From Kristina.Carlton@JICompanies.com Thu Sep 2 15:35:22 2004 From: Kristina.Carlton@JICompanies.com (Kristina Carlton) Date: Thu, 2 Sep 2004 09:35:22 -0500 Subject: converting numerics Message-ID: <157670322ACDBB4D82690A2A917E6E1B0345A214@jiexch01.jicompanies.com> 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_01C490FA.145821A0 Content-Type: text/plain; charset="iso-8859-1" Hi All, I am writing out data via Quiz extracts. Due to the size of the files we are using the subfiles versus the textfiles and I noticed that my number formats are gone in the subfiles. I tried the PICTURE command but it basically ignores it. So for example, 114.56 looks like +00000011456. When we read this data into our Progress database, it comes in as 11,456. Is the only way to get around this in Quiz to convert the numerics to ASCII and then physically place the decimal? If so, do I have to check for values between 0.01 and 0.09 (my 0.01 came across as ". 1") or am I missing something here? Would it be easier to just use a Progress procedure to convert the numerics before reading them in? Any input is greatly appreciated. Thanks, -Kristina Ext. 2459 ------_=_NextPart_001_01C490FA.145821A0 Content-Type: text/html; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable

Hi All,

 

=

I am writing out data via Quiz extracts. Due to the size of the files we = are using the subfiles versus the textfiles and I noticed that my number = formats are gone in the subfiles. I tried the PICTURE command but it basically = ignores it. So for example, 114.56 looks like +00000011456. When we read this = data into our Progress database, it comes in as 11,456. =

 

=

Is the only way to get around this in Quiz to convert the numerics to = ASCII and then physically place the decimal? If so, do I have to check for values = between 0.01 and 0.09 (my 0.01 came across as “. 1”) or am I = missing something here? Would it be easier to just use a Progress procedure to convert the = numerics before reading them in? Any input is greatly = appreciated.

 

=

Thanks,<= /span>

<= span class=3DEmailStyle15>-Kristina

=

Ext. 2459

 

------_=_NextPart_001_01C490FA.145821A0-- From chris.sharman@ccagroup.co.uk Thu Sep 2 15:59:59 2004 From: chris.sharman@ccagroup.co.uk (Chris Sharman) Date: Thu, 02 Sep 2004 15:59:59 +0100 Subject: [MAYBE SPAM] converting numerics In-Reply-To: <200409021546015389696@equus.ccagroup.co.uk> References: <200409021546015389696@equus.ccagroup.co.uk> Message-ID: <4137356F.1040608@ccagroup.co.uk> Kristina Carlton wrote: > I am writing out data via Quiz extracts. Due to the size of the files we > are using the subfiles versus the textfiles and I noticed that my number > formats are gone in the subfiles. I tried the PICTURE command but it > basically ignores it. So for example, 114.56 looks like +00000011456. > When we read this data into our Progress database, it comes in as 11,456. > > Is the only way to get around this in Quiz to convert the numerics to > ASCII and then physically place the decimal? If so, do I have to check The best ascii conversion routine is formatnumber (present in more recent Powerhouse versions I believe - 7.10F+ for VMS) - this allows you to do just about everything you could want. The ascii function is, as you've noticed, very basic. Chris ----------------------------------------------------------------------- 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.Deskin@Cognos.COM Thu Sep 2 16:13:43 2004 From: Bob.Deskin@Cognos.COM (Deskin, Bob) Date: Thu, 2 Sep 2004 11:13:43 -0400 Subject: [MAYBE SPAM] converting numerics Message-ID: <2B4E386EAEC90947802323BF373DC0EC025920C3@sottemail1.ent.ad.cognos.com> Subfiles are designed to output data to be read by another application (QUIZ, QTP, etc). That's why the data is in an internal format. The external formatting (PIC, etc) is not ignored but is carried along to be used if necessary in a future report. Chris is right about using FORMATNUMBER. It's available in the current PowerHouse Series 8 releases (8.4x). To use it, create a character item DEFINE that uses the FORMATNUMBER and whatever formatting you need. Then put the defined item in the subfile. Bob Deskin Senior 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: powerh-l-admin@lists.sowder.com [mailto:powerh-l-admin@lists.sowder.com] On Behalf Of Chris Sharman Sent: September 2, 2004 11:00 AM To: powerh-l@lists.sowder.com Subject: Re: [MAYBE SPAM] converting numerics Kristina Carlton wrote: > I am writing out data via Quiz extracts. Due to the size of the files > we > are using the subfiles versus the textfiles and I noticed that my number > formats are gone in the subfiles. I tried the PICTURE command but it > basically ignores it. So for example, 114.56 looks like +00000011456. > When we read this data into our Progress database, it comes in as 11,456. > > Is the only way to get around this in Quiz to convert the numerics to > ASCII and then physically place the decimal? If so, do I have to check The best ascii conversion routine is formatnumber (present in more recent Powerhouse versions I believe - 7.10F+ for VMS) - this allows you to do just about everything you could want. The ascii function is, as you've noticed, very basic. Chris ----------------------------------------------------------------------- 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.sowder.com Subscribe: "subscribe" in message body to powerh-l-request@lists.sowder.com Unsubscribe: "unsubscribe " in message body to powerh-l-request@lists.sowder.com http://lists.sowder.com/mailman/listinfo/powerh-l This list is closed, thus to post to the list you must be a subscriber. This message may contain privileged and/or confidential information. If you have received this e-mail in error or are not the intended recipient, you may not use, copy, disseminate or distribute it; do not open any attachments, delete it immediately from your system and notify the sender promptly by e-mail that you have done so. Thank you. From paul_diaconescu@hotmail.com Fri Sep 3 22:23:50 2004 From: paul_diaconescu@hotmail.com (PAUL DIACONESCU) Date: Fri, 03 Sep 2004 17:23:50 -0400 Subject: [MAYBE SPAM] converting numerics Message-ID: You can also use the ascii command. It is easier than you think. For exemple: def d-num = 14356 def d-asc cha*10 = asc(d-num/100) + "." + asc(d-num,16)[15:2] d-asc will be = 143.56 paul diaconescu >From: Chris Sharman >To: powerh-l@lists.sowder.com >Subject: Re: [MAYBE SPAM] converting numerics >Date: Thu, 02 Sep 2004 15:59:59 +0100 > >Kristina Carlton wrote: >>I am writing out data via Quiz extracts. Due to the size of the files we >>are using the subfiles versus the textfiles and I noticed that my number >>formats are gone in the subfiles. I tried the PICTURE command but it >>basically ignores it. So for example, 114.56 looks like +00000011456. When >>we read this data into our Progress database, it comes in as 11,456. >> >>Is the only way to get around this in Quiz to convert the numerics to >>ASCII and then physically place the decimal? If so, do I have to check > >The best ascii conversion routine is formatnumber (present in more recent >Powerhouse versions I believe - 7.10F+ for VMS) - this allows you to do >just about everything you could want. >The ascii function is, as you've noticed, very basic. > >Chris > > > >----------------------------------------------------------------------- > >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.sowder.com >Subscribe: "subscribe" in message body to powerh-l-request@lists.sowder.com >Unsubscribe: "unsubscribe " in message body to >powerh-l-request@lists.sowder.com >http://lists.sowder.com/mailman/listinfo/powerh-l >This list is closed, thus to post to the list you must be a subscriber. _________________________________________________________________ Des mécanismes de contrôle parental puissants permettent à votre enfant de découvrir tout ce qu’Internet a à offrir. http://join.msn.com/?pgmarket=fr-ca&page=features/parental&ST=1&xAPID=1983&DI=2043 Commencez dès maintenant à profiter de tous les avantages de MSN Premium et obtenez les deux premiers mois GRATUITS*. From Kristina.Carlton@JICompanies.com Wed Sep 8 16:24:01 2004 From: Kristina.Carlton@JICompanies.com (Kristina Carlton) Date: Wed, 8 Sep 2004 10:24:01 -0500 Subject: Count Message-ID: <157670322ACDBB4D82690A2A917E6E1B0345A25E@jiexch01.jicompanies.com> 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_01C495B7.DE49D5A8 Content-Type: text/plain; charset="iso-8859-1" Hi All, I am extracting data via Quiz into portable subfiles. The data in the subfiles will then be used to load into a Progress warehouse. I was hoping to get a count of the records I extract with each quiz and write them to a file along with the date. The date is not a problem, but the count is. I was going to access the subfile and simply count the number of records, but got an error message that portable subfiles cannot be accessed. Anybody have a workaround for this? We are on OpenVMS V7.3-2 and Quiz V7.10.G4. Thanks, -Kristina Ext. 2459 ------_=_NextPart_001_01C495B7.DE49D5A8 Content-Type: text/html; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable

Hi All,

 

=

I am extracting data via Quiz into portable subfiles. The data in the = subfiles will then be used to load into a Progress warehouse. I was hoping to = get a count of the records I extract with each quiz and write them to a file = along with the date. The date is not a problem, but the count is. I was going = to access the subfile and simply count the number of records, but got an = error message that portable subfiles cannot be accessed. Anybody have a = workaround for this?

 

=

We are on OpenVMS V7.3-2 and Quiz = V7.10.G4.

 

=

Thanks,<= /span>

<= span class=3DEmailStyle15>-Kristina

=

Ext. 2459

 

------_=_NextPart_001_01C495B7.DE49D5A8-- From dmorrison@mcbrideelectric.com Wed Sep 8 16:42:17 2004 From: dmorrison@mcbrideelectric.com (David Morrison - Corporate) Date: Wed, 8 Sep 2004 08:42:17 -0700 Subject: Count Message-ID: <741A7B80E63EFE498457491B6AE5BBE7262F79@srvrexchange.mcbrideelectric.com> This is a multi-part message in MIME format. ------_=_NextPart_001_01C495BA.6B9527DF Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable Kristina, =20 How about outputting COUNT to each record that you write with your QUIZ = program. The last record written will then have the total count of the = number of records. =20 David Morrison McBride Electric -----Original Message----- From: powerh-l-admin@lists.sowder.com = [mailto:powerh-l-admin@lists.sowder.com]On Behalf Of Kristina Carlton Sent: Wednesday, September 08, 2004 8:24 AM To: 'powerh-l@lists.sowder.com' Subject: Count Hi All, =20 I am extracting data via Quiz into portable subfiles. The data in the = subfiles will then be used to load into a Progress warehouse. I was = hoping to get a count of the records I extract with each quiz and write = them to a file along with the date. The date is not a problem, but the = count is. I was going to access the subfile and simply count the number = of records, but got an error message that portable subfiles cannot be = accessed. Anybody have a workaround for this? =20 We are on OpenVMS V7.3-2 and Quiz V7.10.G4. =20 Thanks, -Kristina Ext. 2459 =20 ------_=_NextPart_001_01C495BA.6B9527DF Content-Type: text/html; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable
Kristina,
 
How = about=20 outputting COUNT to each record that you write with your QUIZ = program.  The=20 last record written will then have the total count of the number of=20 records.
 
David=20 Morrison
McBride=20 Electric
-----Original Message-----
From:=20 powerh-l-admin@lists.sowder.com = [mailto:powerh-l-admin@lists.sowder.com]On=20 Behalf Of Kristina Carlton
Sent: Wednesday, September = 08, 2004=20 8:24 AM
To: 'powerh-l@lists.sowder.com'
Subject:=20 Count

Hi=20 All,

 

I am extracting = data via=20 Quiz into portable subfiles. The data in the subfiles will then be = used to=20 load into a Progress warehouse. I was hoping to get a count of the = records I=20 extract with each quiz and write them to a file along with the date. = The date=20 is not a problem, but the count is. I was going to access the subfile = and=20 simply count the number of records, but got an error message that = portable=20 subfiles cannot be accessed. Anybody have a workaround for=20 this?

 

We are on = OpenVMS V7.3-2=20 and Quiz V7.10.G4.

 

Thanks,

-Kristina

Ext.=20 2459

 

------_=_NextPart_001_01C495BA.6B9527DF-- From Roger.Kesterson@mtdsw.com Wed Sep 8 16:51:38 2004 From: Roger.Kesterson@mtdsw.com (Roger Kesterson) Date: Wed, 8 Sep 2004 08:51:38 -0700 Subject: Count Message-ID: 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_01C495BB.B9B227A0 Content-Type: text/plain; charset="ISO-8859-1" I have never tried this, but can you use PORTABLE and DICTIONARY at the same time to get it to read its own portable subfile? -----Original Message----- From: Kristina Carlton [mailto:Kristina.Carlton@JICompanies.com] Sent: Wednesday, September 08, 2004 8:24 AM To: 'powerh-l@lists.sowder.com' Subject: Count Hi All, I am extracting data via Quiz into portable subfiles. The data in the subfiles will then be used to load into a Progress warehouse. I was hoping to get a count of the records I extract with each quiz and write them to a file along with the date. The date is not a problem, but the count is. I was going to access the subfile and simply count the number of records, but got an error message that portable subfiles cannot be accessed. Anybody have a workaround for this? We are on OpenVMS V7.3-2 and Quiz V7.10.G4. Thanks, -Kristina Ext. 2459 ------_=_NextPart_001_01C495BB.B9B227A0 Content-Type: text/html; charset="ISO-8859-1"
I have never tried this, but can you use PORTABLE and DICTIONARY at the same time to get it to read its own portable subfile?
-----Original Message-----
From: Kristina Carlton [mailto:Kristina.Carlton@JICompanies.com]
Sent: Wednesday, September 08, 2004 8:24 AM
To: 'powerh-l@lists.sowder.com'
Subject: Count

Hi All,

 

I am extracting data via Quiz into portable subfiles. The data in the subfiles will then be used to load into a Progress warehouse. I was hoping to get a count of the records I extract with each quiz and write them to a file along with the date. The date is not a problem, but the count is. I was going to access the subfile and simply count the number of records, but got an error message that portable subfiles cannot be accessed. Anybody have a workaround for this?

 

We are on OpenVMS V7.3-2 and Quiz V7.10.G4.

 

Thanks,

-Kristina

Ext. 2459

 

------_=_NextPart_001_01C495BB.B9B227A0-- From joeboyle_adt@hotmail.com Wed Sep 8 19:01:21 2004 From: joeboyle_adt@hotmail.com (Joe Boyle) Date: Wed, 8 Sep 2004 19:01:21 +0100 Subject: Count In-Reply-To: <157670322ACDBB4D82690A2A917E6E1B0345A25E@jiexch01.jicompanies.com> Message-ID: This is a multi-part message in MIME format. ------=_NextPart_000_0010_01C495D6.3AB4E490 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit it might be worth checking that the .ps and .psd files are in the local directory; if they aren't you can either specify the path in the access statement or set a logical e.g. define /job myportsf sys$scratch:myportsf. Regards, Joe. This e-mail and all information contained in it is confidential and may be legally privileged. If you are not the intended recipient, your access to this e-mail is unauthorized. Any use, dissemination, distribution, publication or copying by you of this e-mail or any of the information contained within it is prohibited and may be unlawful. Do not open any attachments, delete it immediately from your system and notify the sender promptly by e-mail that you have done so. The content of this e-mail and any attachments sent with it may have been altered without the consent or knowledge of the author. _____ From: powerh-l-admin@lists.sowder.com [mailto:powerh-l-admin@lists.sowder.com] On Behalf Of Kristina Carlton Sent: 08 September 2004 16:24 To: 'powerh-l@lists.sowder.com' Subject: Count Hi All, I am extracting data via Quiz into portable subfiles. The data in the subfiles will then be used to load into a Progress warehouse. I was hoping to get a count of the records I extract with each quiz and write them to a file along with the date. The date is not a problem, but the count is. I was going to access the subfile and simply count the number of records, but got an error message that portable subfiles cannot be accessed. Anybody have a workaround for this? We are on OpenVMS V7.3-2 and Quiz V7.10.G4. Thanks, -Kristina Ext. 2459 ------=_NextPart_000_0010_01C495D6.3AB4E490 Content-Type: text/html; charset="us-ascii" Content-Transfer-Encoding: quoted-printable

it might be worth checking that the = ps and .psd files are in the local directory; if they aren’t you can = either specify the path in the access statement or set a logical e.g. define = /job myportsf sys$scratch:myportsf.

Regards, Joe.

This e-mail and all information contained in it is confidential and may be legally privileged. If you are not the = intended recipient, your access to this e-mail is unauthorized. Any use, dissemination, distribution, publication or copying by you of this e-mail or any of the information contained within it is prohibited and may be unlawful. Do = not open any attachments, delete it immediately from your system and notify the = sender promptly by e-mail that you have done so. The content of this e-mail and = any attachments sent with it may have been altered without the consent or = knowledge of the author.

From: powerh-l-admin@lists.sowder.com [mailto:powerh-l-admin@lists.sowder.com] = On Behalf Of Kristina Carlton
Sent: 08 September 2004 = 16:24
To: = 'powerh-l@lists.sowder.com'
Subject: = Count

 

Hi = All,

 

I am extracting = data via Quiz into portable subfiles. The data in the subfiles will then be used = to load into a Progress warehouse. I was hoping to get a count of the records I = extract with each quiz and write them to a file along with the date. The date is = not a problem, but the count is. I was going to access the subfile and simply = count the number of records, but got an error message that portable subfiles = cannot be accessed. Anybody have a workaround for = this?

 

We are on = OpenVMS V7.3-2 and Quiz V7.10.G4.

 

Thanks,

-Kristina

Ext. = 2459

 

------=_NextPart_000_0010_01C495D6.3AB4E490-- From Kristina.Carlton@JICompanies.com Wed Sep 8 21:44:30 2004 From: Kristina.Carlton@JICompanies.com (Kristina Carlton) Date: Wed, 8 Sep 2004 15:44:30 -0500 Subject: reporting items Message-ID: <157670322ACDBB4D82690A2A917E6E1B0345A273@jiexch01.jicompanies.com> 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_01C495E4.A3E0A12C Content-Type: text/plain; charset="iso-8859-1" If I report a defined item in a subfile can I then access the subfile and refer to the previously defined item? I am writing the file name and one of the fields - case_num - to SUBFIL.PS which looks like this: CAS 0000001 CAS 0000002 . . . CAS 0000020 DES 0000001 DES 0000001 . . . and so on I now want to access the subfile and use the previously defined item as a sort item. Is this possible in Quiz? -Kristina Ext. 2459 ------_=_NextPart_001_01C495E4.A3E0A12C Content-Type: text/html; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable

If I report a defined item in a subfile can I then access the subfile and = refer to the previously defined item?

 

=

I am writing the file name and one of the fields – case_num -  to SUBFIL.PS which looks like = this:

 

=

CAS      = 0000001

CAS      = 0000002

.=

.=

.=

CAS      = 0000020

DES      = 0000001

DES      = 0000001

.=

.=

.=

and so on

 

=

I now want to access the subfile and use the previously defined item as a = sort item. Is this possible in Quiz?

 

=

<= span class=3DEmailStyle15>-Kristina

=

Ext. 2459

 

------_=_NextPart_001_01C495E4.A3E0A12C-- From robeconsult@sbcglobal.net Wed Sep 8 22:59:19 2004 From: robeconsult@sbcglobal.net (Robert Edis) Date: Wed, 8 Sep 2004 14:59:19 -0700 (PDT) Subject: PH gig in France Message-ID: <20040908215919.30691.qmail@web81306.mail.yahoo.com> DEC VAX, VMS, Powerhouse, Cobol Developer Company: MCG Job type: Permanent Salary: negotiable salary Location: France Date posted: 7 Sep 2004 16:39 We are seeking a talented programmer to work in France (Lille). The technical skills required for the role include a minimum of 2 years programming experience in Powerhouse and Cobol in a VMS/DCL environment. Ideally you will also have good SAS experience (advantageous but not essential) and you... For more information: http://www.cwjobs.co.uk/joblink.asp?JobID=12354429&GUID=2e6123d4e784102d570218ffa9bef910&InstId=BE6A66892FAB4735B7E5CF875045652D Disclaimer: Blue (me) knows nothing more about this gig other than what's displayed here so please don't email me for more information. OK? From resolutebay@yahoo.com Thu Sep 9 00:05:53 2004 From: resolutebay@yahoo.com (Penney, John) Date: Wed, 8 Sep 2004 16:05:53 -0700 (PDT) Subject: PH gig in France In-Reply-To: <20040908215919.30691.qmail@web81306.mail.yahoo.com> Message-ID: <20040908230553.28070.qmail@web14027.mail.yahoo.com> Blue: And "French as a first language", so all you bilingual QC PowerHousers, go for it! JP A Canuck in Sacto --- Robert Edis wrote: > DEC VAX, VMS, Powerhouse, Cobol Developer > > Company: MCG > Job type: Permanent > Salary: negotiable salary > Location: France > Date posted: 7 Sep 2004 16:39 > > We are seeking a talented programmer to work in > France > (Lille). The > technical skills required for the role include a > minimum of 2 years > programming experience in Powerhouse and Cobol in a > VMS/DCL environment. > Ideally you will also have good SAS experience > (advantageous but not > essential) and you... > > For more information: > http://www.cwjobs.co.uk/joblink.asp?JobID=12354429&GUID=2e6123d4e784102d570218ffa9bef910&InstId=BE6A66892FAB4735B7E5CF875045652D > > Disclaimer: Blue (me) knows nothing more about this > gig other than what's displayed here so please don't > email me for more information. OK? > = = = = = = = = = = = = = = = = = = = = = = = = = = > = = > Mailing list: powerh-l@lists.sowder.com > Subscribe: "subscribe" in message body to > powerh-l-request@lists.sowder.com > Unsubscribe: "unsubscribe " in message > body to powerh-l-request@lists.sowder.com > http://lists.sowder.com/mailman/listinfo/powerh-l > This list is closed, thus to post to the list you > must be a subscriber. > __________________________________ Do you Yahoo!? New and Improved Yahoo! Mail - Send 10MB messages! http://promotions.yahoo.com/new_mail From ppfb@hotmail.com Fri Sep 10 01:48:22 2004 From: ppfb@hotmail.com (Peter Bateman) Date: Thu, 09 Sep 2004 21:48:22 -0300 Subject: Count Message-ID:

You can have a "FINAL FOOTING" as well as a "REPORT SUMMARY"

ACCESS ...

REPORT SUMMARY ...

FINAL FOOTING "DATE " SYSDATE "RECORD COUNT" COUNT NORESET

SET REPORT DEVICE DISC NAME WHATEVER

SET NOHEAD

GO

>From: Roger Kesterson <Roger.Kesterson@mtdsw.com>
>To: "'powerh-l@lists.sowder.com'" <powerh-l@lists.sowder.com>
>Subject: RE: Count
>Date: Wed, 8 Sep 2004 08:51:38 -0700
>
>I have never tried this, but can you use PORTABLE and DICTIONARY at the same
>time to get it to read its own portable subfile?
>-----Original Message-----
>From: Kristina Carlton [mailto:Kristina.Carlton@JICompanies.com]
>Sent: Wednesday, September 08, 2004 8:24 AM
>To: 'powerh-l@lists.sowder.com'
>Subject: Count
>
>
>Hi All,
>
>I am extracting data via Quiz into portable subfiles. The data in the
>subfiles will then be used to load into a Progress warehouse. I was hoping
>to get a count of the records I extract with each quiz and write them to a
>file along with the date. The date is not a problem, but the count is. I was
>going to access the subfile and simply count the number of records, but got
>an error message that portable subfiles cannot be accessed. Anybody have a
>workaround for this?
>
>We are on OpenVMS V7.3-2 and Quiz V7.10.G4.
>
>Thanks,
>-Kristina
>Ext. 2459
>

Take the effort out of being organized with MSN® Calendar. MSN Premium: Join now and get the first two months FREE* From mhooper@mhg.co.za Fri Sep 10 09:28:44 2004 From: mhooper@mhg.co.za (Maeve Hooper) Date: Fri, 10 Sep 2004 10:28:44 +0200 Subject: Record status in quick - resetting Message-ID: <97641B5DAC415F4FB2FF529F792392B903286872@cptexc01.za.mhgad.com> This is a multi-part message in MIME format. ------_=_NextPart_001_01C49710.2F4D0059 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable Hello Does anyone know how to 'fool' quick into thinking that a record has = been written away (put) when it hasn't. I have a program that has a = primary file - various fields are changed on the screen by the user. = Then for various reasons, I use embedded sql to write the record away - = NOT Powerhouse. I don't want the user to get the message "Date changed = but not updated". How do I reset the alteredrecord status ? I've tried = doing a "get" of the record again which resets the status but the values = in the file buffer are 0 - even though the record does exist (I'm in = find mode). I want to leave the file as primary to preserve the screen = functionality. Any ideas would be great as I'm probably going to need to this fairly = often. Maeve Hooper Database Consultant Metropolitan Health Group Ph: (021) 480-6871 Cell: 083 6666 973 Fax: (021) 480 4932 ------_=_NextPart_001_01C49710.2F4D0059 Content-Type: text/html; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable Record status in quick - resetting

Hello

Does anyone know how to 'fool' quick = into thinking that a record has been written away (put) when it hasn't. = I have a program that has a primary file - various fields are changed on = the screen by the user. Then for various reasons, I use embedded sql to = write the record away - NOT Powerhouse. I don't want the user to get the = message "Date changed but not updated". How do I reset the = alteredrecord status ?  I've tried doing a "get" of the = record again which resets the status but the values in the file buffer = are 0 - even though the record does exist (I'm in find mode). I want to = leave the file as primary to preserve the screen = functionality.

Any ideas would be great as I'm = probably going to need to this fairly often.

Maeve Hooper
Database Consultant
Metropolitan Health Group
Ph:  (021) 480-6871
Cell: 083 6666 973
Fax: (021) 480 4932

------_=_NextPart_001_01C49710.2F4D0059-- From oh@scanconsult.dk Fri Sep 10 10:38:25 2004 From: oh@scanconsult.dk (Ole Hansen, ScanConsult) Date: Fri, 10 Sep 2004 11:38:25 +0200 Subject: Record status in quick - resetting In-Reply-To: <97641B5DAC415F4FB2FF529F792392B903286872@cptexc01.za.mhgad.com> Message-ID: <002001c49719$ebea1660$0a0b10ac@scanconsult02.bogus> This is a multi-part message in MIME format. ------=_NextPart_000_0021_01C4972A.AF72E660 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: quoted-printable Hello Paul, =20 I don't know of a way to 'fool' quick like you want, but I would use temporaries for all fields on the screen. =20 Set and display the temporaries in the postfind procedure, and let the = user modify fields as required.=20 =20 Update using your embedded sql - taking the values from the temporaries rather than from the record items. =20 If there are many fields on the screen, this is of cource impractical, = but it will leave the retrieved record status 'unchanged'. =20 Mange hilsner / With kind regards =20 Ole Hansen =20 =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D E-mail: oh@scanconsult.dk Cell Phone: +45 4040 1150 =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D ScanConsult IT-Partners ApS Voice : +45 8738 7474 Jegstrupvej 96A Fax : +45 8738 7475 DK-8361 Hasselager Support : support@scanconsult.dk Denmark Sales : sales@scanconsult.dk http://www.scanconsult.dk =3D ParaSuite : A Suite of Air Cargo Handling solutions =3D =3D ParaSafe : A Suite of IT-Security solutions =3D =3D Partners : Hewlett-Packard, Cognos, Microsoft and more =3D -----Original Message----- From: powerh-l-admin@lists.sowder.com [mailto:powerh-l-admin@lists.sowder.com] On Behalf Of Maeve Hooper Sent: 10. september 2004 10:29 To: powerh-l@lists.sowder.com Cc: Paul Howard (E-mail) Subject: Record status in quick - resetting Hello=20 Does anyone know how to 'fool' quick into thinking that a record has = been written away (put) when it hasn't. I have a program that has a primary = file - various fields are changed on the screen by the user. Then for various reasons, I use embedded sql to write the record away - NOT Powerhouse. I don't want the user to get the message "Date changed but not updated". = How do I reset the alteredrecord status ? I've tried doing a "get" of the record again which resets the status but the values in the file buffer = are 0 - even though the record does exist (I'm in find mode). I want to leave = the file as primary to preserve the screen functionality. Any ideas would be great as I'm probably going to need to this fairly = often. Maeve Hooper=20 Database Consultant=20 Metropolitan Health Group=20 Ph: (021) 480-6871=20 Cell: 083 6666 973=20 Fax: (021) 480 4932=20 ------=_NextPart_000_0021_01C4972A.AF72E660 Content-Type: text/html; charset="us-ascii" Content-Transfer-Encoding: quoted-printable Message
Hello=20 Paul,
 
I=20 don't know of a way to 'fool' quick like you want, but I would use = temporaries=20 for all fields on the screen.
 
Set=20 and display the temporaries in the postfind procedure, and let the user = modify=20 fields as required.
 
Update=20 using your embedded sql - taking the values from the temporaries rather = than=20 from the record items.
 
If=20 there are many fields on the screen, this is of cource impractical, = but it=20 will leave the retrieved record status = 'unchanged'.
 
Mange hilsner / With kind=20 regards
 
Ole Hansen
 
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D
E-mail: oh@scanconsult.dk   =      =20 Cell Phone: +45 4040=20 1150
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D
ScanConsult=20 IT-Partners ApS        =20 Voice   : +45 8738 7474
Jegstrupvej=20 96A           &nbs= p;        =20 Fax     : +45 8738 7475
DK-8361=20 Hasselager         Support : support@scanconsult.dk
Denm= ark           &nbs= p;       =20 Sales   :   sales@scanconsult.dk
 &n= bsp;           &nb= sp;  =20 http://www.scanconsult.dk &n= bsp;           &nb= sp; =20
=3D ParaSuite :  A Suite of Air Cargo Handling=20 solutions    =3D
=3D ParaSafe  :  A Suite of = IT-Security=20 solutions           = =3D
=3D=20 Partners  : Hewlett-Packard, Cognos, Microsoft and more = =3D
-----Original Message-----
From:=20 powerh-l-admin@lists.sowder.com = [mailto:powerh-l-admin@lists.sowder.com] On=20 Behalf Of Maeve Hooper
Sent: 10. september 2004=20 10:29
To: powerh-l@lists.sowder.com
Cc: Paul = Howard=20 (E-mail)
Subject: Record status in quick -=20 resetting



Hello

Does anyone know how to 'fool' quick = into thinking=20 that a record has been written away (put) when it hasn't. I have a = program=20 that has a primary file - various fields are changed on the screen by = the=20 user. Then for various reasons, I use embedded sql to write the record = away -=20 NOT Powerhouse. I don't want the user to get the message "Date changed = but not=20 updated". How do I reset the alteredrecord status ?  I've tried = doing a=20 "get" of the record again which resets the status but the values in = the file=20 buffer are 0 - even though the record does exist (I'm in find mode). I = want to=20 leave the file as primary to preserve the screen = functionality.

Any ideas would be great as I'm = probably going to=20 need to this fairly often.

Maeve Hooper
Database Consultant
Metropolitan=20 Health Group
Ph:  (021)=20 480-6871
Cell: 083 6666 = 973=20
Fax: (021) 480 4932=20

------=_NextPart_000_0021_01C4972A.AF72E660-- From doug.brown@ntlworld.com Fri Sep 10 12:06:28 2004 From: doug.brown@ntlworld.com (Douglas Brown) Date: Fri, 10 Sep 2004 12:06:28 +0100 Subject: Record status in quick - resetting In-Reply-To: <97641B5DAC415F4FB2FF529F792392B903286872@cptexc01.za.mhgad.com> Message-ID: This is a multi-part message in MIME format. ------=_NextPart_000_0026_01C4972E.9A239710 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: 7bit Maeve A trick we used was to do a 'get' in the pre-update procedure thereby clearing the record flags, not sure why your get isn't working. Regards Douglas Brown ---------------- +44.7710-716-147 ------=_NextPart_000_0026_01C4972E.9A239710 Content-Type: text/html; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable
Maeve
 
A=20 trick we used was to do a 'get' in the pre-update procedure thereby = clearing the=20 record flags, not sure why your get isn't working.

Regards
Douglas=20 Brown
----------------
+44.7710-716-147=20

------=_NextPart_000_0026_01C4972E.9A239710-- From joeboyle_adt@hotmail.com Fri Sep 10 12:08:13 2004 From: joeboyle_adt@hotmail.com (Joe Boyle) Date: Fri, 10 Sep 2004 12:08:13 +0100 Subject: Record status in quick - resetting In-Reply-To: <97641B5DAC415F4FB2FF529F792392B903286872@cptexc01.za.mhgad.com> Message-ID: This is a multi-part message in MIME format. ------=_NextPart_000_002A_01C4972E.D9347500 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit how about letting quick do the update in the standard way and then run your SQL - as an update rather than an insert - in the postupdate procedure. Regards, Joe. This e-mail and all information contained in it is confidential and may be legally privileged. If you are not the intended recipient, your access to this e-mail is unauthorized. Any use, dissemination, distribution, publication or copying by you of this e-mail or any of the information contained within it is prohibited and may be unlawful. Do not open any attachments, delete it immediately from your system and notify the sender promptly by e-mail that you have done so. The content of this e-mail and any attachments sent with it may have been altered without the consent or knowledge of the author. _____ From: powerh-l-admin@lists.sowder.com [mailto:powerh-l-admin@lists.sowder.com] On Behalf Of Maeve Hooper Sent: 10 September 2004 09:29 To: powerh-l@lists.sowder.com Cc: Paul Howard (E-mail) Subject: Record status in quick - resetting Hello Does anyone know how to 'fool' quick into thinking that a record has been written away (put) when it hasn't. I have a program that has a primary file - various fields are changed on the screen by the user. Then for various reasons, I use embedded sql to write the record away - NOT Powerhouse. I don't want the user to get the message "Date changed but not updated". How do I reset the alteredrecord status ? I've tried doing a "get" of the record again which resets the status but the values in the file buffer are 0 - even though the record does exist (I'm in find mode). I want to leave the file as primary to preserve the screen functionality. Any ideas would be great as I'm probably going to need to this fairly often. Maeve Hooper Database Consultant Metropolitan Health Group Ph: (021) 480-6871 Cell: 083 6666 973 Fax: (021) 480 4932 ------=_NextPart_000_002A_01C4972E.D9347500 Content-Type: text/html; charset="us-ascii" Content-Transfer-Encoding: quoted-printable Record status in quick - resetting

how about letting quick do the = update in the standard way and then run your SQL - as an update rather than an = insert - in the postupdate procedure.

 

Regards, Joe.

This e-mail and all information contained in it is confidential and may be legally privileged. If you are not the = intended recipient, your access to this e-mail is unauthorized. Any use, dissemination, distribution, publication or copying by you of this e-mail or any of the information contained within it is prohibited and may be unlawful. Do = not open any attachments, delete it immediately from your system and notify the = sender promptly by e-mail that you have done so. The content of this e-mail and = any attachments sent with it may have been altered without the consent or = knowledge of the author.

From: powerh-l-admin@lists.sowder.com [mailto:powerh-l-admin@lists.sowder.com] = On Behalf Of Maeve Hooper
Sent: 10 September 2004 = 09:29
To: = powerh-l@lists.sowder.com
Cc: Paul Howard = (E-mail)
Subject: Record status in = quick - resetting

 

 

Hello

Does anyone know how to 'fool' quick into thinking that a record has been = written away (put) when it hasn't. I have a program that has a primary file - = various fields are changed on the screen by the user. Then for various reasons, = I use embedded sql to write the record away - NOT Powerhouse. I don't want the = user to get the message "Date changed but not updated". How do I = reset the alteredrecord status ?  I've tried doing a "get" of the = record again which resets the status but the values in the file buffer are 0 - = even though the record does exist (I'm in find mode). I want to leave the = file as primary to preserve the screen = functionality.

Any ideas would be great as I'm probably going to need to this fairly = often.

Maeve Hooper
Database Consultant
Metropolitan Health Group
Ph:  (021) 480-6871
Cell: 083 6666 973
Fax: (021) 480 4932

------=_NextPart_000_002A_01C4972E.D9347500-- From Bob.Deskin@Cognos.COM Fri Sep 10 13:11:59 2004 From: Bob.Deskin@Cognos.COM (Deskin, Bob) Date: Fri, 10 Sep 2004 08:11:59 -0400 Subject: Record status in quick - resetting Message-ID: <2B4E386EAEC90947802323BF373DC0EC025920F5@sottemail1.ent.ad.cognos.com> This is a multi-part message in MIME format. ------_=_NextPart_001_01C4972F.5F9FDF20 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable Is it possible that the transaction is not committed at the time you do = your GET? If it isn't then the GET will not "see" the new row. You may = be able to use the COMMIT verb but if you need your transaction to = continue, then I suggest one of the other techniques that have been = mentioned. Bob Deskin =20 Senior Product Manager, Application Development Tools=20 Cognos Inc. 3755 Riverside Drive, Ottawa ON K1G 4K9 CANADA=20 bob.deskin@cognos.com (613) 738-1338 ext 7268=20 -----Original Message----- From: powerh-l-admin@lists.sowder.com = [mailto:powerh-l-admin@lists.sowder.com]On Behalf Of Maeve Hooper Sent: September 10, 2004 4:29 AM To: powerh-l@lists.sowder.com Cc: Paul Howard (E-mail) Subject: Record status in quick - resetting Hello=20 Does anyone know how to 'fool' quick into thinking that a record has = been written away (put) when it hasn't. I have a program that has a = primary file - various fields are changed on the screen by the user. = Then for various reasons, I use embedded sql to write the record away - = NOT Powerhouse. I don't want the user to get the message "Date changed = but not updated". How do I reset the alteredrecord status ? I've tried = doing a "get" of the record again which resets the status but the values = in the file buffer are 0 - even though the record does exist (I'm in = find mode). I want to leave the file as primary to preserve the screen = functionality. Any ideas would be great as I'm probably going to need to this fairly = often.=20 Maeve Hooper=20 Database Consultant=20 Metropolitan Health Group=20 Ph: (021) 480-6871=20 Cell: 083 6666 973=20 Fax: (021) 480 4932=20 =20 This message may contain privileged and/or confidential = information. If you have received this e-mail in error or are not the = intended recipient, you may not use, copy, disseminate or distribute it; = do not open any attachments, delete it immediately from your system and = notify the sender promptly by e-mail that you have done so. Thank you.=20 =20 ------_=_NextPart_001_01C4972F.5F9FDF20 Content-Type: text/html; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable Record status in quick - resetting
Is it=20 possible that the transaction is not committed at the time you do your = GET? If=20 it isn't then the GET will not "see" the new row. You may be able to use = the=20 COMMIT verb but if you need your transaction to continue, then I suggest = one of=20 the other techniques that have been mentioned.

Bob=20 Deskin           &= nbsp; =20
Senior Product=20 Manager, Application Development Tools
Cognos=20 Inc. 3755=20 Riverside Drive, Ottawa ON K1G 4K9 CANADA
bob.deskin@cognos.com = (613) 738-1338 ext 7268=20

-----Original Message-----
From:=20 powerh-l-admin@lists.sowder.com = [mailto:powerh-l-admin@lists.sowder.com]On=20 Behalf Of Maeve Hooper
Sent: September 10, 2004 4:29=20 AM
To: powerh-l@lists.sowder.com
Cc: Paul Howard=20 (E-mail)
Subject: Record status in quick -=20 resetting



Hello

Does anyone know how to 'fool' quick = into thinking=20 that a record has been written away (put) when it hasn't. I have a = program=20 that has a primary file - various fields are changed on the screen by = the=20 user. Then for various reasons, I use embedded sql to write the record = away -=20 NOT Powerhouse. I don't want the user to get the message "Date changed = but not=20 updated". How do I reset the alteredrecord status ?  I've tried = doing a=20 "get" of the record again which resets the status but the values in = the file=20 buffer are 0 - even though the record does exist (I'm in find mode). I = want to=20 leave the file as primary to preserve the screen = functionality.

Any ideas would be great as I'm = probably going to=20 need to this fairly often.

Maeve Hooper
Database Consultant
Metropolitan=20 Health Group
Ph:  (021)=20 480-6871
Cell: 083 6666 = 973=20
Fax: (021) 480 4932=20

 

       This message may contain privileged and/or confidential = information.  If you have received this e-mail in error or are not = the intended recipient, you may not use, copy, disseminate or distribute = it; do not open any attachments, delete it immediately from your system = and notify the sender promptly by e-mail that you have done so.  = Thank you.

------_=_NextPart_001_01C4972F.5F9FDF20-- From ppfb@hotmail.com Fri Sep 10 17:44:48 2004 From: ppfb@hotmail.com (Peter Bateman) Date: Fri, 10 Sep 2004 13:44:48 -0300 Subject: Record status in quick - resetting Message-ID:

Depending on your transaction model and your database you may have to

commit not only the transaction in which you did your non-PowerHouse SQL update but

also Quick's "query" transaction.

This may lead to the lost of the pointer to the next row

and hence a need to restart the find. 

Regards,

Peter Bateman

>From: "Deskin, Bob" <Bob.Deskin@cognos.com>
>To: "Maeve Hooper" <mhooper@mhg.co.za>, <powerh-l@lists.sowder.com>
>CC: "Paul Howard (E-mail)" <paul@synergy.co.za>
>Subject: RE: Record status in quick - resetting
>Date: Fri, 10 Sep 2004 08:11:59 -0400
>
>Is it possible that the transaction is not committed at the time you do your GET? If it isn't then the GET will not "see" the new row. You may be able to use the COMMIT verb but if you need your transaction to continue, then I suggest one of the other techniques that have been mentioned.
>Bob Deskin
>Senior Product Manager, Application Development Tools
>Cognos Inc. 3755 Riverside Drive, Ottawa ON K1G 4K9 CANADA
>bob.deskin@cognos.com (613) 738-1338 ext 7268
>
>-----Original Message-----
>From: powerh-l-admin@lists.sowder.com [mailto:powerh-l-admin@lists.sowder.com]On Behalf Of Maeve Hooper
>Sent: September 10, 2004 4:29 AM
>To: powerh-l@lists.sowder.com
>Cc: Paul Howard (E-mail)
>Subject: Record status in quick - resetting
>
>
>
>
>
>Hello
>
>Does anyone know how to 'fool' quick into thinking that a record has been written away (put) when it hasn't. I have a program that has a primary file - various fields are changed on the screen by the user. Then for various reasons, I use embedded sql to write the record away - NOT Powerhouse. I don't want the user to get the message "Date changed but not updated". How do I reset the alteredrecord status ?  I've tried doing a "get" of the record again which resets the status but the values in the file buffer are 0 - even though the record does exist (I'm in find mode). I want to leave the file as primary to preserve the screen functionality.
>
>Any ideas would be great as I'm probably going to need to this fairly often.
>
>Maeve Hooper
>Database Consultant
>Metropolitan Health Group
>Ph:  (021) 480-6871
>Cell: 083 6666 973
>Fax: (021) 480 4932
>
>        This message may contain privileged and/or confidential information.  If you have received this e-mail in error or are not the intended recipient, you may not use, copy, disseminate or distribute it; do not open any attachments, delete it immediately from your system and notify the sender promptly by e-mail that you have done so.  Thank you.
>

Share a single photo or an entire slide show right inside your e-mail with MSN Premium: Join now and get the first two months FREE* From mhooper@mhg.co.za Mon Sep 13 14:23:31 2004 From: mhooper@mhg.co.za (Maeve Hooper) Date: Mon, 13 Sep 2004 15:23:31 +0200 Subject: powerh-l digest, Vol 1 #1046 - 3 msgs Message-ID: <97641B5DAC415F4FB2FF529F792392B903286882@cptexc01.za.mhgad.com> Thanks to all for the suggestions. -----Original Message----- From: powerh-l-admin@lists.sowder.com [mailto:powerh-l-admin@lists.sowder.com]On Behalf Of powerh-l-request@lists.sowder.com Sent: Saturday, September 11, 2004 1:10 PM To: powerh-l@lists.sowder.com Subject: powerh-l digest, Vol 1 #1046 - 3 msgs Send powerh-l mailing list submissions to powerh-l@lists.sowder.com To subscribe or unsubscribe via the World Wide Web, visit http://lists.sowder.com/mailman/listinfo/powerh-l or, via email, send a message with subject or body 'help' to powerh-l-request@lists.sowder.com You can reach the person managing the list at powerh-l-admin@lists.sowder.com When replying, please edit your Subject line so it is more specific than "Re: Contents of powerh-l digest..." Today's Topics: 1. RE:Record status in quick - resetting (Joe Boyle) 2. RE:Record status in quick - resetting (Deskin, Bob) 3. RE:Record status in quick - resetting (Peter Bateman) --__--__-- Message: 1 From: "Joe Boyle" To: "'Maeve Hooper'" , Cc: "'Paul Howard \(E-mail\)'" Subject: RE: Record status in quick - resetting Date: Fri, 10 Sep 2004 12:08:13 +0100 This is a multi-part message in MIME format. ------=_NextPart_000_002A_01C4972E.D9347500 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit how about letting quick do the update in the standard way and then run your SQL - as an update rather than an insert - in the postupdate procedure. Regards, Joe. This e-mail and all information contained in it is confidential and may be legally privileged. If you are not the intended recipient, your access to this e-mail is unauthorized. Any use, dissemination, distribution, publication or copying by you of this e-mail or any of the information contained within it is prohibited and may be unlawful. Do not open any attachments, delete it immediately from your system and notify the sender promptly by e-mail that you have done so. The content of this e-mail and any attachments sent with it may have been altered without the consent or knowledge of the author. _____ From: powerh-l-admin@lists.sowder.com [mailto:powerh-l-admin@lists.sowder.com] On Behalf Of Maeve Hooper Sent: 10 September 2004 09:29 To: powerh-l@lists.sowder.com Cc: Paul Howard (E-mail) Subject: Record status in quick - resetting Hello Does anyone know how to 'fool' quick into thinking that a record has been written away (put) when it hasn't. I have a program that has a primary file - various fields are changed on the screen by the user. Then for various reasons, I use embedded sql to write the record away - NOT Powerhouse. I don't want the user to get the message "Date changed but not updated". How do I reset the alteredrecord status ? I've tried doing a "get" of the record again which resets the status but the values in the file buffer are 0 - even though the record does exist (I'm in find mode). I want to leave the file as primary to preserve the screen functionality. Any ideas would be great as I'm probably going to need to this fairly often. Maeve Hooper Database Consultant Metropolitan Health Group Ph: (021) 480-6871 Cell: 083 6666 973 Fax: (021) 480 4932 ------=_NextPart_000_002A_01C4972E.D9347500 Content-Type: text/html; charset="us-ascii" Content-Transfer-Encoding: quoted-printable Record status in quick - resetting

how about letting quick do the = update in the standard way and then run your SQL - as an update rather than an = insert - in the postupdate procedure.

 

Regards, Joe.

This e-mail and all information contained in it is confidential and may be legally privileged. If you are not the = intended recipient, your access to this e-mail is unauthorized. Any use, dissemination, distribution, publication or copying by you of this e-mail or any of the information contained within it is prohibited and may be unlawful. Do = not open any attachments, delete it immediately from your system and notify the = sender promptly by e-mail that you have done so. The content of this e-mail and = any attachments sent with it may have been altered without the consent or = knowledge of the author.

From: powerh-l-admin@lists.sowder.com [mailto:powerh-l-admin@lists.sowder.com] = On Behalf Of Maeve Hooper
Sent: 10 September 2004 = 09:29
To: = powerh-l@lists.sowder.com
Cc: Paul Howard = (E-mail)
Subject: Record status in = quick - resetting

 

 

Hello

Does anyone know how to 'fool' quick into thinking that a record has been = written away (put) when it hasn't. I have a program that has a primary file - = various fields are changed on the screen by the user. Then for various reasons, = I use embedded sql to write the record away - NOT Powerhouse. I don't want the = user to get the message "Date changed but not updated". How do I = reset the alteredrecord status ?  I've tried doing a "get" of the = record again which resets the status but the values in the file buffer are 0 - = even though the record does exist (I'm in find mode). I want to leave the = file as primary to preserve the screen = functionality.

Any ideas would be great as I'm probably going to need to this fairly = often.

Maeve Hooper
Database Consultant
Metropolitan Health Group
Ph:  (021) 480-6871
Cell: 083 6666 973
Fax: (021) 480 4932

------=_NextPart_000_002A_01C4972E.D9347500-- --__--__-- Message: 2 Subject: RE: Record status in quick - resetting Date: Fri, 10 Sep 2004 08:11:59 -0400 From: "Deskin, Bob" To: "Maeve Hooper" , Cc: "Paul Howard \(E-mail\)" This is a multi-part message in MIME format. ------_=_NextPart_001_01C4972F.5F9FDF20 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable Is it possible that the transaction is not committed at the time you do = your GET? If it isn't then the GET will not "see" the new row. You may = be able to use the COMMIT verb but if you need your transaction to = continue, then I suggest one of the other techniques that have been = mentioned. Bob Deskin =20 Senior Product Manager, Application Development Tools=20 Cognos Inc. 3755 Riverside Drive, Ottawa ON K1G 4K9 CANADA=20 bob.deskin@cognos.com (613) 738-1338 ext 7268=20 -----Original Message----- From: powerh-l-admin@lists.sowder.com = [mailto:powerh-l-admin@lists.sowder.com]On Behalf Of Maeve Hooper Sent: September 10, 2004 4:29 AM To: powerh-l@lists.sowder.com Cc: Paul Howard (E-mail) Subject: Record status in quick - resetting Hello=20 Does anyone know how to 'fool' quick into thinking that a record has = been written away (put) when it hasn't. I have a program that has a = primary file - various fields are changed on the screen by the user. = Then for various reasons, I use embedded sql to write the record away - = NOT Powerhouse. I don't want the user to get the message "Date changed = but not updated". How do I reset the alteredrecord status ? I've tried = doing a "get" of the record again which resets the status but the values = in the file buffer are 0 - even though the record does exist (I'm in = find mode). I want to leave the file as primary to preserve the screen = functionality. Any ideas would be great as I'm probably going to need to this fairly = often.=20 Maeve Hooper=20 Database Consultant=20 Metropolitan Health Group=20 Ph: (021) 480-6871=20 Cell: 083 6666 973=20 Fax: (021) 480 4932=20 =20 This message may contain privileged and/or confidential = information. If you have received this e-mail in error or are not the = intended recipient, you may not use, copy, disseminate or distribute it; = do not open any attachments, delete it immediately from your system and = notify the sender promptly by e-mail that you have done so. Thank you.=20 =20 ------_=_NextPart_001_01C4972F.5F9FDF20 Content-Type: text/html; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable Record status in quick - resetting
Is it=20 possible that the transaction is not committed at the time you do your = GET? If=20 it isn't then the GET will not "see" the new row. You may be able to use = the=20 COMMIT verb but if you need your transaction to continue, then I suggest = one of=20 the other techniques that have been mentioned.

Bob=20 Deskin           &= nbsp; =20
Senior Product=20 Manager, Application Development Tools
Cognos=20 Inc. 3755=20 Riverside Drive, Ottawa ON K1G 4K9 CANADA
bob.deskin@cognos.com = (613) 738-1338 ext 7268=20

-----Original Message-----
From:=20 powerh-l-admin@lists.sowder.com = [mailto:powerh-l-admin@lists.sowder.com]On=20 Behalf Of Maeve Hooper
Sent: September 10, 2004 4:29=20 AM
To: powerh-l@lists.sowder.com
Cc: Paul Howard=20 (E-mail)
Subject: Record status in quick -=20 resetting



Hello

Does anyone know how to 'fool' quick = into thinking=20 that a record has been written away (put) when it hasn't. I have a = program=20 that has a primary file - various fields are changed on the screen by = the=20 user. Then for various reasons, I use embedded sql to write the record = away -=20 NOT Powerhouse. I don't want the user to get the message "Date changed = but not=20 updated". How do I reset the alteredrecord status ?  I've tried = doing a=20 "get" of the record again which resets the status but the values in = the file=20 buffer are 0 - even though the record does exist (I'm in find mode). I = want to=20 leave the file as primary to preserve the screen = functionality.

Any ideas would be great as I'm = probably going to=20 need to this fairly often.

Maeve Hooper
Database Consultant
Metropolitan=20 Health Group
Ph:  (021)=20 480-6871
Cell: 083 6666 = 973=20
Fax: (021) 480 4932=20

 

       This message may contain privileged and/or confidential = information.  If you have received this e-mail in error or are not = the intended recipient, you may not use, copy, disseminate or distribute = it; do not open any attachments, delete it immediately from your system = and notify the sender promptly by e-mail that you have done so.  = Thank you.

------_=_NextPart_001_01C4972F.5F9FDF20-- --__--__-- Message: 3 From: "Peter Bateman" To: powerh-l@lists.sowder.com Cc: paul@synergy.co.za Subject: RE: Record status in quick - resetting Date: Fri, 10 Sep 2004 13:44:48 -0300

Depending on your transaction model and your database you may have to

commit not only the transaction in which you did your non-PowerHouse SQL update but

also Quick's "query" transaction.

This may lead to the lost of the pointer to the next row

and hence a need to restart the find. 

Regards,

Peter Bateman

>From: "Deskin, Bob" <Bob.Deskin@cognos.com>
>To: "Maeve Hooper" <mhooper@mhg.co.za>, <powerh-l@lists.sowder.com>
>CC: "Paul Howard (E-mail)" <paul@synergy.co.za>
>Subject: RE: Record status in quick - resetting
>Date: Fri, 10 Sep 2004 08:11:59 -0400
>
>Is it possible that the transaction is not committed at the time you do your GET? If it isn't then the GET will not "see" the new row. You may be able to use the COMMIT verb but if you need your transaction to continue, then I suggest one of the other techniques that have been mentioned.
>Bob Deskin
>Senior Product Manager, Application Development Tools
>Cognos Inc. 3755 Riverside Drive, Ottawa ON K1G 4K9 CANADA
>bob.deskin@cognos.com (613) 738-1338 ext 7268
>
>-----Original Message-----
>From: powerh-l-admin@lists.sowder.com [mailto:powerh-l-admin@lists.sowder.com]On Behalf Of Maeve Hooper
>Sent: September 10, 2004 4:29 AM
>To: powerh-l@lists.sowder.com
>Cc: Paul Howard (E-mail)
>Subject: Record status in quick - resetting
>
>
>
>
>
>Hello
>
>Does anyone know how to 'fool' quick into thinking that a record has been written away (put) when it hasn't. I have a program that has a primary file - various fields are changed on the screen by the user. Then for various reasons, I use embedded sql to write the record away - NOT Powerhouse. I don't want the user to get the message "Date changed but not updated". How do I reset the alteredrecord status ?  I've tried doing a "get" of the record again which resets the status but the values in the file buffer are 0 - even though the record does exist (I'm in find mode). I want to leave the file as primary to preserve the screen functionality.
>
>Any ideas would be great as I'm probably going to need to this fairly often.
>
>Maeve Hooper
>Database Consultant
>Metropolitan Health Group
>Ph:  (021) 480-6871
>Cell: 083 6666 973
>Fax: (021) 480 4932
>
>        This message may contain privileged and/or confidential information.  If you have received this e-mail in error or are not the intended recipient, you may not use, copy, disseminate or distribute it; do not open any attachments, delete it immediately from your system and notify the sender promptly by e-mail that you have done so.  Thank you.
>

Share a single photo or an entire slide show right inside your e-mail with MSN Premium: Join now and get the first two months FREE* --__--__-- = = = = = = = = = = = = = = = = = = = = = = = = = = = = Mailing list: powerh-l@lists.sowder.com Subscribe: "subscribe" in message body to powerh-l-request@lists.sowder.com Unsubscribe: "unsubscribe " in message body to powerh-l-request@lists.sowder.com http://lists.sowder.com/mailman/listinfo/powerh-l This list is closed, thus to post to the list you must be a subscriber. End of powerh-l Digest From Kristina.Carlton@JICompanies.com Mon Sep 13 17:01:30 2004 From: Kristina.Carlton@JICompanies.com (Kristina Carlton) Date: Mon, 13 Sep 2004 11:01:30 -0500 Subject: Hotfiles in Impromptu Message-ID: <157670322ACDBB4D82690A2A917E6E1B0345A2D5@jiexch01.jicompanies.com> 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_01C499AA.EF3520C8 Content-Type: text/plain; charset="iso-8859-1" Can someone help me with changing a hot file in Impromptu? We are running Cognos Impromptu Administrator version 7.1.339.0 I have a hot file that needs to be updated. When I try to open the hot file with impromptu I don't have an option to open .ims files (the hotfile). So I created a new one with the same columns, the same file name, etc. and copied it to the directory where the current hot file is saved. When I run the report I get the following error message: DMS-E-RBI-COLULMN, Column c21 was not found. I tried an 'open with' on the file and selected Impromptu but get an error message that the file is not a report. So my dilemma is I can't really look at the old hot file to see exactly what it looks like and creating a new one does not seem to work with existing reports - only new ones. Any suggestions are appreciated. Thanks, -Kristina Ext. 2459 ------_=_NextPart_001_01C499AA.EF3520C8 Content-Type: text/html; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable

Can someone help me with changing a = hot file in Impromptu?

We are running Cognos Impromptu = Administrator version 7.1.339.0

 

I have a hot file that needs to be = updated.

When I try to open the hot file = with impromptu I don’t have an option to open .ims files (the = hotfile).

So I created a new one with the = same columns, the same file name, etc. and copied it to the directory where the = current hot file is saved.

When I run the report I get the = following error message:

 

=

DMS-E-RBI-COLULMN, Column c21 was = not found.

 

=

I tried an ‘open with’ on the file and selected Impromptu but = get an error message that the file is not a report. So my dilemma is I can’t = really look at the old hot file to see exactly what it looks like and creating a new = one does not seem to work with existing reports – only new = ones.

 <= /span>

Any suggestions are appreciated.

