintenance, then you won't have to change the FOR statement. -----Original Message----- From: Daniel Mielke [mailto:DMielke@aicfunds.com] Sent: Wednesday, May 23, 2001 11:45 AM To: Powerhouse List (E-mail) Subject: AS/400 "Refresh" of a defined item as part of a cluster Hello all, I want to build a function key to toggle the display of a 35 character field in the available 20 character display. Error Error Field Name Nbr Description xxxxxxxxxxxxxxxxxxxx xxxx xxxxxxxxxxxxxxxxxxxx xxxxxxxxxxxxxxxxxxxx xxxx xxxxxxxxxxxxxxxxxxxx xxxxxxxxxxxxxxxxxxxx xxxx xxxxxxxxxxxxxxxxxxxx xxxxxxxxxxxxxxxxxxxx xxxx xxxxxxxxxxxxxxxxxxxx xxxxxxxxxxxxxxxxxxxx xxxx xxxxxxxxxxxxxxxxxxxx xxxxxxxxxxxxxxxxxxxx xxxx xxxxxxxxxxxxxxxxxxxx xxxxxxxxxxxxxxxxxxxx xxxx xxxxxxxxxxxxxxxxxxxx xxxxxxxxxxxxxxxxxxxx 1234 PROBLEM EXISTS BETWE to... xxxxxxxxxxxxxxxxxxxx 1234 BETWEEN CHAIR&KEYBRD The source below seems to work, but only for the first item in the cluster. How is it coded for multiple items? I'd like to see the entire block toggled. >key 2 local label "Desc.Toggle" ACTION AND DATA DESIGNER TOGGLE >temporary show char*1 initial "B" reset at mode >define description char*20 = substring(DESCR-35,1,20) & > if show = "B" & > else substring(DESCR-35,16,20) & > if show = "E" >PROCEDURE DESIGNER TOGGLE > BEGIN > if SHOW = "B" > then let SHOW = "E" > else let SHOW = "B" > display description > END > BUY. HOLD. AND PROSPER. > > Daniel Mielke > Senior User Analyst > AIC Limited > 1-888-710-4242 (4614) > dmielke@aicfunds.com > www.aicfunds.com > = = = = = = = = = = = = = = = = = = = = = = = = = = = = Mailing list: powerh-l@lists.swau.edu Subscribe: "subscribe" in message body to powerh-l-request@lists.swau.edu Unsubscribe: "unsubscribe" in message body to powerh-l-request@lists.swau.edu http://lists.swau.edu/mailman/listinfo/powerh-l This list is closed, thus to post to the list you must be a subscriber. From DMielke@aicfunds.com Wed, 23 May 2001 14:59:34 -0400 Date: Wed, 23 May 2001 14:59:34 -0400 From: Daniel Mielke DMielke@aicfunds.com Subject: AS/400 "Refresh" of a defined item as part of a cluster Thanks - works like a charm... -----Original Message----- From: Jones Allen (Van) [mailto:Allen.Jones@vpgroup.com] Sent: Wednesday, May 23, 2001 2:57 PM To: Daniel Mielke; Powerhouse List (E-mail) Subject: RE: AS/400 "Refresh" of a defined item as part of a cluster You need a FOR statement to run the DISPLAY for each occurrence of your cluster: ... FOR DISPLAY description ... Use the file name in the FOR statement to specify the number of times the FOR loop should execute, in case you change the occurrence in later screen maintenance, then you won't have to change the FOR statement. -----Original Message----- From: Daniel Mielke [mailto:DMielke@aicfunds.com] Sent: Wednesday, May 23, 2001 11:45 AM To: Powerhouse List (E-mail) Subject: AS/400 "Refresh" of a defined item as part of a cluster Hello all, I want to build a function key to toggle the display of a 35 character field in the available 20 character display. Error Error Field Name Nbr Description xxxxxxxxxxxxxxxxxxxx xxxx xxxxxxxxxxxxxxxxxxxx xxxxxxxxxxxxxxxxxxxx xxxx xxxxxxxxxxxxxxxxxxxx xxxxxxxxxxxxxxxxxxxx xxxx xxxxxxxxxxxxxxxxxxxx xxxxxxxxxxxxxxxxxxxx xxxx xxxxxxxxxxxxxxxxxxxx xxxxxxxxxxxxxxxxxxxx xxxx xxxxxxxxxxxxxxxxxxxx xxxxxxxxxxxxxxxxxxxx xxxx xxxxxxxxxxxxxxxxxxxx xxxxxxxxxxxxxxxxxxxx xxxx xxxxxxxxxxxxxxxxxxxx xxxxxxxxxxxxxxxxxxxx 1234 PROBLEM EXISTS BETWE to... xxxxxxxxxxxxxxxxxxxx 1234 BETWEEN CHAIR&KEYBRD The source below seems to work, but only for the first item in the cluster. How is it coded for multiple items? I'd like to see the entire block toggled. >key 2 local label "Desc.Toggle" ACTION AND DATA DESIGNER TOGGLE >temporary show char*1 initial "B" reset at mode >define description char*20 = substring(DESCR-35,1,20) & > if show = "B" & > else substring(DESCR-35,16,20) & > if show = "E" >PROCEDURE DESIGNER TOGGLE > BEGIN > if SHOW = "B" > then let SHOW = "E" > else let SHOW = "B" > display description > END > BUY. HOLD. AND PROSPER. > > Daniel Mielke > Senior User Analyst > AIC Limited > 1-888-710-4242 (4614) > dmielke@aicfunds.com > www.aicfunds.com > = = = = = = = = = = = = = = = = = = = = = = = = = = = = Mailing list: powerh-l@lists.swau.edu Subscribe: "subscribe" in message body to powerh-l-request@lists.swau.edu Unsubscribe: "unsubscribe" in message body to powerh-l-request@lists.swau.edu http://lists.swau.edu/mailman/listinfo/powerh-l This list is closed, thus to post to the list you must be a subscriber. From PICKERIJ@norbord.com Wed, 23 May 2001 15:08:30 -0400 Date: Wed, 23 May 2001 15:08:30 -0400 From: Pickering, John (NORBORD) PICKERIJ@norbord.com Subject: AS/400 "Refresh" of a defined item as part of a cluster First, does the as/400 version of PH have scrolling fields?? This would allow you to use the standard Quick interface to move the contents left and right. Second, to improve your method. To "toggle" the entire block I'd probably do it in the OUTPUT procedure. Something like: procedure output descr-35 begin if "E" = show then let fieldtext = descr-35[16:20] else let fieldtext = descr-35[1:20] end Notice I made the beginning of descr-35 the default (i.e. any value other than "E" for your item show). You'll also need a loop somewhere to redisplay the field after you change the value of show. Something like: procedure designer toggle begin if "E" = show then let show = "B" else let show = "E" for [the file or item that occurs] display descr-35 end Regards, JWP, Toronto > -----Original Message----- > From: Daniel Mielke [SMTP:DMielke@aicfunds.com] > Sent: Wednesday, May 23, 2001 2:45 PM > To: Powerhouse List (E-mail) > Subject: AS/400 "Refresh" of a defined item as part of a cluster > > Hello all, > > I want to build a function key to toggle the display of a 35 character > field > in the available 20 character display. > > Error Error > Field Name Nbr Description > xxxxxxxxxxxxxxxxxxxx xxxx xxxxxxxxxxxxxxxxxxxx > xxxxxxxxxxxxxxxxxxxx xxxx xxxxxxxxxxxxxxxxxxxx > xxxxxxxxxxxxxxxxxxxx xxxx xxxxxxxxxxxxxxxxxxxx > xxxxxxxxxxxxxxxxxxxx xxxx xxxxxxxxxxxxxxxxxxxx > xxxxxxxxxxxxxxxxxxxx xxxx xxxxxxxxxxxxxxxxxxxx > xxxxxxxxxxxxxxxxxxxx xxxx xxxxxxxxxxxxxxxxxxxx > xxxxxxxxxxxxxxxxxxxx xxxx xxxxxxxxxxxxxxxxxxxx > xxxxxxxxxxxxxxxxxxxx 1234 PROBLEM EXISTS BETWE > to... > xxxxxxxxxxxxxxxxxxxx 1234 BETWEEN CHAIR&KEYBRD > > The source below seems to work, but only for the first item in the > cluster. > > > How is it coded for multiple items? I'd like to see the entire block > toggled. > > >key 2 local label "Desc.Toggle" ACTION AND DATA DESIGNER TOGGLE > > >temporary show char*1 initial "B" reset at mode > >define description char*20 = substring(DESCR-35,1,20) & > > if show = "B" & > > else substring(DESCR-35,16,20) & > > if show = "E" > > >PROCEDURE DESIGNER TOGGLE > > BEGIN > > if SHOW = "B" > > then let SHOW = "E" > > else let SHOW = "B" > > display description > > END > > From bob.edis@fleetpride.com Thu, 24 May 2001 08:32:23 -0500 Date: Thu, 24 May 2001 08:32:23 -0500 From: Edis, Bob bob.edis@fleetpride.com Subject: COMMIT G'day Roger Some RDBMS's give a lot of user control over transactions. E.g. in Oracle Rdb you can define a custom transaction separate from the default query and update transactions in order to keep display data up-to-date on a screen. Say you have a main screen with several subscreens under it. Each sub screen represents a different entity or aspect of the primary entity. On the main screen you have flags that tell you if the subscreens have been used, i.e. data inserted. A custom transaction can keep these flags up-to-date through careful use of a COMMIT on that transaction. This COMMIT would have no impact on the main default transactions. Regards, Blue -----Original Message----- From: Kesterson, Roger [mailto:Roger.Kesterson@mtdsw.com] Sent: Wednesday, May 23, 2001 9:54 AM To: PowerHouse List Server (E-mail) Subject: COMMIT This might seem like a silly question, but when might one use a COMMIT command in QUICK? = = = = = = = = = = = = = = = = = = = = = = = = = = = = Mailing list: powerh-l@lists.swau.edu Subscribe: "subscribe" in message body to powerh-l-request@lists.swau.edu Unsubscribe: "unsubscribe" in message body to powerh-l-request@lists.swau.edu http://lists.swau.edu/mailman/listinfo/powerh-l This list is closed, thus to post to the list you must be a subscriber. From Roger.Kesterson@mtdsw.com Thu, 24 May 2001 08:11:55 -0700 Date: Thu, 24 May 2001 08:11:55 -0700 From: Kesterson, Roger Roger.Kesterson@mtdsw.com Subject: COMMIT Thanks for all the responses. It wasn't clear to me from reading the manual that it is only for relational. We only use PH with RMS, so I guess that's all I needed to know. I am experiencing another strange problem. One thing we are using PH for (PH 7.10.G2, OpenVMS 7.1) is to run a QUICK program in batch that acts as a "server" for some data collection systems running on remote nodes. The communication is via a combination of mailboxes and DECNET transparent task-to-task. This setup, so far, appears to be very robust, and seems to recover quickly and cleanly from network "glitches" which we experience on a regular basis since one of our nodes is in Mexico, where the power grid and data lines are quite unstable. But I digress. The error is: "Data access error. (LOAD_LISTS*01)?