Spectrum Show
Everybody, The Spectrum show was small (200-ish) but very active. I got to spend time with Janet Oswald and Adam Van Beek (both of whom are IBM people who have commited time to the new international U2 User's group). Both my booth (Key Ally/Moose Enterprises) and the IBM booth gave out U2UG.org sticker and flyers. To all of you who could not be there, you missed a good show. There were several U2 specific classes and offerings. The next Spectrum, Sept. 2004, will be in London. The October show will be in Australia. If you can manage either show, it is well worth attending. -- Sincerely, Charles Barouch www.KeyAlly.com [EMAIL PROTECTED] -- u2-users mailing list [EMAIL PROTECTED] http://www.oliver.com/mailman/listinfo/u2-users
Re: LOCATE Command - Help :(
This is a similar issue with the LOCATE and understanding what level it's trying to locate in. For example, let's say you just read in a REC that has attributes, values and sub values. If one of the attributes may contain the value "FRED" then the reference for REC would be: LOCATE "FRED" IN REC SETTING POS... and POS would be the attribute FRED was in. Notice REC is alone without any <>. If attribute 5 could have a whole value of TOM, then the reference would be: LOCATE "TOM" IN REC<5> SETTING POS. and POS would be the value TOM was in. If a rare circumstance has "BILL" in a value in REC<7,10>, then it's LOCATE "BILL" IN REC<7,10> SETTING POS and POS would be the subvalue BILL was in. Basically it's one level up. If you READV A5 and are LOCATING for a simple multi-value of "SUE", then you use LOCATE "SUE" IN A5<1> SETTING POS not LOCATE "SUE" IN A5 SETTING POS Reason: The LOCATE doesn't know that A5 was the result of a READV. It very well could be the start of something attributed later but now just has a lot of multi-values in attribute 1. my 3 cents. - Original Message - From: "Kate" <[EMAIL PROTECTED]> To: "U2 Users Discussion List" <[EMAIL PROTECTED]> Sent: Friday, March 19, 2004 10:42 PM Subject: LOCATE Command - Help :( > Hi! > > I am trying to accomplish the following > > 1) Build a header for all possible dates for a person with certain data in it. e.g. one person might have 3 dates where they had activity but another person might have 5 dates with activity, but I want the header to have all possible dates with activity. So the Header Date variable contains all possible dates. And that prints fine. > > LOCATE V.DATES IN X.DATE.LIST<1,1> BY "AR" SETTING LOC ELSE > INS V.DATES BEFORE X.DATE.LIST<1,LOC> > END > > But then for each person, I want to have the values print under the correct date headings. And it changes for each page. > > 2) For each record: > a) obtain the # of dates and print the values to print under the correct dates for each person > > LOCATE X.AB.DATE IN X.DATE.LIST<1,1> BY "AR" SETTING POS THEN > IF VALUE = 'B' THEN > X.AB.PRINT<1,POS> = TRIM(X.AB.TYPE) > END > END > > Then I remove the @VM's and print X.AB.PRINT. It prints the values but all in one sequence. Not in the proper position. > > Can anyone tell me what I am doing wrong? > > Thanks for any help you can pass my way. > > Kate > > > > > -- > u2-users mailing list > [EMAIL PROTECTED] > http://www.oliver.com/mailman/listinfo/u2-users -- u2-users mailing list [EMAIL PROTECTED] http://www.oliver.com/mailman/listinfo/u2-users
RE: LOCATE Command - Help :(
The reason that the following line does not work is this: LOCATE V.DATES IN X.DATE.LIST<1,1> BY "AR" SETTING LOC ELSE You are telling it to locate V.DATES in X.DATE.LIST attribute 1, multivalue 1, but you are doing your insert in multivalue 2 then multivalue 3. If you switch your code to what Roger suggests you will be looking in only attribute 1. This is the correct way to do this. In Rogers example, the first locate looks for 13219 in X.DATE.LIST<1,1,1>. It is actually looking for the subvalue it belongs in. When you do the second locate, LOC gets set to 2 because X.DATE.LIST<1,1,1> has a value of 13219 which is less than 13223. The INS statement then says put 13223 before X.DATE.LIST<1,2> even though we were actually looking in X.DATE.LIST<1,1,2>. Then when you do the locate the third time it does the same thing. It looks in X.DATE.LIST<1,1> at the first subvalue, which has a value of 13219 and sets LOC to 2. When you do an INS before X.DATE.LIST<1,2> it puts the 13224 before the 13223. Nick Cipollina Systems Analyst SuperValu Eastern Region (804)746-6068 [EMAIL PROTECTED] Roger Glenfield <[EMAIL PROTECTED]To: U2 Users Discussion List <[EMAIL PROTECTED]> ne.net> cc: Sent by: Subject: RE: LOCATE Command - Help :( u2-users-bounces@ oliver.com 03/19/2004 11:57 PM Please respond to U2 Users Discussion List Suggestion in place. I didn't look/test the second locate. But you'll probably have the same problem. Hope this helps. Roger Glenfield > -Original Message- > From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] > Behalf Of Kate > Sent: Friday, March 19, 2004 10:42 PM > To: U2 Users Discussion List > Subject: LOCATE Command - Help :( > > Hi! > > I am trying to accomplish the following > > 1) Build a header for all possible dates for a person with > certain data in it. e.g. one person might have 3 dates where > they had activity but another person might have 5 dates with > activity, but I want the header to have all possible dates with > activity. So the Header Date variable contains all possible > dates. And that prints fine. > > LOCATE V.DATES IN X.DATE.LIST<1,1> BY "AR" SETTING LOC ELSE try LOCATE V.DATES IN X.DATE.LIST<1> BY "AR" SETTING LOC ELSE ---on Pick flavor at least, your code doesn't work --I feed the code this sequence of dates, 13219, 13224, 13223 and they don't come out 'ar' sequenced. --switch to my code and I then get 13219, 13223, 13224 > INS V.DATES BEFORE X.DATE.LIST<1,LOC> > END > > But then for each person, I want to have the values print under > the correct date headings. And it changes for each page. > > 2) For each record: > a) obtain the # of dates and print the values to print > under the correct dates for each person > > LOCATE X.AB.DATE IN X.DATE.LIST<1,1> BY "AR" SETTING POS THEN > IF VALUE = 'B' THEN > X.AB.PRINT<1,POS> = TRIM(X.AB.TYPE) > END > END > > -- u2-users mailing list [EMAIL PROTECTED] http://www.oliver.com/mailman/listinfo/u2-users -- u2-users mailing list [EMAIL PROTECTED] http://www.oliver.co
