buggnubg - Bugs: bug #17937, Pythons "match()" gives...

 
 

bug #17937: Pythons "match()" gives inconsistent results in TUI after "analyse move" failure in MG

Submitted by:  Gunnar Nick Bluth <bluthg>
Submitted on:  Sat 07 Oct 2006 03:16:58 PM UTC  
 
Originator Name: Gunnar "Nick" BluthItem Group: Other bug
Category: Command line interfaceRelease: 0.14.3-devel
Privacy: PublicOperating System: Linux
OS Type: NoneProcessor Type: None
Severity: 3 - NormalOpen/Closed: Closed
Assigned to: NonePriority: 5 - Normal
Status: Fixed

Add a New Comment(Rich Markup)
   

You are not logged in

Please log in, so followups can be emailed to you.

 

Thu 03 Apr 2008 11:23:10 AM UTC, comment #5:

Happens with virtually any MG .mat file I feed into it...
zoloft | buy propecia | cialis

Anonymous
Sat 21 Oct 2006 11:25:11 AM UTC, comment #4:

Remark: the same happens when I just call gnubg with "-t". So it's not a compile time issue, but seems to be related to some "if (fX)" clause....

Gunnar Nick Bluth <bluthg>
Mon 16 Oct 2006 10:29:37 PM UTC, comment #3:

Could this be related to ImportMat() calling GTKThaw() calling GTKSetMoveRecord()?
GTKSetMoveRecord() seems to fiddle around with the records a lot. Could be the reason why plLastMove->plNext->p is not set when USE_GTK is inactive?

Gunnar Nick Bluth <bluthg>
Sun 08 Oct 2006 12:15:30 PM UTC, comment #2:

reproduce.py attached (a stripped version of what I use).

"get_move_diff" is the function that causes the trouble.
"newm" is totally different from "m" when the "Cannat analyse move" messages appeared.

Happens with virtually any MG .mat file I feed into it...

Gunnar Nick Bluth <bluthg>
Sun 08 Oct 2006 08:06:46 AM UTC, comment #1:

Could you attach a minimal example of a match and python commands causing the problems.

Christian Anthon <c_anthon>
Project Administrator
Sat 07 Oct 2006 03:16:58 PM UTC, original submission:

(Not sure if it's one bug or two...)

I have written a script to selectively compare move analysis results for different plies.

Everything is fine so far in GUI mode, but in TUI, "analyse
move" gives me "Sorry, cannot analyse move!" quite often (well, always in MG) and as a result, Pythons "match()" returns some crippled match data.

I've tracked the first issue ("Cannot analyse move") down to
plLastMove->plNext->p
being False in analysis.c, but only for Money Game sessions in text mode.
The same .mat's imported into the GUI work as expected, but I planned to run this in batch mode on a headless machine, so X is not really an option...

No idea why the match data is crippled afterwards. The ['game'] list is only 1 or two elements long then...

System:
- Mandriva 2007, gcc version 4.1.1 20060724 (prerelease) (4.1.1-3mdk) (verified on RH AS 4)
- Current CVS checkout

GUI build:
./configure --prefix=/usr --enable-sse --enable-bearoff --with-board3d --with-gtk --with-python --with-readline --with-x --without-sound --with-timecontrol
TUI build:
./configure --prefix=/usr --enable-sse --enable-bearoff --without-board3d --without-gtk --with-python --with-readline --without-x --without-sound --with-timecontrol

Any hint would be strongly appreciated!

Nick

Gunnar Nick Bluth <bluthg>

 

(Note: upload size limit is set to 16384 kB, after insertion of the required escape characters.)

Attach File(s):
   
   
Comment:
   

Attached Files
file #10919:  1003843.mat added by bluthg (7KiB - text/plain - One example of .mat file this crashes upon....)
file #10918:  reproduce.py added by bluthg (9KiB - application/x-python)

 

Depends on the following items: None found

Items that depend on this one: None found

 

Carbon-Copy List
  • -unavailable- added by c_anthon (Updated the item)
  •  

    Do you think this task is very important?
    If so, you can click here to add your encouragement to it.
    This task has 0 encouragements so far.

    Only logged-in users can vote.

     

    Please enter the title of George Orwell's famous dystopian book (it's a date):

     

     

    Follow 5 latest changes.

    Date Changed By Updated Field Previous Value => Replaced By
    Sun 29 Mar 2009 07:29:51 PM UTCc_anthonOpen/ClosedOpen=>Closed
      StatusConfirmed=>Fixed
    Sun 12 Aug 2007 10:22:32 AM UTCc_anthonStatusNone=>Confirmed
    Sun 08 Oct 2006 12:21:18 PM UTCbluthgAttached File-=>Added 1003843.mat, #10919
    Sun 08 Oct 2006 12:15:30 PM UTCbluthgAttached File-=>Added reproduce.py, #10918

    Back to the top


    Powered by Savane 3.1-cleanup1