 <= /span>

Thanks,

<= span class=3DEmailStyle15>-Kristina

=

Ext. 2459

 

------_=_NextPart_001_01C499AA.EF3520C8-- From Jari.Kaljunen@teliasonera.com Wed Sep 15 16:40:30 2004 From: Jari.Kaljunen@teliasonera.com (Jari.Kaljunen@teliasonera.com) Date: Wed, 15 Sep 2004 17:40:30 +0200 Subject: write to file with quick Message-ID: Hello, I am writing to File: ASC600 Organization: SEQUENTIAL Type: RMS Open: ASC600 Record Format: Variable Supersede: No Records & Sizes: ASC600 600 bytes Item Type Size Occ Offset REC600 CHARACTER 600 in quick using let and put as follows let rec600 = "" + asc(arnr of arenden,8) + asc(parmblnr,2) & + atgkod of atgkoder + "" put asc600 reset The performance is very poor. The administrator believs that quick closes and opens the file all the time. Can that be true? What should I do? We are creating an xml-file (it will be about 5 GB) with data from several tables and our solution with qtp needed too many subfiles and sortings using lots of space. /Jari From Harold.A.Johnson@gems1.gov.bc.ca Wed Sep 15 16:49:39 2004 From: Harold.A.Johnson@gems1.gov.bc.ca (Johnson, Harold A EDUC:EX) Date: Wed, 15 Sep 2004 08:49:39 -0700 Subject: write to file with quick Message-ID: Are you sure the performance problem is with the writing and not with the reading of the original data? I would suspect a problem there instead. -----Original Message----- From: powerh-l-admin@lists.sowder.com [mailto:powerh-l-admin@lists.sowder.com]On Behalf Of Jari.Kaljunen@teliasonera.com Sent: 2004 September 15 8:41 AM To: powerh-l@lists.sowder.com Subject: write to file with quick Hello, I am writing to File: ASC600 Organization: SEQUENTIAL Type: RMS Open: ASC600 Record Format: Variable Supersede: No Records & Sizes: ASC600 600 bytes Item Type Size Occ Offset REC600 CHARACTER 600 in quick using let and put as follows let rec600 = "" + asc(arnr of arenden,8) + asc(parmblnr,2) & + atgkod of atgkoder + "" put asc600 reset The performance is very poor. The administrator believs that quick closes and opens the file all the time. Can that be true? What should I do? We are creating an xml-file (it will be about 5 GB) with data from several tables and our solution with qtp needed too many subfiles and sortings using lots of space. /Jari = = = = = = = = = = = = = = = = = = = = = = = = = = = = Mailing list: powerh-l@lists.sowder.com Subscribe: "subscribe" in message body to powerh-l-request@lists.sowder.com Unsubscribe: "unsubscribe " in message body to powerh-l-request@lists.sowder.com http://lists.sowder.com/mailman/listinfo/powerh-l This list is closed, thus to post to the list you must be a subscriber. From Bob.Deskin@Cognos.COM Wed Sep 15 19:07:57 2004 From: Bob.Deskin@Cognos.COM (Deskin, Bob) Date: Wed, 15 Sep 2004 14:07:57 -0400 Subject: write to file with quick Message-ID: <2B4E386EAEC90947802323BF373DC0EC0259212B@sottemail1.ent.ad.cognos.com> Without seeing the code, we can't tell exactly what's happening. We do not open and close the file for every write by default, however we may be posting an end of file. There are also file locks to consider. What you may be getting is file contention in a multi-user environment. Bob Deskin Senior Product Manager, Application Development Tools Cognos Inc. 3755 Riverside Drive, Ottawa ON K1G 4K9 CANADA bob.deskin@cognos.com (613) 738-1338 ext 7268 -----Original Message----- From: powerh-l-admin@lists.sowder.com [mailto:powerh-l-admin@lists.sowder.com]On Behalf Of Johnson, Harold A EDUC:EX Sent: September 15, 2004 11:50 AM To: 'Jari.Kaljunen@teliasonera.com'; powerh-l@lists.sowder.com Subject: RE: write to file with quick Are you sure the performance problem is with the writing and not with the reading of the original data? I would suspect a problem there instead. -----Original Message----- From: powerh-l-admin@lists.sowder.com [mailto:powerh-l-admin@lists.sowder.com]On Behalf Of Jari.Kaljunen@teliasonera.com Sent: 2004 September 15 8:41 AM To: powerh-l@lists.sowder.com Subject: write to file with quick Hello, I am writing to File: ASC600 Organization: SEQUENTIAL Type: RMS Open: ASC600 Record Format: Variable Supersede: No Records & Sizes: ASC600 600 bytes Item Type Size Occ Offset REC600 CHARACTER 600 in quick using let and put as follows let rec600 = "" + asc(arnr of arenden,8) + asc(parmblnr,2) & + atgkod of atgkoder + "" put asc600 reset The performance is very poor. The administrator believs that quick closes and opens the file all the time. Can that be true? What should I do? We are creating an xml-file (it will be about 5 GB) with data from several tables and our solution with qtp needed too many subfiles and sortings using lots of space. /Jari = = = = = = = = = = = = = = = = = = = = = = = = = = = = Mailing list: powerh-l@lists.sowder.com Subscribe: "subscribe" in message body to powerh-l-request@lists.sowder.com Unsubscribe: "unsubscribe " in message body to powerh-l-request@lists.sowder.com http://lists.sowder.com/mailman/listinfo/powerh-l This list is closed, thus to post to the list you must be a subscriber. = = = = = = = = = = = = = = = = = = = = = = = = = = = = Mailing list: powerh-l@lists.sowder.com Subscribe: "subscribe" in message body to powerh-l-request@lists.sowder.com Unsubscribe: "unsubscribe " in message body to powerh-l-request@lists.sowder.com http://lists.sowder.com/mailman/listinfo/powerh-l This list is closed, thus to post to the list you must be a subscriber. This message may contain privileged and/or confidential information. If you have received this e-mail in error or are not the intended recipient, you may not use, copy, disseminate or distribute it; do not open any attachments, delete it immediately from your system and notify the sender promptly by e-mail that you have done so. Thank you. From Jari.Kaljunen@teliasonera.com Wed Sep 15 19:19:40 2004 From: Jari.Kaljunen@teliasonera.com (Jari.Kaljunen@teliasonera.com) Date: Wed, 15 Sep 2004 20:19:40 +0200 Subject: SV: write to file with quick Message-ID: I am not sure of anything. We have not been running this kind of quick in batch before. I am reading a subfile sequential and get the data from RDB with indexes which are normally very fast in qtp and quiz. Anyhow my senior OpenVms administrator believes (after examination of system while execution of the batch) that the file is opened and closed all the time and normally he has right. It is possible to see both the used and allocated disc space with "dir /size=all". Normally when creating a subfile only the allocated disc space is shown until the qtp or quiz is ready and closes the file. /Jari -----Ursprungligt meddelande----- Från: Johnson, Harold A EDUC:EX [mailto:Harold.A.Johnson@gems1.gov.bc.ca] Skickat: den 15 september 2004 17:50 Till: Kaljunen, Jari S. /TSS - Sergel Kredittjänster /+46-8-4526435; powerh-l@lists.sowder.com Ämne: RE: write to file with quick Are you sure the performance problem is with the writing and not with the reading of the original data? I would suspect a problem there instead. -----Original Message----- From: powerh-l-admin@lists.sowder.com [mailto:powerh-l-admin@lists.sowder.com]On Behalf Of Jari.Kaljunen@teliasonera.com Sent: 2004 September 15 8:41 AM To: powerh-l@lists.sowder.com Subject: write to file with quick Hello, I am writing to File: ASC600 Organization: SEQUENTIAL Type: RMS Open: ASC600 Record Format: Variable Supersede: No Records & Sizes: ASC600 600 bytes Item Type Size Occ Offset REC600 CHARACTER 600 in quick using let and put as follows let rec600 = "" + asc(arnr of arenden,8) + asc(parmblnr,2) & + atgkod of atgkoder + "" put asc600 reset The performance is very poor. The administrator believs that quick closes and opens the file all the time. Can that be true? What should I do? We are creating an xml-file (it will be about 5 GB) with data from several tables and our solution with qtp needed too many subfiles and sortings using lots of space. /Jari = = = = = = = = = = = = = = = = = = = = = = = = = = = = Mailing list: powerh-l@lists.sowder.com Subscribe: "subscribe" in message body to powerh-l-request@lists.sowder.com Unsubscribe: "unsubscribe " in message body to powerh-l-request@lists.sowder.com http://lists.sowder.com/mailman/listinfo/powerh-l This list is closed, thus to post to the list you must be a subscriber. From Jari.Kaljunen@teliasonera.com Thu Sep 16 11:22:20 2004 From: Jari.Kaljunen@teliasonera.com (Jari.Kaljunen@teliasonera.com) Date: Thu, 16 Sep 2004 12:22:20 +0200 Subject: SV: write to file with quick Message-ID: Hello again, a little test without reading database still very slow (actually no difference...). It takes 49 s to run this with 100 records on subfile. The result file is 20100 lines. I even wrote a test quiz reading the same subfile and having 200 defines, each including systime and some more, each defined item written on an own line. It takes quiz 3 s to write these 20000 lines. /Jari screen test_writing_file menu stopscreen file *ik006002 designer file asc600 designer temp t_count procedure internal write begin for 200 begin let t_count = t_count + 1 let rec600 = asc(t_count) & + " " + asc(systime) + " procedure internal write " put asc600 reset end end procedure initialize begin let t_count = 0 while retr ik006002 seq begin let rec600 = asc(t_count) & + " " + asc(systime) + " procedure initialize " put asc600 reset do internal write end return end build -----Ursprungligt meddelande----- Från: Johnson, Harold A EDUC:EX [mailto:Harold.A.Johnson@gems1.gov.bc.ca] Skickat: den 15 september 2004 17:50 Till: Kaljunen, Jari S. /TSS - Sergel Kredittjänster /+46-8-4526435; powerh-l@lists.sowder.com Ämne: RE: write to file with quick Are you sure the performance problem is with the writing and not with the reading of the original data? I would suspect a problem there instead. -----Original Message----- From: powerh-l-admin@lists.sowder.com [mailto:powerh-l-admin@lists.sowder.com]On Behalf Of Jari.Kaljunen@teliasonera.com Sent: 2004 September 15 8:41 AM To: powerh-l@lists.sowder.com Subject: write to file with quick Hello, I am writing to File: ASC600 Organization: SEQUENTIAL Type: RMS Open: ASC600 Record Format: Variable Supersede: No Records & Sizes: ASC600 600 bytes Item Type Size Occ Offset REC600 CHARACTER 600 in quick using let and put as follows let rec600 = "" + asc(arnr of arenden,8) + asc(parmblnr,2) & + atgkod of atgkoder + "" put asc600 reset The performance is very poor. The administrator believs that quick closes and opens the file all the time. Can that be true? What should I do? We are creating an xml-file (it will be about 5 GB) with data from several tables and our solution with qtp needed too many subfiles and sortings using lots of space. /Jari = = = = = = = = = = = = = = = = = = = = = = = = = = = = Mailing list: powerh-l@lists.sowder.com Subscribe: "subscribe" in message body to powerh-l-request@lists.sowder.com Unsubscribe: "unsubscribe " in message body to powerh-l-request@lists.sowder.com http://lists.sowder.com/mailman/listinfo/powerh-l This list is closed, thus to post to the list you must be a subscriber. From ppfb@hotmail.com Thu Sep 16 20:01:38 2004 From: ppfb@hotmail.com (Peter Bateman) Date: Thu, 16 Sep 2004 16:01:38 -0300 Subject: SV: write to file with quick Message-ID:

I would try 'file asc600 designer open append' It may be a machine specific option.

But I didn't see it in the manual though.

Regards,

Peter

>From: <Jari.Kaljunen@teliasonera.com>
>To: <powerh-l@lists.sowder.com>
>Subject: SV: write to file with quick
>Date: Thu, 16 Sep 2004 12:22:20 +0200
>
>Hello again,
>
>a little test without reading database still very slow (actually no difference...). It takes 49 s to run this with 100 records on subfile. The result file is 20100 lines.
>
>I even wrote a test quiz reading the same subfile and having 200 defines, each including systime and some more, each defined item written on an own line. It takes quiz 3 s to write these 20000 lines.
>
>
>/Jari
>
>
>screen test_writing_file menu stopscreen
>
>file *ik006002 designer
>file asc600 designer
>
>temp t_count
>
>procedure internal write
>   begin
>      for 200
>         begin
>            let t_count = t_count + 1
>            let rec600 = asc(t_count) &
>                 + " " + asc(systime) + " procedure internal write "
>            put asc600 reset
>         end
>   end
>
>procedure initialize
>   begin
>    let t_count = 0
>    while retr ik006002 seq
>      begin
>        let rec600 = asc(t_count) &
>                 + " " + asc(systime) + " procedure initialize "
>        put asc600 reset
>        do internal write
>      end
>      return
>   end
>
>build
>
>-----Ursprungligt meddelande-----
>Från: Johnson, Harold A EDUC:EX
>[mailto:Harold.A.Johnson@gems1.gov.bc.ca]
>Skickat: den 15 september 2004 17:50
>Till: Kaljunen, Jari S. /TSS - Sergel Kredittjänster /+46-8-4526435;
>powerh-l@lists.sowder.com
>Ämne: RE: write to file with quick
>
>
>Are you sure the performance problem is with the writing and not with the
>reading of the original data?   I would suspect a problem there instead.
>
>
>-----Original Message-----
>From: powerh-l-admin@lists.sowder.com
>[mailto:powerh-l-admin@lists.sowder.com]On Behalf Of
>Jari.Kaljunen@teliasonera.com
>Sent: 2004 September 15 8:41 AM
>To: powerh-l@lists.sowder.com
>Subject: write to file with quick
>
>
>Hello,
>
>I am writing to
>
>File: ASC600
>Organization: SEQUENTIAL
>Type: RMS
>Open: ASC600
>Record Format: Variable
>Supersede: No
>Records & Sizes: ASC600 600 bytes
>Item Type Size Occ Offset
>REC600 CHARACTER 600
>
>in quick using let and put as follows
>
>let rec600 = "<HeaderRef2>" + asc(arnr of arenden,8) + asc(parmblnr,2) &
>+ atgkod of atgkoder + "</HeaderRef2>"
>
>put asc600 reset
>
>
>The performance is very poor. The administrator believs that quick closes
>and opens the file all the time.
>Can that be true?
>What should I do?
>
>We are creating an xml-file (it will be about 5 GB) with data from several
>tables and our solution with qtp needed too many subfiles and sortings using
>lots of space.
>
>/Jari
>
>= = = = = = = = = = = = = = = = = = = = = = = = = = = =
>Mailing list: powerh-l@lists.sowder.com
>Subscribe: "subscribe" in message body to powerh-l-request@lists.sowder.com
>Unsubscribe: "unsubscribe <password>" in message body to
>powerh-l-request@lists.sowder.com
>http://lists.sowder.com/mailman/listinfo/powerh-l
>This list is closed, thus to post to the list you must be a subscriber.
>
>= = = = = = = = = = = = = = = = = = = = = = = = = = = =
>Mailing list: powerh-l@lists.sowder.com
>Subscribe: "subscribe" in message body to powerh-l-request@lists.sowder.com
>Unsubscribe: "unsubscribe <password>" in message body to powerh-l-request@lists.sowder.com
>http://lists.sowder.com/mailman/listinfo/powerh-l
>This list is closed, thus to post to the list you must be a subscriber.

Don't just Search. Find! The new MSN Search: Fast. Clear. Easy. From Jari.Kaljunen@teliasonera.com Thu Sep 16 20:41:55 2004 From: Jari.Kaljunen@teliasonera.com (Jari.Kaljunen@teliasonera.com) Date: Thu, 16 Sep 2004 21:41:55 +0200 Subject: SV: SV: write to file with quick Message-ID: This is a multi-part message in MIME format. ------_=_NextPart_001_01C49C25.38F9D8E4 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable You mean file asc600 designer open append ? =20 Well, it didn't help.. =20 /Jari =20 =20 -----Ursprungligt meddelande----- Fr=E5n: Peter Bateman [mailto:ppfb@hotmail.com] Skickat: den 16 september 2004 21:02 Till: powerh-l@lists.sowder.com =C4mne: RE: SV: write to file with quick I would try 'file asc600 designer open append' It may be a machine = specific option. But I didn't see it in the manual though. Regards, Peter >From: =20 >To: =20 >Subject: SV: write to file with quick=20 >Date: Thu, 16 Sep 2004 12:22:20 +0200=20 >=20 >Hello again,=20 >=20 >a little test without reading database still very slow (actually no = difference...). It takes 49 s to run this with 100 records on subfile. = The result file is 20100 lines.=20 >=20 >I even wrote a test quiz reading the same subfile and having 200 = defines, each including systime and some more, each defined item written = on an own line. It takes quiz 3 s to write these 20000 lines.=20 >=20 >=20 >/Jari=20 >=20 >=20 >screen test_writing_file menu stopscreen=20 >=20 >file *ik006002 designer=20 >file asc600 designer=20 >=20 >temp t_count=20 >=20 >procedure internal write=20 > begin=20 > for 200=20 > begin=20 > let t_count =3D t_count + 1=20 > let rec600 =3D asc(t_count) &=20 > + " " + asc(systime) + " procedure internal write "=20 > put asc600 reset=20 > end=20 > end=20 >=20 >procedure initialize=20 > begin=20 > let t_count =3D 0=20 > while retr ik006002 seq=20 > begin=20 > let rec600 =3D asc(t_count) &=20 > + " " + asc(systime) + " procedure initialize "=20 > put asc600 reset=20 > do internal write=20 > end=20 > return=20 > end=20 >=20 >build=20 >=20 >-----Ursprungligt meddelande-----=20 >Fr=E5n: Johnson, Harold A EDUC:EX=20 >[mailto:Harold.A.Johnson@gems1.gov.bc.ca]=20 >Skickat: den 15 september 2004 17:50=20 >Till: Kaljunen, Jari S. /TSS - Sergel Kredittj=E4nster /+46-8-4526435;=20 >powerh-l@lists.sowder.com=20 >=C4mne: RE: write to file with quick=20 >=20 >=20 >Are you sure the performance problem is with the writing and not with = the=20 >reading of the original data? I would suspect a problem there = instead.=20 >=20 >=20 >-----Original Message-----=20 >From: powerh-l-admin@lists.sowder.com=20 >[mailto:powerh-l-admin@lists.sowder.com]On Behalf Of=20 >Jari.Kaljunen@teliasonera.com=20 >Sent: 2004 September 15 8:41 AM=20 >To: powerh-l@lists.sowder.com=20 >Subject: write to file with quick=20 >=20 >=20 >Hello,=20 >=20 >I am writing to=20 >=20 >File: ASC600=20 >Organization: SEQUENTIAL=20 >Type: RMS=20 >Open: ASC600=20 >Record Format: Variable=20 >Supersede: No=20 >Records & Sizes: ASC600 600 bytes=20 >Item Type Size Occ Offset=20 >REC600 CHARACTER 600=20 >=20 >in quick using let and put as follows=20 >=20 >let rec600 =3D "" + asc(arnr of arenden,8) + = asc(parmblnr,2) &=20 >+ atgkod of atgkoder + ""=20 >=20 >put asc600 reset=20 >=20 >=20 >The performance is very poor. The administrator believs that quick = closes=20 >and opens the file all the time.=20 >Can that be true?=20 >What should I do?=20 >=20 >We are creating an xml-file (it will be about 5 GB) with data from = several=20 >tables and our solution with qtp needed too many subfiles and sortings = using=20 >lots of space.=20 >=20 >/Jari=20 >=20 >=3D =3D =3D =3D =3D =3D =3D =3D =3D =3D =3D =3D =3D =3D =3D =3D =3D =3D = =3D =3D =3D =3D =3D =3D =3D =3D =3D =3D=20 >Mailing list: powerh-l@lists.sowder.com=20 >Subscribe: "subscribe" in message body to = powerh-l-request@lists.sowder.com=20 >Unsubscribe: "unsubscribe " in message body to=20 >powerh-l-request@lists.sowder.com=20 >http://lists.sowder.com/mailman/listinfo/powerh-l=20 >This list is closed, thus to post to the list you must be a subscriber. = >=20 >=3D =3D =3D =3D =3D =3D =3D =3D =3D =3D =3D =3D =3D =3D =3D =3D =3D =3D = =3D =3D =3D =3D =3D =3D =3D =3D =3D =3D=20 >Mailing list: powerh-l@lists.sowder.com=20 >Subscribe: "subscribe" in message body to = powerh-l-request@lists.sowder.com=20 >Unsubscribe: "unsubscribe " in message body to = powerh-l-request@lists.sowder.com=20 >http://lists.sowder.com/mailman/listinfo/powerh-l=20 >This list is closed, thus to post to the list you must be a subscriber. = _____ =20 Don't just Search. Find! The new MSN Search: = Fast. Clear. Easy. =3D =3D = =3D =3D =3D =3D =3D =3D =3D =3D =3D =3D =3D =3D =3D =3D =3D =3D =3D =3D = =3D =3D =3D =3D =3D =3D =3D =3D Mailing list: powerh-l@lists.sowder.com = Subscribe: "subscribe" in message body to = powerh-l-request@lists.sowder.com Unsubscribe: "unsubscribe " in message = body to powerh-l-request@lists.sowder.com = http://lists.sowder.com/mailman/listinfo/powerh-l This list is closed, = thus to post to the list you must be a subscriber.=20 ------_=_NextPart_001_01C49C25.38F9D8E4 Content-Type: text/html; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable
You mean file asc600 designer open append = ?
&nbs= p;
Well, it = didn't=20 help..
&nbs= p;
/Jari

 

 
-----Ursprungligt meddelande-----
Fr=E5n: Peter = Bateman=20 [mailto:ppfb@hotmail.com]
Skickat: den 16 september 2004=20 21:02
Till: powerh-l@lists.sowder.com
=C4mne: RE: = SV: write=20 to file with quick

I would try 'file asc600 designer open append' It may be a machine = specific=20 option.

But I didn't see it in the manual though.

Regards,

Peter

>From: <Jari.Kaljunen@teliasonera.com>=20
>To: <powerh-l@lists.sowder.com>=20
>Subject: SV: write to file with quick=20
>Date: Thu, 16 Sep 2004 12:22:20 +0200=20
>=20
>Hello again,=20
>=20
>a little test without reading database still very slow=20 (actually no difference...). It takes 49 s to run this with 100 = records on=20 subfile. The result file is 20100 lines.=20
>=20
>I even wrote a test quiz reading the same subfile and = having=20 200 defines, each including systime and some more, each defined item = written=20 on an own line. It takes quiz 3 s to write these 20000 lines.=20
>=20
>=20
>/Jari=20
>=20
>=20
>screen test_writing_file menu stopscreen=20
>=20
>file *ik006002 designer=20
>file asc600 designer=20
>=20
>temp t_count=20
>=20
>procedure internal write=20
>   begin=20
>      for 200=20
>         begin=20 =
>         &nbs= p;  let=20 t_count =3D t_count + 1=20 =
>         &nbs= p;  let=20 rec600 =3D asc(t_count) &=20 =
>         &nbs= p;      =20 + " " + asc(systime) + " procedure internal write "=20 =
>         &nbs= p;  put=20 asc600 reset=20
>         end=20
>   end=20
>=20
>procedure initialize=20
>   begin=20
>    let t_count =3D 0=20
>    while retr ik006002 seq=20
>      begin=20
>        let = rec600 =3D=20 asc(t_count) &=20 =
>         &nbs= p;      =20 + " " + asc(systime) + " procedure initialize "=20
>        put = asc600=20 reset=20
>        do = internal=20 write=20
>      end=20
>      return=20
>   end=20
>=20
>build=20
>=20
>-----Ursprungligt meddelande-----=20
>Fr=E5n: Johnson, Harold A EDUC:EX=20
>[mailto:Harold.A.Johnson@gems1.gov.bc.ca]=20
>Skickat: den 15 september 2004 17:50=20
>Till: Kaljunen, Jari S. /TSS - Sergel Kredittj=E4nster=20 /+46-8-4526435;=20
>powerh-l@lists.sowder.com=20
>=C4mne: RE: write to file with quick=20
>=20
>=20
>Are you sure the performance problem is with the = writing and=20 not with the=20
>reading of the original data?   I would = suspect a=20 problem there instead.=20
>=20
>=20
>-----Original Message-----=20
>From: powerh-l-admin@lists.sowder.com=20
>[mailto:powerh-l-admin@lists.sowder.com]On Behalf Of=20
>Jari.Kaljunen@teliasonera.com=20
>Sent: 2004 September 15 8:41 AM=20
>To: powerh-l@lists.sowder.com=20
>Subject: write to file with quick=20
>=20
>=20
>Hello,=20
>=20
>I am writing to=20
>=20
>File: ASC600=20
>Organization: SEQUENTIAL=20
>Type: RMS=20
>Open: ASC600=20
>Record Format: Variable=20
>Supersede: No=20
>Records & Sizes: ASC600 600 bytes=20
>Item Type Size Occ Offset=20
>REC600 CHARACTER 600=20
>=20
>in quick using let and put as follows=20
>=20
>let rec600 =3D "<HeaderRef2>" + asc(arnr of = arenden,8) +=20 asc(parmblnr,2) &=20
>+ atgkod of atgkoder + "</HeaderRef2>"=20
>=20
>put asc600 reset=20
>=20
>=20
>The performance is very poor. The administrator believs = that=20 quick closes=20
>and opens the file all the time.=20
>Can that be true?=20
>What should I do?=20
>=20
>We are creating an xml-file (it will be about 5 GB) = with data=20 from several=20
>tables and our solution with qtp needed too many = subfiles and=20 sortings using=20
>lots of space.=20
>=20
>/Jari=20
>=20
>=3D =3D =3D =3D =3D =3D =3D =3D =3D =3D =3D =3D =3D =3D = =3D =3D =3D =3D =3D =3D =3D =3D =3D =3D =3D =3D =3D =3D=20
>Mailing list: powerh-l@lists.sowder.com=20
>Subscribe: "subscribe" in message body to=20 powerh-l-request@lists.sowder.com=20
>Unsubscribe: "unsubscribe <password>" in message = body to=20
>powerh-l-request@lists.sowder.com=20
>http://lists.sowder.com/mailman/listinfo/powerh-l=20
>This list is closed, thus to post to the list you must = be a=20 subscriber.=20
>=20
>=3D =3D =3D =3D =3D =3D =3D =3D =3D =3D =3D =3D =3D =3D = =3D =3D =3D =3D =3D =3D =3D =3D =3D =3D =3D =3D =3D =3D=20
>Mailing list: powerh-l@lists.sowder.com=20
>Subscribe: "subscribe" in message body to=20 powerh-l-request@lists.sowder.com=20
>Unsubscribe: "unsubscribe <password>" in message = body to=20 powerh-l-request@lists.sowder.com=20
>http://lists.sowder.com/mailman/listinfo/powerh-l=20
>This list is closed, thus to post to the list you must = be a=20 subscriber.=20