RE: LOCATE Command - Help :(
Suggestion in place. I didn't look/test the second locate. But you'll probably have the same problem. Hope this helps. Roger Glenfield > -Original Message- > From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] > Behalf Of Kate > Sent: Friday, March 19, 2004 10:42 PM > To: U2 Users Discussion List > Subject: LOCATE Command - Help :( > > > Hi! > > I am trying to accomplish the following > > 1) Build a header for all possible dates for a person with > certain data in it. e.g. one person might have 3 dates where > they had activity but another person might have 5 dates with > activity, but I want the header to have all possible dates with > activity. So the Header Date variable contains all possible > dates. And that prints fine. > > LOCATE V.DATES IN X.DATE.LIST<1,1> BY "AR" SETTING LOC ELSE try LOCATE V.DATES IN X.DATE.LIST<1> BY "AR" SETTING LOC ELSE ---on Pick flavor at least, your code doesn't work --I feed the code this sequence of dates, 13219, 13224, 13223 and they don't come out 'ar' sequenced. --switch to my code and I then get 13219, 13223, 13224 > INS V.DATES BEFORE X.DATE.LIST<1,LOC> > END > > But then for each person, I want to have the values print under > the correct date headings. And it changes for each page. > > 2) For each record: > a) obtain the # of dates and print the values to print > under the correct dates for each person > > LOCATE X.AB.DATE IN X.DATE.LIST<1,1> BY "AR" SETTING POS THEN > IF VALUE = 'B' THEN > X.AB.PRINT<1,POS> = TRIM(X.AB.TYPE) > END > END > > -- u2-users mailing list [EMAIL PROTECTED] http://www.oliver.com/mailman/listinfo/u2-users
LOCATE Command - Help :(
Hi! I am trying to accomplish the following 1) Build a header for all possible dates for a person with certain data in it. e.g. one person might have 3 dates where they had activity but another person might have 5 dates with activity, but I want the header to have all possible dates with activity. So the Header Date variable contains all possible dates. And that prints fine. LOCATE V.DATES IN X.DATE.LIST<1,1> BY "AR" SETTING LOC ELSE INS V.DATES BEFORE X.DATE.LIST<1,LOC> END But then for each person, I want to have the values print under the correct date headings. And it changes for each page. 2) For each record: a) obtain the # of dates and print the values to print under the correct dates for each person LOCATE X.AB.DATE IN X.DATE.LIST<1,1> BY "AR" SETTING POS THEN IF VALUE = 'B' THEN X.AB.PRINT<1,POS> = TRIM(X.AB.TYPE) END END Then I remove the @VM's and print X.AB.PRINT. It prints the values but all in one sequence. Not in the proper position. Can anyone tell me what I am doing wrong? Thanks for any help you can pass my way. Kate -- u2-users mailing list [EMAIL PROTECTED] http://www.oliver.com/mailman/listinfo/u2-users
[UV] txlg distributed files
UV (HPUX 11i, UV10.0.16) allows you to supposedly activate / deactivate a distributed file, i.e., the top level conglomerate structure, but I *think* the action is totally superfluous; that is, you must activate / deactivate each partfile independently. I *think* acting on the type-27 distributed file itself has no meaning whatsoever and is _B_E_N_I_G_N_. Can anyone verify that ? ? ? Stated again with a little more detail : If you have a distributed file, XYZ, composed of partfiles XYZ_1 & XYZ_2 you can tell Universe to activate/deactivate XYZ for logging and (a) the corresponding record for XYZ in UV.TRANS will have Status set "Active"/null, respectively; and (b) the date-time-modified stamp on the little unix file that defines the partfiles to XYZ is also updated. But none of the partfiles or their own corresponding UV.TRANS records are touched. And it does not matter whether an application opens & writes to XYZ or whether it explicitly opens & writes to individual partfiles, XYZ_1, or XYZ_2; logging will occur / not occur depending solely on how XYZ_1 & XYZ_2 are each individually set, not on what the logging subsystem says it knows about the distributed file, XYZ.. Right? Wrong? If right, is this by design or oversight? Just making sure I haven't missed something vital. Thanks, Chuck Stevenson -- u2-users mailing list [EMAIL PROTECTED] http://www.oliver.com/mailman/listinfo/u2-users
Re: [UV] SP.EDIT & SP.ASSIGN HS
If you are the owner of all of these print jobs type SP.EDIT (MD If you are not the owner of all the jobs, log in as administrator and type SP.EDIT (UMD If entry are present in the spool file that you do not want to delete the user the spooler id number with the command: SP.EDIT 01200 (MD Ron White --- [ Eckel certifies this E-mail to be virus free. ] -- u2-users mailing list [EMAIL PROTECTED] http://www.oliver.com/mailman/listinfo/u2-users
Re: D3 Automatic Faxing
Karl L Pearson wrote: I don't know if it runs on D3 or not. Ask Esker Software (U.S. 800-556-4874). We use this product over 2 faxes. We routinely send between 150 to 300 faxes a day and over 600 at Month-End. I monitor it with a 'roll-my-own' package I wrote and it seems quite okay so far. There are NT-based (NTFS required) monitoring and admin tools that are supposed to be nice, but since I run Linux as my desktop OS, I haven't bothered using those tools. I don't know if anyone else is using this product or not. We're in the process of implementing Esker's VSI-FAX product. It's running on its own linux box and I'm getting the faxes to it via an NFS share mounted to our UV server. Just a matter of writing a flat file with some tags at the top. You can also email faxes to it. It's very customizable. The Windows client interface is convenient. You can start/stop the server, monitor queues, etc. -John -- u2-users mailing list [EMAIL PROTECTED] http://www.oliver.com/mailman/listinfo/u2-users
Re: [UV] SP.EDIT & SP.ASSIGN HS
