I found at least a very strange thing: If the path to the *.tex file is an UNC-address like
| //SBS2011/path/to/file I'm getting the error. If the path to the *.tex file ist just something like | D:/DATA/filename.tex it works even from Emacs when I run pdfsizeopt.exe. Aha. Hm. So either pdfsizeopt.exe or Emacs have an issue with UNC-Pathes. So this is definitely no AUCTeX issue, as it seems. Am Mittwoch, 2. August 2017, 12:51:38 CEST schrieb AW: > Hi! > > Strange Emacs and maybe AUCTeX issue. > > Emacs-Version: GNU Emacs 25.1.1 (x86_64-w64-mingw32), AUCTeX: 11.91.0. We > are on Windows 8.1. > > I'm using pdfsizeopt to shrink the PDFs produced with AUCTeX. I installed > various times pdfsizeopt according to the manual, see here: > https://github.com/pts/pdfsizeopt > > pdfsizeopt works, but _only_ if called from cmd.exe. Output is this: > > #+BEGIN_SRC > > R:\Bilder\...>pdfsizeopt.exe 170802-XXX.pdf 170802-XXX.pdf > info: This is pdfsizeopt ZIP rUNKNOWN size=114611. > info: prepending to PATH: C:\pdfsizeopt\pdfsizeopt_win32exec > info: loading PDF from: 170802-XXX.pdf > info: loaded PDF of 840447 bytes > info: found 83 obj offsets and 1 obj streams in xref stream > info: separated to 81 objs + xref + trailer > warning: cannot parse obj 19: pdfsizeopt.main.PdfIndirectLengthError: > missing ob > j for indirect /Length 40 0 R at ofs=20930 > ... [many more messages] > info: found 6 Type1 fonts loaded > info: writing Type1CConverter (417100 font bytes) to: > psotmp.3232.conv.tmp.ps info: using Ghostscript > "C:\pdfsizeopt\pdfsizeopt_win32exec\pdfsizeopt_gswin \gsw > in32c.exe": GPL Ghostscript 9.02 (2011-03-30) > info: executing Type1CConverter with Ghostscript: "C:\pdfsizeopt > \pdfsizeopt_win3 > 2exec\pdfsizeopt_gswin\gswin32c.exe" -q -P- -dNOPAUSE -dBATCH - > sDEVICE=pdfwrite > -dPDFSETTINGS=/printer -dColorConversionStrategy=/LeaveColorUnchanged - > sOutputFi > le=psotmp.3232.conv.tmp.pdf -f psotmp.3232.conv.tmp.ps > Type1CConverter: using interpreter GPL Ghostscript 902 20110330 > Type1CConverter: converting font /TBQTVH+LinBiolinum to /Obj0000000070 > Type1CConverter: converting font /DAOZLD+LinBiolinumI to /Obj0000000072 > Type1CConverter: converting font /SEISLV+LinLibertineB to /Obj0000000074 > Type1CConverter: converting font /MNLYJI+LinLibertine to /Obj0000000076 > Type1CConverter: converting font /WBBRJQ+LinLibertineI to /Obj0000000078 > Type1CConverter: converting font /MKHCTZ+LMMathSymbols8-Regular to / > Obj000000008 > 0 > Type1CConverter: all OK > info: loading PDF from: psotmp.3232.conv.tmp.pdf > info: loaded PDF of 28886 bytes > info: separated to 28 objs + xref + trailer > info: found 6 fonts in GS output > info: optimized total Type1 font size 417168 to Type1C font size 20657 (5%) > info: optimized Type1 font XObject 70,69: new size=7149 (9%) > info: optimized Type1 font XObject 72,71: new size=3093 (5%) > info: optimized Type1 font XObject 74,73: new size=2665 (3%) > info: optimized Type1 font XObject 76,75: new size=7473 (7%) > info: optimized Type1 font XObject 78,77: new size=1646 (2%) > info: optimized Type1 font XObject 80,79: new size=658 (25%) > info: found 6 Type1C fonts loaded > info: writing Type1CParser (20533 font bytes) to: psotmp. > 3232.conv.parse.tmp.ps > info: executing Type1CParser with Ghostscript: "C:\pdfsizeopt > \pdfsizeopt_win32ex > ec\pdfsizeopt_gswin\gswin32c.exe" -q -P- -dNOPAUSE -dBATCH -sDEVICE=nullpage > - sD > ataFile=psotmp.3232.conv.parsedata.tmp.ps -f psotmp.3232.conv.parse.tmp.ps > Type1CParser: using interpreter GPL Ghostscript 902 20110330 > Type1CParser: all OK > info: parsed 6 Type1C