[Kst] Re: eparse.cpp, memory leak, and lex

Peter Kümmel syntheticpp at gmx.net
Sun Jan 16 16:10:55 CET 2011


On 15.01.2011 21:16, Barth Netterfield wrote:
> Hi,
>
> There is a memory leak in escan.cpp, and I'm looking for some advice:
>
> -It looks like escan.cpp was at some time generated from escan.l
> -It looks like eparse.cpp was at some time generated from eparse.y
> -These cpp files were committed by fidler in commit 726751.
> -changes to the .l and .y files were committed after this by fidler
> (731128), but it is possible the .cpp and .h files were generated from
> these.  Or maybe not.  Most of the changes are much older than this.
> -I don't think that lex/yacc have been part of the build system for a very
> long time, if ever.
> -kuemmel made some changes to the cpp files recently to allow it to build
> under windows.
>
> We are now unable to re-generate the cpp files from the .l/.y files without
> hand editing the output.
>
> 1) Is this OK and so should we just continue to modify the cpp files for bug
> fixes as they come along?  If not... ?
>
> 2) eparse.cpp and escan.cpp are is listed in the cmake files as
> kst_dont_merge.  What does kst_dont_merge do/mean?

'kst_dont_merge' is only active when the option 'kst_merge_files' is used,
then for each library a files is generated which includes all source files
of the library, this speeds up compilation about factor 5.

'kst_dont_merge' prevents the adding of the listed files to the generated
file, all listed files will be compiled separately. Sometimes it is
needed because of too much compiler errors, and it is simpler not to
merge the file.

Peter



>
>
>
>
> _______________________________________________
> Kst mailing list
> Kst at kde.org
> https://mail.kde.org/mailman/listinfo/kst


More information about the Kst mailing list