First, I'm unable to "kill" the Entrys in SP.EDIT.There is a bunch of crap we don't need (about 15 entries spooled) I read on the syntax of the SP.EDIT command but I must have it wrong. >SP.EDIT 01200 MD >Entry # 01200 : Is Ready to Output. It doesn't delete it but just shows it Ready again. - Original Message - From: "Ray Wurlod" <[EMAIL PROTECTED]> To: "U2 Users Discussion List" <[EMAIL PROTECTED]> Sent: Thursday, March 18, 2004 8:12 PM Subject: Re: [UV] SP.EDIT & SP.ASSIGN HS > SP.ASSIGN (HS sends the file to the spooler, with HOLD and RETAIN attributes asserted. It does not send files to the &HOLD& file. You will find your files in the directory referred to by your UVSPOOL configuration parameter. You can use the SPOOL -LIST command to see that they are held from printing. > > You can use the SP.EDIT command to print them and manage them (for example "kill" (de-queue) them). > > If you want to send output to &HOLD& the correct command to use is SETPTR with the sixth argument set to 3. For example: SETPTR 0,132,2,0,0,3,BANNER MYFILE,BRIEF (this command directs the default print channel to a record called MYFILE in the &HOLD& file. > > If you examine the SP.ASSIGN verb in the VOC file, and trace it back to source, you will find that SP.ASSIGN (HS actually ends up invoking the SETPTR command, but in mode 1 (spooler), for example SETPTR 0,132,60,3,3,1,HOLD,RETAIN > > > - Original Message - > From: "Cyndi Calvin" <[EMAIL PROTECTED]> > Date: Thu, 18 Mar 2004 15:39:19 -0800 > To: "U2 Users Discussion List" <[EMAIL PROTECTED]> > Subject: SP.EDIT & SP.ASSIGN HS > > > Hi, I'm trying to use the 'SP.ASSIGN HS' to send the output to file under UNIX &HOLD& directory. > > > > I'm doing and seeing the following -- but no file ever shows up under &HOLD& directory. Any clues or thoughts? It gets to the queue? But I don't know what to answer to get it to file. Thanks!!! > > > > All these "Ready to Output" should be deleted but I can't > > > > SP.EDIT before I run the SP.ASSIGN > > >SP.EDIT > > Entry # 01200 : Is Ready to Output. > > Entry # 01201 : Is Ready to Output. > > Entry # 01202 : Is Ready to Output. > > Entry # 01203 : Is Ready to Output. > > Entry # 01211 : Is Ready to Output. > > Entry # 01227 : Is Ready to Output. > > Entry # 01230 : Is Ready to Output. > > Entry # 01231 : Is Ready to Output. > > Entry # 01234 : Is Ready to Output. > > Entry # 01236 : Is Ready to Output. > > Entry # 01238 : Is Ready to Output. > > Entry # 01239 : Is Ready to Output. > > Entry # 01259 : Is Ready to Output. > > Entry # 01260 : Is Ready to Output. > > Entry # 01264 : Is Ready to Output. > > Entry # 01267 : Is Ready to Output. > > Entry # 01277 : Is Ready to Output. > > Entry # 01282 : Is Ready to Output. > > Entry # 01364 : Is Ready to Output. > > Entry # 01370 : Is Ready to Output. > > Entry # 01371 : Is Ready to Output. > > >SP.ASSIGN HS > > >LIST.ITEM CUST (P > > Spooler Entry #2877 > > >SP.EDIT > > Entry # 01200 : Is Ready to Output. > > Entry # 01201 : Is Ready to Output. > > Entry # 01202 : Is Ready to Output. > > Entry # 01203 : Is Ready to Output. > > Entry # 01211 : Is Ready to Output. > > Entry # 01227 : Is Ready to Output. > > Entry # 01230 : Is Ready to Output. > > Entry # 01231 : Is Ready to Output. > > Entry # 01234 : Is Ready to Output. > > Entry # 01236 : Is Ready to Output. > > Entry # 01238 : Is Ready to Output. > > Entry # 01239 : Is Ready to Output. > > Entry # 01259 : Is Ready to Output. > > Entry # 01260 : Is Ready to Output. > > Entry # 01264 : Is Ready to Output. > > Entry # 01267 : Is Ready to Output. > > Entry # 01277 : Is Ready to Output. > > Entry # 01282 : Is Ready to Output. > > Entry # 01364 : Is Ready to Output. > > Entry # 01370 : Is Ready to Output. > > Entry # 01371 : Is Ready to Output. > > Entry # 02877Status : Hold*Size: 422305 > > Priority: 88 Copies : 1Delay : > > Pages : All Eject : Yes Delete : Yes > > Form: 1Printer : [Any]Options : None > > Display (Y/N/S/D/X/) ? N (if I say Y it does show on the screen) > > Spool (Y/N=CR) ? Y > > Pages to output : ALL > > > > -- > > u2-users mailing list > > [EMAIL PROTECTED] > > http://www.oliver.com/mailman/listinfo/u2-users > > -- > u2-users mailing list > [EMAIL PROTECTED] > http://www.oliver.com/mailman/listinfo/u2-users > -- u2-users mailing list [EMAIL PROTECTED] http://www.oliver.com/mailman/listinfo/u2-users
RE: UNCLASSIFIED RE: Unidata "Flashbasic"
Ah yes, RAID :-( Unfortunately you have to get the program into BEBUG state before you can start the process of profiling, then enter a "$" or "#" at the ever-popular user-friendly "::" prompt, then "C", then when the program executes a "STOP" it goes back to "::" again and you have to enter "C" again to get back to TCL. At least 3 user interactions and the user has to have TCL access by the look of it (I haven't tested that). And you couldn't do it at all in a PHANTOM. Whereas for the UD model, to get coarse-level timings, you just change your menu program (or your UniObjects host?) to add a "-G" to the "RUN ..." statements it emits, or even put one in the LOGIN Paragraph. 'Look Ma, no hands!', the user doesn't even have to be aware of what's going on, and you can profile a PHANTOM job. I have found very-low-level profiling useful in a development or test environment, but often only after coarse-level profiling in Production has identified which little beastie is soaking up all the CPU! HTH Mike -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Leroy Dreyfuss Sent: Saturday, 20 March 2004 08:56 To: U2 Users Discussion List Subject: Re: UNCLASSIFIED RE: Unidata "Flashbasic" Mike, UniVerse does have some capabilities you may have forgotten about. Using RAID, you can generate program/subroutine timings (#) and instruction counting ($). Take a look at the RAID information in the BASIC book to see if they help. Regards, LeRoy F. Dreyfuss Advanced Technical Services - UniVerse IBM U2 Data Management Solutions Tel: 303-672-1254 Fax: 303-294-4832 Mobile: 720-341-4317 External email: [EMAIL PROTECTED] -- u2-users mailing list [EMAIL PROTECTED] http://www.oliver.com/mailman/listinfo/u2-users The information contained in this Internet Email message is intended for the addressee only and may contain privileged information, but not necessarily the official views or opinions of the New Zealand Defence Force. If you are not the intended recipient you must not use, disclose, copy or distribute this message or the information in it. If you have received this message in error, please Email or telephone the sender immediately. -- u2-users mailing list [EMAIL PROTECTED] http://www.oliver.com/mailman/listinfo/u2-users
Re: UNCLASSIFIED RE: Unidata "Flashbasic"
