[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Bug#584350: refcard: FTBFS: xmlroff segfaults on japanese PDF



tag 584350 help
thanks

Hi,

Wolfgang, how did you succeed in building the japanese PDF ?
xmlroff version did not change in sid since your upload.

On Thu, Jun 03, 2010 at 12:10:10PM +0200, Lucas Nussbaum wrote:
> Source: refcard
> Version: 5.0.5-3
> Severity: serious
> Tags: squeeze sid
> User: debian-qa@lists.debian.org
> Usertags: qa-ftbfs-20100602 qa-ftbfs
> Justification: FTBFS on amd64
> 
> During a rebuild of all packages in sid, your package failed to build on
> amd64.
> 
> Relevant part:
[..]
> > xsltproc --nonet --novalid preproc.xsl entries-ja.dbk > refcard-ja.dbk
> > xsltproc --nonet --novalid fo.xsl refcard-ja.dbk > refcard-ja-a4.fo
> > Making portrait pages on USletter paper (99mmx210mm)
> > L=`echo refcard-ja-a4.s.pdf | sed 's/.*-\(.*\)-.*/\1/'`; \
> > 	USE_XR=`echo " ar cs el he hi ja ml sk zh_CN zh_TW " | grep " $L "`; \
> > 	if [ -n "$USE_XR" ]; then \
> > 	    USE_GP=`echo " ar eu he hi ja sv " | grep " $L "`; \
> > 	    if [ -n "$USE_GP" ]; then \
> > 		xmlroff --backend gp -o refcard-ja-a4.s.pdf refcard-ja-a4.fo; \

For japanese, gnomeprint backend is used, but xmlroff segaults.

> > 	    else \
> > 		xmlroff --backend cairo -o refcard-ja-a4.s.pdf refcard-ja-a4.fo; \
> > 	    fi; \

Using cairo backend, japanese refcard can be built (while it failed by
the past, see #492597), but I cannot get correct correct letters in the
generated PDF (while metadata are correctly endoded).

> > 	else \
> > 	    if [ -n "$USE_DBLATEX" ]; then \
> > 		SRC=`echo refcard-ja-a4.fo|sed 's/-\(a4\|lt\).fo/.dbk/'`; \
> > 		FMT=`echo refcard-ja-a4.fo|sed 's/.*-\(a4\|lt\).fo/\1/'`; \
> > 		dblatex --backend=xetex --texstyle=refcard.sty --xsl-user=dblatex.xsl --output=refcard-ja-a4.s.pdf --param=format=$FMT $SRC; \
> > 	    else \
> > 		fop refcard-ja-a4.fo refcard-ja-a4.s.pdf; \
> > 	    fi; \
> > 	fi

No more success with the japanese PDF using fop or dblatex (which means
#492350 is to be reopened ?).
(I will provide soon generated the several test cases).

> > (xmlroff:24762): libfo-WARNING **: fo-area-page-error: Child area overflows parent area
> > Object path: /FoAreaTree[1]/FoAreaPage[4]/FoAreaViewportReference[1]
> > 
> > (xmlroff:24762): libfo-WARNING **: fo-area-page-error: Child area overflows parent area
> > Object path: /FoAreaTree[1]/FoAreaPage[4]/FoAreaViewportReference[1]
> > 
> > (xmlroff:24762): libfo-WARNING **: fo-area-page-error: Child area overflows parent area
> > Object path: /FoAreaTree[1]/FoAreaPage[4]/FoAreaViewportReference[1]
> > 
> > ** (xmlroff:24762): WARNING **: IPP request failed with status 1280
> > 
> > ** (xmlroff:24762): WARNING **: IPP request failed with status 1280
> > 
> > (xmlroff:24762): GnomePrint-CRITICAL **: gnome_rfont_get_glyph_stdadvance: assertion `glyph < GRF_NUM_GLYPHS (rfont)' failed
[..]
> > (xmlroff:24762): GnomePrint-CRITICAL **: gnome_rfont_get_glyph_stdadvance: assertion `glyph < GRF_NUM_GLYPHS (rfont)' failed
> > Segmentation fault
> > make[1]: *** [refcard-ja-a4.s.pdf] Error 139
> 
> The full build log is available from:
>    http://people.debian.org/~lucas/logs/2010/06/02/refcard_5.0.5-3_lsid64.buildlog
 
For the moment, I see no way of fixing other than disabling japanese
build.

-- 
Simon Paillard



Reply to: