poppler issueshttps://gitlab.freedesktop.org/poppler/poppler/-/issues2018-08-21T11:07:20Zhttps://gitlab.freedesktop.org/poppler/poppler/-/issues/520pdftohtml: RTL text generated backwards2018-08-21T11:07:20ZBugzilla Migration Userpdftohtml: RTL text generated backwards## Submitted by Nezmer
Assigned to **poppler-bugs**
**[Link to original bug (#28076)](https://bugs.freedesktop.org/show_bug.cgi?id=28076)**
## Description
"pdftohtml" seems to generate RTL text backwards. It's like (abc) is genera...## Submitted by Nezmer
Assigned to **poppler-bugs**
**[Link to original bug (#28076)](https://bugs.freedesktop.org/show_bug.cgi?id=28076)**
## Description
"pdftohtml" seems to generate RTL text backwards. It's like (abc) is generated (cba). You can read the generated text from LTR but that's not convenient ;)
"pdftotext" is behaving correctly.https://gitlab.freedesktop.org/poppler/poppler/-/issues/519Unexpected end of file in flate stream2018-10-11T08:22:38ZBugzilla Migration UserUnexpected end of file in flate stream## Submitted by cla..@..eat.dk
Assigned to **poppler-bugs**
**[Link to original bug (#104454)](https://bugs.freedesktop.org/show_bug.cgi?id=104454)**
## Description
Created attachment 136490
test pdf
pdfimages -list test.pdf
*...## Submitted by cla..@..eat.dk
Assigned to **poppler-bugs**
**[Link to original bug (#104454)](https://bugs.freedesktop.org/show_bug.cgi?id=104454)**
## Description
Created attachment 136490
test pdf
pdfimages -list test.pdf
**Attachment 136490**, "test pdf":
[Microsoft_Outlook_-_Notatformat.pdf](/uploads/5aee2257d3bb5e41257e0f9731daf636/Microsoft_Outlook_-_Notatformat.pdf)https://gitlab.freedesktop.org/poppler/poppler/-/issues/518Quality of convertion pdf to html2018-08-21T11:07:12ZBugzilla Migration UserQuality of convertion pdf to html## Submitted by isaric
Assigned to **poppler-bugs**
**[Link to original bug (#43297)](https://bugs.freedesktop.org/show_bug.cgi?id=43297)**
## Description
I am French, sorry for my translation.
I use poppler-utils 0.16.7-2ubuntu2 ...## Submitted by isaric
Assigned to **poppler-bugs**
**[Link to original bug (#43297)](https://bugs.freedesktop.org/show_bug.cgi?id=43297)**
## Description
I am French, sorry for my translation.
I use poppler-utils 0.16.7-2ubuntu2 (oneiric)
I use "pdftohtml -c *.pdf"
for exemple isaric.cof.free.fr/PDFtoHTML/5-1d-a-afk-1-15-18-55-62.pdf
give http://isaric.cof.free.fr/PDFtoHTML/5-1d-a-afk_ind.html with shifts on the level of spaces
The columns don't use the same ones spaces.
in advance thankshttps://gitlab.freedesktop.org/poppler/poppler/-/issues/517Poppler does not draw caret annotations if they don't have an /AP dictionary2018-08-21T11:07:07ZBugzilla Migration UserPoppler does not draw caret annotations if they don't have an /AP dictionary## Submitted by Germán Poo-Caamaño
Assigned to **poppler-bugs**
**[Link to original bug (#74028)](https://bugs.freedesktop.org/show_bug.cgi?id=74028)**
## Description
Created attachment 92736
PDF test case with caret annotations
...## Submitted by Germán Poo-Caamaño
Assigned to **poppler-bugs**
**[Link to original bug (#74028)](https://bugs.freedesktop.org/show_bug.cgi?id=74028)**
## Description
Created attachment 92736
PDF test case with caret annotations
This is part of a report in https://bugzilla.gnome.org/show_bug.cgi?id=626982
In the attached page, there are plenty of caret annotations, which are
rendered in Acroread with the symbol up tack ⊥ (u+22a5) or short up tack
(u+2ae0). However, in poppler there is nothing rendered.