HENDERSON MICHAEL MR <[EMAIL PROTECTED]> wrote on 03/19/2004 02:46:32 PM: > > > > Piece of cake! Compile and run with the -G option. > > Pity that's a UniData-only feature. > It'd be really nice to have the same functionality in UniVerse as well! You'll get no argument from me. I optimize application code on both U2 platforms, and UniData's profiling is a really grand shortcut to identifying bottlenecks. Maybe if enough UniVerse users request it, it will be added. Tim Snyder IBM Data Management Solutions Consulting I/T Specialist , U2 Professional Services [EMAIL PROTECTED] -- u2-users mailing list [EMAIL PROTECTED] http://www.oliver.com/mailman/listinfo/u2-users
Re: UNCLASSIFIED RE: Unidata "Flashbasic"
Mike, UniVerse does have some capabilities you may have forgotten about. Using RAID, you can generate program/subroutine timings (#) and instruction counting ($). Take a look at the RAID information in the BASIC book to see if they help. Regards, LeRoy F. Dreyfuss Advanced Technical Services - UniVerse IBM U2 Data Management Solutions Tel: 303-672-1254 Fax: 303-294-4832 Mobile: 720-341-4317 External email: [EMAIL PROTECTED] WWW: http://www.ibm.com/software/data/u2/support www.ibm.com/software/data/u2/support - Open, Query, Update, Search - Online! HENDERSON MICHAEL MR "'U2 Users Discussion List'" Sent by: <[EMAIL PROTECTED]> u2-users-bounces@ cc oliver.com Subject UNCLASSIFIED RE: Unidata 03/19/2004 12:46 "Flashbasic" PM Please respond to U2 Users Discussion List Tim, -Original Message- > From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Timothy Snyder > Sent: Saturday, 20 March 2004 05:59 > To: U2 Users Discussion List > Subject: RE: Unidata "Flashbasic" [snip] > > Piece of cake! Compile and run with the -G option. Pity that's a UniData-only feature. It'd be really nice to have the same functionality in UniVerse as well! Mike > It's documented in the Administering UniData manuals in > the chapter on Monitoring and Tuning UniData. > > > Tim Snyder > IBM Data Management Solutions > Consulting I/T Specialist , U2 Professional Services > The information contained in this Internet Email message is intended for the addressee only and may contain privileged information, but not necessarily the official views or opinions of the New Zealand Defence Force. If you are not the intended recipient you must not use, disclose, copy or distribute this message or the information in it. If you have received this message in error, please Email or telephone the sender immediately. -- u2-users mailing list [EMAIL PROTECTED] http://www.oliver.com/mailman/listinfo/u2-users -- u2-users mailing list [EMAIL PROTECTED] http://www.oliver.com/mailman/listinfo/u2-users
UNCLASSIFIED RE: Unidata "Flashbasic"
Tim, -Original Message- > From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Timothy Snyder > Sent: Saturday, 20 March 2004 05:59 > To: U2 Users Discussion List > Subject: RE: Unidata "Flashbasic" [snip] > > Piece of cake! Compile and run with the -G option. Pity that's a UniData-only feature. It'd be really nice to have the same functionality in UniVerse as well! Mike > It's documented in the Administering UniData manuals in > the chapter on Monitoring and Tuning UniData. > > > Tim Snyder > IBM Data Management Solutions > Consulting I/T Specialist , U2 Professional Services > The information contained in this Internet Email message is intended for the addressee only and may contain privileged information, but not necessarily the official views or opinions of the New Zealand Defence Force. If you are not the intended recipient you must not use, disclose, copy or distribute this message or the information in it. If you have received this message in error, please Email or telephone the sender immediately. -- u2-users mailing list [EMAIL PROTECTED] http://www.oliver.com/mailman/listinfo/u2-users
RE: Unidata "Flashbasic"
It was easier to document than I thought... 1) Decide what program(s) to 'profile': You would probably only profile the header program FIRST, then see what called subroutines are taking a while in total, then switch your profiling to THAT subroutine only. The profiler generates A LOT of data, so you don't want to profile all 35,000,000 lines of code you've written in a single sitting! 2) Compile the programs you want to target with a -G option BASIC BP PROGTOTEST -G 3) Run the program using the -G option. RUN BP PROGTOTEST -G Note: If your PROGTOTEST is an embedded subroutine, no problem, you would type RUN BP MASTERPROG -G , and the -G option would start profiling on the MASTERPROG in 'summary', but would do 'detail' profiling on the program PROGTOTEST as you compiled it with the -G option. 4) Shell to the account you are working in and find files marked 'profile.NNN' and 'profile.elapse.NNN', where NNN = PID. These are NT or *nix files - not VOC entries. So you can't see them until you go to to OS level for the account you are in. The profile.NNN shows the time in actual CPU usage... It shows %time of time spent in the listed routine, cumulative time in secs(*more on this in a moment), seconds spent on this routine, calls to this routine, routine name The profile.elapse.NNN shows the time on the CLOCK, so if you had to key data in to a subroutine, you can expect to see it near the top of the % of time occupied. %time spent in the listed routine, cumulative seconds spent to date on the funtions (*more on this in a moment), Raw Seconds spent, # calls to this routine, routine name. The Cumulative time is not sequential: You may see data like this: %time CumSec Seconds Calls Name 10.03.00 3.00 1d:\account\_ROUTINEC 10.06.00 3.00 1d:\account\_ROUTINEA 6.78.00 2.00 1d:\account\_ROUTINEB See - the CumSecs is not the time elapsed for the ORDER the routines processed in, but the Cumulative time for the routines in the order of %time spent. Also - if you have not compiled a routine with a -G option, the WHOLE ROUTINE shows up on one line, as shown above. However, if you had keyed ROUTINEC with a -G option, you would see this instead: %time CumSec Seconds Calls Name 10.03.00 3.00 1d:\account\_ROUTINEA 6.75.00 2.00 1d:\account\_ROUTINEB 5.06.50 1.50 6d:\account\_ROUTINEC:1000 2.57.25 .75 1d:\account\_ROUTINEC:10 2.58.00 .75 1d:\account\_ROUTINEC:100 Where the # after the colon is the internal subroutine label being exectuted. In this case, I can see subroutine 1000 is called 6 times in the program, so that works out to be .25 seconds a use. Not much to fix there compared to 10 and 100 at .75 each, UNLESS I didn't know subroutine 1000 was being called 6 times! As I said, you can