Don't just Search. Find! The new MSN = Search: Fast.=20 Clear. Easy. =3D =3D =3D =3D =3D =3D =3D =3D =3D =3D =3D =3D =3D =3D = =3D =3D =3D =3D =3D =3D =3D =3D =3D =3D =3D =3D =3D =3D Mailing=20 list: powerh-l@lists.sowder.com Subscribe: "subscribe" in message body = to=20 powerh-l-request@lists.sowder.com Unsubscribe: "unsubscribe = " in=20 message body to powerh-l-request@lists.sowder.com=20 http://lists.sowder.com/mailman/listinfo/powerh-l This list is closed, = thus to=20 post to the list you must be a subscriber.
------_=_NextPart_001_01C49C25.38F9D8E4-- From wilbert.vissers@hp.com Fri Sep 17 03:17:24 2004 From: wilbert.vissers@hp.com (Vissers, Wilbert) Date: Fri, 17 Sep 2004 14:17:24 +1200 Subject: write to file with quick Message-ID: Hi Jari, these options may help, we think from your ref to RMS that you may be working on a Vax, so this is from Michael and I: Options: (a) Pre-allocate disc space for the file. The reason being that on a VAX the file may be being continuously extended by the operating system 1 record at a time because that's what quick would be writing. That's what the system administrator could have been seeing. (b) Write to an indexed file instead of a sequential file, then extract the data part afterwards with a utility or Quiz. Set a simple key to the record number to keep the index clean. The reason for this idea is that indexed files are more normal and there may well be something unusual about what quick needs to do when working with sequential files. Couldn't resist doing this at lunchtime, so we had a little play and ran some tests on a very slow old machine we have here, for 20,000 records: (1) Quiz to subfile = 25 seconds (2) Quick to sequential file = 32 minutes yes minutes (3) Quick to indexed file = 7 minutes Quiz code: ---------- ACCESS GL-BATCH REPORT SUMMARY ALL SET REPORT LIMIT 20000 SET SUBFILE GO Quick code: ----------- LET WK-COUNT = 0 INFO = "STARTING " + ASCII(SYSTIME,8) NOW RESPONSE WHILE WK-COUNT < 20000 BEGIN LET WK-COUNT = WK-COUNT + 1 LET BATCH-KEY OF GL-BATCH = ASCII(WK-COUNT) PUT GL-BATCH RESET ; <---- DESIGNER FILE END INFO="FINISHED " + ASCII(SYSTIME,8) NOW RESPONSE Hope something in this helps, best regards. Wilbert From wilbert.vissers@hp.com Fri Sep 17 03:22:12 2004 From: wilbert.vissers@hp.com (Vissers, Wilbert) Date: Fri, 17 Sep 2004 14:22:12 +1200 Subject: write to file with quick Message-ID: Hi Jari, a postnote : The quick code we did was outside the UPDATE procedure, to avoid quick having to create the rollback buffers. cheers. From Bob.Deskin@Cognos.COM Fri Sep 17 15:38:58 2004 From: Bob.Deskin@Cognos.COM (Deskin, Bob) Date: Fri, 17 Sep 2004 10:38:58 -0400 Subject: write to file with quick Message-ID: <2B4E386EAEC90947802323BF373DC0EC0259213A@sottemail1.ent.ad.cognos.com> This is consistent with my understanding of what happens. For direct files, we post the end-of-file marker on each write. That's because if we don't and the file is shared, the record just written may not be available to other users. This would be the case if the records are blocked. Normally, the block isn't posted until full or the file is closed. This causes additional output to the file. In QUIZ, the file is opened for output only and there is no need to do this posting, or any locks around the output. Bob Deskin Senior 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: powerh-l-admin@lists.sowder.com [mailto:powerh-l-admin@lists.sowder.com] On Behalf Of Vissers, Wilbert Sent: September 16, 2004 10:17 PM To: powerh-l@lists.sowder.com Subject: RE:write to file with quick Hi Jari, these options may help, we think from your ref to RMS that you may be working on a Vax, so this is from Michael and I: Options: (a) Pre-allocate disc space for the file. The reason being that on a VAX the file may be being continuously extended by the operating system 1 record at a time because that's what quick would be writing. That's what the system administrator could have been seeing. (b) Write to an indexed file instead of a sequential file, then extract the data part afterwards with a utility or Quiz. Set a simple key to the record number to keep the index clean. The reason for this idea is that indexed files are more normal and there may well be something unusual about what quick needs to do when working with sequential files. Couldn't resist doing this at lunchtime, so we had a little play and ran some tests on a very slow old machine we have here, for 20,000 records: (1) Quiz to subfile = 25 seconds (2) Quick to sequential file = 32 minutes yes minutes (3) Quick to indexed file = 7 minutes Quiz code: ---------- ACCESS GL-BATCH REPORT SUMMARY ALL SET REPORT LIMIT 20000 SET SUBFILE GO Quick code: ----------- LET WK-COUNT = 0 INFO = "STARTING " + ASCII(SYSTIME,8) NOW RESPONSE WHILE WK-COUNT < 20000 BEGIN LET WK-COUNT = WK-COUNT + 1 LET BATCH-KEY OF GL-BATCH = ASCII(WK-COUNT) PUT GL-BATCH RESET ; <---- DESIGNER FILE END INFO="FINISHED " + ASCII(SYSTIME,8) NOW RESPONSE Hope something in this helps, best regards. Wilbert = = = = = = = = = = = = = = = = = = = = = = = = = = = = Mailing list: powerh-l@lists.sowder.com Subscribe: "subscribe" in message body to powerh-l-request@lists.sowder.com Unsubscribe: "unsubscribe " in message body to powerh-l-request@lists.sowder.com http://lists.sowder.com/mailman/listinfo/powerh-l This list is closed, thus to post to the list you must be a subscriber. This message may contain privileged and/or confidential information. If you have received this e-mail in error or are not the intended recipient, you may not use, copy, disseminate or distribute it; do not open any attachments, delete it immediately from your system and notify the sender promptly by e-mail that you have done so. Thank you. From John.Pickering@norbord.com Fri Sep 17 16:42:51 2004 From: John.Pickering@norbord.com (Pickering, John (NORBORD)) Date: Fri, 17 Sep 2004 11:42:51 -0400 Subject: write to file with quick Message-ID: <611340310619D711AA4000306E1CC512027979FD@TORHEXCH> I'm certainly no Vax expert so I'll ask what may be a stupid question. Is there an open option available to Quick that would force it to open the file in the same manner as Quiz? That is, open it with exclusive write access as in the case of this app, nothing else wants to read the file while it is being created. Then perhaps Quick could approach the times that Wilbert Vissers reported. Could it be as simple as "file whatever designer open append exclusive" ? Regards, JWP -----Original Message----- From: Deskin, Bob [mailto:Bob.Deskin@Cognos.COM] Sent: Friday, September 17, 2004 10:39 AM To: powerh-l@lists.sowder.com Subject: RE: write to file with quick This is consistent with my understanding of what happens. For direct files, we post the end-of-file marker on each write. That's because if we don't and the file is shared, the record just written may not be available to other users. This would be the case if the records are blocked. Normally, the block isn't posted until full or the file is closed. This causes additional output to the file. In QUIZ, the file is opened for output only and there is no need to do this posting, or any locks around the output. Bob Deskin Senior 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 From Bob.Deskin@Cognos.COM Fri Sep 17 16:49:33 2004 From: Bob.Deskin@Cognos.COM (Deskin, Bob) Date: Fri, 17 Sep 2004 11:49:33 -0400 Subject: write to file with quick Message-ID: <2B4E386EAEC90947802323BF373DC0EC0259213C@sottemail1.ent.ad.cognos.com> Yes, QUICK allows you to open the file exclusively. However, this might be a problem in a multi-user environment. Bob -----Original Message----- From: Pickering, John (NORBORD) [mailto:John.Pickering@norbord.com] Sent: September 17, 2004 11:43 AM To: Deskin, Bob; powerh-l@lists.sowder.com Subject: RE: write to file with quick I'm certainly no Vax expert so I'll ask what may be a stupid question. Is there an open option available to Quick that would force it to open the file in the same manner as Quiz? That is, open it with exclusive write access as in the case of this app, nothing else wants to read the file while it is being created. Then perhaps Quick could approach the times that Wilbert Vissers reported. Could it be as simple as "file whatever designer open append exclusive" ? Regards, JWP -----Original Message----- From: Deskin, Bob [mailto:Bob.Deskin@Cognos.COM] Sent: Friday, September 17, 2004 10:39 AM To: powerh-l@lists.sowder.com Subject: RE: write to file with quick This is consistent with my understanding of what happens. For direct files, we post the end-of-file marker on each write. That's because if we don't and the file is shared, the record just written may not be available to other users. This would be the case if the records are blocked. Normally, the block isn't posted until full or the file is closed. This causes additional output to the file. In QUIZ, the file is opened for output only and there is no need to do this posting, or any locks around the output. Bob Deskin Senior 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 This message may contain privileged and/or confidential information. If you have received this e-mail in error or are not the intended recipient, you may not use, copy, disseminate or distribute it; do not open any attachments, delete it immediately from your system and notify the sender promptly by e-mail that you have done so. Thank you. From Tracy.Johnson@msiusa.com Fri Sep 17 17:01:02 2004 From: Tracy.Johnson@msiusa.com (Johnson, Tracy) Date: Fri, 17 Sep 2004 12:01:02 -0400 Subject: write to file with quick Message-ID: <970F565A5733E64686BC3A10FBBAE87C56EBF9@msi03-mail.noam.msiusa.com> I didn't know Quiz had write capability outside of subfiles! BT Tracy Johnson MSI Schaevitz Sensors > -----Original Message----- > From: powerh-l-admin@lists.sowder.com > [mailto:powerh-l-admin@lists.sowder.com]On Behalf Of Pickering, John > (NORBORD) > Sent: Friday, September 17, 2004 11:43 AM > To: 'Deskin, Bob'; powerh-l@lists.sowder.com > Subject: RE: write to file with quick > > > I'm certainly no Vax expert so I'll ask what may be a stupid question. > > Is there an open option available to Quick that would force > it to open the > file in the same manner as Quiz? That is, open it with exclusive write > access as in the case of this app, nothing else wants to read > the file while > it is being created. Then perhaps Quick could approach the times that > Wilbert Vissers reported. > > Could it be as simple as "file whatever designer open append > exclusive" ? > > Regards, > JWP > > -----Original Message----- > From: Deskin, Bob [mailto:Bob.Deskin@Cognos.COM] > Sent: Friday, September 17, 2004 10:39 AM > To: powerh-l@lists.sowder.com > Subject: RE: write to file with quick > > > This is consistent with my understanding of what happens. For direct > files, we post the end-of-file marker on each write. That's because if > we don't and the file is shared, the record just written may not be > available to other users. This would be the case if the records are > blocked. Normally, the block isn't posted until full or the file is > closed. This causes additional output to the file. In QUIZ, > the file is > opened for output only and there is no need to do this posting, or any > locks around the output. > > Bob Deskin > Senior 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 > > = = = = = = = = = = = = = = = = = = = = = = = = = = = = > Mailing list: powerh-l@lists.sowder.com > Subscribe: "subscribe" in message body to > powerh-l-request@lists.sowder.com > Unsubscribe: "unsubscribe " in message body to > powerh-l-request@lists.sowder.com > http://lists.sowder.com/mailman/listinfo/powerh-l > This list is closed, thus to post to the list you must be a > subscriber. > From joeboyle_adt@hotmail.com Fri Sep 17 22:28:10 2004 From: joeboyle_adt@hotmail.com (Joe Boyle) Date: Fri, 17 Sep 2004 22:28:10 +0100 Subject: SV: write to file with quick In-Reply-To: Message-ID: This is a multi-part message in MIME format. ------=_NextPart_000_0003_01C49D05.9CE46E70 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable why not allow the performance of quiz to generate the subfile and then = use the DCL command 'CONVERT/FDL' to convert the file to the required = sequential format. It may even be the case, that if the 'nodict' qualifier is used = when creating the subfile, then the subfile need only be renamed and copied = to the required directory. =20 I am guessing that a 'nodict' subfile will have a sequential rather than direct record format; you could always use the analyse command to = examine the subfile's record format if you aren=92t sure. =20 Presumably, once the file is built, subsequent updates will be writing = fewer records into the file, and the slow performance of Quick may not be as = great an issue =96 or you could even go back to Qtp. Regards, Joe. This e-mail and all information contained in it is confidential and may = be legally privileged. If you are not the intended recipient, your access = to this e-mail is unauthorized. Any use, dissemination, distribution, publication or copying by you of this e-mail or any of the information contained within it is prohibited and may be unlawful. Do not open any attachments, delete it immediately from your system and notify the = sender promptly by e-mail that you have done so. The content of this e-mail and = any attachments sent with it may have been altered without the consent or knowledge of the author. _____ =20 From: powerh-l-admin@lists.sowder.com [mailto:powerh-l-admin@lists.sowder.com] On Behalf Of Jari.Kaljunen@teliasonera.com Sent: 16 September 2004 20:42 To: powerh-l@lists.sowder.com Subject: SV: SV: write to file with quick =20 You mean file asc600 designer open append ? =20 Well, it didn't help.. =20 /Jari =20 =20 -----Ursprungligt meddelande----- Fr=E5n: Peter Bateman [mailto:ppfb@hotmail.com] Skickat: den 16 september 2004 21:02 Till: powerh-l@lists.sowder.com =C4mne: RE: SV: write to file with quick I would try 'file asc600 designer open append' It may be a machine = specific option. But I didn't see it in the manual though. Regards, Peter >From: =20 >To: =20 >Subject: SV: write to file with quick=20 >Date: Thu, 16 Sep 2004 12:22:20 +0200=20 >=20 >Hello again,=20 >=20 >a little test without reading database still very slow (actually no difference...). It takes 49 s to run this with 100 records on subfile. = The result file is 20100 lines.=20 >=20 >I even wrote a test quiz reading the same subfile and having 200 = defines, each including systime and some more, each defined item written on an = own line. It takes quiz 3 s to write these 20000 lines.=20 >=20 >=20 >/Jari=20 >=20 >=20 >screen test_writing_file menu stopscreen=20 >=20 >file *ik006002 designer=20 >file asc600 designer=20 >=20 >temp t_count=20 >=20 >procedure internal write=20 > begin=20 > for 200=20 > begin=20 > let t_count =3D t_count + 1=20 > let rec600 =3D asc(t_count) &=20 > + " " + asc(systime) + " procedure internal write "=20 > put asc600 reset=20 > end=20 > end=20 >=20 >procedure initialize=20 > begin=20 > let t_count =3D 0=20 > while retr ik006002 seq=20 > begin=20 > let rec600 =3D asc(t_count) &=20 > + " " + asc(systime) + " procedure initialize "=20 > put asc600 reset=20 > do internal write=20 > end=20 > return=20 > end=20 >=20 >build=20 >=20 >-----Ursprungligt meddelande-----=20 >Fr=E5n: Johnson, Harold A EDUC:EX=20 >[mailto:Harold.A.Johnson@gems1.gov.bc.ca]=20 >Skickat: den 15 september 2004 17:50=20 >Till: Kaljunen, Jari S. /TSS - Sergel Kredittj=E4nster /+46-8-4526435;=20 >powerh-l@lists.sowder.com=20 >=C4mne: RE: write to file with quick=20 >=20 >=20 >Are you sure the performance problem is with the writing and not with = the=20 >reading of the original data? I would suspect a problem there = instead.=20 >=20 >=20 >-----Original Message-----=20 >From: powerh-l-admin@lists.sowder.com=20 >[mailto:powerh-l-admin@lists.sowder.com]On Behalf Of=20 >Jari.Kaljunen@teliasonera.com=20 >Sent: 2004 September 15 8:41 AM=20 >To: powerh-l@lists.sowder.com=20 >Subject: write to file with quick=20 >=20 >=20 >Hello,=20 >=20 >I am writing to=20 >=20 >File: ASC600=20 >Organization: SEQUENTIAL=20 >Type: RMS=20 >Open: ASC600=20 >Record Format: Variable=20 >Supersede: No=20 >Records & Sizes: ASC600 600 bytes=20 >Item Type Size Occ Offset=20 >REC600 CHARACTER 600=20 >=20 >in quick using let and put as follows=20 >=20 >let rec600 =3D "" + asc(arnr of arenden,8) + = asc(parmblnr,2) &=20 >+ atgkod of atgkoder + ""=20 >=20 >put asc600 reset=20 >=20 >=20 >The performance is very poor. The administrator believs that quick = closes=20 >and opens the file all the time.=20 >Can that be true?=20 >What should I do?=20 >=20 >We are creating an xml-file (it will be about 5 GB) with data from = several=20 >tables and our solution with qtp needed too many subfiles and sortings using=20 >lots of space.=20 >=20 >/Jari=20 >=20 >=3D =3D =3D =3D =3D =3D =3D =3D =3D =3D =3D =3D =3D =3D =3D =3D =3D =3D = =3D =3D =3D =3D =3D =3D =3D =3D =3D =3D=20 >Mailing list: powerh-l@lists.sowder.com=20 >Subscribe: "subscribe" in message body to = powerh-l-request@lists.sowder.com >Unsubscribe: "unsubscribe " in message body to=20 >powerh-l-request@lists.sowder.com=20 >http://lists.sowder.com/mailman/listinfo/powerh-l=20 >This list is closed, thus to post to the list you must be a subscriber. = >=20 >=3D =3D =3D =3D =3D =3D =3D =3D =3D =3D =3D =3D =3D =3D =3D =3D =3D =3D = =3D =3D =3D =3D =3D =3D =3D =3D =3D =3D=20 >Mailing list: powerh-l@lists.sowder.com=20 >Subscribe: "subscribe" in message body to = powerh-l-request@lists.sowder.com >Unsubscribe: "unsubscribe " in message body to powerh-l-request@lists.sowder.com=20 >http://lists.sowder.com/mailman/listinfo/powerh-l=20 >This list is closed, thus to post to the list you must be a subscriber. = _____ =20 Don't just Search. Find! The new MSN Search: Fast. Clear. Easy. =3D =3D = =3D =3D =3D =3D =3D =3D =3D =3D =3D =3D =3D =3D =3D =3D =3D =3D =3D =3D =3D =3D =3D =3D =3D = =3D =3D =3D Mailing list: powerh-l@lists.sowder.com Subscribe: "subscribe" in message body to powerh-l-request@lists.sowder.com Unsubscribe: "unsubscribe " in message body to powerh-l-request@lists.sowder.com http://lists.sowder.com/mailman/listinfo/powerh-l This list is closed, = thus to post to the list you must be a subscriber.=20 ------=_NextPart_000_0003_01C49D05.9CE46E70 Content-Type: text/html; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable

why not allow = the performance of quiz to generate the subfile and then use the DCL command = 'CONVERT/FDL' to convert the file to the required sequential format. It may even be = the case, that if the 'nodict' qualifier is used when creating the subfile, then = the subfile need only be renamed and copied to the required = directory.

 =

I am guessing = that a 'nodict' subfile will have a sequential rather than direct record format; you = could always use the analyse command to examine the subfile's record format if = you aren’t sure.

 

Presumably, once the file is built, subsequent updates will be writing fewer records into the file, and the = slow performance of Quick may not be as great an issue – or you could = even go back to Qtp.

Regards, Joe.

This e-mail and all information contained in it is = confidential and may be legally privileged. If you are not the = intended recipient, your access to this e-mail is unauthorized. Any use, dissemination, distribution, publication or copying by you of this e-mail or any of the information contained within it is prohibited and may be unlawful. Do = not open any attachments, delete it immediately from your system and notify the = sender promptly by e-mail that you have done so. The content of this e-mail and = any attachments sent with it may have been altered without the consent or = knowledge of the author.

From: powerh-l-admin@lists.sowder.com [mailto:powerh-l-admin@lists.sowder.com] = On Behalf Of = Jari.Kaljunen@teliasonera.com
Sent: 16 September 2004 = 20:42
To: = powerh-l@lists.sowder.com
Subject: SV: SV: write to = file with quick

 

You mean file asc600 designer open append = ?

 

Well, it didn't = help..

 

/Jari

 

 

-----Ursprungligt = meddelande-----
Fr=E5n: Peter Bateman [mailto:ppfb@hotmail.com]
Skickat: den 16 september = 2004 21:02
Till: = powerh-l@lists.sowder.com
=C4mne: RE: SV: write to = file with quick

I would try 'file asc600 designer open append' It may be a machine specific = option.

But I didn't see it in the manual though.

Regards,

Peter

>From: <Jari.Kaljunen@teliasonera.com> =

>To: <powerh-l@lists.sowder.com> =

>Subject: SV: write to file with quick =

>Date: Thu, 16 Sep 2004 12:22:20 +0200 =

>

>Hello again,

>

>a little test without reading database still very slow = (actually no difference...). It takes 49 s to run this with 100 records on subfile. = The result file is 20100 lines.

>

>I even wrote a test quiz reading the same subfile and having = 200 defines, each including systime and some more, each defined item written = on an own line. It takes quiz 3 s to write these 20000 lines. =

>

>

>/Jari

>

>

>screen test_writing_file menu stopscreen =

>

>file *ik006002 designer

>file asc600 designer

>

>temp t_count

>

>procedure internal write

>   begin

>      for 200 =

>         begin =

>          &= nbsp; let t_count =3D t_count + 1

>          &= nbsp; let rec600 =3D asc(t_count) &

>          &= nbsp;      + " " + asc(systime) + " procedure internal write " =

>          &= nbsp; put asc600 reset

>         end =

>   end

>

>procedure initialize

>   begin

>    let t_count =3D 0 =

>    while retr ik006002 seq =

>      begin =

>        let rec600 = =3D asc(t_count) &

>          &= nbsp;      + " " + asc(systime) + " procedure initialize " =

>        put asc600 = reset

>        do internal = write

>      end =

>      return =

>   end

>

>build

>

>-----Ursprungligt meddelande----- =

>Fr=E5n: Johnson, Harold A EDUC:EX =

>[mailto:Harold.A.Johnson@gems1.gov.bc.ca] =

>Skickat: den 15 september 2004 17:50 =

>Till: Kaljunen, Jari S. /TSS - Sergel Kredittj=E4nster /+46-8-4526435;

>powerh-l@lists.sowder.com

>=C4mne: RE: write to file with quick =

>

>

>Are you sure the performance problem is with the writing and = not with the

>reading of the original data?   I would suspect a = problem there instead.

>

>

>-----Original Message-----

>From: powerh-l-admin@lists.sowder.com =

>[mailto:powerh-l-admin@lists.sowder.com]On Behalf Of =

>Jari.Kaljunen@teliasonera.com

>Sent: 2004 September 15 8:41 AM =

>To: powerh-l@lists.sowder.com

>Subject: write to file with quick =

>

>

>Hello,

>

>I am writing to

>

>File: ASC600

>Organization: SEQUENTIAL

>Type: RMS

>Open: ASC600

>Record Format: Variable

>Supersede: No

>Records & Sizes: ASC600 600 bytes =

>Item Type Size Occ Offset

>REC600 CHARACTER 600

>

>in quick using let and put as follows =

>

>let rec600 =3D "<HeaderRef2>" + asc(arnr of arenden,8) + asc(parmblnr,2) &

>+ atgkod of atgkoder + "</HeaderRef2>" =

>

>put asc600 reset

>

>

>The performance is very poor. The administrator believs that = quick closes

>and opens the file all the time. =

>Can that be true?

>What should I do?

>

>We are creating an xml-file (it will be about 5 GB) with = data from several

>tables and our solution with qtp needed too many subfiles = and sortings using

>lots of space.

>

>/Jari

>

>=3D =3D =3D =3D =3D =3D =3D =3D =3D =3D =3D =3D =3D =3D =3D = =3D =3D =3D =3D =3D =3D =3D =3D =3D =3D =3D =3D =3D =

>Mailing list: powerh-l@lists.sowder.com =

>Subscribe: "subscribe" in message body to powerh-l-request@lists.sowder.com

>Unsubscribe: "unsubscribe <password>" in = message body to

>powerh-l-request@lists.sowder.com =

>http://lists.sowder.com/mailman/listinfo/powerh-l =

>This list is closed, thus to post to the list you must be a subscriber.

>

>=3D =3D =3D =3D =3D =3D =3D =3D =3D =3D =3D =3D =3D =3D =3D = =3D =3D =3D =3D =3D =3D =3D =3D =3D =3D =3D =3D =3D =

>Mailing list: powerh-l@lists.sowder.com =

>Subscribe: "subscribe" in message body to powerh-l-request@lists.sowder.com

>Unsubscribe: "unsubscribe <password>" in = message body to powerh-l-request@lists.sowder.com

>http://lists.sowder.com/mailman/listinfo/powerh-l =

>This list is closed, thus to post to the list you must be a subscriber.


Don't just Search. Find! The new MSN Search: = Fast. Clear. Easy. =3D =3D =3D =3D =3D =3D =3D =3D =3D =3D =3D =3D =3D =3D =3D = =3D =3D =3D =3D =3D =3D =3D =3D =3D =3D =3D =3D =3D Mailing list: powerh-l@lists.sowder.com Subscribe: "subscribe" in = message body to powerh-l-request@lists.sowder.com Unsubscribe: "unsubscribe = " in message body to powerh-l-request@lists.sowder.com http://lists.sowder.com/mailman/listinfo/powerh-l This list is closed, = thus to post to the list you must be a subscriber.

------=_NextPart_000_0003_01C49D05.9CE46E70-- From Brett@Kinetix.co.nz Sun Sep 19 23:44:11 2004 From: Brett@Kinetix.co.nz (Brett Greiner (Kinetix)) Date: Mon, 20 Sep 2004 10:44:11 +1200 Subject: freeform not working as expected Message-ID: <9D966695054CFA4A8FD712F96673CCFB11F83C@Afid> This is a multi-part message in MIME format. ------_=_NextPart_001_01C49E9A.2E4D506C Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: quoted-printable We are using Ph 8.13.D2 on an AIX. I have a data file to read in which has amounts specified as <1 char sign>00000031.17 ie. 00000031.17 for $31.17. When I setup the element as FREEFORM its only displayed as 31. If I remove the decimal place from the data, it comes out correctly as 31.17. =20 The manual for Freeform says its works with leadings/trailing spaces and an imbedded decimal point. Here is the PDL code of the element; Element CALTEX-VOLUME NUMERIC Size 12 & Decimal 2 & Leading Sign " " & picture " ^^^^^^^^.^^" & Default Item Datatype FREEFORM SIZE 12 Can anyone help? Thanks in advance, Brett. Brett Greiner,=20 Kinetix Group Ltd., ph. +64 6 759 1929 fx. +64 6 759 1928 ------_=_NextPart_001_01C49E9A.2E4D506C Content-Type: text/html; charset="us-ascii" Content-Transfer-Encoding: quoted-printable Message

We are using Ph 8.13.D2 on an = AIX.

I have a data file to read in which = has=20 amounts specified as <1 char sign>00000031.17 ie. 00000031.17 for = $31.17.=20 When I setup the element as FREEFORM its only displayed as 31. = If I remove the decimal place from the = data, it=20 comes out correctly as 31.17. 

The manual for Freeform says its = works with=20 leadings/trailing spaces and an imbedded decimal point.

Here is the PDL code of the=20 element;

Element=20 CALTEX-VOLUME          =       =20 NUMERIC   Size=20 12            = ;         =20 &
    Decimal 2 &
    = Leading Sign "=20 " &
    picture " ^^^^^^^^.^^" = &
   =20 Default Item Datatype FREEFORM SIZE 12

Can anyone help?

Thanks in advance,

Brett.

Brett Greiner,=20
Kinetix Group = Ltd.,
ph. +64 6 759 = 1929
fx.  +64 6 759=20 1928
=00 ------_=_NextPart_001_01C49E9A.2E4D506C-- From joeboyle_adt@hotmail.com Mon Sep 20 11:22:52 2004 From: joeboyle_adt@hotmail.com (Joe Boyle) Date: Mon, 20 Sep 2004 11:22:52 +0100 Subject: freeform not working as expected In-Reply-To: <9D966695054CFA4A8FD712F96673CCFB11F83C@Afid> Message-ID: This is a multi-part message in MIME format. ------=_NextPart_000_0011_01C49F04.2B841AE0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit I suspect that as the decimal part is stored/present in the item you will need to add syntax 'output scale 2 ' to the element syntax. This multiplies the value up to an integer ( same as * 100 ) - only the integer part is moved into the picture when a field is displayed. Regards, Joe. This e-mail and all information contained in it is confidential and may be legally privileged. If you are not the intended recipient, your access to this e-mail is unauthorized. Any use, dissemination, distribution, publication or copying by you of this e-mail or any of the information contained within it is prohibited and may be unlawful. Do not open any attachments, delete it immediately from your system and notify the sender promptly by e-mail that you have done so. The content of this e-mail and any attachments sent with it may have been altered without the consent or knowledge of the author. _____ From: powerh-l-admin@lists.sowder.com [mailto:powerh-l-admin@lists.sowder.com] On Behalf Of Brett Greiner (Kinetix) Sent: 19 September 2004 23:44 To: powerh-l@lists.sowder.com Subject: freeform not working as expected We are using Ph 8.13.D2 on an AIX. I have a data file to read in which has amounts specified as <1 char sign>00000031.17 ie. 00000031.17 for $31.17. When I setup the element as FREEFORM its only displayed as 31. If I remove the decimal place from the data, it comes out correctly as 31.17. The manual for Freeform says its works with leadings/trailing spaces and an imbedded decimal point. Here is the PDL code of the element; Element CALTEX-VOLUME NUMERIC Size 12 & Decimal 2 & Leading Sign " " & picture " ^^^^^^^^.^^" & Default Item Datatype FREEFORM SIZE 12 Can anyone help? Thanks in advance, Brett. Brett Greiner, Kinetix Group Ltd., ph. +64 6 759 1929 fx. +64 6 759 1928 ------=_NextPart_000_0011_01C49F04.2B841AE0 Content-Type: text/html; charset="us-ascii" Content-Transfer-Encoding: quoted-printable Message

I suspect that as the decimal part = is stored/present in the item you will need to add syntax ‘output = scale 2 ‘ to the element syntax.

 

This multiplies the value up to an = integer ( same as * 100 ) – only the integer part is moved into the = picture when a field is displayed.

 

Regards, Joe.

This e-mail and all information contained in it is confidential and may be legally privileged. If you are not the = intended recipient, your access to this e-mail is unauthorized. Any use, dissemination, distribution, publication or copying by you of this e-mail or any of the = information contained within it is prohibited and may be unlawful. Do not open any attachments, delete it immediately from your system and notify the = sender promptly by e-mail that you have done so. The content of this e-mail and = any attachments sent with it may have been altered without the consent or = knowledge of the author.

From: = powerh-l-admin@lists.sowder.com [mailto:powerh-l-admin@lists.sowder.com] On Behalf Of Brett Greiner (Kinetix)
Sent: 19 September 2004 = 23:44
To: = powerh-l@lists.sowder.com
Subject: freeform not = working as expected

 

We are using Ph 8.13.D2 on an = AIX.

I have a data file to read in which has = amounts specified as <1 char sign>00000031.17 ie. 00000031.17 for $31.17. = When I setup the element as FREEFORM its only displayed as 31. If I remove = the decimal place from the data, it comes out correctly as 31.17.  =

The manual for Freeform says its works = with leadings/trailing spaces and an imbedded decimal = point.

Here is the PDL code of the = element;

Element CALTEX-VOLUME          =        NUMERIC   Size 12            = ;          &
    Decimal 2 &
    Leading Sign " " &
    picture " ^^^^^^^^.^^" &
    Default Item Datatype FREEFORM SIZE = 12

Can anyone = help?

Thanks in = advance,

Brett.

Brett Greiner,

Kinetix Group = Ltd.,

ph. +64 6 759 = 1929

fx.  +64 6 759 = 1928

------=_NextPart_000_0011_01C49F04.2B841AE0-- From robert.mills@windsong-services.co.uk Mon Sep 20 11:53:16 2004 From: robert.mills@windsong-services.co.uk (Robert Mills) Date: Mon, 20 Sep 2004 11:53:16 +0100 Subject: Upgrading to 8.49D Message-ID: <08F03A7FCBC8AD4AAB87773F25282729A2384A@whgexh01.whgl.co.uk> This is a multi-part message in MIME format. ------_=_NextPart_001_01C49F00.083CC987 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable Greeting to the list, We are about to upgrade from 8.19D7 and 8.39C to 8.49D and am interested = in comments by anybody that has done this. Have you had any problems and if so what. regards, =20 Robert W.Mills Systems Development Manager Windsong Services (020) 8309 3604 ------_=_NextPart_001_01C49F00.083CC987 Content-Type: text/html; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable Upgrading to 8.49D

Greeting to the list,

We are about to upgrade from = 8.19D7 and 8.39C to 8.49D and am interested in comments by anybody that = has done this.

Have you had any problems and = if so what.

