Nancy, Take a look, also, at a quick summary of debugging commands I compiled for GTM here:
http://openforum.worldvista.org/~forum/index.php?title=Debugging_with_GT.M It doesn't answer your qestion about dumping the stack (I can't remember how to do that either right now), but it gives other helpful tips. Kevin --- Nancy Anthracite <[EMAIL PROTECTED]> wrote: > Frankly, I am very glad I asked this question > because it is revealing of more > than just my ignorance. > > How do you dump the stack? And the error trap your > refer to is the one seen in > D ^XTER? > > On Wednesday 25 May 2005 10:23 am, Greg Woodhouse > wrote: > > FWIW, I think you (Nancy) did a pretty good job of > picking up on the > > relevant details. It might be useful to think of > it this way: Kernel > > keeps track of session specific state (such as > characteristics of the > > current output device) using various standard > variables in the process > > symbol table. Particularly "famous" variables > include DUZ (who's signed > > in) and the IO* variables. In this case, the issue > with IOSL was > > secondary (having to do with checking to see if > the error message will > > fit on the current screen/page). A good technique > (if the error is > > logged in the error trap) is to look at the stack > dump, so that you can > > trace your way back to the ultimate source of the > trouble. > > > > --- [EMAIL PROTECTED] wrote: > > > > I would like to know how to chase this down > for future reference. > > > > > > Am I on the > > > > > > > right track and now what? How do I find out > more about IOSL? > > > > > > > > This is what I did so far: > > > > > > > > I found the 32nd line after the tag WR in the > routine DDSU.m. the > > > > > > line with > > > > > > > the WR tag is 126. The lines below are > 149-159 and there were no > > > > > > line wraps > > > > > > > above to screw things up that I could see. It > seems that WR+32 is > > > > > > probably > > > > > > > line 157, so WR+32 must count the line with WR > as 1, correct? > > > > > > Nancy, WR+32 is the line 32 lines later than WR. > (WR counts as WR+0) > > > to see the line directly, you can just type: > > > MUMPS> WRITE $TEXT(WR+32^DDSU),! > > > . X DDSID > > > > > > which explains the headache, since it is an > Xecute command on the > > > local > > > variable DDSID (which gets changed at run-time). > As you can see in > > > your > > > code fragment below, the variable IOSL > (Input-Output-Screen-Length) > > > is > > > only referenced on line WR+31 if the variable > DDS is defined, either > > > with > > > sub-nodes, or a value itself. $D(DDS)= 1 , 10 or > 11 (Self, > > > SubnodesOnly, > > > Self+Subnodes) > > > > > > This kind of problem is hard to debug, since it > depends on the > > > late-bound > > > value of a variable. Usually, the only way to > find out the problem > > > is to > > > use the ZWRITE DDSID command immediately after > you see the error. > > > > > > > W DDH(A0,A4) > > > > I $D(DDH("ID")) D S:$D(DUOUT) DIY=U > > > > . N DDD,DIY,DDSID > > > > . S DDSID=DDH("ID") > > > > . S:$D(DDH("ID",1))#2 DDSID(1)=DDH("ID",1) > > > > . N DDH > > > > . S:$D(DDSID(1))#2 DDH("ID",1)=DDSID(1) K > DDSID(1) > > > > . S Y=A4 > > > > . S:$D(DDS) > DDQ=$S(DY>(IOSL-1):IOSL-1,1:DY)_U_$X > > > > . X DDSID > > > > Q > > > > > > > > > > > > IOSL is not newed and is used early in the > routine,and gives me a > > > > > > value, so I > > > > > > > assume it is the undefined global variable. > This is immediately > > > > > > after I > > > > > > > started up with D P^DI, which I think wipes > the local variables, > > > > > > correct? > > > > > > > GTM>W IOSL > > > > 24 > > > > > > > > I asked for that section in the global, I > think. > > > > > > > > GTM>ZWR ^DD(200.07475,.01,0) > > > > > ^DD(200.07475,.01,0)="CERTIFICATION^MP747.9'^QA(747.9,^0;1^Q" > > > > > > > > Then I looked in the Data Dictionary: > > > > > > > > ****************** > > > > Select OPTION: 8 DATA DICTIONARY UTILITIES > > > > Select DATA DICTIONARY UTILITY OPTION: ^ > > > > Select OPTION: ^ > > > > GTM>D Q^DI > > > > VA FileMan 22.0 > > > > Select OPTION: 8 DATA DICTIONARY UTILITIES > > > > Select DATA DICTIONARY UTILITY OPTION: 1 LIST > FILE ATTRIBUTES > > > > START WITH WHAT FILE: NEW PERSON// > > > > GO TO WHAT FILE: NEW PERSON// > > > > Select SUB-FILE: > > > > Select LISTING FORMAT: STANDARD// > > > > Start with field: FIRST// Certification > > > > Go to field: > > > > DEVICE: TELNET > > > > STANDARD DATA DICTIONARY #200 -- NEW PERSON > FILE > > > > > MAY > > > > > > 25,[EMAIL PROTECTED]:04:05 PAGE > > > > > > > 1 > > > > STORED IN ^VA(200, (1154 ENTRIES) SITE: > CAMP MASTER UCI: > > > > > > VAH,ROU (VERSION > > > > > > > 8.0) > > > > > > > > DATA NAME GLOBAL > DATA > > > > ELEMENT TITLE LOCATION > TYPE > > > > > --------------------------------------------------------------------------- > >---- > > > > > > 200,747.5 CERTIFICATION QAR2;0 > POINTER Multiple > > > > > > #200.07475 > > > > > > > DESCRIPTION: This field > allows you to enter > > > > > > specialties in > > > > > > > which the > practitioner is Board > > > > > > Certified. > > > > > > > 200.07475,.01 CERTIFICATION 0;1 > POINTER ** TO AN > > > > > > UNDEFINED FILE ** > > > > > > > (Multiply > asked) > > > > LAST EDITED: FEB 01, 1991 > > > > HELP-PROMPT: Enter the > specialties in which > > > > > > the > === message truncated === __________________________________ Yahoo! Mail Mobile Take Yahoo! Mail with you! Check email on your mobile phone. http://mobile.yahoo.com/learn/mail ------------------------------------------------------- SF.Net email is sponsored by: GoToMeeting - the easiest way to collaborate online with coworkers and clients while avoiding the high cost of travel and communications. There is no equipment to buy and you can meet as often as you want. Try it free.http://ads.osdn.com/?ad_id=7402&alloc_id=16135&op=click _______________________________________________ Hardhats-members mailing list Hardhats-members@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/hardhats-members