**Attachment 92736**, "PDF test case with caret annotations":
[page-36.pdf](/uploads/cb75de8a25ee952d2281bc49d8e4dbc3/page-36.pdf)
### See also
* [Bug 626982](https://bugzilla.gnome.org/show_bug.cgi?id=626982)https://gitlab.freedesktop.org/poppler/poppler/-/issues/515Disable bilinear filtering of images at native resolution -- Patch supplied2018-08-21T11:05:46ZBugzilla Migration UserDisable bilinear filtering of images at native resolution -- Patch supplied## Submitted by Charles Hyder
Assigned to **poppler-bugs**
**[Link to original bug (#68360)](https://bugs.freedesktop.org/show_bug.cgi?id=68360)**
## Description
Here's a PDF file that contains scanned (raster) b/w image @ 400dpi,...## Submitted by Charles Hyder
Assigned to **poppler-bugs**
**[Link to original bug (#68360)](https://bugs.freedesktop.org/show_bug.cgi?id=68360)**
## Description
Here's a PDF file that contains scanned (raster) b/w image @ 400dpi, together with the result of its rendering with pdftoppm from two poppler releases: 0.20.5 & 0.24.0:
http://ge.tt/56fxgSp/v/0?c
The latter (newer version) produces a fuzzy image. In fact, I've also tried version 0.22.x and it also gives fuzzy results. The effect is not specific to the PDF file: it is reproduced on any PDF file that contains raster b/w images.
Meanwhile, pdfimages (ImageOutputDev) works just fine. So I figure it must be the SplashOutputDevhttps://gitlab.freedesktop.org/poppler/poppler/-/issues/514Vector graphics in pdf file are displayed with wrong linewidth2018-08-21T11:05:37ZBugzilla Migration UserVector graphics in pdf file are displayed with wrong linewidth## Submitted by Germán Poo-Caamaño
Assigned to **poppler-bugs**
**[Link to original bug (#97995)](https://bugs.freedesktop.org/show_bug.cgi?id=97995)**
## Description
As reported in https://bugzilla.gnome.org/show_bug.cgi?id=74525...## Submitted by Germán Poo-Caamaño
Assigned to **poppler-bugs**
**[Link to original bug (#97995)](https://bugs.freedesktop.org/show_bug.cgi?id=97995)**
## Description
As reported in https://bugzilla.gnome.org/show_bug.cgi?id=745258
"You can open these two files with Evince
https://sites.google.com/site/espinozahg/notes/linux/linewidth-0.1.pdf
https://sites.google.com/site/espinozahg/notes/linux/linewidth-0.3.pdf
Evince displays both images very similarly and with a quite thick linewidth.
Okular displays them much better and you can notice the difference in linewidth.
Printing to paper works correctly in Evince."https://gitlab.freedesktop.org/poppler/poppler/-/issues/513pdfimages - can't extract images2018-10-11T09:00:00ZBugzilla Migration Userpdfimages - can't extract images## Submitted by cla..@..eat.dk
Assigned to **poppler-bugs**
**[Link to original bug (#104134)](https://bugs.freedesktop.org/show_bug.cgi?id=104134)**
## Description
In the following two pdf files pdfimages can't extract images
pd...## Submitted by cla..@..eat.dk
Assigned to **poppler-bugs**
**[Link to original bug (#104134)](https://bugs.freedesktop.org/show_bug.cgi?id=104134)**
## Description
In the following two pdf files pdfimages can't extract images
pdfimages -png file.pdf /tmphttps://gitlab.freedesktop.org/poppler/poppler/-/issues/512Bounding boxes of text selection marks are much too large in some cases.2018-08-21T11:05:30ZBugzilla Migration UserBounding boxes of text selection marks are much too large in some cases.## Submitted by tho..@..ner.at
Assigned to **poppler-bugs**
**[Link to original bug (#40556)](https://bugs.freedesktop.org/show_bug.cgi?id=40556)**
## Description
Created attachment 50817
This is the sample pdf illustrating the bu...## Submitted by tho..@..ner.at
Assigned to **poppler-bugs**
**[Link to original bug (#40556)](https://bugs.freedesktop.org/show_bug.cgi?id=40556)**
## Description
Created attachment 50817
This is the sample pdf illustrating the bug
When I select text on the first page of the sample document, everything is ok and looks as expected. But text on the second page has way too large bounding boxes associated for the selection marks when selecting it (tested in evince and okular). In acroread, selection looks as expected.
I created the document with LaTeX by including the second page via \includepdf.
**Attachment 50817**, "This is the sample pdf illustrating the bug":
[test.pdf](/uploads/20442195bcb7a58ed15e8ff6d2cbf02c/test.pdf)https://gitlab.freedesktop.org/poppler/poppler/-/issues/511Huge simple PDF displayed blank in poppler-glib-demo2018-08-21T11:05:17ZBugzilla Migration UserHuge simple PDF displayed blank in poppler-glib-demo## Submitted by Maxim Iorsh
Assigned to **poppler-bugs**
**[Link to original bug (#56858)](https://bugs.freedesktop.org/show_bug.cgi?id=56858)**
## Description
A very large PDF produced by a Xerox wide scanner is displayed blank.
...## Submitted by Maxim Iorsh
Assigned to **poppler-bugs**
**[Link to original bug (#56858)](https://bugs.freedesktop.org/show_bug.cgi?id=56858)**
## Description
A very large PDF produced by a Xerox wide scanner is displayed blank.
File location: https://docs.google.com/file/d/14hFFrjSSbiEcfML1sgOttcT865GC7tHqv27wiuEfQ-KlvuRfU67Dkj9E9JaM/edit
Basically it contains nothing but a 21590 x 161385 b/w bitmap
Note that Acrobat Reader fails to display it too. Okular displays properly, but slowly. PDF-XChange Viewer for Windows displays properly and very fast.https://gitlab.freedesktop.org/poppler/poppler/-/issues/510pdfimages 0.62 extract image at low resolution than embedded in PDF2018-10-11T20:17:54ZBugzilla Migration Userpdfimages 0.62 extract image at low resolution than embedded in PDF## Submitted by Valerio Messina
Assigned to **poppler-bugs**
**[Link to original bug (#104684)](https://bugs.freedesktop.org/show_bug.cgi?id=104684)**
## Description
Created attachment 136828
sample PDF with 4 pages
using pdfimag...## Submitted by Valerio Messina
Assigned to **poppler-bugs**
**[Link to original bug (#104684)](https://bugs.freedesktop.org/show_bug.cgi?id=104684)**
## Description
Created attachment 136828
sample PDF with 4 pages
using pdfimages and extracting the images from the attached 4 pages PDF, generate tens of small useless files and 4 real images, but also those images really at very low resolution, so text is unreadable.
$ pdfimages -all FPGA_CQFP352adapter_Aldec_orig.pdf FPGA_CQFP352adapter_Aldec
platform:
Linux64 and Win64
**Attachment 136828**, "sample PDF with 4 pages":
[FPGA_CQFP352adapter_Aldec_orig.pdf](/uploads/e864ea34e5141e468188394fe7866310/FPGA_CQFP352adapter_Aldec_orig.pdf)https://gitlab.freedesktop.org/poppler/poppler/-/issues/509recursive xref loop2021-11-05T10:59:56ZBugzilla Migration Userrecursive xref loop## Submitted by pdknsk
Assigned to **poppler-bugs**
**[Link to original bug (#106112)](https://bugs.freedesktop.org/show_bug.cgi?id=106112)**
## Description
This hit a timeout.
#20 0x67474d in XRef::fetch(int, int, int) poppl...## Submitted by pdknsk
Assigned to **poppler-bugs**
**[Link to original bug (#106112)](https://bugs.freedesktop.org/show_bug.cgi?id=106112)**
## Description
This hit a timeout.
#20 0x67474d in XRef::fetch(int, int, int) poppler/poppler/XRef.cc:1175:26
#21 0x66d9ac in Object::fetch(XRef*, int) const poppler/poppler/Object.cc:125:16
#22 0x67006a in Dict::lookup(char const*, int) const poppler/poppler/Dict.cc:260:35
#23 0x686193 in Stream::addFilters(Dict*, int) poppler/poppler/Stream.cc:182:17
#24 0x6d581c in Parser::makeStream(Object&&, unsigned char*, CryptAlgorithm, int, int, int, int, bool) poppler/poppler/Parser.cc:272:14
#25 0x6d4978 in Parser::getObj(bool, unsigned char*, CryptAlgorithm, int, int, int, int, bool) poppler/poppler/Parser.cc:138:34
#26 0x67474d in XRef::fetch(int, int, int) poppler/poppler/XRef.cc:1175:26
#27 0x66d9ac in Object::fetch(XRef*, int) const poppler/poppler/Object.cc:125:16
#28 0x67006a in Dict::lookup(char const*, int) const poppler/poppler/Dict.cc:260:35
#29 0x686193 in Stream::addFilters(Dict*, int) poppler/poppler/Stream.cc:182:17
#30 0x6d581c in Parser::makeStream(Object&&, unsigned char*, CryptAlgorithm, int, int, int, int, bool) poppler/poppler/Parser.cc:272:14
#31 0x6d4978 in Parser::getObj(bool, unsigned char*, CryptAlgorithm, int, int, int, int, bool) poppler/poppler/Parser.cc:138:34
#32 0x6d4522 in Parser::getObj(bool, unsigned char*, CryptAlgorithm, int, int, int, int, bool) poppler/poppler/Parser.cc:95:21
#33 0x6d467b in Parser::getObj(bool, unsigned char*, CryptAlgorithm, int, int, int, int, bool) poppler/poppler/Parser.cc:123:16
#34 0x67474d in XRef::fetch(int, int, int) poppler/poppler/XRef.cc:1175:26
I expected a stack-overflow, but even after manually running for a few minutes it didn't occur.https://gitlab.freedesktop.org/poppler/poppler/-/issues/508Differing number of items returned from get_text{,layout}2018-08-21T11:05:01ZBugzilla Migration UserDiffering number of items returned from get_text{,layout}## Submitted by Peter Waller
Assigned to **poppler-bugs**
**[Link to original bug (#73885)](https://bugs.freedesktop.org/show_bug.cgi?id=73885)**
## Description
Created attachment 92530
Single page PDF document describing the issu...## Submitted by Peter Waller
Assigned to **poppler-bugs**
**[Link to original bug (#73885)](https://bugs.freedesktop.org/show_bug.cgi?id=73885)**
## Description
Created attachment 92530
Single page PDF document describing the issue
As reported here, I've found a PDF which returns a differing number of rectangles from poppler_get_page_layout as from poppler_get_text. I've reproduced this with the glib demo app on the master branch at c8a845cf7c7752d3b7dad06013d3154812c66c92.