find out SO much about your code running this - you may realize that optimizing one or two sections of a program will have a large impact. As well, once you get a printout of what programs are using what amount of time, it will lead you to do a -G on them - in the example above, I can see I should also profile ROUTINEA to see if I can hammer out some time from the routine. Once you are done with profiling, you should recompile the program WITHOUT the -G option - this will produce the smaller object code again since it will be built without the 'profile' handles. Hope that helps someone! David Wolverton -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of David Wolverton Sent: Friday, March 19, 2004 12:30 PM To: 'U2 Users Discussion List' Subject: RE: Unidata "Flashbasic" I'll do it before the weekend is over... It's *very* cool, and useful too - what a great combo! DW -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Wendy Smoak Sent: Friday, March 19, 2004 10:49 AM To: U2 Users Discussion List Subject: RE: Unidata "Flashbasic" David Wolverton wrote: > if you > compile a routine with UniData's Profiler Option 'on', the really > interesting thing is that you can see where the system is spending a > lot of time, or doing a lot of recursion - you can see clock time > spent, CPU time spent, and number of times the subroutine was called. Would you be willing to post a HOWTO? Where do you set this option, what does the output look like? This sounds very interesting but is yet another thing that would take me half a day to figure out. If you sketch out the basics I'll fill in the details and put it on the Wiki. -- Wendy Smoak Application Systems Analyst, Sr. ASU IA Information Resources Management -- u2-users mailing list [EMAIL PROTECTED] http://www.oliver.com/mailman/listinfo/u2-users -- u2-users mailing list [EMAIL PROTECTED] http://www.oliver.com/mailman/listinfo/u2-users -- u2-users mailing list [EMAIL PROTECTED] http://www.oliver.com/mailman/listinfo/u2-users
RE: Unidata "Flashbasic"
www . pickwiki . com -Original Message- Behalf Of Marlene Yokoyama Whats the Wiki?? -- u2-users mailing list [EMAIL PROTECTED] http://www.oliver.com/mailman/listinfo/u2-users
Wiki (was RE: Unidata "Flashbasic")
Marlene Yokoyama > Whats the Wiki?? http://www.pickwiki.com Anyone can edit any of the pages or add new ones. Just SquashWordsTogether and it will create a link. ;) -- Wendy Smoak Application Systems Analyst, Sr. ASU IA Information Resources Management -- u2-users mailing list [EMAIL PROTECTED] http://www.oliver.com/mailman/listinfo/u2-users
RE: Unidata "Flashbasic"
I'll do it before the weekend is over... It's *very* cool, and useful too - what a great combo! DW -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Wendy Smoak Sent: Friday, March 19, 2004 10:49 AM To: U2 Users Discussion List Subject: RE: Unidata "Flashbasic" David Wolverton wrote: > if you > compile a routine with UniData's Profiler Option 'on', the really > interesting thing is that you can see where the system is spending a > lot of time, or doing a lot of recursion - you can see clock time > spent, CPU time spent, and number of times the subroutine was called. Would you be willing to post a HOWTO? Where do you set this option, what does the output look like? This sounds very interesting but is yet another thing that would take me half a day to figure out. If you sketch out the basics I'll fill in the details and put it on the Wiki. -- Wendy Smoak Application Systems Analyst, Sr. ASU IA Information Resources Management -- u2-users mailing list [EMAIL PROTECTED] http://www.oliver.com/mailman/listinfo/u2-users -- u2-users mailing list [EMAIL PROTECTED] http://www.oliver.com/mailman/listinfo/u2-users
RE: Unidata "Flashbasic"
Whats the Wiki?? >>> [EMAIL PROTECTED] 3/19/2004 8:49:02 AM >>> David Wolverton wrote: > if you > compile a routine with UniData's Profiler Option 'on', the really > interesting thing is that you can see where the system is > spending a lot of > time, or doing a lot of recursion - you can see clock time > spent, CPU time > spent, and number of times the subroutine was called. Would you be willing to post a HOWTO? Where do you set this option, what does the output look like? This sounds very interesting but is yet another thing that would take me half a day to figure out. If you sketch out the basics I'll fill in the details and put it on the Wiki. -- Wendy Smoak Application Systems Analyst, Sr. ASU IA Information Resources Management -- u2-users mailing list [EMAIL PROTECTED] http://www.oliver.com/mailman/listinfo/u2-users -- u2-users mailing list [EMAIL PROTECTED] http://www.oliver.com/mailman/listinfo/u2-users
RE: Unidata "Flashbasic"
Wendy Smoak wrote on 03/19/2004 11:49:02 AM: > David Wolverton wrote: > > if you > > compile a routine with UniData's Profiler Option 'on', the really > > interesting thing is that you can see where the system is > > spending a lot of > > time, or doing a lot of recursion - you can see clock time > > spent, CPU time > > spent, and number of times the subroutine was called. > > Would you be willing to post a HOWTO? Where do you set this option, > what does the output look like? This sounds very interesting but is yet > another thing that would take me half a day to figure out. > > If you sketch out the basics I'll fill in the details and put it on the > Wiki. > Piece of cake! Compile and run with the -G option. It's documented in the Administering UniData manuals in the chapter on Monitoring and Tuning UniData. Tim Snyder IBM Data Management Solutions Consulting I/T Specialist , U2 Professional Services Office (717) 545-6403 (rolls to cell phone) [EMAIL PROTECTED] -- u2-users mailing list [EMAIL PROTECTED] http://www.oliver.com/mailman/listinfo/u2-users
RE: Unidata "Flashbasic"
I'd be interested in this, too. It sounds very useful. Thanks to you both! Dave -Original Message- Behalf Of Wendy Smoak Would you be willing to post a HOWTO? Where do you set this option, what does the output look like? This sounds very interesting but is yet another thing that would take me half a day to figure out. If you sketch out the basics I'll fill in the details and put it on the Wiki. -- u2-users mailing list [EMAIL PROTECTED] http://www.oliver.com/mailman/listinfo/u2-users
RE: Crazy intermittent problem. Need help as soon as possible.