fonts > info: writing Type1CGenerator (58243 bytes in 6 fonts) to: psotmp. > 3232.conv.gen. > tmp.ps > info: executing Type1CGenerator with Ghostscript: "C:\pdfsizeopt > \pdfsizeopt_win3 > 2exec\pdfsizeopt_gswin\gswin32c.exe" -q -P- -dNOPAUSE -dBATCH - > sDEVICE=pdfwrite > -dPDFSETTINGS=/printer -dColorConversionStrategy=/LeaveColorUnchanged - > sOutputFi > le=psotmp.3232.conv.gen.tmp.pdf -f psotmp.3232.conv.gen.tmp.ps > Type1CGenerator: using interpreter GPL Ghostscript 902 20110330 > Type1CGenerator: all OK > info: loading PDF from: psotmp.3232.conv.gen.tmp.pdf > info: loaded PDF of 28079 bytes > info: separated to 34 objs + xref + trailer > info: found 6 fonts loaded > info: optimized Type1C fonts to form 21874 bytes to 21862 bytes (100%) > info: eliminated 23 unused objs in 23 classes > warning: obj 19 missing, referenced by objs [16]... > warning: obj 20 missing, referenced by objs [16]... > ...[many more messages] > info: compressed 2 streams, kept 2 of them uncompressed > info: saving PDF with 37 objs to: 170802-XXX.pdf > info: generated object stream of 2287 bytes in 27 objects (17%) > info: generated 38966 bytes (5%) > > R:\Bilder\... > > #+END_SRC > > But it does not work, if I call it from Emacs! > > Command: > #+BEGIN_SRC > pdfsizeopt.exe "170802-XXX".pdf "170802-XXX".pdf > #+END_SRC > > Output: > #+BEGIN_SRC > Running `pdfsizeopt' on `170802-XXX' with ``pdfsizeopt.exe "170802-XXX".pdf > "170802-XXX".pdf'' > info: This is pdfsizeopt ZIP rUNKNOWN size=114611. > info: prepending to PATH: C:\pdfsizeopt\pdfsizeopt_win32exec > info: loading PDF from: 170802-XXX.pdf > info: loaded PDF of 38966 bytes > info: found 39 obj offsets and 1 obj streams in xref stream > info: separated to 37 objs + xref + trailer > info: found 0 Type1 fonts loaded > info: found 6 Type1C fonts loaded > info: writing Type1CParser (20521 font bytes) to: psotmp. > 5260.conv.parse.tmp.ps > Traceback (most recent call last): > File "C:\pdfsizeopt\pdfsizeopt_win32exec\python26.zip\runpy.py", line 122, > in _run_module_as_main > File "C:\pdfsizeopt\pdfsizeopt_win32exec\python26.zip\runpy.py", line 34, > in _run_code > File "C:\pdfsizeopt\pdfsizeopt.single\__main__.py", line 1, in <module> > File "C:\pdfsizeopt\pdfsizeopt.single\mainrun.py", line 10, in <module> > File "C:\pdfsizeopt\pdfsizeopt.single\pdfsizeopt\main.py", line 8919, in > main > File "C:\pdfsizeopt\pdfsizeopt.single\pdfsizeopt\main.py", line 6438, in > OptimizeType1CFonts > File "C:\pdfsizeopt\pdfsizeopt.single\pdfsizeopt\main.py", line 6084, in > _ProcessType1CFonts > File "C:\pdfsizeopt\pdfsizeopt.single\pdfsizeopt\main.py", line 5571, in > ParseType1CFonts > File "C:\pdfsizeopt\pdfsizeopt.single\pdfsizeopt\main.py", line 140, in > GetGsCommand > AssertionError: Could not find a working Ghostscript. > > TeX Output exited abnormally with code 1 at Wed Aug 2 12:35:31 > > #+END_SRC > > So why on Earth does pdfsizeopt within Emacs not find "a working > ghostscript", while called from cmd.exe inside any folder of the network > here runs smoothly? > > Please note that I asked this question at stackexchange whithout answer and > on the website of pdfsizeopt. So if you can rule out that it is an AUCTeX > issue, my last resort would be the Emacs-Developer mailing list... > > Thanks for any help! > > Kind Regards, > > Alexander > > > _______________________________________________ > auctex mailing list > [email protected] > https://lists.gnu.org/mailman/listinfo/auctex _______________________________________________ auctex mailing list [email protected] https://lists.gnu.org/mailman/listinfo/auctex