http://lists.freedesktop.org/archives/poppler/2014-January/010793.html
I've attached a page which reproduces the issue.
**Attachment 92530**, "Single page PDF document describing the issue":
[2014-01-17-broken.pdf](/uploads/6be3c7dbe5cf6cf93596d10c33bca27e/2014-01-17-broken.pdf)https://gitlab.freedesktop.org/poppler/poppler/-/issues/506Pdftoppm Fontconfig Error Causes Compressed Text2018-10-11T20:19:14ZBugzilla Migration UserPdftoppm Fontconfig Error Causes Compressed Text## Submitted by Cory
Assigned to **poppler-bugs**
**[Link to original bug (#78626)](https://bugs.freedesktop.org/show_bug.cgi?id=78626)**
## Description
Created attachment 98938
rendered text
On Ubuntu and OS X, using Poppler 0.2...## Submitted by Cory
Assigned to **poppler-bugs**
**[Link to original bug (#78626)](https://bugs.freedesktop.org/show_bug.cgi?id=78626)**
## Description
Created attachment 98938
rendered text
On Ubuntu and OS X, using Poppler 0.26.0
The pdf text displays properly with Acrobat, but the following error happens when I use pdftoppm and then the text is really hard to read.
Fontconfig error: "/usr/local/etc/fonts/conf.d/10-scale-bitmap-fonts.conf", line 70: non-double matrix element
Fontconfig error: "/usr/local/etc/fonts/conf.d/10-scale-bitmap-fonts.conf", line 70: non-double matrix element
Fontconfig warning: "/usr/local/etc/fonts/conf.d/10-scale-bitmap-fonts.conf", line 78: saw unknown, expected number
My command is:
pdftoppm some_text_compresses.pdf -r 150 /tmp/rastered
Attached is a screenshot of the result.
The text renders properly with ghostscript.
The PDF is sensitive, so I will need to e-mail it to a developer directly.
Thank you so much for all of your help!
**Attachment 98938**, "rendered text":
![rendered_text](/uploads/24df6b15a8da9f151c642106e4190ec2/rendered_text.png)https://gitlab.freedesktop.org/poppler/poppler/-/issues/505Evince garbles text selection2018-08-21T11:04:44ZBugzilla Migration UserEvince garbles text selection## Submitted by Jason Crain
Assigned to **poppler-bugs**
**[Link to original bug (#94902)](https://bugs.freedesktop.org/show_bug.cgi?id=94902)**
## Description
Created attachment 122875
evince-selection-bug.pdf
Forwarding from ht...## Submitted by Jason Crain
Assigned to **poppler-bugs**
**[Link to original bug (#94902)](https://bugs.freedesktop.org/show_bug.cgi?id=94902)**
## Description
Created attachment 122875
evince-selection-bug.pdf
Forwarding from https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=817239
----------
Sascha Brawer wrote:
Package: evince
Version: 3.14.1-2
Severity: normal
When selecting the text of the attched PDF in evince 3.14.1,
some characters appear garbled. The _unselected_ rendering is fine,
the garbling only happens upon text selection.
----------
I can confirm this on current cairo and evince master. In the attached PDF, selecting certain characters makes them disappear or turn into squares.