Good thought. I haven't, but it's a good way to go. I'll let you know what I find. Thanks, Dave -Original Message- Behalf Of Piers Angliss Have you (can you) examined the Schedule1 table to verify that the corruption is occurring between Schedule1 and SCHED_DATA and not between screen and Schedule1 ? -- u2-users mailing list [EMAIL PROTECTED] http://www.oliver.com/mailman/listinfo/u2-users
RE: Unidata "Flashbasic"
David Wolverton wrote: > if you > compile a routine with UniData's Profiler Option 'on', the really > interesting thing is that you can see where the system is > spending a lot of > time, or doing a lot of recursion - you can see clock time > spent, CPU time > spent, and number of times the subroutine was called. Would you be willing to post a HOWTO? Where do you set this option, what does the output look like? This sounds very interesting but is yet another thing that would take me half a day to figure out. If you sketch out the basics I'll fill in the details and put it on the Wiki. -- Wendy Smoak Application Systems Analyst, Sr. ASU IA Information Resources Management -- u2-users mailing list [EMAIL PROTECTED] http://www.oliver.com/mailman/listinfo/u2-users
RE: Crazy intermittent problem. Need help as soon as possible.
Dave, Have you (can you) examined the Schedule1 table to verify that the corruption is occurring between Schedule1 and SCHED_DATA and not between screen and Schedule1 ? just a straw to clutch at ;-) -- u2-users mailing list [EMAIL PROTECTED] http://www.oliver.com/mailman/listinfo/u2-users
RE: Unidata "Flashbasic"
No - there's no additional compile step I'm aware of... You're forced to use 'best practices' to optimize a program - but if you compile a routine with UniData's Profiler Option 'on', the really interesting thing is that you can see where the system is spending a lot of time, or doing a lot of recursion - you can see clock time spent, CPU time spent, and number of times the subroutine was called. I found it immensely useful to find areas of programs I could optimize further to get the most 'bang for my buck' - it also led me to find a subroutine I called many times that had a file open within it - by fixing that one issue, I saved 30% on one routine... YMMV... David W. -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Raymond de Bourbon Sent: Friday, March 19, 2004 8:39 AM To: 'U2 Users' Subject: Unidata "Flashbasic" Does Unidata have an option similar to D3 that enables one to "Flashcompile" a basic program into native machine code? I have a program that I need to squeeze the optimal performance out of.. Regards Raymond de Bourbon -- u2-users mailing list [EMAIL PROTECTED] http://www.oliver.com/mailman/listinfo/u2-users -- u2-users mailing list [EMAIL PROTECTED] http://www.oliver.com/mailman/listinfo/u2-users
RE: Crazy intermittent problem. Need help as soon as possible.
Are these internal in the UniObjects interface? I haven't encountered them before. - Dave -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] Behalf Of Donald Kibbey Sent: Friday, March 19, 2004 9:48 AM To: [EMAIL PROTECTED] Subject: RE: Crazy intermittent problem. Need help as soon aspossible. streamwriter and textwriter are .Net objects used to write "stuff" to files. I've encountered some interesting issues when I forgot to flush them Don Kibbey Financial Systems Manager Finnegan, Henderson, Farabow, Garrett & Dunner LLP >>> [EMAIL PROTECTED] 03/19/04 09:17AM >>> Hi, I'm using UniObjects to write the data, too, using SchedFile.Write(). I've posted some code in my last note. I haven't heard of the streamwriter and/or textwriter objects. Could you please tell me more? Thanks, Dave -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] Behalf Of Donald Kibbey Sent: Thursday, March 18, 2004 4:30 PM To: [EMAIL PROTECTED] Subject: Re: Crazy intermittent problem. Need help as soon aspossible. You don't say what your using to write the data. On the PC side, look for the flush() function of the streamwriter and textwriter objects. Are you maintaining a constant connection with the UniData machine? Really need some code to look at to help. Thanks, Don Kibbey Financial Systems Manager Finnegan, Henderson, Farabow, Garrett & Dunner LLP >>> [EMAIL PROTECTED] 03/18/04 04:17PM >>> Hi, Everyone. I'm having trouble with a scheduling program I'm writing in VB.net, on a WinXPpro system, using UniObjects, that extracts and writes data to a file in our Unidata 6.0.3 system on an AIX 5.1L system. The trouble that I'm having is that the data is getting corrupted. As the screen doesn't rebuild after a data save, I don't 'see' the problem until I restart the program the next time around, and data is re-read from the database. I've been through my program 3 times, and can't find an error. I'm assuming at this point that the error is outside my program, as I ran the same test data through the program 5 times, and it only corrupted the data 3 times. The other two it worked flawlessly. If it was an error in my program, it shouldn't ever work right, should it? Also, the program treats every record equally, but the data is intermittently corrupted, some records process fine, others get corrupted. Some fields are lost all together. Fortunately it's still in test mode, so we haven't lost any real data yet. The data is doing things like this: Record 1: Record 2: DelDate:03/18/0403/18/04 <--OK SeqNo: 3 3<--OK Type: F 0
Defunct processes
A few weeks ago, we upgraded to Unidata 6.0. Since then, we are getting several defunct processes. The pattern is the same every week. We stop/start unidata on Sunday. On every other night, we only pause/resume. Since we upgraded, on Thursday and Friday of each week, several dozen processes go into a defunct state, with smm as the parent process. These processes die when unidata is stopped on Sunday. Other than occupying space in the process table, they do not seem to be causing any problems. Is anybody else getting a lot of defunct processes? Charlie Rubeor Unix/Database Admin The Wiremold Company 800.338.1315 x3498 860.523.3690 fax > À votre service > -- u2-users mailing list [EMAIL PROTECTED] http://www.oliver.com/mailman/listinfo/u2-users
RE: Crazy intermittent problem. Need help as soon as possible.
