Summary: | source/fitz/colorspace.c:1486: fast_rgb_to_cmyk: Assertion `"This should never happen" == NULL' failed. | ||
---|---|---|---|
Product: | MuPDF | Reporter: | fiesh <fiesh> |
Component: | fitz | Assignee: | MuPDF bugs <mupdf-bugs> |
Status: | RESOLVED FIXED | ||
Severity: | normal | CC: | sebastian.rasmussen |
Priority: | P4 | ||
Version: | 1.12.0 | ||
Hardware: | PC | ||
OS: | Linux | ||
Customer: | Word Size: | --- | |
Attachments: | PDF triggering assertion |
Description
fiesh
2018-10-05 06:50:58 UTC
Please attach the document with which you reproduce this. Without that file it is really hard to figure out what went wrong. Also, having read the other bug, are you building mupdf with the bundled (and patched) lcms2 or are you building with the system version of lcms2? Created attachment 15777 [details]
PDF triggering assertion
Sorry, accidentally omitted the attachment. I'm building it with the system lcms2. Thank you! I attempted to reproduce this locally and it renders fine. Do you mind try building mupdf with the bundled lcms2 version? You are likely correct in that this is indeed a duplicate of 699262, but it would be nice to confirm that. Building with the system version of lcms2 is not supported because we patch and require lcms2 to support multithreading. Indeed compiling it with the provided lcms fixes the issue! So I suppose this should be seen as a gentoo bug? That it attempts to use the system liblcms, when really it is a different, incompatible library? Alright, then we know that this is not really a bug with mupdf itself. That means I'll close the bug. Yes, I would say so. You can also see this discussion over at fedora where they concluded as much: https://bugzilla.redhat.com/show_bug.cgi?id=1553915#c7 Reported as such at https://bugs.gentoo.org/667788 |