**Attachment 122875**, "evince-selection-bug.pdf":
[evince-selection-bug.pdf](/uploads/35d8bd9597b3da0092504bc20e928a1e/evince-selection-bug.pdf)https://gitlab.freedesktop.org/poppler/poppler/-/issues/504endless loop resulting OOM2018-10-11T20:33:08ZBugzilla Migration Userendless loop resulting OOM## Submitted by Hui Peng
Assigned to **poppler-bugs**
**[Link to original bug (#104798)](https://bugs.freedesktop.org/show_bug.cgi?id=104798)**
## Description
Created attachment 136967
a tar.gz file containing the testcase
when u...## Submitted by Hui Peng
Assigned to **poppler-bugs**
**[Link to original bug (#104798)](https://bugs.freedesktop.org/show_bug.cgi?id=104798)**
## Description
Created attachment 136967
a tar.gz file containing the testcase
when using tools like pdftohtml, pdftoppm, pdftops, pdftotext
on the uploaded testcases, the parser gets stuck in endless loop
resulting OOM.
This is the stacktrace of pdftohtml:
```
#0 sysmalloc (nb=nb@entry=0x8590, av=0x7ffff7792c20 <main_arena>) at malloc.c:2768
#1 0x00007ffff7444645 in _int_malloc (av=av@entry=0x7ffff7792c20 <main_arena>, bytes=bytes@entry=0x8580) at malloc.c:4135
#2 0x00007ffff7446f3e in __GI___libc_malloc (bytes=0x8580) at malloc.c:3086
#3 0x00007ffff7828458 in operator new(unsigned long) () from /usr/lib/x86_64-linux-gnu/libstdc++.so.6
#4 0x00007ffff7ca55c9 in Stream::makeFilter (this=this@entry=0x555576634210, name=<optimized out>, str=str@entry=0x555576634210, params=params@entry=0x7fffffffc2e0, recursion=recursion@entry=0x3, dict=dict@entry=0x0) at /home/huip/tmp/tfuzz_eval/poppler-0.62.0/poppler/Stream.cc:324
#5 0x00007ffff7ca5ccd in Stream::addFilters (this=this@entry=0x555576634210, dict=<optimized out>, recursion=recursion@entry=0x3) at /home/huip/tmp/tfuzz_eval/poppler-0.62.0/poppler/Stream.cc:198
#6 0x00007ffff7c95688 in Parser::makeStream(Object&&, unsigned char*, CryptAlgorithm, int, int, int, int, bool) (this=this@entry=0x5555555ccb30, dict=dict@entry=<unknown type in /home/huip/tmp/tfuzz_eval/poppler-0.62.0/build/libpoppler.so.73, CU 0x22f494, DIE 0x2330c7>, fileKey=fileKey@entry=0x0, encAlgorithm=encAlgorithm@entry=cryptNone, keyLength=keyLength@entry=0x30cb, objNum=objNum@entry=0x4, objGen=0x0, recursion=0x3, strict=0x0) at /home/huip/tmp/tfuzz_eval/poppler-0.62.0/poppler/Parser.cc:269
#7 0x00007ffff7c95e95 in Parser::getObj (this=this@entry=0x5555555ccb30, simpleOnly=simpleOnly@entry=0x0, fileKey=fileKey@entry=0x0, encAlgorithm=encAlgorithm@entry=cryptNone, keyLength=keyLength@entry=0x30cb, objNum=0x4, objGen=0x0, recursion=0x2, strict=0x0) at /home/huip/tmp/tfuzz_eval/poppler-0.62.0/poppler/Parser.cc:135
#8 0x00007ffff7c95ba8 in Parser::getObj (this=this@entry=0x5555555ccb30, simpleOnly=simpleOnly@entry=0x0, fileKey=fileKey@entry=0x0, encAlgorithm=encAlgorithm@entry=cryptNone, keyLength=keyLength@entry=0x30cb, objNum=0x4, objGen=0x0, recursion=0x1, strict=0x0) at /home/huip/tmp/tfuzz_eval/poppler-0.62.0/poppler/Parser.cc:93
#9 0x00007ffff7c95cf2 in Parser::getObj (this=this@entry=0x5555555ccb30, simpleOnly=simpleOnly@entry=0x0, fileKey=0x0, encAlgorithm=cryptNone, keyLength=0x30cb, objNum=0x4, objGen=0x0, recursion=0x0, strict=0x0) at /home/huip/tmp/tfuzz_eval/poppler-0.62.0/poppler/Parser.cc:120
#10 0x00007ffff7cb1dc6 in XRef::fetch (this=0x5555555ccd30, num=<optimized out>, gen=0x0, recursion=recursion@entry=0x0) at /home/huip/tmp/tfuzz_eval/poppler-0.62.0/poppler/XRef.cc:1171
#11 0x00007ffff7c8ead6 in Object::fetch (this=this@entry=0x5555555d1838, xref=<optimized out>, recursion=recursion@entry=0x0) at /home/huip/tmp/tfuzz_eval/poppler-0.62.0/poppler/Object.cc:125