Hi, and thanks for the assistance. > Timothy Snyder wrote: > I'm just trying to isolate whether the problem with your program is in the update or > display routine. Sorry that I didn't explain this better. What I'm actually seeing is that everything on the screen looks fine. I'll move a few SchedBoxes around, and everything still looks fine. Then when I shut down the program, and restart it, the data displayed on the screen is incorrect compared to the way that I left it. The issue is that the data isn't being written during the 'shutdown' process. It's written individually each time that something changes. When I go back to check the record in Unidata, using a LIST command or the AE editor, I can verify that the data that is actually in the file, is incorrect. So, the data is being corrupted before I even know about it. Probably the reason that it's hard to track this down, is that I'm displaying data in textboxes that are part of a custom control. These aren't directly connected to the DataTable where the data resides. Basically, you read the SCHED_DATA record to the SchedFile object. Then, that data is added to the 'Schedule1' table, one record at a time. After the Schedule1 table is filled, then the SchedBox objects are created, each using the data from one Schedule1 record. Then it's a reverse process when you write back to the file. You make changes to the SchedBox object. This in turn updates the Schedule1 DataTable. Then the program has to put the DataTable record into the SchedFile object, where it's written back to the SCHED_DATA Unidata file. (It would have been a lot easier, if I'd hadn't had to create an array of custom controls (SchedBoxes). Each SchedBox object contains 10 textboxes and 8 other data members. Maybe I'm just doing too many steps. Is there a better way to keep the SchedBox custom control's data sync'ed to the Unidata file? Thanks, Dave -- u2-users mailing list [EMAIL PROTECTED] http://www.oliver.com/mailman/listinfo/u2-users
RFID tracking...
Has anyone here worked with the RFID readers? We are looking to track attendance at meetings. Options are Barcoded badges, Mag stripe Cards - both require action to scan. I was wondering if RFID might be an option. To imbed a RF tag on a badge, which when walking past a RFID reader would sense the tag and read it's ID. Anyone know of any good MFG's/Sites that discusses RFID implementation? One product I read had inches as the distance from reader and tag, can this be increased to feet? Thanks George George Gallen Senior Programmer/Analyst Accounting/Data Division [EMAIL PROTECTED] ph:856.848.1000 Ext 220 SLACK Incorporated - An innovative information, education and management company http://www.slackinc.com -- u2-users mailing list [EMAIL PROTECTED] http://www.oliver.com/mailman/listinfo/u2-users
RE: Crazy intermittent problem. Need help as soon as possible.
Dave Tabor wrote on 03/19/2004 09:39:29 AM: > Yes the records are being stored erroneously. I don't actually > notice the problem until the data is read when opening the program > the second time. It sounds like you're basing this on the way your program is displaying things. Have you looked at the data in any other way (i.e.: UniQuery, an editor, a straight BASIC program, etc.)? I'm just trying to isolate whether the problem with your program is in the update or display routine. Tim Snyder IBM Data Management Solutions Consulting I/T Specialist , U2 Professional Services [EMAIL PROTECTED] -- u2-users mailing list [EMAIL PROTECTED] http://www.oliver.com/mailman/listinfo/u2-users
RE: Crazy intermittent problem. Need help as soon as possible.
streamwriter and textwriter are .Net objects used to write "stuff" to files. I've encountered some interesting issues when I forgot to flush them Don Kibbey Financial Systems Manager Finnegan, Henderson, Farabow, Garrett & Dunner LLP >>> [EMAIL PROTECTED] 03/19/04 09:17AM >>> Hi, I'm using UniObjects to write the data, too, using SchedFile.Write(). I've posted some code in my last note. I haven't heard of the streamwriter and/or textwriter objects. Could you please tell me more? Thanks, Dave -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] Behalf Of Donald Kibbey Sent: Thursday, March 18, 2004 4:30 PM To: [EMAIL PROTECTED] Subject: Re: Crazy intermittent problem. Need help as soon aspossible. You don't say what your using to write the data. On the PC side, look for the flush() function of the streamwriter and textwriter objects. Are you maintaining a constant connection with the UniData machine? Really need some code to look at to help. Thanks, Don Kibbey Financial Systems Manager Finnegan, Henderson, Farabow, Garrett & Dunner LLP >>> [EMAIL PROTECTED] 03/18/04 04:17PM >>> Hi, Everyone. I'm having trouble with a scheduling program I'm writing in VB.net, on a WinXPpro system, using UniObjects, that extracts and writes data to a file in our Unidata 6.0.3 system on an AIX 5.1L system. The trouble that I'm having is that the data is getting corrupted. As the screen doesn't rebuild after a data save, I don't 'see' the problem until I restart the program the next time around, and data is re-read from the database. I've been through my program 3 times, and can't find an error. I'm assuming at this point that the error is outside my program, as I ran the same test data through the program 5 times, and it only corrupted the data 3 times. The other two it worked flawlessly. If it was an error in my program, it shouldn't ever work right, should it? Also, the program treats every record equally, but the data is intermittently corrupted, some records process fine, others get corrupted. Some fields are lost all together. Fortunately it's still in test mode, so we haven't lost any real data yet. The data is doing things like this: Record 1: Record 2: DelDate:03/18/0403/18/04 <--OK SeqNo: 3 3<--OK Type: F 0
Unidata "Flashbasic"
Does Unidata have an option similar to D3 that enables one to "Flashcompile" a basic program into native machine code? I have a program that I need to squeeze the optimal performance out of.. Regards Raymond de Bourbon -- u2-users mailing list [EMAIL PROTECTED] http://www.oliver.com/mailman/listinfo/u2-users
RE: Crazy intermittent problem. Need help as soon as possible.