%RMS-F-FAC, record operation not permitted by specified file access (FAC)" The LOAD_LISTS file is defined as: FILE LOAD_LISTS DESIGNER OPEN WRITE SHARE FILE LOAD_LISTS DESIGNER ALIAS LOAD_LISTS_REF OPEN READ SHARE Even though the error message says (LOAD_LISTS*01), it is occurring in a procedure that is trying to read from LOAD_LISTS_REF. It seems to only occur after it writes a record to LOAD_LISTS for one request from the data collection system, and then the very next request causes it to attempt to read the same record from LOAD_LISTS_REF. Here are the code snipets: PROCEDURE INTERNAL PROCESS_TRC BEGIN LET T-TRAILER = (MBX_BUF OF MBXOUT)[1:9] LET T-LOAD-LIST = (MBX_BUF OF MBXOUT)[10:6] LET LOAD_LIST_NUMBER OF LOAD_LISTS = T-LOAD-LIST LET TRAILER_NO OF LOAD_LISTS = T-TRAILER LET DATE_TIME_CREATED OF LOAD_LISTS = & ASCII(SYSDATE,8) + ASCII(SYSTIME,8) PUT LOAD_LISTS RESET END PROCEDURE INTERNAL PROCESS_CMP BEGIN LET T-LOAD-LIST = (MBX_BUF OF MBXOUT)[1:6] GET LOAD_LISTS_REF VIA LOAD_LIST_NUMBER & USING T-LOAD-LIST OPTIONAL IF ACCESSOK THEN BEGIN . . . END END So basically, PROCESS_TRC is called, followed fairly quickly by PROCESS_CMP, using the same value for T-LOAD-LIST, and the program blows on the GET with the above error. It seems like it might be a timing issue, because when the program comes back up, the PROCESS_CMP works fine on the same load list number. Thanks, Roger > -----Original Message----- > From: Kesterson, Roger [mailto:Roger.Kesterson@mtdsw.com] > Sent: Wednesday, May 23, 2001 9:54 AM > To: PowerHouse List Server (E-mail) > Subject: COMMIT > > > This might seem like a silly question, but when might one use a COMMIT > command in QUICK? > > = = = = = = = = = = = = = = = = = = = = = = = = = = = = > Mailing list: powerh-l@lists.swau.edu > Subscribe: "subscribe" in message body to > powerh-l-request@lists.swau.edu > Unsubscribe: "unsubscribe" in message body to > powerh-l-request@lists.swau.edu > http://lists.swau.edu/mailman/listinfo/powerh-l > This list is closed, thus to post to the list you must be a > subscriber. > From ntietz@MCARE.MED.UMICH.EDU Thu, 24 May 2001 11:16:30 -0400 Date: Thu, 24 May 2001 11:16:30 -0400 From: Nancy Tietz ntietz@MCARE.MED.UMICH.EDU Subject: COMMIT You could try making the Reference file a Designer instead, since you are coding the GET yourself anyway. Also, check that the TEMP field doesn't get reset. You can do that with the RESET AT STARTUP option on the TEMP statement. If you use DEBUG in Qdesign, you could see what was happening to your temp and define fields. I hope that helps, Nancy -----Original Message----- From: Kesterson, Roger [mailto:Roger.Kesterson@mtdsw.com] Sent: Thursday, May 24, 2001 11:12 AM To: 'powerh-l@list.swau.edu' Subject: RE: COMMIT Thanks for all the responses. It wasn't clear to me from reading the manual that it is only for relational. We only use PH with RMS, so I guess that's all I needed to know. I am experiencing another strange problem. One thing we are using PH for (PH 7.10.G2, OpenVMS 7.1) is to run a QUICK program in batch that acts as a "server" for some data collection systems running on remote nodes. The communication is via a combination of mailboxes and DECNET transparent task-to-task. This setup, so far, appears to be very robust, and seems to recover quickly and cleanly from network "glitches" which we experience on a regular basis since one of our nodes is in Mexico, where the power grid and data lines are quite unstable. But I digress. The error is: "Data access error. (LOAD_LISTS*01)?%RMS-F-FAC, record operation not permitted by specified file access (FAC)" The LOAD_LISTS file is defined as: FILE LOAD_LISTS DESIGNER OPEN WRITE SHARE FILE LOAD_LISTS DESIGNER ALIAS LOAD_LISTS_REF OPEN READ SHARE Even though the error message says (LOAD_LISTS*01), it is occurring in a procedure that is trying to read from LOAD_LISTS_REF. It seems to only occur after it writes a record to LOAD_LISTS for one request from the data collection system, and then the very next request causes it to attempt to read the same record from LOAD_LISTS_REF. Here are the code snipets: PROCEDURE INTERNAL PROCESS_TRC BEGIN LET T-TRAILER = (MBX_BUF OF MBXOUT)[1:9] LET T-LOAD-LIST = (MBX_BUF OF MBXOUT)[10:6] LET LOAD_LIST_NUMBER OF LOAD_LISTS = T-LOAD-LIST LET TRAILER_NO OF LOAD_LISTS = T-TRAILER LET DATE_TIME_CREATED OF LOAD_LISTS = & ASCII(SYSDATE,8) + ASCII(SYSTIME,8) PUT LOAD_LISTS RESET END PROCEDURE INTERNAL PROCESS_CMP BEGIN LET T-LOAD-LIST = (MBX_BUF OF MBXOUT)[1:6] GET LOAD_LISTS_REF VIA LOAD_LIST_NUMBER & USING T-LOAD-LIST OPTIONAL IF ACCESSOK THEN BEGIN . . . END END So basically, PROCESS_TRC is called, followed fairly quickly by PROCESS_CMP, using the same value for T-LOAD-LIST, and the program blows on the GET with the above error. It seems like it might be a timing issue, because when the program comes back up, the PROCESS_CMP works fine on the same load list number. Thanks, Roger > -----Original Message----- > From: Kesterson, Roger [mailto:Roger.Kesterson@mtdsw.com] > Sent: Wednesday, May 23, 2001 9:54 AM > To: PowerHouse List Server (E-mail) > Subject: COMMIT > > > This might seem like a silly question, but when might one use a COMMIT > command in QUICK? > > = = = = = = = = = = = = = = = = = = = = = = = = = = = = > Mailing list: powerh-l@lists.swau.edu > Subscribe: "subscribe" in message body to > powerh-l-request@lists.swau.edu > Unsubscribe: "unsubscribe" in message body to > powerh-l-request@lists.swau.edu > http://lists.swau.edu/mailman/listinfo/powerh-l > This list is closed, thus to post to the list you must be a > subscriber. > = = = = = = = = = = = = = = = = = = = = = = = = = = = = Mailing list: powerh-l@lists.swau.edu Subscribe: "subscribe" in message body to powerh-l-request@lists.swau.edu Unsubscribe: "unsubscribe" in message body to powerh-l-request@lists.swau.edu http://lists.swau.edu/mailman/listinfo/powerh-l This list is closed, thus to post to the list you must be a subscriber. From rendell@agbus.com Thu, 24 May 2001 08:27:15 -0700 Date: Thu, 24 May 2001 08:27:15 -0700 From: Rendell Requiro rendell@agbus.com Subject: AS/400 "Refresh" of a defined item as part of a cluster This is a multi-part message in MIME format. ------=_NextPart_000_000F_01C0E42B.56A42AE0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable I thought that in an AS400 CUA screen you could just put your cursor in = the field to toggle and press F20 or F19 to 'scroll' the larger field = data in a smaller dispay field. Is this not the case? We use this = scrolling field technique often. Rendell Requiro AG Business Management, Inc. rendell@agbus.com 831-455-2355 x101 voice 831-455-2264 fax www.agbus.com ------=_NextPart_000_000F_01C0E42B.56A42AE0 Content-Type: text/html; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable
I thought that in an AS400 CUA screen you could just = put your=20 cursor in the field to toggle and press F20 or F19 to 'scroll' the = larger field=20 data in a smaller dispay field.   Is this not the case?  = We use=20 this scrolling field technique often.