#12 0x00007ffff7c290aa in Dict::lookup (this=this@entry=0x5555555d1700, key=key@entry=0x7ffff7d11493 "FontDescriptor", recursion=recursion@entry=0x0) at /home/huip/tmp/tfuzz_eval/poppler-0.62.0/poppler/Dict.cc:260
#13 0x00007ffff7c532f4 in GfxFont::getFontType (xref=xref@entry=0x5555555ccd30, fontDict=fontDict@entry=0x5555555d1700, embID=embID@entry=0x7fffffffc8b8) at /home/huip/tmp/tfuzz_eval/poppler-0.62.0/poppler/GfxFont.cc:343
#14 0x00007ffff7c58f5e in GfxFont::makeFont (xref=xref@entry=0x5555555ccd30, tagA=0x5555555d1520 "F1", idA=idA@entry=..., fontDict=fontDict@entry=0x5555555d1700) at /home/huip/tmp/tfuzz_eval/poppler-0.62.0/poppler/GfxFont.cc:222
#15 0x00007ffff7c5917f in GfxFontDict::GfxFontDict (this=0x5555555d1580, xref=0x5555555ccd30, fontDictRef=0x0, fontDict=0x5555555d14c0) at /home/huip/tmp/tfuzz_eval/poppler-0.62.0/poppler/GfxFont.cc:2457
#16 0x00007ffff7c3c09b in GfxResources::GfxResources (this=0x5555555cd240, xref=0x5555555ccd30, resDictA=<optimized out>, nextA=0x0) at /home/huip/tmp/tfuzz_eval/poppler-0.62.0/poppler/Gfx.cc:338
#17 0x00007ffff7c47efb in Gfx::Gfx (this=0x5555555d12f0, docA=<optimized out>, outA=0x5555555cd4b0, pageNum=0x1, resDict=0x5555555ce1a0, hDPI=108, vDPI=108, box=0x7fffffffcb50, cropBox=0x0, rotate=0x0, abortCheckCbkA=0x0, abortCheckCbkDataA=0x0, xrefA=0x5555555ccd30) at /home/huip/tmp/tfuzz_eval/poppler-0.62.0/poppler/Gfx.cc:541
#18 0x00007ffff7c941a6 in Page::createGfx (this=this@entry=0x5555555d1220, out=out@entry=0x5555555cd4b0, hDPI=hDPI@entry=108, vDPI=vDPI@entry=108, rotate=rotate@entry=0x0, useMediaBox=useMediaBox@entry=0x1, crop=<optimized out>, crop@entry=0x0, sliceX=sliceX@entry=0xffffffff, sliceY=0xffffffff, sliceW=0xffffffff, sliceH=0xffffffff, printing=0x0, abortCheckCbk=0x0, abortCheckCbkData=0x0, xrefA=0x5555555ccd30) at /home/huip/tmp/tfuzz_eval/poppler-0.62.0/poppler/Page.cc:521
#19 0x00007ffff7c9443a in Page::displaySlice (this=0x5555555d1220, out=0x5555555cd4b0, hDPI=108, vDPI=108, rotate=0x0, useMediaBox=0x1, crop=0x0, sliceX=sliceX@entry=0xffffffff, sliceY=0xffffffff, sliceW=0xffffffff, sliceH=0xffffffff, printing=0x0, abortCheckCbk=0x0, abortCheckCbkData=0x0, annotDisplayDecideCbk=0x0, annotDisplayDecideCbkData=0x0, copyXRef=0x0) at /home/huip/tmp/tfuzz_eval/poppler-0.62.0/poppler/Page.cc:552
#20 0x00007ffff7c94708 in Page::display (this=<optimized out>, out=<optimized out>, hDPI=<optimized out>, vDPI=<optimized out>, rotate=<optimized out>, useMediaBox=<optimized out>, crop=<optimized out>, printing=<optimized out>, abortCheckCbk=0x0, abortCheckCbkData=0x0, annotDisplayDecideCbk=0x0, annotDisplayDecideCbkData=0x0, copyXRef=0x0) at /home/huip/tmp/tfuzz_eval/poppler-0.62.0/poppler/Page.cc:481
#21 0x00007ffff7c98e29 in PDFDoc::displayPages (this=0x5555555cc4b0, out=0x5555555cd4b0, firstPage=<optimized out>, lastPage=0x1, hDPI=108, vDPI=108, rotate=0x0, useMediaBox=0x1, crop=0x0, printing=0x0, abortCheckCbk=0x0, abortCheckCbkData=0x0, annotDisplayDecideCbk=0x0, annotDisplayDecideCbkData=0x0) at /home/huip/tmp/tfuzz_eval/poppler-0.62.0/poppler/PDFDoc.cc:513
#22 0x000055555555ef20 in main (argc=<optimized out>, argc@entry=0x2, argv=argv@entry=0x7fffffffcf78) at /home/huip/tmp/tfuzz_eval/poppler-0.62.0/utils/pdftohtml.cc:392
#23 0x00007ffff73d91c1 in __libc_start_main (main=0x55555555e4b0 <main(int, char**)>, argc=0x2, argv=0x7fffffffcf78, init=<optimized out>, fini=<optimized out>, rtld_fini=<optimized out>, stack_end=0x7fffffffcf68) at ../csu/libc-start.c:308
#24 0x000055555555f1aa in _start ()
```