regards,
 
Robert W.Mills
Systems Development = Manager
Windsong Services
(020) = 8309 3604

------_=_NextPart_001_01C49F00.083CC987-- From Jari.Kaljunen@teliasonera.com Mon Sep 20 14:26:38 2004 From: Jari.Kaljunen@teliasonera.com (Jari.Kaljunen@teliasonera.com) Date: Mon, 20 Sep 2004 15:26:38 +0200 Subject: SV: write to file with quick Message-ID: Hello and thank for the suggestions. I have now learned that the reason is an old bug in quick. The output file should be declared as direct in dictionary and the record length should be fixed. The performance is still not very goot but it will do just now. /Jari -----Ursprungligt meddelande----- Från: Vissers, Wilbert [mailto:wilbert.vissers@hp.com] Skickat: den 17 september 2004 04:17 Till: powerh-l@lists.sowder.com Ämne: RE:write to file with quick Hi Jari, these options may help, we think from your ref to RMS that you may be working on a Vax, so this is from Michael and I: Options: (a) Pre-allocate disc space for the file. The reason being that on a VAX the file may be being continuously extended by the operating system 1 record at a time because that's what quick would be writing. That's what the system administrator could have been seeing. (b) Write to an indexed file instead of a sequential file, then extract the data part afterwards with a utility or Quiz. Set a simple key to the record number to keep the index clean. The reason for this idea is that indexed files are more normal and there may well be something unusual about what quick needs to do when working with sequential files. Couldn't resist doing this at lunchtime, so we had a little play and ran some tests on a very slow old machine we have here, for 20,000 records: (1) Quiz to subfile = 25 seconds (2) Quick to sequential file = 32 minutes yes minutes (3) Quick to indexed file = 7 minutes Quiz code: ---------- ACCESS GL-BATCH REPORT SUMMARY ALL SET REPORT LIMIT 20000 SET SUBFILE GO Quick code: ----------- LET WK-COUNT = 0 INFO = "STARTING " + ASCII(SYSTIME,8) NOW RESPONSE WHILE WK-COUNT < 20000 BEGIN LET WK-COUNT = WK-COUNT + 1 LET BATCH-KEY OF GL-BATCH = ASCII(WK-COUNT) PUT GL-BATCH RESET ; <---- DESIGNER FILE END INFO="FINISHED " + ASCII(SYSTIME,8) NOW RESPONSE Hope something in this helps, best regards. Wilbert = = = = = = = = = = = = = = = = = = = = = = = = = = = = Mailing list: powerh-l@lists.sowder.com Subscribe: "subscribe" in message body to powerh-l-request@lists.sowder.com Unsubscribe: "unsubscribe " in message body to powerh-l-request@lists.sowder.com http://lists.sowder.com/mailman/listinfo/powerh-l This list is closed, thus to post to the list you must be a subscriber. From Jari.Kaljunen@teliasonera.com Mon Sep 20 14:36:47 2004 From: Jari.Kaljunen@teliasonera.com (Jari.Kaljunen@teliasonera.com) Date: Mon, 20 Sep 2004 15:36:47 +0200 Subject: SV: SV: write to file with quick /using quiz instead Message-ID: Hi, there will be "report" from 4 different tables. Of course we could have several quiz-runs producing files with additional sorting fields and then merge the files omitting sorting fields. Then there would be 2 problems left: disc space for the intermediate files and conversion of characters &,<,>,'," to xml-entities which is not very practical i quiz. /Jari -----Ursprungligt meddelande----- Från: Joe Boyle [mailto:joeboyle_adt@hotmail.com] Skickat: den 17 september 2004 23:28 Till: Kaljunen, Jari S. /TSS - Sergel Kredittjänster /+46-8-4526435; powerh-l@lists.sowder.com Ämne: RE: SV: write to file with quick why not allow the performance of quiz to generate the subfile and then use the DCL command 'CONVERT/FDL' to convert the file to the required sequential format. It may even be the case, that if the 'nodict' qualifier is used when creating the subfile, then the subfile need only be renamed and copied to the required directory. I am guessing that a 'nodict' subfile will have a sequential rather than direct record format; you could always use the analyse command to examine the subfile's record format if you aren't sure. Presumably, once the file is built, subsequent updates will be writing fewer records into the file, and the slow performance of Quick may not be as great an issue - or you could even go back to Qtp. Regards, Joe. This e-mail and all information contained in it is confidential and may be legally privileged. If you are not the intended recipient, your access to this e-mail is unauthorized. Any use, dissemination, distribution, publication or copying by you of this e-mail or any of the information contained within it is prohibited and may be unlawful. Do not open any attachments, delete it immediately from your system and notify the sender promptly by e-mail that you have done so. The content of this e-mail and any attachments sent with it may have been altered without the consent or knowledge of the author. From: powerh-l-admin@lists.sowder.com [mailto:powerh-l-admin@lists.sowder.com] On Behalf Of Jari.Kaljunen@teliasonera.com Sent: 16 September 2004 20:42 To: powerh-l@lists.sowder.com Subject: SV: SV: write to file with quick You mean file asc600 designer open append ? Well, it didn't help.. /Jari -----Ursprungligt meddelande----- Från: Peter Bateman [mailto:ppfb@hotmail.com] Skickat: den 16 september 2004 21:02 Till: powerh-l@lists.sowder.com Ämne: RE: SV: write to file with quick I would try 'file asc600 designer open append' It may be a machine specific option. But I didn't see it in the manual though. Regards, Peter >From: >To: >Subject: SV: write to file with quick >Date: Thu, 16 Sep 2004 12:22:20 +0200 > >Hello again, > >a little test without reading database still very slow (actually no difference...). It takes 49 s to run this with 100 records on subfile. The result file is 20100 lines. > >I even wrote a test quiz reading the same subfile and having 200 defines, each including systime and some more, each defined item written on an own line. It takes quiz 3 s to write these 20000 lines. > > >/Jari > > >screen test_writing_file menu stopscreen > >file *ik006002 designer >file asc600 designer > >temp t_count > >procedure internal write > begin > for 200 > begin > let t_count = t_count + 1 > let rec600 = asc(t_count) & > + " " + asc(systime) + " procedure internal write " > put asc600 reset > end > end > >procedure initialize > begin > let t_count = 0 > while retr ik006002 seq > begin > let rec600 = asc(t_count) & > + " " + asc(systime) + " procedure initialize " > put asc600 reset > do internal write > end > return > end > >build > >-----Ursprungligt meddelande----- >Från: Johnson, Harold A EDUC:EX >[mailto:Harold.A.Johnson@gems1.gov.bc.ca] >Skickat: den 15 september 2004 17:50 >Till: Kaljunen, Jari S. /TSS - Sergel Kredittjänster /+46-8-4526435; >powerh-l@lists.sowder.com >Ämne: RE: write to file with quick > > >Are you sure the performance problem is with the writing and not with the >reading of the original data? I would suspect a problem there instead. > > >-----Original Message----- >From: powerh-l-admin@lists.sowder.com >[mailto:powerh-l-admin@lists.sowder.com]On Behalf Of >Jari.Kaljunen@teliasonera.com >Sent: 2004 September 15 8:41 AM >To: powerh-l@lists.sowder.com >Subject: write to file with quick > > >Hello, > >I am writing to > >File: ASC600 >Organization: SEQUENTIAL >Type: RMS >Open: ASC600 >Record Format: Variable >Supersede: No >Records & Sizes: ASC600 600 bytes >Item Type Size Occ Offset >REC600 CHARACTER 600 > >in quick using let and put as follows > >let rec600 = "" + asc(arnr of arenden,8) + asc(parmblnr,2) & >+ atgkod of atgkoder + "" > >put asc600 reset > > >The performance is very poor. The administrator believs that quick closes >and opens the file all the time. >Can that be true? >What should I do? > >We are creating an xml-file (it will be about 5 GB) with data from several >tables and our solution with qtp needed too many subfiles and sortings using >lots of space. > >/Jari > >= = = = = = = = = = = = = = = = = = = = = = = = = = = = >Mailing list: powerh-l@lists.sowder.com >Subscribe: "subscribe" in message body to powerh-l-request@lists.sowder.com >Unsubscribe: "unsubscribe " in message body to >powerh-l-request@lists.sowder.com >http://lists.sowder.com/mailman/listinfo/powerh-l >This list is closed, thus to post to the list you must be a subscriber. > >= = = = = = = = = = = = = = = = = = = = = = = = = = = = >Mailing list: powerh-l@lists.sowder.com >Subscribe: "subscribe" in message body to powerh-l-request@lists.sowder.com >Unsubscribe: "unsubscribe " in message body to powerh-l-request@lists.sowder.com >http://lists.sowder.com/mailman/listinfo/powerh-l >This list is closed, thus to post to the list you must be a subscriber. Don't just Search. Find! The new MSN Search: Fast. Clear. Easy. = = = = = = = = = = = = = = = = = = = = = = = = = = = = Mailing list: powerh-l@lists.sowder.com Subscribe: "subscribe" in message body to powerh-l-request@lists.sowder.com Unsubscribe: "unsubscribe " in message body to powerh-l-request@lists.sowder.com http://lists.sowder.com/mailman/listinfo/powerh-l This list is closed, thus to post to the list you must be a subscriber. From Bob.Deskin@Cognos.COM Mon Sep 20 14:44:29 2004 From: Bob.Deskin@Cognos.COM (Deskin, Bob) Date: Mon, 20 Sep 2004 09:44:29 -0400 Subject: write to file with quick Message-ID: <2B4E386EAEC90947802323BF373DC0EC0259214D@sottemail1.ent.ad.cognos.com> I checked the bug in QUICK. It dates back to the mid-90s. There was a problem with sequential files where the last buffer load was not being posted in certain cases. To solve this problem, the buffer was posted on every write. This, of course, resulted in a performance issue where large numbers of records were written. Unfortunately, it seems to be a case of catch 22 and the bug was never resolved completely. Most files declared as SEQUENTIAL should actually be declared as DIRECT because they're used that way. SEQUENTIAL is actually used for things like tape files that cannot be updated in place. Bob Deskin Senior 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: powerh-l-admin@lists.sowder.com [mailto:powerh-l-admin@lists.sowder.com] On Behalf Of Jari.Kaljunen@teliasonera.com Sent: September 20, 2004 9:27 AM To: wilbert.vissers@hp.com; powerh-l@lists.sowder.com Subject: SV: write to file with quick Hello and thank for the suggestions. I have now learned that the reason is an old bug in quick. The output file should be declared as direct in dictionary and the record length should be fixed. The performance is still not very goot but it will do just now. /Jari -----Ursprungligt meddelande----- Från: Vissers, Wilbert [mailto:wilbert.vissers@hp.com] Skickat: den 17 september 2004 04:17 Till: powerh-l@lists.sowder.com Ämne: RE:write to file with quick Hi Jari, these options may help, we think from your ref to RMS that you may be working on a Vax, so this is from Michael and I: Options: (a) Pre-allocate disc space for the file. The reason being that on a VAX the file may be being continuously extended by the operating system 1 record at a time because that's what quick would be writing. That's what the system administrator could have been seeing. (b) Write to an indexed file instead of a sequential file, then extract the data part afterwards with a utility or Quiz. Set a simple key to the record number to keep the index clean. The reason for this idea is that indexed files are more normal and there may well be something unusual about what quick needs to do when working with sequential files. Couldn't resist doing this at lunchtime, so we had a little play and ran some tests on a very slow old machine we have here, for 20,000 records: (1) Quiz to subfile = 25 seconds (2) Quick to sequential file = 32 minutes yes minutes (3) Quick to indexed file = 7 minutes Quiz code: ---------- ACCESS GL-BATCH REPORT SUMMARY ALL SET REPORT LIMIT 20000 SET SUBFILE GO Quick code: ----------- LET WK-COUNT = 0 INFO = "STARTING " + ASCII(SYSTIME,8) NOW RESPONSE WHILE WK-COUNT < 20000 BEGIN LET WK-COUNT = WK-COUNT + 1 LET BATCH-KEY OF GL-BATCH = ASCII(WK-COUNT) PUT GL-BATCH RESET ; <---- DESIGNER FILE END INFO="FINISHED " + ASCII(SYSTIME,8) NOW RESPONSE Hope something in this helps, best regards. Wilbert = = = = = = = = = = = = = = = = = = = = = = = = = = = = Mailing list: powerh-l@lists.sowder.com Subscribe: "subscribe" in message body to powerh-l-request@lists.sowder.com Unsubscribe: "unsubscribe " in message body to powerh-l-request@lists.sowder.com http://lists.sowder.com/mailman/listinfo/powerh-l This list is closed, thus to post to the list you must be a subscriber. = = = = = = = = = = = = = = = = = = = = = = = = = = = = Mailing list: powerh-l@lists.sowder.com Subscribe: "subscribe" in message body to powerh-l-request@lists.sowder.com Unsubscribe: "unsubscribe " in message body to powerh-l-request@lists.sowder.com http://lists.sowder.com/mailman/listinfo/powerh-l This list is closed, thus to post to the list you must be a subscriber. This message may contain privileged and/or confidential information. If you have received this e-mail in error or are not the intended recipient, you may not use, copy, disseminate or distribute it; do not open any attachments, delete it immediately from your system and notify the sender promptly by e-mail that you have done so. Thank you. From macljohn@isu.edu Mon Sep 20 17:41:13 2004 From: macljohn@isu.edu (John MacLerran) Date: Mon, 20 Sep 2004 10:41:13 -0600 Subject: QKGO files from MPE to Unix? Message-ID: <414F0829.7070400@isu.edu> Dear PowerHouse List We're on the road to migrate from PowerHouse on MPE to PowerHouse on HP-UX. Can anyone tell me if QKGO files can be converted from one format to the other, and if so, how? I tried just FTP'ing a file, then creating a new blank one in QKGO on unix, then CONV to attempt to convert it -- it didn't give any error messages, but my DFK entries didn't come through. Thanks! -- ---------------------------------------------------------------------- John MacLerran IT Systems Analyst email: macljohn@isu.edu Idaho State University V(208) 282-2954 http://www.isu.edu/~macljohn F(208) 282-3673 ---------------------------------------------------------------------- From Bob.Deskin@Cognos.COM Mon Sep 20 18:34:01 2004 From: Bob.Deskin@Cognos.COM (Deskin, Bob) Date: Mon, 20 Sep 2004 13:34:01 -0400 Subject: QKGO files from MPE to Unix? Message-ID: <2B4E386EAEC90947802323BF373DC0EC02592156@sottemail1.ent.ad.cognos.com> Unless someone's developed a method for converting, there's no official method from Cognos. Since QKGO file contents are usually fairly straightforward, conversion rarely comes up. As for copying, the TIC files are indexed and are not transferable. Bob Deskin Senior 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: powerh-l-admin@lists.sowder.com [mailto:powerh-l-admin@lists.sowder.com] On Behalf Of John MacLerran Sent: September 20, 2004 12:41 PM To: PowerHouse Mail List (E-mail) Subject: QKGO files from MPE to Unix? Dear PowerHouse List We're on the road to migrate from PowerHouse on MPE to PowerHouse on HP-UX. Can anyone tell me if QKGO files can be converted from one format to the other, and if so, how? I tried just FTP'ing a file, then creating a new blank one in QKGO on unix, then CONV to attempt to convert it -- it didn't give any error messages, but my DFK entries didn't come through. Thanks! -- ---------------------------------------------------------------------- John MacLerran IT Systems Analyst email: macljohn@isu.edu Idaho State University V(208) 282-2954 http://www.isu.edu/~macljohn F(208) 282-3673 ---------------------------------------------------------------------- = = = = = = = = = = = = = = = = = = = = = = = = = = = = Mailing list: powerh-l@lists.sowder.com Subscribe: "subscribe" in message body to powerh-l-request@lists.sowder.com Unsubscribe: "unsubscribe " in message body to powerh-l-request@lists.sowder.com http://lists.sowder.com/mailman/listinfo/powerh-l This list is closed, thus to post to the list you must be a subscriber. This message may contain privileged and/or confidential information. If you have received this e-mail in error or are not the intended recipient, you may not use, copy, disseminate or distribute it; do not open any attachments, delete it immediately from your system and notify the sender promptly by e-mail that you have done so. Thank you. From joeboyle_adt@hotmail.com Tue Sep 21 11:03:06 2004 From: joeboyle_adt@hotmail.com (Joe Boyle) Date: Tue, 21 Sep 2004 11:03:06 +0100 Subject: SV: write to file with quick /using quiz instead In-Reply-To: Message-ID: personally, I would prefer to code the requirements that you list below and then use 'convert', rather than wait for quick to write 9 million 600 byte records ( 5 gig ); especially as you say that performance still isn't great even when using the direct format. Regards, Joe. This e-mail and all information contained in it is confidential and may be legally privileged. If you are not the intended recipient, your access to this e-mail is unauthorized. Any use, dissemination, distribution, publication or copying by you of this e-mail or any of the information contained within it is prohibited and may be unlawful. Do not open any attachments, delete it immediately from your system and notify the sender promptly by e-mail that you have done so. The content of this e-mail and any attachments sent with it may have been altered without the consent or knowledge of the author. -----Original Message----- From: Jari.Kaljunen@teliasonera.com [mailto:Jari.Kaljunen@teliasonera.com] Sent: 20 September 2004 14:37 To: joeboyle_adt@hotmail.com; powerh-l@lists.sowder.com Subject: SV: SV: write to file with quick /using quiz instead Hi, there will be "report" from 4 different tables. Of course we could have several quiz-runs producing files with additional sorting fields and then merge the files omitting sorting fields. Then there would be 2 problems left: disc space for the intermediate files and conversion of characters &,<,>,'," to xml-entities which is not very practical i quiz. /Jari -----Ursprungligt meddelande----- Från: Joe Boyle [mailto:joeboyle_adt@hotmail.com] Skickat: den 17 september 2004 23:28 Till: Kaljunen, Jari S. /TSS - Sergel Kredittjänster /+46-8-4526435; powerh-l@lists.sowder.com Ämne: RE: SV: write to file with quick why not allow the performance of quiz to generate the subfile and then use the DCL command 'CONVERT/FDL' to convert the file to the required sequential format. It may even be the case, that if the 'nodict' qualifier is used when creating the subfile, then the subfile need only be renamed and copied to the required directory. I am guessing that a 'nodict' subfile will have a sequential rather than direct record format; you could always use the analyse command to examine the subfile's record format if you aren't sure. Presumably, once the file is built, subsequent updates will be writing fewer records into the file, and the slow performance of Quick may not be as great an issue - or you could even go back to Qtp. Regards, Joe. This e-mail and all information contained in it is confidential and may be legally privileged. If you are not the intended recipient, your access to this e-mail is unauthorized. Any use, dissemination, distribution, publication or copying by you of this e-mail or any of the information contained within it is prohibited and may be unlawful. Do not open any attachments, delete it immediately from your system and notify the sender promptly by e-mail that you have done so. The content of this e-mail and any attachments sent with it may have been altered without the consent or knowledge of the author. From: powerh-l-admin@lists.sowder.com [mailto:powerh-l-admin@lists.sowder.com] On Behalf Of Jari.Kaljunen@teliasonera.com Sent: 16 September 2004 20:42 To: powerh-l@lists.sowder.com Subject: SV: SV: write to file with quick You mean file asc600 designer open append ? Well, it didn't help.. /Jari -----Ursprungligt meddelande----- Från: Peter Bateman [mailto:ppfb@hotmail.com] Skickat: den 16 september 2004 21:02 Till: powerh-l@lists.sowder.com Ämne: RE: SV: write to file with quick I would try 'file asc600 designer open append' It may be a machine specific option. But I didn't see it in the manual though. Regards, Peter >From: >To: >Subject: SV: write to file with quick >Date: Thu, 16 Sep 2004 12:22:20 +0200 > >Hello again, > >a little test without reading database still very slow (actually no difference...). It takes 49 s to run this with 100 records on subfile. The result file is 20100 lines. > >I even wrote a test quiz reading the same subfile and having 200 defines, each including systime and some more, each defined item written on an own line. It takes quiz 3 s to write these 20000 lines. > > >/Jari > > >screen test_writing_file menu stopscreen > >file *ik006002 designer >file asc600 designer > >temp t_count > >procedure internal write > begin > for 200 > begin > let t_count = t_count + 1 > let rec600 = asc(t_count) & > + " " + asc(systime) + " procedure internal write " > put asc600 reset > end > end > >procedure initialize > begin > let t_count = 0 > while retr ik006002 seq > begin > let rec600 = asc(t_count) & > + " " + asc(systime) + " procedure initialize " > put asc600 reset > do internal write > end > return > end > >build > >-----Ursprungligt meddelande----- >Från: Johnson, Harold A EDUC:EX >[mailto:Harold.A.Johnson@gems1.gov.bc.ca] >Skickat: den 15 september 2004 17:50 >Till: Kaljunen, Jari S. /TSS - Sergel Kredittjänster /+46-8-4526435; >powerh-l@lists.sowder.com >Ämne: RE: write to file with quick > > >Are you sure the performance problem is with the writing and not with the >reading of the original data? I would suspect a problem there instead. > > >-----Original Message----- >From: powerh-l-admin@lists.sowder.com >[mailto:powerh-l-admin@lists.sowder.com]On Behalf Of >Jari.Kaljunen@teliasonera.com >Sent: 2004 September 15 8:41 AM >To: powerh-l@lists.sowder.com >Subject: write to file with quick > > >Hello, > >I am writing to > >File: ASC600 >Organization: SEQUENTIAL >Type: RMS >Open: ASC600 >Record Format: Variable >Supersede: No >Records & Sizes: ASC600 600 bytes >Item Type Size Occ Offset >REC600 CHARACTER 600 > >in quick using let and put as follows > >let rec600 = "" + asc(arnr of arenden,8) + asc(parmblnr,2) & >+ atgkod of atgkoder + "" > >put asc600 reset > > >The performance is very poor. The administrator believs that quick closes >and opens the file all the time. >Can that be true? >What should I do? > >We are creating an xml-file (it will be about 5 GB) with data from several >tables and our solution with qtp needed too many subfiles and sortings using >lots of space. > >/Jari > >= = = = = = = = = = = = = = = = = = = = = = = = = = = = >Mailing list: powerh-l@lists.sowder.com >Subscribe: "subscribe" in message body to powerh-l-request@lists.sowder.com >Unsubscribe: "unsubscribe " in message body to >powerh-l-request@lists.sowder.com >http://lists.sowder.com/mailman/listinfo/powerh-l >This list is closed, thus to post to the list you must be a subscriber. > >= = = = = = = = = = = = = = = = = = = = = = = = = = = = >Mailing list: powerh-l@lists.sowder.com >Subscribe: "subscribe" in message body to powerh-l-request@lists.sowder.com >Unsubscribe: "unsubscribe " in message body to powerh-l-request@lists.sowder.com >http://lists.sowder.com/mailman/listinfo/powerh-l >This list is closed, thus to post to the list you must be a subscriber. Don't just Search. Find! The new MSN Search: Fast. Clear. Easy. = = = = = = = = = = = = = = = = = = = = = = = = = = = = Mailing list: powerh-l@lists.sowder.com Subscribe: "subscribe" in message body to powerh-l-request@lists.sowder.com Unsubscribe: "unsubscribe " in message body to powerh-l-request@lists.sowder.com http://lists.sowder.com/mailman/listinfo/powerh-l This list is closed, thus to post to the list you must be a subscriber. From Jari.Kaljunen@teliasonera.com Tue Sep 21 11:30:41 2004 From: Jari.Kaljunen@teliasonera.com (Jari.Kaljunen@teliasonera.com) Date: Tue, 21 Sep 2004 12:30:41 +0200 Subject: SV: SV: write to file with quick /using quiz instead Message-ID: Hi the output record format is not critical, the result file will be used by other system that don't care. But what do you recommend to convert the characters &,<,>,' and " to xml-entities if quiz or qtp is used produce the file? We don't have controls for these characters in DB now and we get files from many sources. We already had a solution with qtp that produced the file but the problem with these characters was not solved. A collegue succested to use quick for that purpose (first separate the data from xml-tags with binary characters and then only check the data part) but the quick performance problem still remains. DCL was far too slow. Management don't want to install Perl (I believe that Perl using regular expressions would be fast enough), collegues don't like the idea of using external C-program... We appreciate any suggestions. /Jari -----Ursprungligt meddelande----- Från: Joe Boyle [mailto:joeboyle_adt@hotmail.com] Skickat: den 21 september 2004 12:03 Till: Kaljunen, Jari S. /TSS - Sergel Kredittjänster /+46-8-4526435; powerh-l@lists.sowder.com Ämne: RE: SV: write to file with quick /using quiz instead personally, I would prefer to code the requirements that you list below and then use 'convert', rather than wait for quick to write 9 million 600 byte records ( 5 gig ); especially as you say that performance still isn't great even when using the direct format. Regards, Joe. This e-mail and all information contained in it is confidential and may be legally privileged. If you are not the intended recipient, your access to this e-mail is unauthorized. Any use, dissemination, distribution, publication or copying by you of this e-mail or any of the information contained within it is prohibited and may be unlawful. Do not open any attachments, delete it immediately from your system and notify the sender promptly by e-mail that you have done so. The content of this e-mail and any attachments sent with it may have been altered without the consent or knowledge of the author. -----Original Message----- From: Jari.Kaljunen@teliasonera.com [mailto:Jari.Kaljunen@teliasonera.com] Sent: 20 September 2004 14:37 To: joeboyle_adt@hotmail.com; powerh-l@lists.sowder.com Subject: SV: SV: write to file with quick /using quiz instead Hi, there will be "report" from 4 different tables. Of course we could have several quiz-runs producing files with additional sorting fields and then merge the files omitting sorting fields. Then there would be 2 problems left: disc space for the intermediate files and conversion of characters &,<,>,'," to xml-entities which is not very practical i quiz. /Jari -----Ursprungligt meddelande----- Från: Joe Boyle [mailto:joeboyle_adt@hotmail.com] Skickat: den 17 september 2004 23:28 Till: Kaljunen, Jari S. /TSS - Sergel Kredittjänster /+46-8-4526435; powerh-l@lists.sowder.com Ämne: RE: SV: write to file with quick why not allow the performance of quiz to generate the subfile and then use the DCL command 'CONVERT/FDL' to convert the file to the required sequential format. It may even be the case, that if the 'nodict' qualifier is used when creating the subfile, then the subfile need only be renamed and copied to the required directory. I am guessing that a 'nodict' subfile will have a sequential rather than direct record format; you could always use the analyse command to examine the subfile's record format if you aren't sure. Presumably, once the file is built, subsequent updates will be writing fewer records into the file, and the slow performance of Quick may not be as great an issue - or you could even go back to Qtp. Regards, Joe. This e-mail and all information contained in it is confidential and may be legally privileged. If you are not the intended recipient, your access to this e-mail is unauthorized. Any use, dissemination, distribution, publication or copying by you of this e-mail or any of the information contained within it is prohibited and may be unlawful. Do not open any attachments, delete it immediately from your system and notify the sender promptly by e-mail that you have done so. The content of this e-mail and any attachments sent with it may have been altered without the consent or knowledge of the author. From: powerh-l-admin@lists.sowder.com [mailto:powerh-l-admin@lists.sowder.com] On Behalf Of Jari.Kaljunen@teliasonera.com Sent: 16 September 2004 20:42 To: powerh-l@lists.sowder.com Subject: SV: SV: write to file with quick You mean file asc600 designer open append ? Well, it didn't help.. /Jari -----Ursprungligt meddelande----- Från: Peter Bateman [mailto:ppfb@hotmail.com] Skickat: den 16 september 2004 21:02 Till: powerh-l@lists.sowder.com Ämne: RE: SV: write to file with quick I would try 'file asc600 designer open append' It may be a machine specific option. But I didn't see it in the manual though. Regards, Peter >From: >To: >Subject: SV: write to file with quick >Date: Thu, 16 Sep 2004 12:22:20 +0200 > >Hello again, > >a little test without reading database still very slow (actually no difference...). It takes 49 s to run this with 100 records on subfile. The result file is 20100 lines. > >I even wrote a test quiz reading the same subfile and having 200 defines, each including systime and some more, each defined item written on an own line. It takes quiz 3 s to write these 20000 lines. > > >/Jari > > >screen test_writing_file menu stopscreen > >file *ik006002 designer >file asc600 designer > >temp t_count > >procedure internal write > begin > for 200 > begin > let t_count = t_count + 1 > let rec600 = asc(t_count) & > + " " + asc(systime) + " procedure internal write " > put asc600 reset > end > end > >procedure initialize > begin > let t_count = 0 > while retr ik006002 seq > begin > let rec600 = asc(t_count) & > + " " + asc(systime) + " procedure initialize " > put asc600 reset > do internal write > end > return > end > >build > >-----Ursprungligt meddelande----- >Från: Johnson, Harold A EDUC:EX >[mailto:Harold.A.Johnson@gems1.gov.bc.ca] >Skickat: den 15 september 2004 17:50 >Till: Kaljunen, Jari S. /TSS - Sergel Kredittjänster /+46-8-4526435; >powerh-l@lists.sowder.com >Ämne: RE: write to file with quick > > >Are you sure the performance problem is with the writing and not with the >reading of the original data? I would suspect a problem there instead. > > >-----Original Message----- >From: powerh-l-admin@lists.sowder.com >[mailto:powerh-l-admin@lists.sowder.com]On Behalf Of >Jari.Kaljunen@teliasonera.com >Sent: 2004 September 15 8:41 AM >To: powerh-l@lists.sowder.com >Subject: write to file with quick > > >Hello, > >I am writing to > >File: ASC600 >Organization: SEQUENTIAL >Type: RMS >Open: ASC600 >Record Format: Variable >Supersede: No >Records & Sizes: ASC600 600 bytes >Item Type Size Occ Offset >REC600 CHARACTER 600 > >in quick using let and put as follows > >let rec600 = "" + asc(arnr of arenden,8) + asc(parmblnr,2) & >+ atgkod of atgkoder + "" > >put asc600 reset > > >The performance is very poor. The administrator believs that quick closes >and opens the file all the time. >Can that be true? >What should I do? > >We are creating an xml-file (it will be about 5 GB) with data from several >tables and our solution with qtp needed too many subfiles and sortings using >lots of space. > >/Jari > >= = = = = = = = = = = = = = = = = = = = = = = = = = = = >Mailing list: powerh-l@lists.sowder.com >Subscribe: "subscribe" in message body to powerh-l-request@lists.sowder.com >Unsubscribe: "unsubscribe " in message body to >powerh-l-request@lists.sowder.com >http://lists.sowder.com/mailman/listinfo/powerh-l >This list is closed, thus to post to the list you must be a subscriber. > >= = = = = = = = = = = = = = = = = = = = = = = = = = = = >Mailing list: powerh-l@lists.sowder.com >Subscribe: "subscribe" in message body to powerh-l-request@lists.sowder.com >Unsubscribe: "unsubscribe " in message body to powerh-l-request@lists.sowder.com >http://lists.sowder.com/mailman/listinfo/powerh-l >This list is closed, thus to post to the list you must be a subscriber. Don't just Search. Find! The new MSN Search: Fast. Clear. Easy. = = = = = = = = = = = = = = = = = = = = = = = = = = = = Mailing list: powerh-l@lists.sowder.com Subscribe: "subscribe" in message body to powerh-l-request@lists.sowder.com Unsubscribe: "unsubscribe " in message body to powerh-l-request@lists.sowder.com http://lists.sowder.com/mailman/listinfo/powerh-l This list is closed, thus to post to the list you must be a subscriber. From joeboyle_adt@hotmail.com Tue Sep 21 11:49:33 2004 From: joeboyle_adt@hotmail.com (Joe Boyle) Date: Tue, 21 Sep 2004 11:49:33 +0100 Subject: SV: write to file with quick /using quiz instead In-Reply-To: Message-ID: I may have misunderstood your comments below, I imagined that you were simply looking to replace the occurrences of these characters in your data with other characters. If this is not the case it might be better to give an example of what your current qdesign source code is doing, and it should be possible to find a quiz equivalent. Regards, Joe. This e-mail and all information contained in it is confidential and may be legally privileged. If you are not the intended recipient, your access to this e-mail is unauthorized. Any use, dissemination, distribution, publication or copying by you of this e-mail or any of the information contained within it is prohibited and may be unlawful. Do not open any attachments, delete it immediately from your system and notify the sender promptly by e-mail that you have done so. The content of this e-mail and any attachments sent with it may have been altered without the consent or knowledge of the author. -----Original Message----- From: powerh-l-admin@lists.sowder.com [mailto:powerh-l-admin@lists.sowder.com] On Behalf Of Jari.Kaljunen@teliasonera.com Sent: 21 September 2004 11:31 To: powerh-l@lists.sowder.com Subject: SV: SV: write to file with quick /using quiz instead Hi the output record format is not critical, the result file will be used by other system that don't care. But what do you recommend to convert the characters &,<,>,' and " to xml-entities if quiz or qtp is used produce the file? We don't have controls for these characters in DB now and we get files from many sources. We already had a solution with qtp that produced the file but the problem with these characters was not solved. A collegue succested to use quick for that purpose (first separate the data from xml-tags with binary characters and then only check the data part) but the quick performance problem still remains. DCL was far too slow. Management don't want to install Perl (I believe that Perl using regular expressions would be fast enough), collegues don't like the idea of using external C-program... We appreciate any suggestions. /Jari -----Ursprungligt meddelande----- Från: Joe Boyle [mailto:joeboyle_adt@hotmail.com] Skickat: den 21 september 2004 12:03 Till: Kaljunen, Jari S. /TSS - Sergel Kredittjänster /+46-8-4526435; powerh-l@lists.sowder.com Ämne: RE: SV: write to file with quick /using quiz instead personally, I would prefer to code the requirements that you list below and then use 'convert', rather than wait for quick to write 9 million 600 byte records ( 5 gig ); especially as you say that performance still isn't great even when using the direct format. Regards, Joe. This e-mail and all information contained in it is confidential and may be legally privileged. If you are not the intended recipient, your access to this e-mail is unauthorized. Any use, dissemination, distribution, publication or copying by you of this e-mail or any of the information contained within it is prohibited and may be unlawful. Do not open any attachments, delete it immediately from your system and notify the sender promptly by e-mail that you have done so. The content of this e-mail and any attachments sent with it may have been altered without the consent or knowledge of the author. -----Original Message----- From: Jari.Kaljunen@teliasonera.com [mailto:Jari.Kaljunen@teliasonera.com] Sent: 20 September 2004 14:37 To: joeboyle_adt@hotmail.com; powerh-l@lists.sowder.com Subject: SV: SV: write to file with quick /using quiz instead Hi, there will be "report" from 4 different tables. Of course we could have several quiz-runs producing files with additional sorting fields and then merge the files omitting sorting fields. Then there would be 2 problems left: disc space for the intermediate files and conversion of characters &,<,>,'," to xml-entities which is not very practical i quiz. /Jari -----Ursprungligt meddelande----- Från: Joe Boyle [mailto:joeboyle_adt@hotmail.com] Skickat: den 17 september 2004 23:28 Till: Kaljunen, Jari S. /TSS - Sergel Kredittjänster /+46-8-4526435; powerh-l@lists.sowder.com Ämne: RE: SV: write to file with quick why not allow the performance of quiz to generate the subfile and then use the DCL command 'CONVERT/FDL' to convert the file to the required sequential format. It may even be the case, that if the 'nodict' qualifier is used when creating the subfile, then the subfile need only be renamed and copied to the required directory. I am guessing that a 'nodict' subfile will have a sequential rather than direct record format; you could always use the analyse command to examine the subfile's record format if you aren't sure. Presumably, once the file is built, subsequent updates will be writing fewer records into the file, and the slow performance of Quick may not be as great an issue - or you could even go back to Qtp. Regards, Joe. This e-mail and all information contained in it is confidential and may be legally privileged. If you are not the intended recipient, your access to this e-mail is unauthorized. Any use, dissemination, distribution, publication or copying by you of this e-mail or any of the information contained within it is prohibited and may be unlawful. Do not open any attachments, delete it immediately from your system and notify the sender promptly by e-mail that you have done so. The content of this e-mail and any attachments sent with it may have been altered without the consent or knowledge of the author. From: powerh-l-admin@lists.sowder.com [mailto:powerh-l-admin@lists.sowder.com] On Behalf Of Jari.Kaljunen@teliasonera.com Sent: 16 September 2004 20:42 To: powerh-l@lists.sowder.com Subject: SV: SV: write to file with quick You mean file asc600 designer open append ? Well, it didn't help.. /Jari -----Ursprungligt meddelande----- Från: Peter Bateman [mailto:ppfb@hotmail.com] Skickat: den 16 september 2004 21:02 Till: powerh-l@lists.sowder.com Ämne: RE: SV: write to file with quick I would try 'file asc600 designer open append' It may be a machine specific option. But I didn't see it in the manual though. Regards, Peter >From: >To: >Subject: SV: write to file with quick >Date: Thu, 16 Sep 2004 12:22:20 +0200 > >Hello again, > >a little test without reading database still very slow (actually no difference...). It takes 49 s to run this with 100 records on subfile. The result file is 20100 lines. > >I even wrote a test quiz reading the same subfile and having 200 defines, each including systime and some more, each defined item written on an own line. It takes quiz 3 s to write these 20000 lines. > > >/Jari > > >screen test_writing_file menu stopscreen > >file *ik006002 designer >file asc600 designer > >temp t_count > >procedure internal write > begin > for 200 > begin > let t_count = t_count + 1 > let rec600 = asc(t_count) & > + " " + asc(systime) + " procedure internal write " > put asc600 reset > end > end > >procedure initialize > begin > let t_count = 0 > while retr ik006002 seq > begin > let rec600 = asc(t_count) & > + " " + asc(systime) + " procedure initialize " > put asc600 reset > do internal write > end > return > end > >build > >-----Ursprungligt meddelande----- >Från: Johnson, Harold A EDUC:EX >[mailto:Harold.A.Johnson@gems1.gov.bc.ca] >Skickat: den 15 september 2004 17:50 >Till: Kaljunen, Jari S. /TSS - Sergel Kredittjänster /+46-8-4526435; >powerh-l@lists.sowder.com >Ämne: RE: write to file with quick > > >Are you sure the performance problem is with the writing and not with the >reading of the original data? I would suspect a problem there instead. > > >-----Original Message----- >From: powerh-l-admin@lists.sowder.com >[mailto:powerh-l-admin@lists.sowder.com]On Behalf Of >Jari.Kaljunen@teliasonera.com >Sent: 2004 September 15 8:41 AM >To: powerh-l@lists.sowder.com >Subject: write to file with quick > > >Hello, > >I am writing to > >File: ASC600 >Organization: SEQUENTIAL >Type: RMS >Open: ASC600 >Record Format: Variable >Supersede: No >Records & Sizes: ASC600 600 bytes >Item Type Size Occ Offset >REC600 CHARACTER 600 > >in quick using let and put as follows > >let rec600 = "" + asc(arnr of arenden,8) + asc(parmblnr,2) & >+ atgkod of atgkoder + "" > >put asc600 reset > > >The performance is very poor. The administrator believs that quick closes >and opens the file all the time. >Can that be true? >What should I do? > >We are creating an xml-file (it will be about 5 GB) with data from several >tables and our solution with qtp needed too many subfiles and sortings using >lots of space. > >/Jari > >= = = = = = = = = = = = = = = = = = = = = = = = = = = = >Mailing list: powerh-l@lists.sowder.com >Subscribe: "subscribe" in message body to powerh-l-request@lists.sowder.com >Unsubscribe: "unsubscribe " in message body to >powerh-l-request@lists.sowder.com >http://lists.sowder.com/mailman/listinfo/powerh-l >This list is closed, thus to post to the list you must be a subscriber. > >= = = = = = = = = = = = = = = = = = = = = = = = = = = = >Mailing list: powerh-l@lists.sowder.com >Subscribe: "subscribe" in message body to powerh-l-request@lists.sowder.com >Unsubscribe: "unsubscribe " in message body to powerh-l-request@lists.sowder.com >http://lists.sowder.com/mailman/listinfo/powerh-l >This list is closed, thus to post to the list you must be a subscriber. Don't just Search. Find! The new MSN Search: Fast. Clear. Easy. = = = = = = = = = = = = = = = = = = = = = = = = = = = = Mailing list: powerh-l@lists.sowder.com Subscribe: "subscribe" in message body to powerh-l-request@lists.sowder.com Unsubscribe: "unsubscribe " in message body to powerh-l-request@lists.sowder.com http://lists.sowder.com/mailman/listinfo/powerh-l This list is closed, thus to post to the list you must be a subscriber. = = = = = = = = = = = = = = = = = = = = = = = = = = = = Mailing list: powerh-l@lists.sowder.com Subscribe: "subscribe" in message body to powerh-l-request@lists.sowder.com Unsubscribe: "unsubscribe " in message body to powerh-l-request@lists.sowder.com http://lists.sowder.com/mailman/listinfo/powerh-l This list is closed, thus to post to the list you must be a subscriber. From Jari.Kaljunen@teliasonera.com Tue Sep 21 12:14:03 2004 From: Jari.Kaljunen@teliasonera.com (Jari.Kaljunen@teliasonera.com) Date: Tue, 21 Sep 2004 13:14:03 +0200 Subject: SV: SV: write to file with quick /using quiz instead Message-ID: below is what we do with fields that might have characters that has to be converted to xml-entities. It would be nice to have a function to use standard regular expressions in PH, wouldn't it? /Jari procedure internal fixa_tecken ; to be optimized begin let t_langd = size(trunc(t_kvar)) let t_klar = 0 let t_klart = "" while t_langd > t_klar begin let t_tecken = t_kvar[t_klar + 1 : 1] if t_tecken = "&" then begin let t_tecken_2 = "&" end else if t_tecken = "<" then begin let t_tecken_2 = "<" end else if t_tecken = '"' then begin let t_tecken_2 = """ end else if t_tecken = ">" then begin let t_tecken_2 = ">" end else if t_tecken = "'" then begin let t_tecken_2 = "'" end else begin let t_tecken_2 = t_tecken end let t_klart = trunc(t_klart) + trunc(t_tecken_2) let t_klar = t_klar + 1 let t_tecken = t_kvar[t_klar + 1 : 1] end end procedure internal kolla_t_kvar ; to be optimized begin let t_ix_amp = index (t_kvar, '&') let t_ix_apos = index (t_kvar, "'") let t_ix_quot = index (t_kvar, '"') let t_ix_gt = index (t_kvar, '>') let t_ix_lt = index (t_kvar, '<') if t_ix_amp <> 0 or t_ix_apos <> 0 or t_ix_quot <> 0 or t_ix_gt <> 0 or t_ix_lt <> 0 then do internal fixa_tecken else begin let t_klart = t_kvar end end let t_kvar = anamn1 of arenden ; suspect field do internal kolla_t_kvar let mperec600 = "" + trunc(t_klart) + "" put mpe600 reset -----Ursprungligt meddelande----- Från: Joe Boyle [mailto:joeboyle_adt@hotmail.com] Skickat: den 21 september 2004 12:50 Till: Kaljunen, Jari S. /TSS - Sergel Kredittjänster /+46-8-4526435; powerh-l@lists.sowder.com Ämne: RE: SV: write to file with quick /using quiz instead I may have misunderstood your comments below, I imagined that you were simply looking to replace the occurrences of these characters in your data with other characters. If this is not the case it might be better to give an example of what your current qdesign source code is doing, and it should be possible to find a quiz equivalent. Regards, Joe. This e-mail and all information contained in it is confidential and may be legally privileged. If you are not the intended recipient, your access to this e-mail is unauthorized. Any use, dissemination, distribution, publication or copying by you of this e-mail or any of the information contained within it is prohibited and may be unlawful. Do not open any attachments, delete it immediately from your system and notify the sender promptly by e-mail that you have done so. The content of this e-mail and any attachments sent with it may have been altered without the consent or knowledge of the author. -----Original Message----- From: powerh-l-admin@lists.sowder.com [mailto:powerh-l-admin@lists.sowder.com] On Behalf Of Jari.Kaljunen@teliasonera.com Sent: 21 September 2004 11:31 To: powerh-l@lists.sowder.com Subject: SV: SV: write to file with quick /using quiz instead Hi the output record format is not critical, the result file will be used by other system that don't care. But what do you recommend to convert the characters &,<,>,' and " to xml-entities if quiz or qtp is used produce the file? We don't have controls for these characters in DB now and we get files from many sources. We already had a solution with qtp that produced the file but the problem with these characters was not solved. A collegue succested to use quick for that purpose (first separate the data from xml-tags with binary characters and then only check the data part) but the quick performance problem still remains. DCL was far too slow. Management don't want to install Perl (I believe that Perl using regular expressions would be fast enough), collegues don't like the idea of using external C-program... We appreciate any suggestions. /Jari -----Ursprungligt meddelande----- Från: Joe Boyle [mailto:joeboyle_adt@hotmail.com] Skickat: den 21 september 2004 12:03 Till: Kaljunen, Jari S. /TSS - Sergel Kredittjänster /+46-8-4526435; powerh-l@lists.sowder.com Ämne: RE: SV: write to file with quick /using quiz instead personally, I would prefer to code the requirements that you list below and then use 'convert', rather than wait for quick to write 9 million 600 byte records ( 5 gig ); especially as you say that performance still isn't great even when using the direct format. Regards, Joe. This e-mail and all information contained in it is confidential and may be legally privileged. If you are not the intended recipient, your access to this e-mail is unauthorized. Any use, dissemination, distribution, publication or copying by you of this e-mail or any of the information contained within it is prohibited and may be unlawful. Do not open any attachments, delete it immediately from your system and notify the sender promptly by e-mail that you have done so. The content of this e-mail and any attachments sent with it may have been altered without the consent or knowledge of the author. -----Original Message----- From: Jari.Kaljunen@teliasonera.com [mailto:Jari.Kaljunen@teliasonera.com] Sent: 20 September 2004 14:37 To: joeboyle_adt@hotmail.com; powerh-l@lists.sowder.com Subject: SV: SV: write to file with quick /using quiz instead Hi, there will be "report" from 4 different tables. Of course we could have several quiz-runs producing files with additional sorting fields and then merge the files omitting sorting fields. Then there would be 2 problems left: disc space for the intermediate files and conversion of characters &,<,>,'," to xml-entities which is not very practical i quiz. /Jari -----Ursprungligt meddelande----- Från: Joe Boyle [mailto:joeboyle_adt@hotmail.com] Skickat: den 17 september 2004 23:28 Till: Kaljunen, Jari S. /TSS - Sergel Kredittjänster /+46-8-4526435; powerh-l@lists.sowder.com Ämne: RE: SV: write to file with quick why not allow the performance of quiz to generate the subfile and then use the DCL command 'CONVERT/FDL' to convert the file to the required sequential format. It may even be the case, that if the 'nodict' qualifier is used when creating the subfile, then the subfile need only be renamed and copied to the required directory. I am guessing that a 'nodict' subfile will have a sequential rather than direct record format; you could always use the analyse command to examine the subfile's record format if you aren't sure. Presumably, once the file is built, subsequent updates will be writing fewer records into the file, and the slow performance of Quick may not be as great an issue - or you could even go back to Qtp. Regards, Joe. This e-mail and all information contained in it is confidential and may be legally privileged. If you are not the intended recipient, your access to this e-mail is unauthorized. Any use, dissemination, distribution, publication or copying by you of this e-mail or any of the information contained within it is prohibited and may be unlawful. Do not open any attachments, delete it immediately from your system and notify the sender promptly by e-mail that you have done so. The content of this e-mail and any attachments sent with it may have been altered without the consent or knowledge of the author. From: powerh-l-admin@lists.sowder.com [mailto:powerh-l-admin@lists.sowder.com] On Behalf Of Jari.Kaljunen@teliasonera.com Sent: 16 September 2004 20:42 To: powerh-l@lists.sowder.com Subject: SV: SV: write to file with quick You mean file asc600 designer open append ? Well, it didn't help.. /Jari -----Ursprungligt meddelande----- Från: Peter Bateman [mailto:ppfb@hotmail.com] Skickat: den 16 september 2004 21:02 Till: powerh-l@lists.sowder.com Ämne: RE: SV: write to file with quick I would try 'file asc600 designer open append' It may be a machine specific option. But I didn't see it in the manual though. Regards, Peter >From: >To: >Subject: SV: write to file with quick >Date: Thu, 16 Sep 2004 12:22:20 +0200 > >Hello again, > >a little test without reading database still very slow (actually no difference...). It takes 49 s to run this with 100 records on subfile. The result file is 20100 lines. > >I even wrote a test quiz reading the same subfile and having 200 defines, each including systime and some more, each defined item written on an own line. It takes quiz 3 s to write these 20000 lines. > > >/Jari > > >screen test_writing_file menu stopscreen > >file *ik006002 designer >file asc600 designer > >temp t_count > >procedure internal write > begin > for 200 > begin > let t_count = t_count + 1 > let rec600 = asc(t_count) & > + " " + asc(systime) + " procedure internal write " > put asc600 reset > end > end > >procedure initialize > begin > let t_count = 0 > while retr ik006002 seq > begin > let rec600 = asc(t_count) & > + " " + asc(systime) + " procedure initialize " > put asc600 reset > do internal write > end > return > end > >build > >-----Ursprungligt meddelande----- >Från: Johnson, Harold A EDUC:EX >[mailto:Harold.A.Johnson@gems1.gov.bc.ca] >Skickat: den 15 september 2004 17:50 >Till: Kaljunen, Jari S. /TSS - Sergel Kredittjänster /+46-8-4526435; >powerh-l@lists.sowder.com >Ämne: RE: write to file with quick > > >Are you sure the performance problem is with the writing and not with the >reading of the original data? I would suspect a problem there instead. > > >-----Original Message----- >From: powerh-l-admin@lists.sowder.com >[mailto:powerh-l-admin@lists.sowder.com]On Behalf Of >Jari.Kaljunen@teliasonera.com >Sent: 2004 September 15 8:41 AM >To: powerh-l@lists.sowder.com >Subject: write to file with quick > > >Hello, > >I am writing to > >File: ASC600 >Organization: SEQUENTIAL >Type: RMS >Open: ASC600 >Record Format: Variable >Supersede: No >Records & Sizes: ASC600 600 bytes >Item Type Size Occ Offset >REC600 CHARACTER 600 > >in quick using let and put as follows > >let rec600 = "" + asc(arnr of arenden,8) + asc(parmblnr,2) & >+ atgkod of atgkoder + "" > >put asc600 reset > > >The performance is very poor. The administrator believs that quick closes >and opens the file all the time. >Can that be true? >What should I do? > >We are creating an xml-file (it will be about 5 GB) with data from several >tables and our solution with qtp needed too many subfiles and sortings using >lots of space. > >/Jari > >= = = = = = = = = = = = = = = = = = = = = = = = = = = = >Mailing list: powerh-l@lists.sowder.com >Subscribe: "subscribe" in message body to powerh-l-request@lists.sowder.com >Unsubscribe: "unsubscribe " in message body to >powerh-l-request@lists.sowder.com >http://lists.sowder.com/mailman/listinfo/powerh-l >This list is closed, thus to post to the list you must be a subscriber. > >= = = = = = = = = = = = = = = = = = = = = = = = = = = = >Mailing list: powerh-l@lists.sowder.com >Subscribe: "subscribe" in message body to powerh-l-request@lists.sowder.com >Unsubscribe: "unsubscribe " in message body to powerh-l-request@lists.sowder.com >http://lists.sowder.com/mailman/listinfo/powerh-l >This list is closed, thus to post to the list you must be a subscriber. Don't just Search. Find! The new MSN Search: Fast. Clear. Easy. = = = = = = = = = = = = = = = = = = = = = = = = = = = = Mailing list: powerh-l@lists.sowder.com Subscribe: "subscribe" in message body to powerh-l-request@lists.sowder.com Unsubscribe: "unsubscribe " in message body to powerh-l-request@lists.sowder.com http://lists.sowder.com/mailman/listinfo/powerh-l This list is closed, thus to post to the list you must be a subscriber. = = = = = = = = = = = = = = = = = = = = = = = = = = = = Mailing list: powerh-l@lists.sowder.com Subscribe: "subscribe" in message body to powerh-l-request@lists.sowder.com Unsubscribe: "unsubscribe " in message body to powerh-l-request@lists.sowder.com http://lists.sowder.com/mailman/listinfo/powerh-l This list is closed, thus to post to the list you must be a subscriber. From joeboyle_adt@hotmail.com Tue Sep 21 12:16:09 2004 From: joeboyle_adt@hotmail.com (Joe Boyle) Date: Tue, 21 Sep 2004 12:16:09 +0100 Subject: SV: write to file with quick /using quiz instead In-Reply-To: Message-ID: alternatively, and assuming that quiz is around 50 times faster than quick when writing the records( or at least sufficiently faster for you to prefer to use quiz), you could allow quiz to write the subfile, and then access the subfile in quick and run it through a while loop processing each record as you are now. I am confident that updates are always significantly faster than writes, and over 9 million records the time saving should be significant. I think it's the case that the data part of the subfile should be in direct format already, so it's possible that you needn't use the convert function at all. Regards, Joe. This e-mail and all information contained in it is confidential and may be legally privileged. If you are not the intended recipient, your access to this e-mail is unauthorized. Any use, dissemination, distribution, publication or copying by you of this e-mail or any of the information contained within it is prohibited and may be unlawful. Do not open any attachments, delete it immediately from your system and notify the sender promptly by e-mail that you have done so. The content of this e-mail and any attachments sent with it may have been altered without the consent or knowledge of the author. -----Original Message----- From: powerh-l-admin@lists.sowder.com [mailto:powerh-l-admin@lists.sowder.com] On Behalf Of Jari.Kaljunen@teliasonera.com Sent: 21 September 2004 11:31 To: powerh-l@lists.sowder.com Subject: SV: SV: write to file with quick /using quiz instead Hi the output record format is not critical, the result file will be used by other system that don't care. But what do you recommend to convert the characters &,<,>,' and " to xml-entities if quiz or qtp is used produce the file? We don't have controls for these characters in DB now and we get files from many sources. We already had a solution with qtp that produced the file but the problem with these characters was not solved. A collegue succested to use quick for that purpose (first separate the data from xml-tags with binary characters and then only check the data part) but the quick performance problem still remains. DCL was far too slow. Management don't want to install Perl (I believe that Perl using regular expressions would be fast enough), collegues don't like the idea of using external C-program... We appreciate any suggestions. /Jari -----Ursprungligt meddelande----- Från: Joe Boyle [mailto:joeboyle_adt@hotmail.com] Skickat: den 21 september 2004 12:03 Till: Kaljunen, Jari S. /TSS - Sergel Kredittjänster /+46-8-4526435; powerh-l@lists.sowder.com Ämne: RE: SV: write to file with quick /using quiz instead personally, I would prefer to code the requirements that you list below and then use 'convert', rather than wait for quick to write 9 million 600 byte records ( 5 gig ); especially as you say that performance still isn't great even when using the direct format. Regards, Joe. This e-mail and all information contained in it is confidential and may be legally privileged. If you are not the intended recipient, your access to this e-mail is unauthorized. Any use, dissemination, distribution, publication or copying by you of this e-mail or any of the information contained within it is prohibited and may be unlawful. Do not open any attachments, delete it immediately from your system and notify the sender promptly by e-mail that you have done so. The content of this e-mail and any attachments sent with it may have been altered without the consent or knowledge of the author. -----Original Message----- From: Jari.Kaljunen@teliasonera.com [mailto:Jari.Kaljunen@teliasonera.com] Sent: 20 September 2004 14:37 To: joeboyle_adt@hotmail.com; powerh-l@lists.sowder.com Subject: SV: SV: write to file with quick /using quiz instead Hi, there will be "report" from 4 different tables. Of course we could have several quiz-runs producing files with additional sorting fields and then merge the files omitting sorting fields. Then there would be 2 problems left: disc space for the intermediate files and conversion of characters &,<,>,'," to xml-entities which is not very practical i quiz. /Jari -----Ursprungligt meddelande----- Från: Joe Boyle [mailto:joeboyle_adt@hotmail.com] Skickat: den 17 september 2004 23:28 Till: Kaljunen, Jari S. /TSS - Sergel Kredittjänster /+46-8-4526435; powerh-l@lists.sowder.com Ämne: RE: SV: write to file with quick why not allow the performance of quiz to generate the subfile and then use the DCL command 'CONVERT/FDL' to convert the file to the required sequential format. It may even be the case, that if the 'nodict' qualifier is used when creating the subfile, then the subfile need only be renamed and copied to the required directory. I am guessing that a 'nodict' subfile will have a sequential rather than direct record format; you could always use the analyse command to examine the subfile's record format if you aren't sure. Presumably, once the file is built, subsequent updates will be writing fewer records into the file, and the slow performance of Quick may not be as great an issue - or you could even go back to Qtp. Regards, Joe. This e-mail and all information contained in it is confidential and may be legally privileged. If you are not the intended recipient, your access to this e-mail is unauthorized. Any use, dissemination, distribution, publication or copying by you of this e-mail or any of the information contained within it is prohibited and may be unlawful. Do not open any attachments, delete it immediately from your system and notify the sender promptly by e-mail that you have done so. The content of this e-mail and any attachments sent with it may have been altered without the consent or knowledge of the author. From: powerh-l-admin@lists.sowder.com [mailto:powerh-l-admin@lists.sowder.com] On Behalf Of Jari.Kaljunen@teliasonera.com Sent: 16 September 2004 20:42 To: powerh-l@lists.sowder.com Subject: SV: SV: write to file with quick You mean file asc600 designer open append ? Well, it didn't help.. /Jari -----Ursprungligt meddelande----- Från: Peter Bateman [mailto:ppfb@hotmail.com] Skickat: den 16 september 2004 21:02 Till: powerh-l@lists.sowder.com Ämne: RE: SV: write to file with quick I would try 'file asc600 designer open append' It may be a machine specific option. But I didn't see it in the manual though. Regards, Peter >From: >To: >Subject: SV: write to file with quick >Date: Thu, 16 Sep 2004 12:22:20 +0200 > >Hello again, > >a little test without reading database still very slow (actually no difference...). It takes 49 s to run this with 100 records on subfile. The result file is 20100 lines. > >I even wrote a test quiz reading the same subfile and having 200 defines, each including systime and some more, each defined item written on an own line. It takes quiz 3 s to write these 20000 lines. > > >/Jari > > >screen test_writing_file menu stopscreen > >file *ik006002 designer >file asc600 designer > >temp t_count > >procedure internal write > begin > for 200 > begin > let t_count = t_count + 1 > let rec600 = asc(t_count) & > + " " + asc(systime) + " procedure internal write " > put asc600 reset > end > end > >procedure initialize > begin > let t_count = 0 > while retr ik006002 seq > begin > let rec600 = asc(t_count) & > + " " + asc(systime) + " procedure initialize " > put asc600 reset > do internal write > end > return > end > >build > >-----Ursprungligt meddelande----- >Från: Johnson, Harold A EDUC:EX >[mailto:Harold.A.Johnson@gems1.gov.bc.ca] >Skickat: den 15 september 2004 17:50 >Till: Kaljunen, Jari S. /TSS - Sergel Kredittjänster /+46-8-4526435; >powerh-l@lists.sowder.com >Ämne: RE: write to file with quick > > >Are you sure the performance problem is with the writing and not with the >reading of the original data? I would suspect a problem there instead. > > >-----Original Message----- >From: powerh-l-admin@lists.sowder.com >[mailto:powerh-l-admin@lists.sowder.com]On Behalf Of >Jari.Kaljunen@teliasonera.com >Sent: 2004 September 15 8:41 AM >To: powerh-l@lists.sowder.com >Subject: write to file with quick > > >Hello, > >I am writing to > >File: ASC600 >Organization: SEQUENTIAL >Type: RMS >Open: ASC600 >Record Format: Variable >Supersede: No >Records & Sizes: ASC600 600 bytes >Item Type Size Occ Offset >REC600 CHARACTER 600 > >in quick using let and put as follows > >let rec600 = "" + asc(arnr of arenden,8) + asc(parmblnr,2) & >+ atgkod of atgkoder + "" > >put asc600 reset > > >The performance is very poor. The administrator believs that quick closes >and opens the file all the time. >Can that be true? >What should I do? > >We are creating an xml-file (it will be about 5 GB) with data from several >tables and our solution with qtp needed too many subfiles and sortings using >lots of space. > >/Jari > >= = = = = = = = = = = = = = = = = = = = = = = = = = = = >Mailing list: powerh-l@lists.sowder.com >Subscribe: "subscribe" in message body to powerh-l-request@lists.sowder.com >Unsubscribe: "unsubscribe " in message body to >powerh-l-request@lists.sowder.com >http://lists.sowder.com/mailman/listinfo/powerh-l >This list is closed, thus to post to the list you must be a subscriber. > >= = = = = = = = = = = = = = = = = = = = = = = = = = = = >Mailing list: powerh-l@lists.sowder.com >Subscribe: "subscribe" in message body to powerh-l-request@lists.sowder.com >Unsubscribe: "unsubscribe " in message body to powerh-l-request@lists.sowder.com >http://lists.sowder.com/mailman/listinfo/powerh-l >This list is closed, thus to post to the list you must be a subscriber. Don't just Search. Find! The new MSN Search: Fast. Clear. Easy. = = = = = = = = = = = = = = = = = = = = = = = = = = = = Mailing list: powerh-l@lists.sowder.com Subscribe: "subscribe" in message body to powerh-l-request@lists.sowder.com Unsubscribe: "unsubscribe " in message body to powerh-l-request@lists.sowder.com http://lists.sowder.com/mailman/listinfo/powerh-l This list is closed, thus to post to the list you must be a subscriber. = = = = = = = = = = = = = = = = = = = = = = = = = = = = Mailing list: powerh-l@lists.sowder.com Subscribe: "subscribe" in message body to powerh-l-request@lists.sowder.com Unsubscribe: "unsubscribe " in message body to powerh-l-request@lists.sowder.com http://lists.sowder.com/mailman/listinfo/powerh-l This list is closed, thus to post to the list you must be a subscriber. From joeboyle_adt@hotmail.com Wed Sep 22 17:14:59 2004 From: joeboyle_adt@hotmail.com (Joe Boyle) Date: Wed, 22 Sep 2004 17:14:59 +0100 Subject: SV: write to file with quick /using quiz instead In-Reply-To: Message-ID: I have included some quiz code below, but as mentioned before I would suggest that you run a timed comparison of the update time required for quick to process an incoming subfile containing unprocessed records, using the syntax you already have. If the combined time is significantly less than when quick does all of the processing including writing the record ( as opposed to simply updating ) obviously I would use the quiz/quick combination. Back to Quiz, do you have an estimate of the maximum number of occurrences of these characters that may exist per record ? If it's very high, clearly the number of defines required by quiz would be high; if the number is low it may be feasible to code them in quiz - example below (although I suspect you need no examples :-) define d'amp1'pos = index (t_kvar, '&') define d'the'txt1 char * 600 = & t_kvar [ 1: d'amp1'pos - 1 ] & + "&" + & t_kvar [ d'amp1'pos + 1 : 600 ] & ; I am assuming 600 will simply add ; the remaining txt and not cause an errorif d'amp1'pos gt 0.5 & else t_kvar define d'amp2'pos = index (d'the'txt1 , '&') define d'the'txt2 char * 600 = & d'the'txt1 [ 1: d'amp2'pos - 1 ] & + "&" + & d'the'txt1 [ d'amp2'pos + 1 : 600 ] & ; I am assuming 600 will simply add ; the remaining txt and not cause an error if d'amp2'pos gt 0.5 & else d'the'txt1 .. Loads more code ... define mperec600 char*600 = & "" + trunc(d'the'txtN) + "" set sub mperec600_sf keep rep summ mperec600 Clearly the code above will have to be repeated for each of the replacement characters, and each must be coded for the maximum number of occurrences that you expect it might occur in the incoming column/s - NASTY. Clearly the quiz code is highly string function intensive and this is likely to have some affect on Quiz's performance. It all depends on how much development time you want to put in compared to the time saving you hope to get out when the process runs. Regards, Joe. This e-mail and all information contained in it is confidential and may be legally privileged. If you are not the intended recipient, your access to this e-mail is unauthorized. Any use, dissemination, distribution, publication or copying by you of this e-mail or any of the information contained within it is prohibited and may be unlawful. Do not open any attachments, delete it immediately from your system and notify the sender promptly by e-mail that you have done so. The content of this e-mail and any attachments sent with it may have been altered without the consent or knowledge of the author. -----Original Message----- From: powerh-l-admin@lists.sowder.com [mailto:powerh-l-admin@lists.sowder.com] On Behalf Of Jari.Kaljunen@teliasonera.com Sent: 21 September 2004 12:14 To: joeboyle_adt@hotmail.com; powerh-l@lists.sowder.com Subject: SV: SV: write to file with quick /using quiz instead below is what we do with fields that might have characters that has to be converted to xml-entities. It would be nice to have a function to use standard regular expressions in PH, wouldn't it? /Jari procedure internal fixa_tecken ; to be optimized begin let t_langd = size(trunc(t_kvar)) let t_klar = 0 let t_klart = "" while t_langd > t_klar begin let t_tecken = t_kvar[t_klar + 1 : 1] if t_tecken = "&" then begin let t_tecken_2 = "&" end else if t_tecken = "<" then begin let t_tecken_2 = "<" end else if t_tecken = '"' then begin let t_tecken_2 = """ end else if t_tecken = ">" then begin let t_tecken_2 = ">" end else if t_tecken = "'" then begin let t_tecken_2 = "'" end else begin let t_tecken_2 = t_tecken end let t_klart = trunc(t_klart) + trunc(t_tecken_2) let t_klar = t_klar + 1 let t_tecken = t_kvar[t_klar + 1 : 1] end end procedure internal kolla_t_kvar ; to be optimized begin let t_ix_amp = index (t_kvar, '&') let t_ix_apos = index (t_kvar, "'") let t_ix_quot = index (t_kvar, '"') let t_ix_gt = index (t_kvar, '>') let t_ix_lt = index (t_kvar, '<') if t_ix_amp <> 0 or t_ix_apos <> 0 or t_ix_quot <> 0 or t_ix_gt <> 0 or t_ix_lt <> 0 then do internal fixa_tecken else begin let t_klart = t_kvar end end let t_kvar = anamn1 of arenden ; suspect field do internal kolla_t_kvar let mperec600 = "" + trunc(t_klart) + "" put mpe600 reset -----Ursprungligt meddelande----- Från: Joe Boyle [mailto:joeboyle_adt@hotmail.com] Skickat: den 21 september 2004 12:50 Till: Kaljunen, Jari S. /TSS - Sergel Kredittjänster /+46-8-4526435; powerh-l@lists.sowder.com Ämne: RE: SV: write to file with quick /using quiz instead I may have misunderstood your comments below, I imagined that you were simply looking to replace the occurrences of these characters in your data with other characters. If this is not the case it might be better to give an example of what your current qdesign source code is doing, and it should be possible to find a quiz equivalent. Regards, Joe. This e-mail and all information contained in it is confidential and may be legally privileged. If you are not the intended recipient, your access to this e-mail is unauthorized. Any use, dissemination, distribution, publication or copying by you of this e-mail or any of the information contained within it is prohibited and may be unlawful. Do not open any attachments, delete it immediately from your system and notify the sender promptly by e-mail that you have done so. The content of this e-mail and any attachments sent with it may have been altered without the consent or knowledge of the author. -----Original Message----- From: powerh-l-admin@lists.sowder.com [mailto:powerh-l-admin@lists.sowder.com] On Behalf Of Jari.Kaljunen@teliasonera.com Sent: 21 September 2004 11:31 To: powerh-l@lists.sowder.com Subject: SV: SV: write to file with quick /using quiz instead Hi the output record format is not critical, the result file will be used by other system that don't care. But what do you recommend to convert the characters &,<,>,' and " to xml-entities if quiz or qtp is used produce the file? We don't have controls for these characters in DB now and we get files from many sources. We already had a solution with qtp that produced the file but the problem with these characters was not solved. A collegue succested to use quick for that purpose (first separate the data from xml-tags with binary characters and then only check the data part) but the quick performance problem still remains. DCL was far too slow. Management don't want to install Perl (I believe that Perl using regular expressions would be fast enough), collegues don't like the idea of using external C-program... We appreciate any suggestions. /Jari -----Ursprungligt meddelande----- Från: Joe Boyle [mailto:joeboyle_adt@hotmail.com] Skickat: den 21 september 2004 12:03 Till: Kaljunen, Jari S. /TSS - Sergel Kredittjänster /+46-8-4526435; powerh-l@lists.sowder.com Ämne: RE: SV: write to file with quick /using quiz instead personally, I would prefer to code the requirements that you list below and then use 'convert', rather than wait for quick to write 9 million 600 byte records ( 5 gig ); especially as you say that performance still isn't great even when using the direct format. Regards, Joe. This e-mail and all information contained in it is confidential and may be legally privileged. If you are not the intended recipient, your access to this e-mail is unauthorized. Any use, dissemination, distribution, publication or copying by you of this e-mail or any of the information contained within it is prohibited and may be unlawful. Do not open any attachments, delete it immediately from your system and notify the sender promptly by e-mail that you have done so. The content of this e-mail and any attachments sent with it may have been altered without the consent or knowledge of the author. -----Original Message----- From: Jari.Kaljunen@teliasonera.com [mailto:Jari.Kaljunen@teliasonera.com] Sent: 20 September 2004 14:37 To: joeboyle_adt@hotmail.com; powerh-l@lists.sowder.com Subject: SV: SV: write to file with quick /using quiz instead Hi, there will be "report" from 4 different tables. Of course we could have several quiz-runs producing files with additional sorting fields and then merge the files omitting sorting fields. Then there would be 2 problems left: disc space for the intermediate files and conversion of characters &,<,>,'," to xml-entities which is not very practical i quiz. /Jari -----Ursprungligt meddelande----- Från: Joe Boyle [mailto:joeboyle_adt@hotmail.com] Skickat: den 17 september 2004 23:28 Till: Kaljunen, Jari S. /TSS - Sergel Kredittjänster /+46-8-4526435; powerh-l@lists.sowder.com Ämne: RE: SV: write to file with quick why not allow the performance of quiz to generate the subfile and then use the DCL command 'CONVERT/FDL' to convert the file to the required sequential format. It may even be the case, that if the 'nodict' qualifier is used when creating the subfile, then the subfile need only be renamed and copied to the required directory. I am guessing that a 'nodict' subfile will have a sequential rather than direct record format; you could always use the analyse command to examine the subfile's record format if you aren't sure. Presumably, once the file is built, subsequent updates will be writing fewer records into the file, and the slow performance of Quick may not be as great an issue - or you could even go back to Qtp. Regards, Joe. This e-mail and all information contained in it is confidential and may be legally privileged. If you are not the intended recipient, your access to this e-mail is unauthorized. Any use, dissemination, distribution, publication or copying by you of this e-mail or any of the information contained within it is prohibited and may be unlawful. Do not open any attachments, delete it immediately from your system and notify the sender promptly by e-mail that you have done so. The content of this e-mail and any attachments sent with it may have been altered without the consent or knowledge of the author. From: powerh-l-admin@lists.sowder.com [mailto:powerh-l-admin@lists.sowder.com] On Behalf Of Jari.Kaljunen@teliasonera.com Sent: 16 September 2004 20:42 To: powerh-l@lists.sowder.com Subject: SV: SV: write to file with quick You mean file asc600 designer open append ? Well, it didn't help.. /Jari -----Ursprungligt meddelande----- Från: Peter Bateman [mailto:ppfb@hotmail.com] Skickat: den 16 september 2004 21:02 Till: powerh-l@lists.sowder.com Ämne: RE: SV: write to file with quick I would try 'file asc600 designer open append' It may be a machine specific option. But I didn't see it in the manual though. Regards, Peter >From: >To: >Subject: SV: write to file with quick >Date: Thu, 16 Sep 2004 12:22:20 +0200 > >Hello again, > >a little test without reading database still very slow (actually no difference...). It takes 49 s to run this with 100 records on subfile. The result file is 20100 lines. > >I even wrote a test quiz reading the same subfile and having 200 defines, each including systime and some more, each defined item written on an own line. It takes quiz 3 s to write these 20000 lines. > > >/Jari > > >screen test_writing_file menu stopscreen > >file *ik006002 designer >file asc600 designer > >temp t_count > >procedure internal write > begin > for 200 > begin > let t_count = t_count + 1 > let rec600 = asc(t_count) & > + " " + asc(systime) + " procedure internal write " > put asc600 reset > end > end > >procedure initialize > begin > let t_count = 0 > while retr ik006002 seq > begin > let rec600 = asc(t_count) & > + " " + asc(systime) + " procedure initialize " > put asc600 reset > do internal write > end > return > end > >build > >-----Ursprungligt meddelande----- >Från: Johnson, Harold A EDUC:EX >[mailto:Harold.A.Johnson@gems1.gov.bc.ca] >Skickat: den 15 september 2004 17:50 >Till: Kaljunen, Jari S. /TSS - Sergel Kredittjänster /+46-8-4526435; >powerh-l@lists.sowder.com >Ämne: RE: write to file with quick > > >Are you sure the performance problem is with the writing and not with the >reading of the original data? I would suspect a problem there instead. > > >-----Original Message----- >From: powerh-l-admin@lists.sowder.com >[mailto:powerh-l-admin@lists.sowder.com]On Behalf Of >Jari.Kaljunen@teliasonera.com >Sent: 2004 September 15 8:41 AM >To: powerh-l@lists.sowder.com >Subject: write to file with quick > > >Hello, > >I am writing to > >File: ASC600 >Organization: SEQUENTIAL >Type: RMS >Open: ASC600 >Record Format: Variable >Supersede: No >Records & Sizes: ASC600 600 bytes >Item Type Size Occ Offset >REC600 CHARACTER 600 > >in quick using let and put as follows > >let rec600 = "" + asc(arnr of arenden,8) + asc(parmblnr,2) & >+ atgkod of atgkoder + "" > >put asc600 reset > > >The performance is very poor. The administrator believs that quick closes >and opens the file all the time. >Can that be true? >What should I do? > >We are creating an xml-file (it will be about 5 GB) with data from several >tables and our solution with qtp needed too many subfiles and sortings using >lots of space. > >/Jari > >= = = = = = = = = = = = = = = = = = = = = = = = = = = = >Mailing list: powerh-l@lists.sowder.com >Subscribe: "subscribe" in message body to powerh-l-request@lists.sowder.com >Unsubscribe: "unsubscribe " in message body to >powerh-l-request@lists.sowder.com >http://lists.sowder.com/mailman/listinfo/powerh-l >This list is closed, thus to post to the list you must be a subscriber. > >= = = = = = = = = = = = = = = = = = = = = = = = = = = = >Mailing list: powerh-l@lists.sowder.com >Subscribe: "subscribe" in message body to powerh-l-request@lists.sowder.com >Unsubscribe: "unsubscribe " in message body to powerh-l-request@lists.sowder.com >http://lists.sowder.com/mailman/listinfo/powerh-l >This list is closed, thus to post to the list you must be a subscriber. Don't just Search. Find! The new MSN Search: Fast. Clear. Easy. = = = = = = = = = = = = = = = = = = = = = = = = = = = = Mailing list: powerh-l@lists.sowder.com Subscribe: "subscribe" in message body to powerh-l-request@lists.sowder.com Unsubscribe: "unsubscribe " in message body to powerh-l-request@lists.sowder.com http://lists.sowder.com/mailman/listinfo/powerh-l This list is closed, thus to post to the list you must be a subscriber. = = = = = = = = = = = = = = = = = = = = = = = = = = = = Mailing list: powerh-l@lists.sowder.com Subscribe: "subscribe" in message body to powerh-l-request@lists.sowder.com Unsubscribe: "unsubscribe " in message body to powerh-l-request@lists.sowder.com http://lists.sowder.com/mailman/listinfo/powerh-l This list is closed, thus to post to the list you must be a subscriber. = = = = = = = = = = = = = = = = = = = = = = = = = = = = Mailing list: powerh-l@lists.sowder.com Subscribe: "subscribe" in message body to powerh-l-request@lists.sowder.com Unsubscribe: "unsubscribe " in message body to powerh-l-request@lists.sowder.com http://lists.sowder.com/mailman/listinfo/powerh-l This list is closed, thus to post to the list you must be a subscriber. From John.Pickering@norbord.com Wed Sep 22 17:41:01 2004 From: John.Pickering@norbord.com (Pickering, John (NORBORD)) Date: Wed, 22 Sep 2004 12:41:01 -0400 Subject: write to file with quick /using quiz instead Message-ID: <611340310619D711AA4000306E1CC51202797A12@TORHEXCH> >But what do you recommend to convert the characters &,<,>,' and " >to xml-entities if quiz or qtp is used produce the file? How about creating the output as a subfile with Quiz since that seems to be quite fast and then using an editor to change the special characters to their xml equivalents? This would seem like a much easier approach. From joeboyle_adt@hotmail.com Wed Sep 22 19:25:36 2004 From: joeboyle_adt@hotmail.com (Joe Boyle) Date: Wed, 22 Sep 2004 19:25:36 +0100 Subject: write to file with quick /using quiz instead In-Reply-To: <611340310619D711AA4000306E1CC51202797A12@TORHEXCH> Message-ID: sadly, you have to use similar code for each char in turn; but on the plus side the revised code below might be more convenient and efficient - well at least it will require fewer 600 char length defines. Once the amps are done, resulting in the creation of defined item d'the'amps, then this defined item would be the target field for the same cycle which might then search and replace the '>' char, and so on... define d'amp1'pos = index (t_kvar, '&') define d'amp2'pos = index (t_kvar[d'amp1'pos + 1, 600 ], '&') & if d'amp1'pos gt 0.5 define d'amp3'pos = index (t_kvar [ d'amp2'pos + 1, 600 ], '&') & if d'amp2'pos gt 0.5 define d'ampn'pos = index (t_kvar [ d'amp3'pos + 1, 600 ], '&') & if d'amp3'pos gt 0.5 define d'the'amps char * 600 = & t_kvar [ 1: d'amp1'pos - 1] & + "&" + & t_kvar [d'amp1'pos + 1: d'amp2'pos - 1] & + "&" + & t_kvar [d'amp2'pos + 1: d'amp3'pos - 1] & + "&" + & t_kvar [d'amp3'pos + 1: d'amp4'pos - 1] & + "&" + & t_kvar [d'amp4'pos + 1:600] & if d'ampn'pos gt 0.5 & .. etc... .. else t_kvar [ 1: d'amp1'pos - 1] & + "&" + & t_kvar [d'amp1'pos + 1: d'amp2'pos - 1] & + "&" + & t_kvar [d'amp2'pos + 1: d'amp3'pos - 1] & + "&" + & t_kvar [d'amp3'pos + 1: d'amp4'pos - 1] & + "&" + & t_kvar [d'amp4'pos + 1:600] & if d'amp4'pos gt 0.5 & .. etc... .. t_kvar [ 1: d'amp1'pos - 1] & + "&" + & t_kvar [ d'amp1'pos + 1 : 600 ] & d'amp1'pos gt 0.5 & else t_kvar The code is untested coz' I don't have access to a compiler - but I have coded similar stuff in the past; and in any case the purpose is simply to convey an idea. I reckon that passing the uncensored subfile through quick is likely to cause least pain all round. P.S. if any one would be kind enough to let me dial in to their server for PH testing it would be most appreciated. Of course I realize how unlikely this is - but if you don't ask .... Regards, Joe. This e-mail and all information contained in it is confidential and may be legally privileged. If you are not the intended recipient, your access to this e-mail is unauthorized. Any use, dissemination, distribution, publication or copying by you of this e-mail or any of the information contained within it is prohibited and may be unlawful. Do not open any attachments, delete it immediately from your system and notify the sender promptly by e-mail that you have done so. The content of this e-mail and any attachments sent with it may have been altered without the consent or knowledge of the author. -----Original Message----- From: powerh-l-admin@lists.sowder.com [mailto:powerh-l-admin@lists.sowder.com] On Behalf Of Pickering, John (NORBORD) Sent: 22 September 2004 17:41 To: Jari.Kaljunen@teliasonera.com; powerh-l@lists.sowder.com Subject: RE:write to file with quick /using quiz instead >But what do you recommend to convert the characters &,<,>,' and " >to xml-entities if quiz or qtp is used produce the file? How about creating the output as a subfile with Quiz since that seems to be quite fast and then using an editor to change the special characters to their xml equivalents? This would seem like a much easier approach. = = = = = = = = = = = = = = = = = = = = = = = = = = = = Mailing list: powerh-l@lists.sowder.com Subscribe: "subscribe" in message body to powerh-l-request@lists.sowder.com Unsubscribe: "unsubscribe " in message body to powerh-l-request@lists.sowder.com http://lists.sowder.com/mailman/listinfo/powerh-l This list is closed, thus to post to the list you must be a subscriber. From a_arjoon@hotmail.com Thu Sep 23 15:33:52 2004 From: a_arjoon@hotmail.com (A'Neil Arjoon) Date: Thu, 23 Sep 2004 09:33:52 -0500 Subject: Jobs in USA Message-ID: Hey Guys Any Cognos Powerhouse jobs out there in the USA for companies who will H1B sponsership from the UK? from A'Neil Arjoon. _________________________________________________________________ Express yourself instantly with MSN Messenger! Download today - it's FREE! hthttp://messenger.msn.click-url.com/go/onm00200471ave/direct/01/ From gammageo@yahoo.com Thu Sep 23 19:06:39 2004 From: gammageo@yahoo.com (George Wen) Date: Thu, 23 Sep 2004 11:06:39 -0700 (PDT) Subject: Jobs in USA In-Reply-To: Message-ID: <20040923180639.13078.qmail@web50402.mail.yahoo.com> --0-731347474-1095962799=:10759 Content-Type: text/plain; charset=us-ascii I'd guess not. There are a lot of IT folks out of work in this country. I know qualified IT guys who are working at Shopko stocking shelves. Hell I'd take a PH job anywhere in the country in second....But Bush says outsourcing/offshoring are good for this country so we're all keeping a stiff uppper lip, eh? A'Neil Arjoon wrote:Hey Guys Any Cognos Powerhouse jobs out there in the USA for companies who will H1B sponsership from the UK? from A'Neil Arjoon. _________________________________________________________________ Express yourself instantly with MSN Messenger! Download today - it's FREE! hthttp://messenger.msn.click-url.com/go/onm00200471ave/direct/01/ = = = = = = = = = = = = = = = = = = = = = = = = = = = = Mailing list: powerh-l@lists.sowder.com Subscribe: "subscribe" in message body to powerh-l-request@lists.sowder.com Unsubscribe: "unsubscribe " in message body to powerh-l-request@lists.sowder.com http://lists.sowder.com/mailman/listinfo/powerh-l This list is closed, thus to post to the list you must be a subscriber. --------------------------------- Do you Yahoo!? vote.yahoo.com - Register online to vote today! --0-731347474-1095962799=:10759 Content-Type: text/html; charset=us-ascii
I'd guess not.  There are a lot of IT folks out of work in this country.  I know qualified IT guys who are working at Shopko stocking shelves.  Hell I'd take a PH job anywhere in the country in second....But Bush says outsourcing/offshoring are good for this country so we're all keeping a stiff uppper lip, eh?

