PR 256281 Editing the FCF ID in summary mode of a location FCF results in app

huangyhg

Editing the FCF ID in summary mode of a location FCF results in application error.
<< Wade Burton  --  08/01/08  12:32:26>>
This was detected by the automated test EditWindowVerifyDataSummaryMode.  I have been able to manually duplicate it, but can be difficult to duplicate.
Using the attached part program, go to the summary mode of the edit window, find FCFLOC1.  Change many of the options for FCFLOC1 in summary mode in the Settings folder, including the ID to FCF100, and then change all the options in summary mode, like graphical analysis, textual analysis, unit type, fit to datums, etc.  Then... go to FCFLOC2 in summary mode in the Settings folder, and change the ID from FCFLOC2 to FCF100.  I get an application error at this point.
I don't get this error if I just immediately go to FCFLOC2 and change the ID, and I can't tell which part of changing FCFLOC1 causes this.  It may be changing the output alignment option, or something else in the summary mode Settings folder, but it is reproducible if I change all the options for FCFLOC1, and then go to change FCFLOC2.
<<END>>

<< Changes made by Neil Kay -- 12/12/08  09:26:06>>
Action:  CMM Group to Yanhua Huang
<<END>>
<< Wade Burton  --  10/20/08  09:52:53>>
I have retested this manually, and I have re-ran all the Edit Window automated tests, including this EditWindowVerifyDataSummaryMode, and I am not seeing this problem anymore.  I do think it was very likely that it was related to the simultaneous evaluation command.  I haven't gone back to a previous version to verify that it was broken, as Don suggested, but I have verified that I'm not getting the problem anymore in the current V4.2 MR2.  So if this passes the code review it can be closed.
<<END>>
<< Don Ruggieri  --  08/22/08  11:24:05>>
I've tried many times, in may variants of PC-DMIS to repeat this, in XP and in Vista, and I cannot repeat the original problem in the pre-fix versions.  I generally want to do this before i declare it "Fixed".  Maybe Wade can perform this procedure?  Very it was broken and then verify that it is fixed?
<<END>>
<< Don Turcotte  --  08/20/08  11:43:11>>
With Tim's authorization I have also merged this into V42R (and V42 beta to keep the beta up to date).
Files inserted to server
------------------------
V42R\DIMENS\SIMULTANEOUS_EVALUATION.CPP
V42\DIMENS\SIMULTANEOUS_EVALUATION.CPP

<<END>>
<< Don Turcotte  --  08/15/08  13:14:14>>
Merged into V431.
Files inserted to server
------------------------
V431\DIMENS\SIMULTANEOUS_EVALUATION.CPP
<<END>>
<< Changes made by Don Turcotte -- 08/15/08  13:14:32>>
Action:  Don Turcotte to CMM Group, Status:  OPEN to REVIEW
<<END>>
<< Changes made by Tim Wernicke -- 08/15/08  10:51:05>>
Action:  CMM Group to Don Turcotte, Status:  REVIEW to OPEN
<<END>>
<< Changes made by Tim Wernicke -- 08/15/08  10:50:19>>
Priority:  Critical to Stop Rel.
<<END>>
<< Don Turcotte  --  08/15/08  11:43:36>>
I have fixed this in V43B and V44B.  
The Simultaneous evaluation command do_math(...) was not establishing the original global_align before calling create_datum_reference_frame(...) on the CPCDdatum_reference object of the CPCDtp_start_location command of the attached FCFs.  The subsequent
((CPCDstart_align *)(datum_ref->m_pAlign))->execute(1,NO_ITER_EXEC);
call causes the global_align to be set to the datum reference frame alignment which is incorrect for the next iteration of the loop.
Files inserted to server
------------------------
V43B\DIMENS\SIMULTANEOUS_EVALUATION.CPP
V44B\DIMENS\SIMULTANEOUS_EVALUATION.CPP
<<END>>
<< Changes made by Don Turcotte -- 08/15/08  11:44:41>>
Action:  Don Turcotte to CMM Group, Status:  OPEN to REVIEW
<<END>>
<< Don Turcotte  --  08/14/08  17:27:38>>
The two FCFs in question are part of a simultaneous evaluation command (FCFLOC1, FCFLOC2) and this is what is causing the problem.  If I delete the simultaneous evaluation command, the crash does not occur.  Now I am trying to figure out why being part of a simultaneous evaluation command is causing this.
<<END>>
<< Don Turcotte  --  08/13/08  17:15:00>>
I am able to reproduce this problem.  Looking for the cause...
<<END>>
<< Changes made by Tim Wernicke -- 08/01/08  13:40:09>>
Priority:  to Critical
<<END>>

huangyhg

m_pPartProgram->GetCurrentSettings()->SetCurrentAlignment(pSaveAlign); // PR256281 - start with the correct global_align