Hi Santiago,
Thanks for your report:
On 24-09-15 19:31:54, Santiago Vila wrote:
> [...]
>
> ======================================================================
> ERROR: test_all_parametred
> (tests.test_libmat2.TestCleaning.test_all_parametred) (case={'name': 'pdf',
> 'parser': <class 'libmat2.pdf.PDFParser'>, 'meta': {'producer':
> 'pdfTeX-1.40.14'}, 'expected_meta': {'creation-date': -1, 'format':
> 'PDF-1.5', 'mod-date': -1}})
> ----------------------------------------------------------------------
> Traceback (most recent call last):
> File "/<<PKGBUILDDIR>>/tests/test_libmat2.py", line 554, in
> test_all_parametred
> self.assertTrue(p1.remove_all())
> ^^^^^^^^^^^^^^^
> File "/<<PKGBUILDDIR>>/libmat2/pdf.py", line 39, in remove_all
> return self.__remove_all_lightweight()
> ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
> File "/<<PKGBUILDDIR>>/libmat2/pdf.py", line 63, in __remove_all_lightweight
> pdf_surface.finish()
> libmat2.pdf.cairo.MemoryError: out of memory
>
> ======================================================================
> ERROR: test_all
> (tests.test_lightweight_cleaning.TestLightWeightCleaning.test_all)
> ----------------------------------------------------------------------
> Traceback (most recent call last):
> File "/<<PKGBUILDDIR>>/tests/test_lightweight_cleaning.py", line 55, in
> test_all
> self.assertTrue(p1.remove_all())
> ^^^^^^^^^^^^^^^
> File "/<<PKGBUILDDIR>>/libmat2/pdf.py", line 39, in remove_all
> return self.__remove_all_lightweight()
> ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
> File "/<<PKGBUILDDIR>>/libmat2/pdf.py", line 63, in __remove_all_lightweight
> pdf_surface.finish()
> libmat2.pdf.cairo.MemoryError: out of memory
>
> ----------------------------------------------------------------------
> Ran 121 tests in 37.711s
>
> FAILED (errors=2)
Are you still able to reproduce this? I've never seen this myself, and
my guess would be that this was caused by a transient error caused by
"something else" happening on the machine you used to build.
I intent to close this -- unless you believe that there is indeed a
problem in regards to the mat2 source code.
Thanks,
Georg