 
Rendell Requiro
AG Business Management, = Inc.
rendell@agbus.com

831-455-2355&n= bsp; =20 x101 =20 voice
831-455-2264        &nbs= p;   =20 fax
www.agbus.com
------=_NextPart_000_000F_01C0E42B.56A42AE0-- From Leonard_Berkowitz@harvardpilgrim.org Thu, 24 May 2001 11:32:17 -0400 Date: Thu, 24 May 2001 11:32:17 -0400 From: Leonard_Berkowitz@harvardpilgrim.org Leonard_Berkowitz@harvardpilgrim.org Subject: Fatal Quick Error HPe3000, PH 8.19 C3 *E* Memory fault (invalid address) .act Type *Fatal Error* *1504* Notify Cognos Customer Support Do you want a QDUMP file produced? (Y/N) One of our users gets this error on a screen, consistently on a Find. We cannot duplicate it on the same system with two different logons. Any suggestions about where to start looking? The user did not generate a QDUMP, but I suppose we can get one. Thanks. =================== Leonard S. Berkowitz Perot Health Care Systems (Harvard Pilgrim Health Care account) voice: 617-509-1212 fax: 617-509-3737 pager: 781-226-2431 From bob.edis@fleetpride.com Thu, 24 May 2001 10:36:17 -0500 Date: Thu, 24 May 2001 10:36:17 -0500 From: Edis, Bob bob.edis@fleetpride.com Subject: Fatal Quick Error Just a guess, but based on past experience I think you have corrupt data in the record the user is trying to find, possible on the index key. Could you have a null value in the field used by the primary (unique) index? Blue -----Original Message----- From: Leonard_Berkowitz@harvardpilgrim.org [mailto:Leonard_Berkowitz@harvardpilgrim.org] Sent: Thursday, May 24, 2001 10:32 AM To: powerh-l@lists.swau.edu Subject: Fatal Quick Error HPe3000, PH 8.19 C3 *E* Memory fault (invalid address) .act Type *Fatal Error* *1504* Notify Cognos Customer Support Do you want a QDUMP file produced? (Y/N) One of our users gets this error on a screen, consistently on a Find. We cannot duplicate it on the same system with two different logons. Any suggestions about where to start looking? The user did not generate a QDUMP, but I suppose we can get one. Thanks. =================== Leonard S. Berkowitz Perot Health Care Systems (Harvard Pilgrim Health Care account) voice: 617-509-1212 fax: 617-509-3737 pager: 781-226-2431 = = = = = = = = = = = = = = = = = = = = = = = = = = = = Mailing list: powerh-l@lists.swau.edu Subscribe: "subscribe" in message body to powerh-l-request@lists.swau.edu Unsubscribe: "unsubscribe" in message body to powerh-l-request@lists.swau.edu http://lists.swau.edu/mailman/listinfo/powerh-l This list is closed, thus to post to the list you must be a subscriber. From Joe.Boyle@cognos.com Thu, 24 May 2001 11:43:10 -0400 Date: Thu, 24 May 2001 11:43:10 -0400 From: Boyle, Joe Joe.Boyle@cognos.com Subject: FW: COMMIT Hi again Roger, there is a second call which found that the screen issuing the error message had the syntax 'ACTIVITIES FIND' coded on the screen statement. I suspect that this can't be the cause of the problem in your case as the error message would be issued every time. In any event, the solution in this case was to change the file syntax to include 'OPEN UPDATE' e.g. "FILE EMPLOYEES DESIGNER OPEN UPDATE". -----Original Message----- From: Boyle, Joe Sent: 24 May 2001 16:25 To: 'Kesterson, Roger' Subject: RE: COMMIT there is a call logged which was resolved after the caller added open numbers to each of the common file opens as it was thought that the record pointers to each of the files were getting mixed up, e.g. current, FILE LOAD_LISTS DESIGNER OPEN WRITE SHARE FILE LOAD_LISTS DESIGNER ALIAS LOAD_LISTS_REF OPEN READ SHARE modify to, FILE LOAD_LISTS DESIGNER OPEN 1 WRITE SHARE FILE LOAD_LISTS DESIGNER ALIAS LOAD_LISTS_REF OPEN 2 READ SHARE regards, Joe Boyle. -----Original Message----- From: Kesterson, Roger [mailto:Roger.Kesterson@mtdsw.com] Sent: 24 May 2001 16:12 To: 'powerh-l@list.swau.edu' Subject: RE: COMMIT Thanks for all the responses. It wasn't clear to me from reading the manual that it is only for relational. We only use PH with RMS, so I guess that's all I needed to know. I am experiencing another strange problem. One thing we are using PH for (PH 7.10.G2, OpenVMS 7.1) is to run a QUICK program in batch that acts as a "server" for some data collection systems running on remote nodes. The communication is via a combination of mailboxes and DECNET transparent task-to-task. This setup, so far, appears to be very robust, and seems to recover quickly and cleanly from network "glitches" which we experience on a regular basis since one of our nodes is in Mexico, where the power grid and data lines are quite unstable. But I digress. The error is: "Data access error. (LOAD_LISTS*01)?%RMS-F-FAC, record operation not permitted by specified file access (FAC)" The LOAD_LISTS file is defined as: FILE LOAD_LISTS DESIGNER OPEN WRITE SHARE FILE LOAD_LISTS DESIGNER ALIAS LOAD_LISTS_REF OPEN READ SHARE Even though the error message says (LOAD_LISTS*01), it is occurring in a procedure that is trying to read from LOAD_LISTS_REF. It seems to only occur after it writes a record to LOAD_LISTS for one request from the data collection system, and then the very next request causes it to attempt to read the same record from LOAD_LISTS_REF. Here are the code snipets: PROCEDURE INTERNAL PROCESS_TRC BEGIN LET T-TRAILER = (MBX_BUF OF MBXOUT)[1:9] LET T-LOAD-LIST = (MBX_BUF OF MBXOUT)[10:6] LET LOAD_LIST_NUMBER OF LOAD_LISTS = T-LOAD-LIST LET TRAILER_NO OF LOAD_LISTS = T-TRAILER LET DATE_TIME_CREATED OF LOAD_LISTS = & ASCII(SYSDATE,8) + ASCII(SYSTIME,8) PUT LOAD_LISTS RESET END PROCEDURE INTERNAL PROCESS_CMP BEGIN LET T-LOAD-LIST = (MBX_BUF OF MBXOUT)[1:6] GET LOAD_LISTS_REF VIA LOAD_LIST_NUMBER & USING T-LOAD-LIST OPTIONAL IF ACCESSOK THEN BEGIN . . . END END So basically, PROCESS_TRC is called, followed fairly quickly by PROCESS_CMP, using the same value for T-LOAD-LIST, and the program blows on the GET with the above error. It seems like it might be a timing issue, because when the program comes back up, the PROCESS_CMP works fine on the same load list number. Thanks, Roger > -----Original Message----- > From: Kesterson, Roger [mailto:Roger.Kesterson@mtdsw.com] > Sent: Wednesday, May 23, 2001 9:54 AM > To: PowerHouse List Server (E-mail) > Subject: COMMIT > > > This might seem like a silly question, but when might one use a COMMIT > command in QUICK? > > = = = = = = = = = = = = = = = = = = = = = = = = = = = = > Mailing list: powerh-l@lists.swau.edu > Subscribe: "subscribe" in message body to > powerh-l-request@lists.swau.edu > Unsubscribe: "unsubscribe" in message body to > powerh-l-request@lists.swau.edu > http://lists.swau.edu/mailman/listinfo/powerh-l > This list is closed, thus to post to the list you must be a > subscriber. > = = = = = = = = = = = = = = = = = = = = = = = = = = = = Mailing list: powerh-l@lists.swau.edu Subscribe: "subscribe" in message body to powerh-l-request@lists.swau.edu Unsubscribe: "unsubscribe" in message body to powerh-l-request@lists.swau.edu http://lists.swau.edu/mailman/listinfo/powerh-l This list is closed, thus to post to the list you must be a subscriber. From Bill.Michael@ipaper.com Thu, 24 May 2001 10:48:45 -0500 Date: Thu, 24 May 2001 10:48:45 -0500 From: Bill D Michael Bill.Michael@ipaper.com Subject: COMMIT I'd suggest making the LOAD_LISTS_REF open clause "OPEN 1 READ SHARE" so you get a separate buffer from the LOAD_LISTS buffer, which by default and your file statement is OPEN 0 WRITE SHARE. Bill "Kesterson, Roger" @cube.swau.edu on 05/24/2001 10:11:55 AM Sent by: powerh-l-admin@cube.swau.edu To: "'powerh-l@list.swau.edu'" cc: Subject: RE: COMMIT The error is: "Data access error. (LOAD_LISTS*01)?