Yes the records are being stored erroneously. I don't actually notice the problem until the data is read when opening the program the second time. My session is opened and closed each time that reading and writing is done. Normally, the connection is closed. I'm basically doing this: 1) Selecting a list of SCHED_DATA records. 2) Getting Records and filling a 'Schedule1' DataTable - Read a record from the SCHED_DATA file. - Use Schedule1.record(x).item() = .record.field(x).stringvalue to fill an individual Schedule1 record. 3) Creating and filling an array of custom SchedBox objects (used for holding scheduling info) 4) Writing individual records when: - Data in a 'SchedBox' is modified - A 'SchedBox' is moved to another day (column) - drag-n-drop. NOTE: When a schedbox is moved, all others below it are moved up one in sequence in the 'from' column, and all others below it are moved down one in sequence in the 'to' column. Any schedboxes that moved are then rewritten to the SCHED_DATA file to store the new SEQ_NO and DEL_DATE. This is when the file is being updated on a mass record basis. - All writing is done by: .RecordID = .write() 5) Allow for the creation of blank 'SchedBox'es, so new schedules can be created. 6) Determine the ID number for the new record in the file. 7) Add the new record to the 'Schedule1' DataTable. 8) Write the new record to the SCHED_DATA file. Data (in the SCHED_DATA file) is being changed this way: __BEFORE__ __AFTER__ DelDate:03/18/0403/18/04 <--OK SeqNo: 3 3<--OK Type: F 0
RE: Crazy intermittent problem. Need help as soon as possible.
Hi, I'm using UniObjects to write the data, too, using SchedFile.Write(). I've posted some code in my last note. I haven't heard of the streamwriter and/or textwriter objects. Could you please tell me more? Thanks, Dave -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] Behalf Of Donald Kibbey Sent: Thursday, March 18, 2004 4:30 PM To: [EMAIL PROTECTED] Subject: Re: Crazy intermittent problem. Need help as soon aspossible. You don't say what your using to write the data. On the PC side, look for the flush() function of the streamwriter and textwriter objects. Are you maintaining a constant connection with the UniData machine? Really need some code to look at to help. Thanks, Don Kibbey Financial Systems Manager Finnegan, Henderson, Farabow, Garrett & Dunner LLP >>> [EMAIL PROTECTED] 03/18/04 04:17PM >>> Hi, Everyone. I'm having trouble with a scheduling program I'm writing in VB.net, on a WinXPpro system, using UniObjects, that extracts and writes data to a file in our Unidata 6.0.3 system on an AIX 5.1L system. The trouble that I'm having is that the data is getting corrupted. As the screen doesn't rebuild after a data save, I don't 'see' the problem until I restart the program the next time around, and data is re-read from the database. I've been through my program 3 times, and can't find an error. I'm assuming at this point that the error is outside my program, as I ran the same test data through the program 5 times, and it only corrupted the data 3 times. The other two it worked flawlessly. If it was an error in my program, it shouldn't ever work right, should it? Also, the program treats every record equally, but the data is intermittently corrupted, some records process fine, others get corrupted. Some fields are lost all together. Fortunately it's still in test mode, so we haven't lost any real data yet. The data is doing things like this: Record 1: Record 2: DelDate:03/18/0403/18/04 <--OK SeqNo: 3 3<--OK Type: F 0
RE: [OT] Symbol PDT6800 Help
Don't remind me :-) Our first office computer was a 25/30 (a 300 chassis with a 250 cpu board). Great machine, but the same size as todays big domestic freezers and that didn't even include the disk drives ... 116Mb :-) Cheers, Wol -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Jefferson, Jim Sent: 18 March 2004 20:08 To: U2 Users Discussion List Subject: RE: [OT] Symbol PDT6800 Help ...or the Prime I-1000's/P400's. Flip switches to set up octal addresses, twist a dial between "fetch" and "store" (?), put your left foot in, put your left foot out... Jim -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Steven M Wagner Sent: Thursday, March 18, 2004 1:54 PM To: U2 Users Discussion List Subject:RE: [OT] Symbol PDT6800 Help Sorta makes me think of the booting instructions on a MicroData Reality. Steve At 02:53 PM 3/18/04 -0500, you wrote: >Not sure exactly, but I've got some cryptic notes here: >I think it's ++ -- u2-users mailing list [EMAIL PROTECTED] http://www.oliver.com/mailman/listinfo/u2-users *** This transmission is intended for the named recipient only. It may contain private and confidential information. If this has come to you in error you must not act on anything disclosed in it, nor must you copy it, modify it, disseminate it in any way, or show it to anyone. Please e-mail the sender to inform us of the transmission error or telephone ECA International immediately and delete the e-mail from your information system. Telephone numbers for ECA International offices are: Sydney +61 (0)2 9911 7799, Hong Kong + 852 2121 2388, London +44 (0)20 7351 5000 and New York +1 212 582 2333. *** -- u2-users mailing list [EMAIL PROTECTED] http://www.oliver.com/mailman/listinfo/u2-users