A'Neil Arjoon <a_arjoon@hotmail.com> wrote:
Hey Guys

Any Cognos Powerhouse jobs out there in the USA for companies who will H1B
sponsership from the UK?



from

A'Neil Arjoon.

_________________________________________________________________
Express yourself instantly with MSN Messenger! Download today - it's FREE!
hthttp://messenger.msn.click-url.com/go/onm00200471ave/direct/01/

= = = = = = = = = = = = = = = = = = = = = = = = = = = =
Mailing list: powerh-l@lists.sowder.com
Subscribe: "subscribe" in message body to powerh-l-request@lists.sowder.com
Unsubscribe: "unsubscribe " in message body to powerh-l-request@lists.sowder.com
http://lists.sowder.com/mailman/listinfo/powerh-l
This list is closed, thus to post to the list you must be a subscriber.

Do you Yahoo!?
vote.yahoo.com - Register online to vote today! --0-731347474-1095962799=:10759-- From gammageo@yahoo.com Thu Sep 23 19:39:03 2004 From: gammageo@yahoo.com (George Wen) Date: Thu, 23 Sep 2004 11:39:03 -0700 (PDT) Subject: Jobs in USA In-Reply-To: Message-ID: <20040923183903.37525.qmail@web50409.mail.yahoo.com> --0-530139467-1095964743=:35555 Content-Type: text/plain; charset=us-ascii Shame on me. Dice is not one of the ones I check often but I will. I'm one of the few PH on the AS400 users (JDE too). Never could figure why it didn't catch on. Thank you for the tip. Roger Kesterson wrote: Have you been checking Dice.com? I have seen PH jobs in New Hampshire, Wisconsin, and other places, on a regular basis. -----Original Message----- From: George Wen [mailto:gammageo@yahoo.com] Sent: Thursday, September 23, 2004 11:07 AM To: powerh-l@lists.sowder.com Subject: Re: Jobs in USA I'd guess not. There are a lot of IT folks out of work in this country. I know qualified IT guys who are working at Shopko stocking shelves. Hell I'd take a PH job anywhere in the country in second....But Bush says outsourcing/offshoring are good for this country so we're all keeping a stiff uppper lip, eh? A'Neil Arjoon wrote: Hey Guys Any Cognos Powerhouse jobs out there in the USA for companies who will H1B sponsership from the UK? from A'Neil Arjoon. _________________________________________________________________ Express yourself instantly with MSN Messenger! Download today - it's FREE! hthttp://messenger.msn.click-url.com/go/onm00200471ave/direct/01/ = = = = = = = = = = = = = = = = = = = = = = = = = = = = Mailing list: powerh-l@lists.sowder.com Subscribe: "subscribe" in message body to powerh-l-request@lists.sowder.com Unsubscribe: "unsubscribe " in message body to powerh-l-request@lists.sowder.com http://lists.sowder.com/mailman/listinfo/powerh-l This list is closed, thus to post to the list you must be a subscriber. --------------------------------- Do you Yahoo!? vote.yahoo.com - Register online to vote today! --------------------------------- Do you Yahoo!? New and Improved Yahoo! Mail - 100MB free storage! --0-530139467-1095964743=:35555 Content-Type: text/html; charset=us-ascii
Shame on me.  Dice is not one of the ones I check often but I will.  I'm one of the few PH on the AS400 users (JDE too).  Never could figure why it didn't catch on.
Thank you for the tip.
 
Roger Kesterson <Roger.Kesterson@mtdsw.com> wrote:
Have you been checking Dice.com?  I have seen PH jobs in New Hampshire, Wisconsin, and other places, on a regular basis.
-----Original Message-----
From: George Wen [mailto:gammageo@yahoo.com]
Sent: Thursday, September 23, 2004 11:07 AM
To: powerh-l@lists.sowder.com
Subject: Re: Jobs in USA

I'd guess not.  There are a lot of IT folks out of work in this country.  I know qualified IT guys who are working at Shopko stocking shelves.  Hell I'd take a PH job anywhere in the country in second....But Bush says outsourcing/offshoring are good for this country so we're all keeping a stiff uppper lip, eh?

A'Neil Arjoon <a_arjoon@hotmail.com> wrote:
Hey Guys

Any Cognos Powerhouse jobs out there in the USA for companies who will H1B
sponsership from the UK?



from

A'Neil Arjoon.

_________________________________________________________________
Express yourself instantly with MSN Messenger! Download today - it's FREE!
hthttp://messenger.msn.click-url.com/go/onm00200471ave/direct/01/

= = = = = = = = = = = = = = = = = = = = = = = = = = = =
Mailing list: powerh-l@lists.sowder.com
Subscribe: "subscribe" in message body to powerh-l-request@lists.sowder.com
Unsubscribe: "unsubscribe " in message body to powerh-l-request@lists.sowder.com
http://lists.sowder.com/mailman/listinfo/powerh-l
This list is closed, thus to post to the list you must be a subscriber.

Do you Yahoo!?
vote.yahoo.com - Register online to vote today!

Do you Yahoo!?
New and Improved Yahoo! Mail - 100MB free storage! --0-530139467-1095964743=:35555-- From joeboyle_adt@hotmail.com Fri Sep 24 16:50:44 2004 From: joeboyle_adt@hotmail.com (Joe Boyle) Date: Fri, 24 Sep 2004 16:50:44 +0100 Subject: write to file with quick /using quiz instead In-Reply-To: Message-ID: the code below is a better example of what I was getting at, many thanks for the pointers/corrections - you know who you are ;-) acc table_in select if table_in_col ne "" define d'amp1'pos = index (table_in_col, '&') define d'amp2'pos = index (table_in_col[d'amp1'pos + 1: 350 ], '&') + d'amp1'pos & if d'amp1'pos gt 0.5 define d'amp3'pos = index (table_in_col [ d'amp2'pos + 1: 350 ], '&') + d'amp2'pos & if d'amp2'pos gt 0.5 define d'amp4'pos = index (table_in_col [ d'amp3'pos + 1: 350 ], '&') + d'amp3'pos & if d'amp3'pos gt 0.5 define d'the'amps char * 350 = & table_in_col [ 1: d'amp1'pos - 1] & + "&" + & table_in_col [d'amp1'pos + 1: d'amp2'pos - 1 - d'amp1'pos] & + "&" + & table_in_col [d'amp2'pos + 1: d'amp3'pos - 1 - d'amp2'pos] & + "&" + & table_in_col [d'amp3'pos + 1: d'amp4'pos - 1 - d'amp3'pos] & + "&" + & table_in_col [d'amp4'pos + 1:350] & if d'amp4'pos ne d'amp3'pos & else table_in_col [ 1: d'amp1'pos - 1] & + "&" + & table_in_col [d'amp1'pos + 1: d'amp2'pos - 1 - d'amp1'pos] & + "&" + & table_in_col [d'amp2'pos + 1: d'amp3'pos - 1 - d'amp2'pos] & + "&" + & table_in_col [d'amp3'pos + 1:350] & if d'amp3'pos ne d'amp2'pos define d'lsthn1'pos = index (d'the'amps, '<') define d'lsthn2'pos = index (d'the'amps[d'lsthn1'pos + 1: 350 ], '<') + d'lsthn1'pos & if d'lsthn1'pos gt 0.5 define d'lsthn3'pos = index (d'the'amps [ d'lsthn2'pos + 1: 350 ], '<') + d'lsthn2'pos & if d'lsthn2'pos gt 0.5 define d'lsthn4'pos = index (d'the'amps [ d'lsthn3'pos + 1: 350 ], '<') + d'lsthn3'pos & if d'lsthn3'pos gt 0.5 define d'the'amps'n'lsthns char * 350 = & d'the'amps [ 1: d'lsthn1'pos - 1] & + "<" + & d'the'amps [d'lsthn1'pos + 1: d'lsthn2'pos - 1 - d'lsthn1'pos] & + "<" + & d'the'amps [d'lsthn2'pos + 1: d'lsthn3'pos - 1 - d'lsthn2'pos] & + "<" + & d'the'amps [d'lsthn3'pos + 1: d'lsthn4'pos - 1 - d'lsthn3'pos] & + "<" + & d'the'amps [d'lsthn4'pos + 1:350] & if d'lsthn4'pos ne d'lsthn3'pos & else d'the'amps [ 1: d'lsthn1'pos - 1] & + "<" + & d'the'amps [d'lsthn1'pos + 1: d'lsthn2'pos - 1 - d'lsthn1'pos] & + "<" + & d'the'amps [d'lsthn2'pos + 1: d'lsthn3'pos - 1 - d'lsthn2'pos] & + "<" + & d'the'amps [d'lsthn3'pos + 1:350] & if d'lsthn3'pos ne d'lsthn2'pos etc... define d'the'amps'n'lsthns'n'gtthns char * 350 = & .. and so on Regards, Joe. This e-mail and all information contained in it is confidential and may be legally privileged. If you are not the intended recipient, your access to this e-mail is unauthorized. Any use, dissemination, distribution, publication or copying by you of this e-mail or any of the information contained within it is prohibited and may be unlawful. Do not open any attachments, delete it immediately from your system and notify the sender promptly by e-mail that you have done so. The content of this e-mail and any attachments sent with it may have been altered without the consent or knowledge of the author. -----Original Message----- From: powerh-l-admin@lists.sowder.com [mailto:powerh-l-admin@lists.sowder.com] On Behalf Of Joe Boyle Sent: 22 September 2004 19:26 To: 'Pickering, John (NORBORD)'; Jari.Kaljunen@teliasonera.com; powerh-l@lists.sowder.com Subject: RE: write to file with quick /using quiz instead sadly, you have to use similar code for each char in turn; but on the plus side the revised code below might be more convenient and efficient - well at least it will require fewer 600 char length defines. Once the amps are done, resulting in the creation of defined item d'the'amps, then this defined item would be the target field for the same cycle which might then search and replace the '>' char, and so on... define d'amp1'pos = index (t_kvar, '&') define d'amp2'pos = index (t_kvar[d'amp1'pos + 1, 600 ], '&') & if d'amp1'pos gt 0.5 define d'amp3'pos = index (t_kvar [ d'amp2'pos + 1, 600 ], '&') & if d'amp2'pos gt 0.5 define d'ampn'pos = index (t_kvar [ d'amp3'pos + 1, 600 ], '&') & if d'amp3'pos gt 0.5 define d'the'amps char * 600 = & t_kvar [ 1: d'amp1'pos - 1] & + "&" + & t_kvar [d'amp1'pos + 1: d'amp2'pos - 1] & + "&" + & t_kvar [d'amp2'pos + 1: d'amp3'pos - 1] & + "&" + & t_kvar [d'amp3'pos + 1: d'amp4'pos - 1] & + "&" + & t_kvar [d'amp4'pos + 1:600] & if d'ampn'pos gt 0.5 & . etc... . else t_kvar [ 1: d'amp1'pos - 1] & + "&" + & t_kvar [d'amp1'pos + 1: d'amp2'pos - 1] & + "&" + & t_kvar [d'amp2'pos + 1: d'amp3'pos - 1] & + "&" + & t_kvar [d'amp3'pos + 1: d'amp4'pos - 1] & + "&" + & t_kvar [d'amp4'pos + 1:600] & if d'amp4'pos gt 0.5 & . etc... . t_kvar [ 1: d'amp1'pos - 1] & + "&" + & t_kvar [ d'amp1'pos + 1 : 600 ] & d'amp1'pos gt 0.5 & else t_kvar The code is untested coz' I don't have access to a compiler - but I have coded similar stuff in the past; and in any case the purpose is simply to convey an idea. I reckon that passing the uncensored subfile through quick is likely to cause least pain all round. P.S. if any one would be kind enough to let me dial in to their server for PH testing it would be most appreciated. Of course I realize how unlikely this is - but if you don't ask .... Regards, Joe. This e-mail and all information contained in it is confidential and may be legally privileged. If you are not the intended recipient, your access to this e-mail is unauthorized. Any use, dissemination, distribution, publication or copying by you of this e-mail or any of the information contained within it is prohibited and may be unlawful. Do not open any attachments, delete it immediately from your system and notify the sender promptly by e-mail that you have done so. The content of this e-mail and any attachments sent with it may have been altered without the consent or knowledge of the author. -----Original Message----- From: powerh-l-admin@lists.sowder.com [mailto:powerh-l-admin@lists.sowder.com] On Behalf Of Pickering, John (NORBORD) Sent: 22 September 2004 17:41 To: Jari.Kaljunen@teliasonera.com; powerh-l@lists.sowder.com Subject: RE:write to file with quick /using quiz instead >But what do you recommend to convert the characters &,<,>,' and " >to xml-entities if quiz or qtp is used produce the file? How about creating the output as a subfile with Quiz since that seems to be quite fast and then using an editor to change the special characters to their xml equivalents? This would seem like a much easier approach. = = = = = = = = = = = = = = = = = = = = = = = = = = = = Mailing list: powerh-l@lists.sowder.com Subscribe: "subscribe" in message body to powerh-l-request@lists.sowder.com Unsubscribe: "unsubscribe " in message body to powerh-l-request@lists.sowder.com http://lists.sowder.com/mailman/listinfo/powerh-l This list is closed, thus to post to the list you must be a subscriber. = = = = = = = = = = = = = = = = = = = = = = = = = = = = Mailing list: powerh-l@lists.sowder.com Subscribe: "subscribe" in message body to powerh-l-request@lists.sowder.com Unsubscribe: "unsubscribe " in message body to powerh-l-request@lists.sowder.com http://lists.sowder.com/mailman/listinfo/powerh-l This list is closed, thus to post to the list you must be a subscriber. From Christina.Hasse@COGNOS.com Mon Sep 27 17:11:07 2004 From: Christina.Hasse@COGNOS.com (Hasse, Christina) Date: Mon, 27 Sep 2004 12:11:07 -0400 Subject: FW: [HP3000-L] Business Analyst opportunity Message-ID: 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_01C4A4AC.98FD979B Content-Type: text/plain FYI -----Original Message----- From: HP-3000 Systems Discussion [mailto:HP3000-L@RAVEN.UTC.EDU] On Behalf Of Simpkins, Terry Sent: Monday, September 27, 2004 9:04 AM To: HP3000-L@RAVEN.UTC.EDU Subject: [HP3000-L] Business Analyst opportunity I am looking for a Business Analyst for a MANMAN-HP shop. This individual will be part of a team that is responsible for day-to-day support in a multi-account, multi-database environment. They will also be involved in and lead projects, as well as new site implementations. We are a growing organization with opportunities around the world. Currently we are involved in active implementations in: - Chicago - Torrance, CA - Paris, France - China We run MANMAN-HP 11.4 and XactMan for a windows front-end. This position is based in Hampton, VA, but travel (domestic and international) is part of the job. The successful candidate will have a thorough understanding of the CORE-4 MANMAN modules as well as experience using: - Cognos Powerhouse - MPEX & Security/3000 The individual selected for the opportunity will work directly on a daily basis with all levels of the organization, from stockroom clerks to the CFO & CEO. If you are looking for an exciting and challenging opportunity, please respond today in confidence to the email address below. ***************************** Terry W. Simpkins Director ISIT Measurement Specialties 757-766-4278 terry.simpkins@msiusa.com ***************************** * To join/leave the list, search archives, change list settings, * * etc., please visit http://raven.utc.edu/archives/hp3000-l.html * This message may contain privileged and/or confidential information. If you have received this e-mail in error or are not the intended recipient, you may not use, copy, disseminate or distribute it; do not open any attachments, delete it immediately from your system and notify the sender promptly by e-mail that you have done so. Thank you. ------_=_NextPart_001_01C4A4AC.98FD979B Content-Type: text/html Content-Transfer-Encoding: quoted-printable FW: [HP3000-L] Business Analyst opportunity

FYI

-----Original Message-----
From: HP-3000 Systems Discussion [mailto:HP3000-L@RAVEN.UTC.EDU= ] On Behalf Of Simpkins, Terry
Sent: Monday, September 27, 2004 9:04 AM
To: HP3000-L@RAVEN.UTC.EDU
Subject: [HP3000-L] Business Analyst = opportunity

I am looking for a Business Analyst for a MANMAN-HP = shop.
This individual will be part of a team that is = responsible for day-to-day support in a multi-account, multi-database = environment.