%RMS-F-FAC, record operation not permitted by specified file access (FAC)" The LOAD_LISTS file is defined as: FILE LOAD_LISTS DESIGNER OPEN WRITE SHARE FILE LOAD_LISTS DESIGNER ALIAS LOAD_LISTS_REF OPEN READ SHARE From Joe.Boyle@cognos.com Thu, 24 May 2001 12:01:21 -0400 Date: Thu, 24 May 2001 12:01:21 -0400 From: Boyle, Joe Joe.Boyle@cognos.com Subject: COMMIT could it be that either of T-LOAD-LIST or MBX_BUF OF MBXOUT do not have data at the first time round so there is no key value present for use with the either the read or write. You could try making T-LOAD-LIST 'reset at startup' ( if it is'nt already ) and make the read/write syntax conditional on there not being a null value in the MBX_BUF OF MBXOUT field. -----Original Message----- From: Kesterson, Roger [mailto:Roger.Kesterson@mtdsw.com] Sent: 24 May 2001 16:12 To: 'powerh-l@list.swau.edu' Subject: RE: COMMIT Thanks for all the responses. It wasn't clear to me from reading the manual that it is only for relational. We only use PH with RMS, so I guess that's all I needed to know. I am experiencing another strange problem. One thing we are using PH for (PH 7.10.G2, OpenVMS 7.1) is to run a QUICK program in batch that acts as a "server" for some data collection systems running on remote nodes. The communication is via a combination of mailboxes and DECNET transparent task-to-task. This setup, so far, appears to be very robust, and seems to recover quickly and cleanly from network "glitches" which we experience on a regular basis since one of our nodes is in Mexico, where the power grid and data lines are quite unstable. But I digress. The error is: "Data access error. (LOAD_LISTS*01)?%RMS-F-FAC, record operation not permitted by specified file access (FAC)" The LOAD_LISTS file is defined as: FILE LOAD_LISTS DESIGNER OPEN WRITE SHARE FILE LOAD_LISTS DESIGNER ALIAS LOAD_LISTS_REF OPEN READ SHARE Even though the error message says (LOAD_LISTS*01), it is occurring in a procedure that is trying to read from LOAD_LISTS_REF. It seems to only occur after it writes a record to LOAD_LISTS for one request from the data collection system, and then the very next request causes it to attempt to read the same record from LOAD_LISTS_REF. Here are the code snipets: PROCEDURE INTERNAL PROCESS_TRC BEGIN LET T-TRAILER = (MBX_BUF OF MBXOUT)[1:9] LET T-LOAD-LIST = (MBX_BUF OF MBXOUT)[10:6] LET LOAD_LIST_NUMBER OF LOAD_LISTS = T-LOAD-LIST LET TRAILER_NO OF LOAD_LISTS = T-TRAILER LET DATE_TIME_CREATED OF LOAD_LISTS = & ASCII(SYSDATE,8) + ASCII(SYSTIME,8) PUT LOAD_LISTS RESET END PROCEDURE INTERNAL PROCESS_CMP BEGIN LET T-LOAD-LIST = (MBX_BUF OF MBXOUT)[1:6] GET LOAD_LISTS_REF VIA LOAD_LIST_NUMBER & USING T-LOAD-LIST OPTIONAL IF ACCESSOK THEN BEGIN . . . END END So basically, PROCESS_TRC is called, followed fairly quickly by PROCESS_CMP, using the same value for T-LOAD-LIST, and the program blows on the GET with the above error. It seems like it might be a timing issue, because when the program comes back up, the PROCESS_CMP works fine on the same load list number. Thanks, Roger > -----Original Message----- > From: Kesterson, Roger [mailto:Roger.Kesterson@mtdsw.com] > Sent: Wednesday, May 23, 2001 9:54 AM > To: PowerHouse List Server (E-mail) > Subject: COMMIT > > > This might seem like a silly question, but when might one use a COMMIT > command in QUICK? > > = = = = = = = = = = = = = = = = = = = = = = = = = = = = > Mailing list: powerh-l@lists.swau.edu > Subscribe: "subscribe" in message body to > powerh-l-request@lists.swau.edu > Unsubscribe: "unsubscribe" in message body to > powerh-l-request@lists.swau.edu > http://lists.swau.edu/mailman/listinfo/powerh-l > This list is closed, thus to post to the list you must be a > subscriber. > = = = = = = = = = = = = = = = = = = = = = = = = = = = = Mailing list: powerh-l@lists.swau.edu Subscribe: "subscribe" in message body to powerh-l-request@lists.swau.edu Unsubscribe: "unsubscribe" in message body to powerh-l-request@lists.swau.edu http://lists.swau.edu/mailman/listinfo/powerh-l This list is closed, thus to post to the list you must be a subscriber. From gabriel.estrada@usa.alcatel.com Thu, 24 May 2001 10:30:58 -0700 Date: Thu, 24 May 2001 10:30:58 -0700 From: Gabriel Estrada gabriel.estrada@usa.alcatel.com Subject: AS/400 "Refresh" of a defined item as part of a cluster I believe you have to set the temporary OCCURS WITH the file. -----Original Message----- From: powerh-l-admin@cube.swau.edu [mailto:powerh-l-admin@cube.swau.edu]On Behalf Of Daniel Mielke Sent: Wednesday, May 23, 2001 11:45 AM To: Powerhouse List (E-mail) Subject: AS/400 "Refresh" of a defined item as part of a cluster Hello all, I want to build a function key to toggle the display of a 35 character field in the available 20 character display. Error Error Field Name Nbr Description xxxxxxxxxxxxxxxxxxxx xxxx xxxxxxxxxxxxxxxxxxxx xxxxxxxxxxxxxxxxxxxx xxxx xxxxxxxxxxxxxxxxxxxx xxxxxxxxxxxxxxxxxxxx xxxx xxxxxxxxxxxxxxxxxxxx xxxxxxxxxxxxxxxxxxxx xxxx xxxxxxxxxxxxxxxxxxxx xxxxxxxxxxxxxxxxxxxx xxxx xxxxxxxxxxxxxxxxxxxx xxxxxxxxxxxxxxxxxxxx xxxx xxxxxxxxxxxxxxxxxxxx xxxxxxxxxxxxxxxxxxxx xxxx xxxxxxxxxxxxxxxxxxxx xxxxxxxxxxxxxxxxxxxx 1234 PROBLEM EXISTS BETWE to... xxxxxxxxxxxxxxxxxxxx 1234 BETWEEN CHAIR&KEYBRD The source below seems to work, but only for the first item in the cluster. How is it coded for multiple items? I'd like to see the entire block toggled. >key 2 local label "Desc.Toggle" ACTION AND DATA DESIGNER TOGGLE >temporary show char*1 initial "B" reset at mode >define description char*20 = substring(DESCR-35,1,20) & > if show = "B" & > else substring(DESCR-35,16,20) & > if show = "E" >PROCEDURE