**Attachment 136967**, "a tar.gz file containing the testcase":
[testcase.tgz](/uploads/ee4e5c9bf76ab3155cf16e85ae9cc89b/testcase.tgz)https://gitlab.freedesktop.org/poppler/poppler/-/issues/503"Unsupported TilingType:3" Avoiding out-of-memory errors2018-08-21T11:04:37ZBugzilla Migration User"Unsupported TilingType:3" Avoiding out-of-memory errors## Submitted by zakiyama01
Assigned to **poppler-bugs**
**[Link to original bug (#102401)](https://bugs.freedesktop.org/show_bug.cgi?id=102401)**
## Description
Creating images from PDF
Certain pattern filling (TilingType = 3) wa...## Submitted by zakiyama01
Assigned to **poppler-bugs**
**[Link to original bug (#102401)](https://bugs.freedesktop.org/show_bug.cgi?id=102401)**
## Description
Creating images from PDF
Certain pattern filling (TilingType = 3) was not rendered properly.
#/usr/bin/pdftocairo -r 174 -scale-to 3000 -cropbox -jpeg pdftocairo.outofmemory.pdf image
SAMPLE PDF
[pdftocairo.outofmemory.pdf](https://github.com/zakiyama01/poppler/files/1251045/pdftocairo.outofmemory.pdf)
And if you set it to a larger image size, you will get out of memory error, rendering will finish with the painting. (Sample PDF is only paint of problem so we can not confirm the stopped image)
#/usr/bin/pdftocairo -r 174 -scale-to 5000 -cropbox -jpeg pdftocairo.outofmemory.pdf image
Internal Error: cairo context error: out of memory`<0a>`
cairo error: out of memory
cairo error: out of memory
I tried poppler-0.57.0. The same symptoms from before.
Because I am in trouble as it is, solve out of memory? Avoidance? I am fixing it.
see patch.
https://github.com/zakiyama01/poppler/commit/b35eba51683fd634142b4607496ac1ea37c2311e
thank youhttps://gitlab.freedesktop.org/poppler/poppler/-/issues/501Horizontal lines appear on blue background on pdf2018-08-21T11:04:29ZBugzilla Migration UserHorizontal lines appear on blue background on pdf## Submitted by Pedro Villavicencio
Assigned to **poppler-bugs**
**[Link to original bug (#19760)](https://bugs.freedesktop.org/show_bug.cgi?id=19760)**
## Description
this report has been filed here:
https://bugs.edge.launchpad....## Submitted by Pedro Villavicencio
Assigned to **poppler-bugs**
**[Link to original bug (#19760)](https://bugs.freedesktop.org/show_bug.cgi?id=19760)**
## Description
this report has been filed here:
https://bugs.edge.launchpad.net/poppler/+bug/318130
"We have a pdf file exported from Open Office Impress. The background is blue and there are horizontal lines through the image. The lines vary somewhat in colour, going from white to the blue.
If the slide show is made in OO Impress, all is well (showing that this isn't a hardware problem). Likewise if the same exported pdf file is shown on Windows XP, all is well."
PDF:
http://launchpadlibrarian.net/21362015/seminar1.pdfhttps://gitlab.freedesktop.org/poppler/poppler/-/issues/500poppler slows down search within some pages of this document2018-08-21T11:04:26ZBugzilla Migration Userpoppler slows down search within some pages of this document## Submitted by Pablo Rodríguez `@ousia`
Assigned to **poppler-bugs**
**[Link to original bug (#28508)](https://bugs.freedesktop.org/show_bug.cgi?id=28508)**
## Description
Making a whole search on http://www.pragma-ade.nl/general...## Submitted by Pablo Rodríguez `@ousia`
Assigned to **poppler-bugs**
**[Link to original bug (#28508)](https://bugs.freedesktop.org/show_bug.cgi?id=28508)**
## Description
Making a whole search on http://www.pragma-ade.nl/general/manuals/mk.pdf, such as searching for "Wiedervereinigung" with evince, I have realized that in some pages (that I guess contain more vector graphics or whatever it might be) searching is much slower.
