cxref self-references itself during the build of the doc subdirectory. On bento, this results in a core dump of cxref. However, none of the doc target gets installed. So, the easiest fix is to not build the doc target. How-To-Repeat: Build cxref on bento.
Lately Chris Knight told: > >Description: > cxref self-references itself during the build of the doc subdirectory. > On bento, this results in a core dump of cxref. > However, none of the doc target gets installed. > So, the easiest fix is to not build the doc target. this is the completely wrong approach. don't ignore the problem, fix it: new file: files/patch-af; bump portrevision (i'd vote for it)? Index: files/patch-af =================================================================== RCS file: files/patch-af diff -N files/patch-af --- /dev/null 1 Jan 1970 00:00:00 -0000 +++ files/patch-af 14 Jan 2003 15:23:10 -0000 @@ -0,0 +1,16 @@ +--- comment.c.orig Tue Nov 16 20:11:38 1999 ++++ comment.c Tue Jan 14 16:17:17 2003 +@@ -158,8 +158,11 @@ + + if(comment) + { +- malloc_comment=Realloc(malloc_comment,strlen(comment)+1); +- strcpy(malloc_comment,comment); ++ if(malloc_comment != comment) ++ { ++ malloc_comment=Realloc(malloc_comment,strlen(comment)+1); ++ strcpy(malloc_comment,comment); ++ } + current_comment=malloc_comment; + } + else dunno why this was never noticed; mebe because of special malloc() behavior on tested platforms. cheers simon -- /"\ http://corecode.ath.cx/#donate \ / \ ASCII Ribbon Campaign / \ Against HTML Mail and News
Howdy, > -----Original Message----- > From: Simon 'corecode' Schubert [mailto:corecode@corecode.ath.cx] > Sent: Wednesday, 15 January 2003 2:28 > To: Chris Knight > Cc: FreeBSD-gnats-submit@FreeBSD.org > Subject: Re: ports/47039: [BENTO FIX] devel/cxref > > [snip] > > this is the completely wrong approach. don't ignore the problem, fix > it: > new file: files/patch-af; bump portrevision (i'd vote for it)? > Well, I wasn't deliberately ignoring the problem - my 5.0-RC (20020112) system built the package just fine without modification. I made the wrong assumption that it was a bento only issue. > [snip] > > dunno why this was never noticed; mebe because of special malloc() > behavior on tested platforms. > My successful build prior to patching used the default malloc settings. > cheers > simon Regards, Chris Knight Systems Administrator AIMS Independent Computer Professionals Tel: +61 3 6334 6664 Fax: +61 3 6331 7032 Mob: +61 419 528 795 Web: http://www.aims.com.au
State Changed From-To: open->closed Committed, thanks. PORTREVISION bump added.