DESIGNER TOGGLE > BEGIN > if SHOW = "B" > then let SHOW = "E" > else let SHOW = "B" > display description > END > BUY. HOLD. AND PROSPER. > > Daniel Mielke > Senior User Analyst > AIC Limited > 1-888-710-4242 (4614) > dmielke@aicfunds.com > www.aicfunds.com > = = = = = = = = = = = = = = = = = = = = = = = = = = = = Mailing list: powerh-l@lists.swau.edu Subscribe: "subscribe" in message body to powerh-l-request@lists.swau.edu Unsubscribe: "unsubscribe" in message body to powerh-l-request@lists.swau.edu http://lists.swau.edu/mailman/listinfo/powerh-l This list is closed, thus to post to the list you must be a subscriber. From Leonard_Berkowitz@harvardpilgrim.org Thu, 24 May 2001 15:13:38 -0400 Date: Thu, 24 May 2001 15:13:38 -0400 From: Leonard_Berkowitz@harvardpilgrim.org Leonard_Berkowitz@harvardpilgrim.org Subject: What can this message indicate? HPe3000 PowerHouse 8.19 C3 "Parser Error: Syntax errorParser Error: Syntax error" Yet the routine worked! I am running the following from a file as redirected input: QUIZ on 05/24/2001 03:26:28 PM To: Leonard Berkowitz/CORP/HPHC@HPHC cc: Subject: RE: What can this message indicate? Hi Leonard, This is a known issue that has been resolved in the later releases of 8.19C. I would recommend contacting Cognos Customer Support to obtain an upgrade. As a workaround, purge the nodict subfile before running. Laura Knowles Cognos Customer Support From PICKERIJ@norbord.com Thu, 24 May 2001 15:33:18 -0400 Date: Thu, 24 May 2001 15:33:18 -0400 From: Pickering, John (NORBORD) PICKERIJ@norbord.com Subject: What can this message indicate? Leonard Yet another wonderfully informative PowerHouse error message! This one means that the temporary file ARCHHIST already exists. Or at least it does in this particular case. And it works because Quiz simply writes over the existing temporary subfile. Obvious isn't it :-) Regards, JWP > -----Original Message----- > From: Leonard_Berkowitz@harvardpilgrim.org > [SMTP:Leonard_Berkowitz@harvardpilgrim.org] > Sent: Thursday, May 24, 2001 3:14 PM > To: powerh-l@lists.swau.edu > Subject: What can this message indicate? > > > > HPe3000 PowerHouse 8.19 C3 > > "Parser Error: Syntax errorParser Error: Syntax error" > > Yet the routine worked! I am running the following from a file as > redirected > input: > > QUIZ > Here is the QUIZIN file: > > SET NOWARN NOPRINT NOSTATISTICS NOLIST > ACCESS PHC-CODE > CHOOSE PHC-KEY "HISTRYE" > DEFINE HISTDATE CHARACTER * 62 = & > "SETVAR ARCHDATE " + PHC-INFO[1:8] > REPORT SUMMARY HISTDATE > SET SUBFILE NAME ARCHHIST & > TEMPORARY NODICTIONARY > GO > EXIT > > Thanks. > =================== > Leonard S. Berkowitz > From bschalme@airspeed.mb.ca Thu, 24 May 2001 14:39:28 -0500 Date: Thu, 24 May 2001 14:39:28 -0500 From: Brian Schalme bschalme@airspeed.mb.ca Subject: What can this message indicate? This is a nuisance error that does not affect data integrity in any way. It happens when you are appending to or overwriting an existing NODICT subfile. If you remove the NODICT from the SUBFILE statement or purge the existing NODICT subfile before executing this code, the error message disappears. This error was fixed in 8.19C5. Brian Schalme Airspeed Consulting bschalme@airspeed.mb.ca Leonard_Berkowitz@harvardpilgrim.org wrote: > > HPe3000 PowerHouse 8.19 C3 > > "Parser Error: Syntax errorParser Error: Syntax error" > > Yet the routine worked! I am running the following from a file as redirected > input: > > QUIZ > Here is the QUIZIN file: > > SET NOWARN NOPRINT NOSTATISTICS NOLIST > ACCESS PHC-CODE > CHOOSE PHC-KEY "HISTRYE" > DEFINE HISTDATE CHARACTER * 62 = & > "SETVAR ARCHDATE " + PHC-INFO[1:8] > REPORT SUMMARY HISTDATE > SET SUBFILE NAME ARCHHIST & > TEMPORARY NODICTIONARY > GO > EXIT > > Thanks. > =================== > Leonard S. Berkowitz > Perot Health Care Systems > (Harvard Pilgrim Health Care account) > voice: 617-509-1212 > fax: 617-509-3737 > pager: 781-226-2431 > > = = = = = = = = = = = = = = = = = = = = = = = = = = = = > Mailing list: powerh-l@lists.swau.edu > Subscribe: "subscribe" in message body to powerh-l-request@lists.swau.edu > Unsubscribe: "unsubscribe" in message body to powerh-l-request@lists.swau.edu > http://lists.swau.edu/mailman/listinfo/powerh-l > This list is closed, thus to post to the list you must be a subscriber. From PICKERIJ@norbord.com Thu, 24 May 2001 15:45:53 -0400 Date: Thu, 24 May 2001 15:45:53 -0400 From: Pickering, John (NORBORD) PICKERIJ@norbord.com Subject: What can this message indicate? As an aside, why bother with the NODICT option? MPE will quite happily "xeq" files of type 644; Quiz and Qtp will quite happily "use" files of type 644 (but you'd need a colon in your example); Reflection et al. will quite happily transfer files of type 644 to your PC (and drop the minidictionary in the process unless you use the labels option). In short, I haven't found a case where I really needed to make a NODICT subfile. In fact, keeping the dictionary around makes it easier to go back months later and figure out just what's in the file. Your mileage may vary. From rendell@agbus.com Fri, 25 May 2001 13:14:56 -0700 Date: Fri, 25 May 2001 13:14:56 -0700 From: Rendell Requiro rendell@agbus.com Subject: AS400 creating subtotals from "Selected" data This is a multi-part message in MIME format. ------=_NextPart_000_0021_01C0E51C.B13EC5E0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable Greetings all. I have a Quick screen that I'd like to tune up a bit. The primary file = occurs ten times on the screen and displays data from a single file. = The user would like to add subtotals at the bottom of each column (easy = enough), but wants to have the subtotals calculated only for the data = that the user has "Selected" with the F10=3DSelect key. Has anyone out there addressed this issue? I can do a WHILE RETRIEVING = for an alias of the primary file in the POSTFIND procedure, but I am not = sure how to evaluate the data in the field(s) that the user is Selecting = on. Any suggestions? Rendell Requiro AG Business Management, Inc. rendell@agbus.com 831-455-2355 x101 voice 831-455-2264 fax www.agbus.com ------=_NextPart_000_0021_01C0E51C.B13EC5E0 Content-Type: text/html; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable
Greetings all.
 
I have a Quick screen that I'd like to tune up a = bit. =20 The primary file occurs ten times on the screen and displays data from a = single=20 file.  The user would like to add subtotals at the bottom of each = column=20 (easy enough), but wants to have the subtotals calculated only for the = data that=20 the user has "Selected" with the F10=3DSelect key.
