Compare the current output against a previous run to determine if there
has been any change since last time, and only run through imagediff if
there has been. For the vector surfaces, we can check the vector output
first and potentially skip the rasterisation. On my machine this reduces
the time for a second run from 6 minutes to 2m30s. As most of the time,
most test output will remain unchanged, so this seems to be a big win. On
unix systems, hard linking is used to reduce the amount of storage space
required - others will see about a three-fold increase in the amount of
disk used. The directory continues to be a stress test for file selectors.
In order to reduce the changes between runs, the current time is no longer
written to the PNG files (justified by that it only exists as a debugging
aid) and the boilerplate tweaks the PS surface so that the creation date
is fixed. To fully realise the benefits here, we need to strip the
creation time from all the reference images...
The biggest problem with using the caches is that different runs of the
test suite can go through different code paths, introducing potential
Heisenbergs. If you suspect that caching is interfering with the test
results, use 'make -C test clean-caches check'.
Whilst writing the base85 stream we advance an extra character if we spot
a '~', however we need to be paranoid that we have not stepped beyond the
end of the data for this write.
libpng changed behaviour in v1.2.30 to call the png_ptr->output_flush_fn
in png_write_end(). If no flush function is provided with
png_set_write_fn(), libpng will use its default fflush() instead - which
assumes that closure passed is a FILE* and not an arbitrary user pointer.
Consequently, we must actually set a dummy output_flush_fn to avoid
segfaulting.
We added cairo_has_show_text_glyphs() before. Since this is really a
surface property, should have the surface method too. Like we added
cairo_surface_show_page()...
The test-suite for win32 shows less than ideal error detection whilst
running on mingw32. Looking at the code, I spotted a few places where the
error propagation could be improved, and lo...
_cairo_cache_remove_random() just returned whether it found an entry to
remove and so the code can be simplified by returning a boolean as opposed
to a status code.
valgrind warns about an uninitialized read after a single char is promoted
to an int when passed to the printf. Silence the warning by using a
explicitly promoting the output byte to a full int.
When searching for a matching font, check the most recently used font
first. This optimizes the common case where pango calls save() and restore()
around rendering each layout, but almost all consecutive layouts use the
same font.