I cannot code, but I guess that poppler might be busy due to other actions than searching. Those actions make sense after finding, not while searching.
I'm not sure whether this would be glib-specific, but this is only the frontend I can check right now.
Thanks for your help,
Pablohttps://gitlab.freedesktop.org/poppler/poppler/-/issues/498NULL pointer dereference in GfxState.cc:61272018-08-21T11:04:18ZBugzilla Migration UserNULL pointer dereference in GfxState.cc:6127## Submitted by foca@salesforce.com
Assigned to **poppler-bugs**
**[Link to original bug (#101504)](https://bugs.freedesktop.org/show_bug.cgi?id=101504)**
## Description
Created attachment 132068
Proof of concept
There is a NULL ...## Submitted by foca@salesforce.com
Assigned to **poppler-bugs**
**[Link to original bug (#101504)](https://bugs.freedesktop.org/show_bug.cgi?id=101504)**
## Description
Created attachment 132068
Proof of concept
There is a NULL dereference in GfxState.cc:6127.
The function drawSoftMaskedImage calls getLine() at CairoOutputDev.cc:2710 which returns NULL, and stores that in pix. The value is than passed on to the getGrayLine function which tries to dereference it, resulting in a null dereference.
2708 for (y = 0; y < maskHeight; y++) {
2709 maskDest = (unsigned char *) (maskBuffer + y * row_stride);
2710 pix = maskImgStr->getLine();
2711 maskColorMap->getGrayLine (pix, maskDest, maskWidth);
2712 }
The reason NULL is returned by getLine due to the following
`ImageStream::getLine`
529 if (unlikely(inputLine == NULL)) {
530 return NULL;
531 }
At the point inp is set to whatever pix was( pix=in), in our case pix was NULL. On line 6127 the dereference takes place and poppler crashes trying to dereference a NULL pointer.
6123 default:
6124 inp = in;
6125 for (j = 0; j < length; j++)
6126 for (i = 0; i < nComps; i++) {
6127 *inp = byte_lookup[*inp * nComps + i];
6128 inp++;
6129 }
A solution could be an additional check at CairoOutputDev.cc:2710 to check the line isn't NULL:
2710 pix = maskImgStr->getLine();
2711 if (pix == NULL) continue;
2712 maskColorMap->getGrayLine (pix, maskDest, maskWidth);
PoC is attached.
This vulnerability has been found by Offensive Research at Salesforce.com:
Alberto Garcia (@algillera), Francisco Oca (@francisco_oca) & Suleman Ali (@Salbei_)
**Attachment 132068**, "Proof of concept":
[PoC.pdf](/uploads/20cc7c8d5fa69ca411fac2617955bc0b/PoC.pdf)https://gitlab.freedesktop.org/poppler/poppler/-/issues/497PDF links -> SVG2018-08-21T11:04:15ZBugzilla Migration UserPDF links -> SVG## Submitted by rdt..@..il.com
Assigned to **poppler-bugs**
**[Link to original bug (#75953)](https://bugs.freedesktop.org/show_bug.cgi?id=75953)**
## Description
I'm using pdf2svg from http://www.cityinthesky.co.uk/opensource/pdf...## Submitted by rdt..@..il.com
Assigned to **poppler-bugs**
**[Link to original bug (#75953)](https://bugs.freedesktop.org/show_bug.cgi?id=75953)**
## Description
I'm using pdf2svg from http://www.cityinthesky.co.uk/opensource/pdf2svg/
toconvert PDF pages to SVG images. The key steps are as follows:
// Open the SVG file
surface = cairo_svg_surface_create(svgFilename, width, height);
drawcontext = cairo_create(surface);
// Render the PDF file into the SVG file
poppler_page_render(page, drawcontext);
cairo_show_page(drawcontext);
For my purposes, this works well, with one exception: clickable
links in the PDF page are not converted into links in the SVG image.
SVG supports simple xlinks and cairo has a function
cairo_surface_set_mime_data with CAIRO_MIME_TYPE_URI as a supported
mime type. I believe links on a PDF page are available using
poppler_page_get_link_mapping.
Seems like this would be a useful enhancement to
poppler_page_render for an svg_surface.