 
Has anyone out there addressed this issue?  I = can do a=20 WHILE RETRIEVING for an alias of the primary file in the POSTFIND = procedure, but=20 I am not sure how to evaluate the data in the field(s) that the user is=20 Selecting on.  Any suggestions?
 
Rendell Requiro
AG Business Management, = Inc.
rendell@agbus.com
831-455-2355&n= bsp; =20 x101 =20 voice
831-455-2264        &nbs= p;   =20 fax
www.agbus.com
------=_NextPart_000_0021_01C0E51C.B13EC5E0-- From stewm@canada.com 26 May 2001 15:48:54 -0700 Date: 26 May 2001 15:48:54 -0700 From: Mark Stewart stewm@canada.com Subject: AS400 creating subtotals from "Selected" data Rendell, What if you had a temporary item that occured with the primary. If the records was selected make the temporary item "Y" or something like that. Subtotal if temporary item = "Y". Mark Stewart Consultants Club Corp. On Fri, 25 May 2001, "Rendell Requiro" wrote: > > > > > > > > >
Greetings all.
>
 
>
I have a Quick screen that I'd like to tune up a bit.  > The primary file occurs ten times on the screen and displays data from a single > file.  The user would like to add subtotals at the bottom of each column > (easy enough), but wants to have the subtotals calculated only for the data that > the user has "Selected" with the F10=Select key.
>
 
>
Has anyone out there addressed this issue?  I can do a > WHILE RETRIEVING for an alias of the primary file in the POSTFIND procedure, but > I am not sure how to evaluate the data in the field(s) that the user is > Selecting on.  Any suggestions?
>
 
>
Rendell Requiro
AG Business Management, Inc.
href="mailto:rendell@agbus.com">rendell@agbus.com
831-455-2355   > x101  > voice
831-455-2264             > fax
href="http://www.agbus.com">www.agbus.com
__________________________________________________________ Get your FREE personalized e-mail at http://www.canada.com From stewm@canada.com 26 May 2001 15:57:28 -0700 Date: 26 May 2001 15:57:28 -0700 From: Mark Stewart stewm@canada.com Subject: COMMIT Roger, Never, it at all possible. It's best just to let PowerHouse handle everything. It does a pretty good job by itself most of the time. Mark Stewart Consultants Club Corp. On Wed, 23 May 2001, "Kesterson, Roger" wrote: > > This might seem like a silly question, but when might one use a COMMIT > command in QUICK? > > = = = = = = = = = = = = = = = = = = = = = = = = = = = = > Mailing list: powerh-l@lists.swau.edu > Subscribe: "subscribe" in message body to powerh-l-request@lists.swau.edu > Unsubscribe: "unsubscribe" in message body to powerh-l-request@lists.swau.edu > http://lists.swau.edu/mailman/listinfo/powerh-l > This list is closed, thus to post to the list you must be a subscriber. __________________________________________________________ Get your FREE personalized e-mail at http://www.canada.com From markus.grossrieder@alba-systems.com Mon, 28 May 2001 18:35:47 +0200 Date: Mon, 28 May 2001 18:35:47 +0200 From: Markus Grossrieder markus.grossrieder@alba-systems.com Subject: Action command S (select) on dates/numbers Esko, that's it ! So easy if you know how ... Thanks, Markus ----- Original Message ----- From: "Esko Lahdensivu" To: "Markus Grossrieder" Cc: Sent: Thursday, May 17, 2001 9:02 AM Subject: Re: Action command S (select) on dates/numbers > > Markus Grossrieder wrote: > > > Thanks to Chris, Joe & Johan for the fast answers ! I'll give the propositions a > > try when I'll have some time. > > > > I was hoping, against all experience, that there is some way, pattern, or holy > > sign to use the 'select processing' for numeric data just as it works for > > strings ... > > > > MPE/iX, PH 8.29. > > If I recall correctly, the original question was about searching dates. well, in > Quick, > dates can be searched ! We use this feature extensively. Say you > have a field "TTASPV OF STTYOTIE DATE DD.MM.YYYY" : to review > all entries from a certain year and month, you would enter (in Select mode) > %##.04.2001 > > % = enter pattern matching mode > ## = any two digits (ie., any day of the month) > the rest is obvious. > > > > > Woudn't it be nice if you could say 'Show me everything from 2001/01/01', or > > This can be done (but only for year 2001) with "%##.##.2001" depending, > of course, from your date format. > > > 'bigger than 10000' ? > > In Quick pattern matching, this would be a nice feature. > > > > > Bob, Conrad, something planned ? > > > > cu, Markus > > > > regards, > Esko > > > From ron@cryptic.rch.unimelb.edu.au Tue, 29 May 2001 12:13:12 +1000 Date: Tue, 29 May 2001 12:13:12 +1000 From: Ron Burnett ron@cryptic.rch.unimelb.edu.au Subject: QTP Questions G'day everyone, Quick questions (sorry, not QUICK, but QTP): What is the maximum width that one can output to a permanent file using QTP (I kinow there is a limit of 255 in QUIZ)? If I use the SET NODICTIONARY construct, does it output a plain (flat) file? For my purposes, I don't want any special file characteristics such as one would get with a normal self-defining subfile. This is for PowerHouse 8.19C running on MPE/iX on an HP3000. Many thanks for your help, Ron Burnett ron@cryptic.rch.unimelb.edu.au From Bob.Deskin@Cognos.COM Tue, 29 May 2001 07:56:06 -0400 Date: Tue, 29 May 2001 07:56:06 -0400 From: Deskin, Bob Bob.Deskin@Cognos.COM Subject: QTP Questions Regarding the maximum width, the limit would be the file system limit. As well, PowerHouse has a limit of 1023 items per file and 31 files per request. With NODICT, yes, you should get a plain flat file with nothing special in the file label (for non-MPE/iXers, we're talking about MPE/iX where subfile info is in the file label). Bob Deskin PowerHouse Web Product Manager, Application Development Tools, Cognos Inc. bob.deskin@cognos.com (613) 738-1338 ext 7268 FAX: (613) 727-1178 3755 Riverside Drive P.O. Box 9707 Stn. T, Ottawa ON K1G 4K9 CANADA -----Original Message----- From: Ron Burnett [mailto:ron@cryptic.rch.unimelb.edu.au] Sent: Monday, May 28, 2001 10:13 PM To: powerh-l@lists.swau.edu Subject: QTP Questions G'day everyone, Quick questions (sorry, not QUICK, but QTP): What is the maximum width that one can output to a permanent file using QTP (I kinow there is a limit of 255 in QUIZ)? If I use the SET NODICTIONARY construct, does it output a plain (flat) file? For my purposes, I don't want any special file characteristics such as one would get with a normal self-defining subfile. This is for PowerHouse 8.19C running on MPE/iX on an HP3000. Many thanks for your help, Ron Burnett ron@cryptic.rch.unimelb.edu.au = = = = = = = = = = = = = = = = = = = = = = = = = = = = Mailing list: powerh-l@lists.swau.edu Subscribe: "subscribe" in message body to powerh-l-request@lists.swau.edu Unsubscribe: "unsubscribe" in message body to powerh-l-request@lists.swau.edu http://lists.swau.edu/mailman/listinfo/powerh-l This list is closed, thus to post to the list you must be a subscriber. From Joe.Boyle@cognos.com Tue, 29 May 2001 08:30:31 -0400 Date: Tue, 29 May 2001 08:30:31 -0400 From: Boyle, Joe Joe.Boyle@cognos.com Subject: QTP Questions I'm not sure how helpful this will be but the pdl manual says, on consecutive pages, that there is a maximum of 4,096 chars per element on p170 and a maximum of 32,767 chars per element on p171 ( and a maximum of 63 files per run QTP ) -----Original Message----- From: Ron Burnett [mailto:ron@cryptic.rch.unimelb.edu.au] Sent: 29 May 2001 03:13 To: powerh-l@lists.swau.edu Subject: QTP Questions G'day everyone, Quick questions (sorry, not QUICK, but QTP): What is the maximum width that one can output to a permanent file using QTP (I kinow there is a limit of 255 in QUIZ)? If I use the SET NODICTIONARY construct, does it output a plain (flat) file? For my purposes, I don't want any special file characteristics such as one would get with a normal self-defining subfile. This is for PowerHouse 8.19C running on MPE/iX on an HP3000. Many thanks for your help, Ron Burnett ron@cryptic.rch.unimelb.edu.au = = = = = = = = = = = = = = = = = = = = = = = = = = = = Mailing list: powerh-l@lists.swau.edu Subscribe: "subscribe" in message body to powerh-l-request@lists.swau.edu Unsubscribe: "unsubscribe" in message body to powerh-l-request@lists.swau.edu http://lists.swau.edu/mailman/listinfo/powerh-l This list is closed, thus to post to the list you must be a subscriber. From robert.mills@windsong-services.co.uk Tue, 29 May 2001 18:00:28 +0100 Date: Tue, 29 May 2001 18:00:28 +0100 From: Robert Mills robert.mills@windsong-services.co.uk