They will also be involved in and lead projects, as = well as new site implementations.

We are a growing organization with opportunities = around the world.
Currently we are involved in active implementations = in:
- Chicago
- Torrance, CA
- Paris, France
- China

We run MANMAN-HP 11.4 and XactMan for a windows = front-end.
This position is based in Hampton, VA, but travel = (domestic and
international) is part of the job.
The successful candidate will have a thorough = understanding of the
CORE-4 MANMAN modules as well as experience = using:
- Cognos Powerhouse
- MPEX & Security/3000

The individual selected for the opportunity will work = directly on a daily basis with all levels of the organization, from = stockroom clerks to the CFO & CEO.

If you are looking for an exciting and challenging = opportunity, please respond today in confidence to the email address = below.

*****************************
Terry W. Simpkins
Director ISIT
Measurement Specialties
757-766-4278
terry.simpkins@msiusa.com
*****************************

* To join/leave the list, search archives, change = list settings, *
* etc., please visit http://raven.utc.edu/archives/hp3000-l.html = *



This message may contain privileged and/or = confidential information.  If you have received this e-mail in = error or are not the intended recipient, you may not use, copy, = disseminate or distribute it; do not open any attachments, delete it = immediately from your system and notify the sender promptly by e-mail = that you have done so.  Thank you.

------_=_NextPart_001_01C4A4AC.98FD979B-- From ggjos@hotmail.com Mon Sep 27 22:18:27 2004 From: ggjos@hotmail.com (greg greg) Date: Mon, 27 Sep 2004 21:18:27 +0000 Subject: Dectect duplicate and issue warning instead of error message(QUICK) Message-ID: HI , I am attempting to get quick to detect a duplicate key and only issue a warning or preferable an information message instead of the error message which prohibits further entry as you would get in the lookup command. Thanks in advance. _________________________________________________________________ Help STOP SPAM with the new MSN 8 and get 2 months FREE* http://join.msn.com/?page=features/junkmail From Brett@Kinetix.co.nz Mon Sep 27 22:33:35 2004 From: Brett@Kinetix.co.nz (Brett Greiner (Kinetix)) Date: Tue, 28 Sep 2004 09:33:35 +1200 Subject: Dectect duplicate and issue warning instead of error message(QUICK) Message-ID: <9D966695054CFA4A8FD712F96673CCFB11F853@Afid> Add optional to the end of the lookup on/noton and the message will only be a warning. Brett. -----Original Message----- From: powerh-l-admin@lists.sowder.com [mailto:powerh-l-admin@lists.sowder.com] On Behalf Of greg greg Sent: Tuesday, 28 September 2004 9:18 a.m. To: powerh-l@lists.sowder.com Subject: Dectect duplicate and issue warning instead of error message(QUICK) HI , I am attempting to get quick to detect a duplicate key and only issue a warning or preferable an information message instead of the error message which prohibits further entry as you would get in the lookup command. Thanks in advance. _________________________________________________________________ Help STOP SPAM with the new MSN 8 and get 2 months FREE* http://join.msn.com/?page=features/junkmail = = = = = = = = = = = = = = = = = = = = = = = = = = = = Mailing list: powerh-l@lists.sowder.com Subscribe: "subscribe" in message body to powerh-l-request@lists.sowder.com Unsubscribe: "unsubscribe " in message body to powerh-l-request@lists.sowder.com http://lists.sowder.com/mailman/listinfo/powerh-l This list is closed, thus to post to the list you must be a subscriber. From brian_matthews_bmw@hotmail.com Tue Sep 28 02:13:15 2004 From: brian_matthews_bmw@hotmail.com (brian_matthewsbrian matthews) Date: Tue, 28 Sep 2004 01:13:15 +0000 Subject: Dectect duplicate and issue warning instead of error message(QUICK) Message-ID: 1. quick would normally re-prompt in the field once you OK the duplicate message 2. why are you coding a lookup noton if you are happy for the key to already exist on file ? 3. what ( in more detail ) do you want to do if a duplicate exists ? 4. if you change the code to remove the error and then update ( and assuming the index is unique in the dictionary ) you will get a similar error message. regards bri >From: "greg greg" >To: powerh-l@lists.sowder.com >Subject: Dectect duplicate and issue warning instead of error >message(QUICK) >Date: Mon, 27 Sep 2004 21:18:27 +0000 > >HI , > I am attempting to get quick to detect a duplicate key and only >issue a warning or preferable an information message instead of the error >message which prohibits further entry as you would get in the lookup >command. > > >Thanks in advance. > >_________________________________________________________________ >Help STOP SPAM with the new MSN 8 and get 2 months FREE* >http://join.msn.com/?page=features/junkmail > >= = = = = = = = = = = = = = = = = = = = = = = = = = = = >Mailing list: powerh-l@lists.sowder.com >Subscribe: "subscribe" in message body to powerh-l-request@lists.sowder.com >Unsubscribe: "unsubscribe " in message body to >powerh-l-request@lists.sowder.com >http://lists.sowder.com/mailman/listinfo/powerh-l >This list is closed, thus to post to the list you must be a subscriber. _________________________________________________________________ It's fast, it's easy and it's free. Get MSN Messenger today! http://www.msn.co.uk/messenger From ggjos@hotmail.com Tue Sep 28 18:00:45 2004 From: ggjos@hotmail.com (greg greg) Date: Tue, 28 Sep 2004 17:00:45 +0000 Subject: Solved RE: Dectect duplicate and issue warning instead of error message(QUICK) Message-ID: Hi Brian, I solved the problem already. It was simply ment to notify a user that another user has or is making an adjustment to the same id and period.The file structure allows for this.I used a procedure with the "if accessok" command, plus some other statements. Thanks. Greg >HI , > I am attempting to get quick to detect a duplicate key and only >issue a warning or preferable an information message instead of the error >message which prohibits further entry as you would get in the lookup >command. > _________________________________________________________________ Protect your PC - get McAfee.com VirusScan Online http://clinic.mcafee.com/clinic/ibuy/campaign.asp?cid=3963 From Kristina.Carlton@JICompanies.com Tue Sep 28 19:19:15 2004 From: Kristina.Carlton@JICompanies.com (Kristina Carlton) Date: Tue, 28 Sep 2004 13:19:15 -0500 Subject: duplicates Message-ID: <157670322ACDBB4D82690A2A917E6E1B0345A45F@jiexch01.jicompanies.com> 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_01C4A587.A9551A59 Content-Type: text/plain; charset="iso-8859-1" Hi, I am running a Quiz extract and I am using the SET NODUPLICATES command. However, I am getting duplicates anyhow. What might I be doing wrong? To clarify further, I run an extract with claim detail information. The claim detail information has the treating physician's FEIN. I access the subfile that is created and link it to our physician file to get detailed physician info. There could be multiple entries in the claim file for each physician obviously, but I was hoping they would not be duplicated in my physician detail extract. Actually, it works for all but one physician (this is a small sample only). I checked that database and do not have duplicates in the physician table. Thanks, -Kristina Ext. 2459 ------_=_NextPart_001_01C4A587.A9551A59 Content-Type: text/html; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable

Hi,

 

=

I am running a Quiz extract and I am using the SET NODUPLICATES command. = However, I am getting duplicates anyhow. What might I be doing = wrong?

 

=

To clarify further, I run an extract with claim detail information. The = claim detail information has the treating physician’s FEIN. I access = the subfile that is created and link it to our physician file to get detailed physician = info. There could be multiple entries in the claim file for each physician = obviously, but I was hoping they would not be duplicated in my physician detail extract. = Actually, it works for all but one physician (this is a small sample only). I = checked that database and do not have duplicates in the physician = table.

 

=

Thanks,<= /span>

<= span class=3DEmailStyle15>-Kristina

=

Ext. 2459

 

------_=_NextPart_001_01C4A587.A9551A59-- From Bob.Deskin@Cognos.COM Tue Sep 28 19:31:09 2004 From: Bob.Deskin@Cognos.COM (Deskin, Bob) Date: Tue, 28 Sep 2004 14:31:09 -0400 Subject: duplicates Message-ID: <2B4E386EAEC90947802323BF373DC0EC025921B8@sottemail1.ent.ad.cognos.com> This is a multi-part message in MIME format. ------_=_NextPart_001_01C4A589.532F18C1 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: quoted-printable SET NODUPLICATES applies to the report line not the record complex data. As the manual states, if the report detail line is identical to the preceding detail line, then it doesn't print if SET NODUPLICATES is specified. If you have a calculated value or a subtotal, this can cause the lines to be different even though the data is the same. Bob Deskin =20 Senior Product Manager, Application Development Tools, Cognos Inc.=20 bob.deskin@cognos.com (613) 738-1338 ext 7268 FAX: (613) 727-1178=20 3755 Riverside Drive P.O. Box 9707 Stn. T, Ottawa ON K1G 4K9 CANADA=20 -----Original Message----- From: powerh-l-admin@lists.sowder.com [mailto:powerh-l-admin@lists.sowder.com] On Behalf Of Kristina Carlton Sent: September 28, 2004 2:19 PM To: 'powerh-l@lists.sowder.com' Subject: duplicates =09 =09 Hi, =20 I am running a Quiz extract and I am using the SET NODUPLICATES command. However, I am getting duplicates anyhow. What might I be doing wrong? =20 To clarify further, I run an extract with claim detail information. The claim detail information has the treating physician's FEIN. I access the subfile that is created and link it to our physician file to get detailed physician info. There could be multiple entries in the claim file for each physician obviously, but I was hoping they would not be duplicated in my physician detail extract. Actually, it works for all but one physician (this is a small sample only). I checked that database and do not have duplicates in the physician table. =20 Thanks, -Kristina Ext. 2459=20 =20 This message may contain privileged and/or confidential = information. If you have received this e-mail in error or are not the = intended recipient, you may not use, copy, disseminate or distribute it; = do not open any attachments, delete it immediately from your system and = notify the sender promptly by e-mail that you have done so. Thank you.=20 =20 ------_=_NextPart_001_01C4A589.532F18C1 Content-Type: text/html; charset="us-ascii" Content-Transfer-Encoding: quoted-printable Message
SET=20 NODUPLICATES applies to the report line not the record complex data. As = the=20 manual states, if the report detail line is identical to the preceding = detail=20 line, then it doesn't print if SET NODUPLICATES is specified. If you = have a=20 calculated value or a subtotal, this can cause the lines to be different = even=20 though the data is the same.

Bob=20 Deskin           &= nbsp; =20
Senior = Product=20 Manager, Application Development Tools, Cognos Inc. =
bob.deskin@cognos.com (613) = 738-1338 ext 7268=20 FAX: (613) 727-1178
3755 Riverside Drive P.O. Box 9707 Stn. T, Ottawa ON K1G 4K9=20 CANADA

-----Original Message-----
From:=20 powerh-l-admin@lists.sowder.com = [mailto:powerh-l-admin@lists.sowder.com] On=20 Behalf Of Kristina Carlton
Sent: September 28, 2004 2:19 = PM
To: 'powerh-l@lists.sowder.com'
Subject:=20 duplicates

Hi,

 

I am running a = Quiz=20 extract and I am using the SET NODUPLICATES command. However, I am = getting=20 duplicates anyhow. What might I be doing=20 wrong?

 

To clarify = further, I run=20 an extract with claim detail information. The claim detail information = has the=20 treating physician’s FEIN. I access the subfile that is created = and link it to=20 our physician file to get detailed physician info. There could be = multiple=20 entries in the claim file for each physician obviously, but I was = hoping they=20 would not be duplicated in my physician detail extract. Actually, it = works for=20 all but one physician (this is a small sample only). I checked that = database=20 and do not have duplicates in the physician=20 table.

 

Thanks,

-Kristina

Ext.=20 2459

 

 

       This message may contain privileged and/or confidential = information.  If you have received this e-mail in error or are not = the intended recipient, you may not use, copy, disseminate or distribute = it; do not open any attachments, delete it immediately from your system = and notify the sender promptly by e-mail that you have done so.  = Thank you.

------_=_NextPart_001_01C4A589.532F18C1-- From brian_matthews_bmw@hotmail.com Tue Sep 28 21:05:20 2004 From: brian_matthews_bmw@hotmail.com (brian_matthewsbrian matthews) Date: Tue, 28 Sep 2004 20:05:20 +0000 Subject: duplicates Message-ID: remove all but the essential source code required to reproduce the problem and then let us know what the remaining code is; someone might then be able to spot the problem. >From: Kristina Carlton >To: "'powerh-l@lists.sowder.com'" >Subject: duplicates >Date: Tue, 28 Sep 2004 13:19:15 -0500 > >Hi, > >I am running a Quiz extract and I am using the SET NODUPLICATES command. >However, I am getting duplicates anyhow. What might I be doing wrong? > >To clarify further, I run an extract with claim detail information. The >claim detail information has the treating physician's FEIN. I access the >subfile that is created and link it to our physician file to get detailed >physician info. There could be multiple entries in the claim file for each >physician obviously, but I was hoping they would not be duplicated in my >physician detail extract. Actually, it works for all but one physician >(this >is a small sample only). I checked that database and do not have duplicates >in the physician table. > >Thanks, >-Kristina >Ext. 2459 > _________________________________________________________________ Stay in touch with absent friends - get MSN Messenger http://www.msn.co.uk/messenger From joeboyle_adt@hotmail.com Wed Sep 29 13:06:00 2004 From: joeboyle_adt@hotmail.com (Joe Boyle) Date: Wed, 29 Sep 2004 13:06:00 +0100 Subject: duplicates In-Reply-To: <157670322ACDBB4D82690A2A917E6E1B0345A45F@jiexch01.jicompanies.com> Message-ID: This is a multi-part message in MIME format. ------=_NextPart_000_0037_01C4A625.10F97080 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit If you have syntax like 'report field1 feild2 etc.' you could try using syntax like 'heading at sortitem field1 feild2 etc.' Regards, Joe. This e-mail and all information contained in it is confidential and may be legally privileged. If you are not the intended recipient, your access to this e-mail is unauthorized. Any use, dissemination, distribution, publication or copying by you of this e-mail or any of the information contained within it is prohibited and may be unlawful. Do not open any attachments, delete it immediately from your system and notify the sender promptly by e-mail that you have done so. The content of this e-mail and any attachments sent with it may have been altered without the consent or knowledge of the author. _____ From: powerh-l-admin@lists.sowder.com [mailto:powerh-l-admin@lists.sowder.com] On Behalf Of Kristina Carlton Sent: 28 September 2004 19:19 To: 'powerh-l@lists.sowder.com' Subject: duplicates Hi, I am running a Quiz extract and I am using the SET NODUPLICATES command. However, I am getting duplicates anyhow. What might I be doing wrong? To clarify further, I run an extract with claim detail information. The claim detail information has the treating physician's FEIN. I access the subfile that is created and link it to our physician file to get detailed physician info. There could be multiple entries in the claim file for each physician obviously, but I was hoping they would not be duplicated in my physician detail extract. Actually, it works for all but one physician (this is a small sample only). I checked that database and do not have duplicates in the physician table. Thanks, -Kristina Ext. 2459 ------=_NextPart_000_0037_01C4A625.10F97080 Content-Type: text/html; charset="us-ascii" Content-Transfer-Encoding: quoted-printable

If you have syntax like = ‘report field1 feild2 etc…’ you could try using syntax like ‘heading = at sortitem field1 feild2 etc…’

 

Regards, Joe.

This e-mail and all information contained in it is confidential and may be legally privileged. If you are not the = intended recipient, your access to this e-mail is unauthorized. Any use, dissemination, distribution, publication or copying by you of this e-mail or any of the information contained within it is prohibited and may be unlawful. Do = not open any attachments, delete it immediately from your system and notify the = sender promptly by e-mail that you have done so. The content of this e-mail and = any attachments sent with it may have been altered without the consent or = knowledge of the author.

From: = powerh-l-admin@lists.sowder.com [mailto:powerh-l-admin@lists.sowder.com] On Behalf Of Kristina Carlton
Sent: 28 September 2004 = 19:19
To: = 'powerh-l@lists.sowder.com'
Subject: = duplicates

 

Hi,

 

I am running a = Quiz extract and I am using the SET NODUPLICATES command. However, I am = getting duplicates anyhow. What might I be doing = wrong?

 

To clarify = further, I run an extract with claim detail information. The claim detail information = has the treating physician’s FEIN. I access the subfile that is created = and link it to our physician file to get detailed physician info. There could be multiple entries in the claim file for each physician obviously, but I = was hoping they would not be duplicated in my physician detail extract. = Actually, it works for all but one physician (this is a small sample only). I = checked that database and do not have duplicates in the physician = table.

 

Thanks,

-Kristina

Ext. = 2459

 

------=_NextPart_000_0037_01C4A625.10F97080-- From Kristina.Carlton@JICompanies.com Wed Sep 29 22:54:36 2004 From: Kristina.Carlton@JICompanies.com (Kristina Carlton) Date: Wed, 29 Sep 2004 16:54:36 -0500 Subject: Linking tables Message-ID: <157670322ACDBB4D82690A2A917E6E1B0345A49B@jiexch01.jicompanies.com> 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_01C4A66E.E9BAE6D9 Content-Type: text/plain; charset="iso-8859-1" Hi List, I am using OpenVMS v7.3-2 and Quiz V7.10.G4. My intention is to link to a table twice but link on different fields. The table is self-referencing and contains physician information. Each physician has a tin (FEIN) and a sub tin (provider within a facility). If the physician works for a clinic he would be listed with the clinic's tin plus a sub tin, but the name field would contain his name. The clinic has no sub tin. In the case of a physician having his own practice he would be listed twice with the same tin, but once without a sub tin and once with a sub tin. Each record will have the same name. This would look as follows: TIN SUB TIN NAME 75-261977 Victory Medical Services 75-261977 001 Dr. Franklin 75-261977 002 Dr. Garcia and so on 75-269999 Dr. Williams 75-269999 001 Dr. Williams I need to link to this file twice in order to get the physician's name as well as the clinic name and also to determine if it is a clinic or an individual. The physician file (PRV) has a combo key on tin/sub tin. The file I am linking to has tin and sub tin as well. I want to link to it once by tin and sub tin and that part is no problem. However, when I attempt to link to it again (giving an alias) with tin only, I get the following error message: Access PRV viaindex PRV_KEY0. ^ *E* Expected: > TO TIN_NUM OF PRV ALIAS PRV1 ^^ Can someone tell me what the correct syntax is? Also, the other file I am linking to is a subfile, not a database table. Thanks, -Kristina Ext. 2459 ------_=_NextPart_001_01C4A66E.E9BAE6D9 Content-Type: text/html; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable Linking tables

Hi List,

I am using OpenVMS v7.3-2 and Quiz V7.10.G4.

My intention is to link to a table twice but link on = different fields. The table is self-referencing and contains physician = information. Each physician has a tin (FEIN) and a sub tin (provider = within a facility). If the physician works for a clinic he would be = listed with the clinic's tin plus a sub tin, but the name field would = contain his name. The clinic has no sub tin.

In the case of a physician having his own practice he = would be listed twice with the same tin, but once without a sub tin and = once with a sub tin. Each record will have the same name. This would = look as follows:

TIN     =         SUB TIN =         NAME
75-261977       =         =         Victory Medical = Services
75-261977       = 001             = Dr. Franklin
75-261977       = 002             = Dr. Garcia
and so on

75-269999       =         =         Dr. Williams
75-269999       = 001             = Dr. Williams

I need to link to this file twice in order to get the = physician's name as well as the clinic name and also to determine if it = is a clinic or an individual. The physician file (PRV) has a combo key = on tin/sub tin. The file I am linking to has tin and sub tin as well. I = want to link to it once by tin and sub tin and that part is no problem. = However, when I attempt to link to it again (giving an alias) with tin = only, I get the following error message:

Access PRV viaindex PRV_KEY0.
          &nb= sp;           &nb= sp;           &nb= sp;           = ^
*E* Expected:  <file>
>        TO = TIN_NUM OF PRV ALIAS PRV1
         = ^^

Can someone tell me what the correct syntax is? Also, = the other file I am linking to is a subfile, not a database = table.

Thanks,
-Kristina
Ext. 2459


------_=_NextPart_001_01C4A66E.E9BAE6D9-- From dmorrison@mcbrideelectric.com Wed Sep 29 23:07:50 2004 From: dmorrison@mcbrideelectric.com (David Morrison - Corporate) Date: Wed, 29 Sep 2004 15:07:50 -0700 Subject: Linking tables Message-ID: <741A7B80E63EFE498457491B6AE5BBE74D26B7@srvrexchange.mcb_corp.mcbrideelectric.com> This is a multi-part message in MIME format. ------_=_NextPart_001_01C4A670.C28C7770 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable Kristina, =20 I'd specify the linkage fields: =20 ACCESS PRV & LINK field1 OF PRV, & field2 OF PRV & TO field1, field OF PRV ALIAS PRV1 OPTIONAL =20 CHOOSE VIAINDEX PRVINDEX =20 David Morrison McBride Electric =20 -----Original Message----- From: powerh-l-admin@lists.sowder.com = [mailto:powerh-l-admin@lists.sowder.com]On Behalf Of Kristina Carlton Sent: Wednesday, September 29, 2004 2:55 PM To: powerh-l@lists.sowder.com Subject: Linking tables Hi List,=20 I am using OpenVMS v7.3-2 and Quiz V7.10.G4.=20 My intention is to link to a table twice but link on different fields. = The table is self-referencing and contains physician information. Each = physician has a tin (FEIN) and a sub tin (provider within a facility). = If the physician works for a clinic he would be listed with the clinic's = tin plus a sub tin, but the name field would contain his name. The = clinic has no sub tin.=20 In the case of a physician having his own practice he would be listed = twice with the same tin, but once without a sub tin and once with a sub = tin. Each record will have the same name. This would look as follows: TIN SUB TIN NAME=20 75-261977 Victory Medical Services=20 75-261977 001 Dr. Franklin=20 75-261977 002 Dr. Garcia=20 and so on=20 75-269999 Dr. Williams=20 75-269999 001 Dr. Williams=20 I need to link to this file twice in order to get the physician's name = as well as the clinic name and also to determine if it is a clinic or an = individual. The physician file (PRV) has a combo key on tin/sub tin. The = file I am linking to has tin and sub tin as well. I want to link to it = once by tin and sub tin and that part is no problem. However, when I = attempt to link to it again (giving an alias) with tin only, I get the = following error message: Access PRV viaindex PRV_KEY0.=20 ^=20 *E* Expected: =20 > TO TIN_NUM OF PRV ALIAS PRV1=20 ^^=20 Can someone tell me what the correct syntax is? Also, the other file I = am linking to is a subfile, not a database table. Thanks,=20 -Kristina=20 Ext. 2459=20 ------_=_NextPart_001_01C4A670.C28C7770 Content-Type: text/html; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable Linking tables
Kristina,
 
I'd = specify the=20 linkage fields:
 
ACCESS PRV=20 &
   LINK=20 field1 OF PRV, &
           = field2 OF=20 PRV  &
  =20 TO   field1, field OF PRV ALIAS PRV1 = OPTIONAL
 
CHOOSE VIAINDEX=20 PRVINDEX
 
David=20 Morrison
McBride=20 Electric
 
-----Original Message-----
From:=20 powerh-l-admin@lists.sowder.com = [mailto:powerh-l-admin@lists.sowder.com]On=20 Behalf Of Kristina Carlton
Sent: Wednesday, September = 29, 2004=20 2:55 PM
To: powerh-l@lists.sowder.com
Subject: = Linking=20 tables

Hi List,

I am using OpenVMS v7.3-2 and Quiz V7.10.G4. =

My intention is to link to a table twice but link on = different=20 fields. The table is self-referencing and contains physician = information. Each=20 physician has a tin (FEIN) and a sub tin (provider within a facility). = If the=20 physician works for a clinic he would be listed with the clinic's tin = plus a=20 sub tin, but the name field would contain his name. The clinic has no = sub tin.=20

In the case of a physician having his own practice = he would be=20 listed twice with the same tin, but once without a sub tin and once = with a sub=20 tin. Each record will have the same name. This would look as=20 follows:

TIN    =20         SUB TIN=20         NAME
75-261977      =20        =20         Victory Medical = Services=20
75-261977      =20 001             = Dr.=20 Franklin
75-261977      =20 002             = Dr.=20 Garcia
and so on

75-269999      =20        =20         Dr. Williams =
75-269999      =20 001             = Dr.=20 Williams

I need to link to this file twice in order to get = the=20 physician's name as well as the clinic name and also to determine if = it is a=20 clinic or an individual. The physician file (PRV) has a combo key on = tin/sub=20 tin. The file I am linking to has tin and sub tin as well. I want to = link to=20 it once by tin and sub tin and that part is no problem. However, when = I=20 attempt to link to it again (giving an alias) with tin only, I get the = following error message:

Access PRV viaindex PRV_KEY0.
          &nbs= p;            = ;            =           =20 ^
*E* Expected:  <file> =
>        TO TIN_NUM OF = PRV ALIAS=20 PRV1
        =20 ^^

Can someone tell me what the correct syntax is? = Also, the=20 other file I am linking to is a subfile, not a database = table.

Thanks,
-Kristina =
Ext. 2459


------_=_NextPart_001_01C4A670.C28C7770-- From brian_matthews_bmw@hotmail.com Thu Sep 30 00:36:18 2004 From: brian_matthews_bmw@hotmail.com (brian_matthewsbrian matthews) Date: Wed, 29 Sep 2004 23:36:18 +0000 Subject: Linking tables Message-ID: or possibly something like ( its difficult without some examples of data ) ACCESS * thesubfile & LINK field1 OF thesubfile , & field2 OF thesubfile & TO field1, field2 OF PRV ALIAS phys OPTIONAL & LINK field1 OF thesubfile & TO field1 of PRV ALIAS clinic OPTIONAL select clinic if field2 of clinic = " " select phys if field2 of phys ne " " rep docname of PRV cinicname of clinic >From: "David Morrison - Corporate" >To: "Kristina Carlton" , > >Subject: RE: Linking tables >Date: Wed, 29 Sep 2004 15:07:50 -0700 > >Kristina, > >I'd specify the linkage fields: > >ACCESS PRV & > LINK field1 OF PRV, & > field2 OF PRV & > TO field1, field OF PRV ALIAS PRV1 OPTIONAL > >CHOOSE VIAINDEX PRVINDEX > >David Morrison >McBride Electric > > >-----Original Message----- >From: powerh-l-admin@lists.sowder.com >[mailto:powerh-l-admin@lists.sowder.com]On Behalf Of Kristina Carlton >Sent: Wednesday, September 29, 2004 2:55 PM >To: powerh-l@lists.sowder.com >Subject: Linking tables > > > >Hi List, > >I am using OpenVMS v7.3-2 and Quiz V7.10.G4. > >My intention is to link to a table twice but link on different fields. The >table is self-referencing and contains physician information. Each >physician has a tin (FEIN) and a sub tin (provider within a facility). If >the physician works for a clinic he would be listed with the clinic's tin >plus a sub tin, but the name field would contain his name. The clinic has >no sub tin. > >In the case of a physician having his own practice he would be listed twice >with the same tin, but once without a sub tin and once with a sub tin. Each >record will have the same name. This would look as follows: > >TIN SUB TIN NAME >75-261977 Victory Medical Services >75-261977 001 Dr. Franklin >75-261977 002 Dr. Garcia >and so on > >75-269999 Dr. Williams >75-269999 001 Dr. Williams > >I need to link to this file twice in order to get the physician's name as >well as the clinic name and also to determine if it is a clinic or an >individual. The physician file (PRV) has a combo key on tin/sub tin. The >file I am linking to has tin and sub tin as well. I want to link to it once >by tin and sub tin and that part is no problem. However, when I attempt to >link to it again (giving an alias) with tin only, I get the following error >message: > >Access PRV viaindex PRV_KEY0. > ^ >*E* Expected: > > TO TIN_NUM OF PRV ALIAS PRV1 > ^^ > >Can someone tell me what the correct syntax is? Also, the other file I am >linking to is a subfile, not a database table. > >Thanks, >-Kristina >Ext. 2459 > > _________________________________________________________________ It's fast, it's easy and it's free. Get MSN Messenger today! http://www.msn.co.uk/messenger From brian_matthews_bmw@hotmail.com Thu Sep 30 00:49:14 2004 From: brian_matthews_bmw@hotmail.com (brian_matthewsbrian matthews) Date: Wed, 29 Sep 2004 23:49:14 +0000 Subject: Linking tables Message-ID: sorry Kristina, I missed the data example. >From: Kristina Carlton >To: powerh-l@lists.sowder.com >Subject: Linking tables >Date: Wed, 29 Sep 2004 16:54:36 -0500 > >Hi List, > >I am using OpenVMS v7.3-2 and Quiz V7.10.G4. > >My intention is to link to a table twice but link on different fields. The >table is self-referencing and contains physician information. Each >physician >has a tin (FEIN) and a sub tin (provider within a facility). If the >physician works for a clinic he would be listed with the clinic's tin plus >a >sub tin, but the name field would contain his name. The clinic has no sub >tin. > >In the case of a physician having his own practice he would be listed twice >with the same tin, but once without a sub tin and once with a sub tin. Each >record will have the same name. This would look as follows: > >TIN SUB TIN NAME >75-261977 Victory Medical Services >75-261977 001 Dr. Franklin >75-261977 002 Dr. Garcia >and so on > >75-269999 Dr. Williams >75-269999 001 Dr. Williams > >I need to link to this file twice in order to get the physician's name as >well as the clinic name and also to determine if it is a clinic or an >individual. The physician file (PRV) has a combo key on tin/sub tin. The >file I am linking to has tin and sub tin as well. I want to link to it once >by tin and sub tin and that part is no problem. However, when I attempt to >link to it again (giving an alias) with tin only, I get the following error >message: > >Access PRV viaindex PRV_KEY0. > ^ >*E* Expected: > > TO TIN_NUM OF PRV ALIAS PRV1 > ^^ > >Can someone tell me what the correct syntax is? Also, the other file I am >linking to is a subfile, not a database table. > >Thanks, >-Kristina >Ext. 2459 > > _________________________________________________________________ Express yourself with cool new emoticons http://www.msn.co.uk/specials/myemo