Summary: | cairo-test.c: make check fails for cairo 1.6.4 on Mac OS X 10.4.11 PPC | ||
---|---|---|---|
Product: | cairo | Reporter: | cfr <reescf> |
Component: | general | Assignee: | Carl Worth <cworth> |
Status: | RESOLVED INVALID | QA Contact: | cairo-bugs mailing list <cairo-bugs> |
Severity: | critical | ||
Priority: | medium | ||
Version: | 1.6.4 | ||
Hardware: | PowerPC | ||
OS: | Mac OS X (All) | ||
Whiteboard: | |||
i915 platform: | i915 features: |
Description
cfr
2008-06-14 07:24:45 UTC
Update: If I edit test/cairo-test.c so that the line "#include <fontconfig/fontconfig.h>" instead reads "#include </usr/local/include/fontconfig/fontconfig.h>", I get past this error. The same error then occurs regarding test/font-options.c and the same edit lets me get past it. This finally gets through make check with 11 of 173 tests failing (ignoring all those requiring the DISPLAY variable to be set which it isn't in this environment). But all those failures look minor to me. So, setting aside the tests which require X11, these edits let me get through the test suite. But I cannot figure out how to get through it without hard-coding the include path for fontconfig.h. I think what is happening is that configure is finding the updated version of fontconfig in /usr/local and make compiles against that, but when I run the test suite, it is using the old version of fontconfig in /usr/X11R6 which does not include the FcFini function and I do not know how to stop it doing this. (Note, I ignore the tests requiring X11 because these parts of Cairo's test suite always fail for me so it seemed better to concentrate on the tests which usually pass.) - cfr This reads like a system configuration issue causing it to compile against one version of the library but link against the other. A nuisance, but hopefully you'll have better success with 1.8.0 as quite a lot of effort went into making the build system more robust across the various architectures (notably win32 and OS/X). You're right to concentrate on only the new failures between releases (though reducing the total number of failures to zero is the ultimate goal, whilst increasing the decision coverage...) So please do try the latest stable release and let us know how it fares. Thanks. |
Use of freedesktop.org services, including Bugzilla, is subject to our Code of Conduct. How we collect and use information is described in our Privacy Policy.