Subject: (no subject) Hi all, The following key statements and procedure appear in a Quick screen. If the user has NOT changed any of the displayed fields then it works OK. If they have changed any filed then a 'Data has been changed but not updated. Repeat the action if this is ok.' message is displayed. How do I get the same function key to perform the requested action? key 5 label "Cancel Change" action designer CANC key 5 label " " data disable procedure designer CANC begin info = "*I* Update of Product Record " + truncate(t-product) + " cancelled." push find end regards, +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ |Robert W.Mills (Systems Development Manager) | |Windsong Services, St. Mary Cray, ENGLAND | |Tel : +44 (0)1689 870622 x3005 | |Fax : +44 (0)1689 899026 | |Email: Robert.Mills@windsong-services.co.uk | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ From robert.mills@windsong-services.co.uk Tue, 29 May 2001 18:15:03 +0100 Date: Tue, 29 May 2001 18:15:03 +0100 From: Robert Mills robert.mills@windsong-services.co.uk Subject: Cancelling changes via a Quick Designer Procedure Sorry about the lack of subject on the original posting. My main client has just been upgraded to Outlook 2000 and the layout is very different. -----Original Message----- From: Robert Mills Sent: 29 May 2001 18:00 To: 'PowerHouse List' Subject: Hi all, The following key statements and procedure appear in a Quick screen. If the user has NOT changed any of the displayed fields then it works OK. If they have changed any filed then a 'Data has been changed but not updated. Repeat the action if this is ok.' message is displayed. How do I get the same function key to perform the requested action? key 5 label "Cancel Change" action designer CANC key 5 label " " data disable procedure designer CANC begin info = "*I* Update of Product Record " + truncate(t-product) + " cancelled." push find end regards, +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ |Robert W.Mills (Systems Development Manager) | |Windsong Services, St. Mary Cray, ENGLAND | |Tel : +44 (0)1689 870622 x3005 | |Fax : +44 (0)1689 899026 | |Email: Robert.Mills@windsong-services.co.uk | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ From PICKERIJ@norbord.com Tue, 29 May 2001 13:26:49 -0400 Date: Tue, 29 May 2001 13:26:49 -0400 From: Pickering, John (NORBORD) PICKERIJ@norbord.com Subject: (no subject) How about intentionally clearing Quick's buffers -- by getting the primary file, either with the current key value or with some bogus value. You might also experiment with severe instead of info but then the push find wouldn't be executed and I doubt that a postcommand on the procedure would work either. Keep playing and let us know what works. Regards, JWP > -----Original Message----- > From: Robert Mills [SMTP:robert.mills@windsong-services.co.uk] > Sent: Tuesday, May 29, 2001 1:00 PM > To: 'PowerHouse List' > Subject: (no subject) > > Hi all, > > The following key statements and procedure appear in a Quick screen. If > the > user has NOT changed any of the displayed fields then it works OK. If they > have changed any filed then a 'Data has been changed but not updated. > Repeat > the action if this is ok.' message is displayed. How do I get the same > function key to perform the requested action? > > key 5 label "Cancel Change" action designer CANC > key 5 label " " data disable > > procedure designer CANC > begin > info = "*I* Update of Product Record " + truncate(t-product) + " > cancelled." > push find > end > > regards, > |Robert W.Mills (Systems Development Manager) | > From Allen.Jones@vpgroup.com Tue, 29 May 2001 10:38:39 -0700 Date: Tue, 29 May 2001 10:38:39 -0700 From: Jones Allen (Van) Allen.Jones@vpgroup.com Subject: Cancelling changes via a Quick Designer Procedure I haven't tried this, but it might work. Check for alteredrecord and if true, push find, then push find again unconditionally. The idea is that if nothing is changed, only one push is done. If one of the files is altered, then two push find's will "repeat the action" as requested. Put your info message last with the "now" keyword so it remains on the screen after all is completed. procedure designer CANC begin if alteredrecord of file1 & or alteredrecord of file2 ... (as many files as you have that could be altered) then push find push find info = "*I* Update of Product Record " + truncate(t-product) + " cancelled." now end Good Luck! -----Original Message----- From: Robert Mills [mailto:robert.mills@windsong-services.co.uk] Sent: Tuesday, May 29, 2001 10:15 AM To: 'PowerHouse List' Subject: Cancelling changes via a Quick Designer Procedure Sorry about the lack of subject on the original posting. My main client has just been upgraded to Outlook 2000 and the layout is very different. -----Original Message----- From: Robert Mills Sent: 29 May 2001 18:00 To: 'PowerHouse List' Subject: Hi all, The following key statements and procedure appear in a Quick screen. If the user has NOT changed any of the displayed fields then it works OK. If they have changed any filed then a 'Data has been changed but not updated. Repeat the action if this is ok.' message is displayed. How do I get the same function key to perform the requested action? key 5 label "Cancel Change" action designer CANC key 5 label " " data disable procedure designer CANC begin info = "*I* Update of Product Record " + truncate(t-product) + " cancelled." push find end regards, +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ |Robert W.Mills (Systems Development Manager) | |Windsong Services, St. Mary Cray, ENGLAND | |Tel : +44 (0)1689 870622 x3005 | |Fax : +44 (0)1689 899026 | |Email: Robert.Mills@windsong-services.co.uk | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ = = = = = = = = = = = = = = = = = = = = = = = = = = = = Mailing list: powerh-l@lists.swau.edu Subscribe: "subscribe" in message body to powerh-l-request@lists.swau.edu Unsubscribe: "unsubscribe" in message body to powerh-l-request@lists.swau.edu http://lists.swau.edu/mailman/listinfo/powerh-l This list is closed, thus to post to the list you must be a subscriber. From Aci.Polajnar@rcc-irc.si Wed, 30 May 2001 07:31:43 +0200 Date: Wed, 30 May 2001 07:31:43 +0200 From: Aci Polajnar Aci.Polajnar@rcc-irc.si Subject: Cancelling changes via a Quick Designer Procedure Hi Robert! As we know, every GET statement in PowerHouse changes record status to NEW, UNCHANGED and UNDELETED so you should read an record from your file which never exists procedure designer CANC begin GET VIA USING 'Tralala' OPTIONAL info = "*I* Update of Product Record " + truncate(t-product) + " cancelled." push find end You should check the record status before and make GET statement conditional if you want. Best regards! Robert Mills wrote: > > Sorry about the lack of subject on the original posting. My main client has > just been upgraded to Outlook 2000 and the layout is very different. > > -----Original Message----- > From: Robert Mills > Sent: 29 May 2001 18:00 > To: 'PowerHouse List' > Subject: > > Hi all, > > The following key statements and procedure appear in a Quick screen. If the > user has NOT changed any of the displayed fields then it works OK. If they > have changed any filed then a 'Data has been changed but not updated. Repeat > the action if this is ok.' message is displayed. How do I get the same > function key to perform the requested action? > > key 5 label "Cancel Change" action designer CANC > key 5 label " " data disable > > procedure designer CANC > begin > info = "*I* Update of Product Record " + truncate(t-product) + " >cancelled." > push find > end > > regards, > +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ > |Robert W.Mills (Systems Development Manager) | > |Windsong Services, St. Mary Cray, ENGLAND | > |Tel : +44 (0)1689 870622 x3005 | > |Fax : +44 (0)1689 899026 | > |Email: Robert.Mills@windsong-services.co.uk | > +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ > > = = = = = = = = = = = = = = = = = = = = = = = = = = = = > Mailing list: powerh-l@lists.swau.edu > Subscribe: "subscribe" in message body to powerh-l-request@lists.swau.edu > Unsubscribe: "unsubscribe" in message body to powerh-l-request@lists.swau.edu > http://lists.swau.edu/mailman/listinfo/powerh-l > This list is closed, thus to post to the list you must be a subscriber. -- _\\///_ (' o-o ') _______________ooOo_( )_oOoo_____________________________________ Aci Polajnar Tel : +386(0)3 4274174 IS Consultant Fax : +386(0)3 4274198 Razvojni Center Celje e-mail: acip@rcc-irc.si 3000 Celje, Slovenia URL : http://www.rcc-irc.si _______________oooO____Oooo______________________________________ ( ) ( ) \ ( ) / \_) (_/ From robert.mills@windsong-services.co.uk Wed, 30 May 2001 12:06:18 +0100 Date: Wed, 30 May 2001 12:06:18 +0100 From: Robert Mills robert.mills@windsong-services.co.uk Subject: Cancelling changes via a Quick Designer Procedure Thanks to all those who responded both ON and OFF (you know who you are) list. As requested (by JWP) I'm posting the solution that I have gone with. It looked the easiest and was the most understandable to whoever came next. regards, Robert Procedure Designer CANC Begin If AlteredRecord of PRODUCT-NO-MSTR Then Begin Get PRODUCT-NO-MSTR Via COMPANY-PRODUCT Using "non existent product" Optional End Information = "*I* Product " + Truncate(t-PROD-CODE) + " update has been cancelled." Push Find End -----Original Message----- From: Aci Polajnar [mailto:Aci.Polajnar@rcc-irc.si] Sent: 30 May 2001 06:32 To: Robert Mills; 'PowerHouse List' Subject: Re: Cancelling changes via a Quick Designer Procedure Hi Robert! As we know, every GET statement in PowerHouse changes record status to NEW, UNCHANGED and UNDELETED so you should read an record from your file which never exists procedure designer CANC begin GET VIA USING 'Tralala' OPTIONAL info = "*I* Update of Product Record " + truncate(t-product) + " cancelled." push find end You should check the record status before and make GET statement conditional if you want. Best regards! Robert Mills wrote: > > Sorry about the lack of subject on the original posting. My main client has > just been upgraded to Outlook 2000 and the layout is very different. > > -----Original Message----- > From: Robert Mills > Sent: 29 May 2001 18:00 > To: 'PowerHouse List' > Subject: > > Hi all, > > The following key statements and procedure appear in a Quick screen. If the > user has NOT changed any of the displayed fields then it works OK. If they > have changed any filed then a 'Data has been changed but not updated. Repeat > the action if this is ok.' message is displayed. How do I get the same > function key to perform the requested action? > > key 5 label "Cancel Change" action designer CANC > key 5 label " " data disable > > procedure designer CANC > begin > info = "*I* Update of Product Record " + truncate(t-product) + " >cancelled." > push find > end > > regards, > +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ > |Robert W.Mills (Systems Development Manager) | > |Windsong Services, St. Mary Cray, ENGLAND | > |Tel : +44 (0)1689 870622 x3005 | > |Fax : +44 (0)1689 899026 | > |Email: Robert.Mills@windsong-services.co.uk | > +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ > > = = = = = = = = = = = = = = = = = = = = = = = = = = = = > Mailing list: powerh-l@lists.swau.edu > Subscribe: "subscribe" in message body to powerh-l-request@lists.swau.edu > Unsubscribe: "unsubscribe" in message body to powerh-l-request@lists.swau.edu > http://lists.swau.edu/mailman/listinfo/powerh-l > This list is closed, thus to post to the list you must be a subscriber. -- _\\///_ (' o-o ') _______________ooOo_( )_oOoo_____________________________________ Aci Polajnar Tel : +386(0)3 4274174 IS Consultant Fax : +386(0)3 4274198 Razvojni Center Celje e-mail: acip@rcc-irc.si 3000 Celje, Slovenia URL : http://www.rcc-irc.si _______________oooO____Oooo______________________________________ ( ) ( ) \ ( ) / \_) (_/ From Leonard_Berkowitz@harvardpilgrim.org Wed, 30 May 2001 16:17:57 -0400 Date: Wed, 30 May 2001 16:17:57 -0400 From: Leonard_Berkowitz@harvardpilgrim.org Leonard_Berkowitz@harvardpilgrim.org Subject: Another message puzzle (for me) HPe3000, PH819C3 "Actual key/index layout does not conform to the expected key/index layout." A little background on this dataset. It was a Manual Master. At the request of the application programmer, I dropped it to a Detail dataset and added two additional paths (keys) tied into existing Manual Master datasets. I've checked each of the keys, and their attributes seem to be correct and correspond to the respective elements in the LINKS TO file. Where should I be looking? Thanks. =================== Leonard S. Berkowitz Perot Health Care Systems (Harvard Pilgrim Health Care account) voice: 617-509-1212 fax: 617-509-3737 pager: 781-226-2431 From PICKERIJ@norbord.com Wed, 30 May 2001 16:42:58 -0400 Date: Wed, 30 May 2001 16:42:58 -0400 From: Pickering, John (NORBORD) PICKERIJ@norbord.com Subject: Another message puzzle (for me) Leonard Sounds like the physical dataset and its definition in the dictionary don't match. Has the dictionary been changed and recompiled to reflect the physical change in the dataset? Has the screen/report/run that gave the message been recompiled with the new dictionary? Regards, JWP > -----Original Message----- > From: Leonard_Berkowitz@harvardpilgrim.org > [SMTP:Leonard_Berkowitz@harvardpilgrim.org] > Sent: Wednesday, May 30, 2001 4:18 PM > To: powerh-l@lists.swau.edu > Subject: Another message puzzle (for me) > > > > HPe3000, PH819C3 > > "Actual key/index layout does not conform to the expected key/index > layout." > > A little background on this dataset. It was a Manual Master. At the > request of > the application programmer, I dropped it to a Detail dataset and added two > additional paths (keys) tied into existing Manual Master datasets. > > I've checked each of the keys, and their attributes seem to be correct and > correspond to the respective elements in the LINKS TO file. > > Where should I be looking? > > Thanks. > =================== > Leonard S. Berkowitz > Perot Health Care Systems > (Harvard Pilgrim Health Care account) > voice: 617-509-1212 > fax: 617-509-3737 > pager: 781-226-2431 > > > > = = = = = = = = = = = = = = = = = = = = = = = = = = = = > Mailing list: powerh-l@lists.swau.edu > Subscribe: "subscribe" in message body to powerh-l-request@lists.swau.edu > Unsubscribe: "unsubscribe" in message body to > powerh-l-request@lists.swau.edu > http://lists.swau.edu/mailman/listinfo/powerh-l > This list is closed, thus to post to the list you must be a subscriber. From joan@intertechsystems.com Thu, 31 May 2001 07:58:40 -0600 Date: Thu, 31 May 2001 07:58:40 -0600 From: Joan Conniff joan@intertechsystems.com Subject: Axiant 4GL & Conditional Compiles We are converting a Powerhouse application from HP3000 to Axiant/SQL Server on Windows 2000. We have conditional compile code throughout various screens. If the conditional compiles are in the screen procedures tab we don't seem to have any problems. We do, however, have some conditional code that has come up on the Layout of the screen in Axiant. We cannot seem to get this code to work on the Layout. Axiant has turned the code into a comment line. Does anyone know how we can make use of this code within the Layout tab? We have two buttons that display under one condition and a different button display under a different condition but are not able to get this to work. Thanks in advance for any help. Joan Conniff Intertech Business Systems (780) 413-0400 From Leonard_Berkowitz@harvardpilgrim.org Thu, 31 May 2001 12:05:01 -0400 Date: Thu, 31 May 2001 12:05:01 -0400 From: Leonard_Berkowitz@harvardpilgrim.org Leonard_Berkowitz@harvardpilgrim.org Subject: Another message puzzle (for me) (solved) First, thanks to the several people who replied, privately and to the list, with suggestions. The suggestions were Compare a QUERY FORM set (or SUPRTOOL) listing with the listing from QSHOW REPORT RECORD Make sure the screen has been recompiled Check file equations to point to the correct PHD Well, I had already done all these things before posting my question to the list, but I did each step again with the same result. At this point, I resorted to Sherlock Holmesian wisdom: "When you have ruled out everything that is impossible, what remains, however improbable, points to the solution." I asked the programmer whether the program had been recompiled. Yes it was. Let's do it again. The programmer recompiled the program, indeed, but forgot the step to copy the object to the account where the screen is executed. Thanks, everyone for your time and trouble. =================== Leonard S. Berkowitz Perot Health Care Systems (Harvard Pilgrim Health Care account) voice: 617-509-1212 fax: 617-509-3737 pager: 781-226-2431