cairo_has_current_point() can be used to determine whether a current
point is defined. We introduce this new symbol with a boolean return
value to avoid the versioning ambiguity of modifying
cairo_get_current_point(). This way we also don't have to map what
should be a routine operation to an error condition as was previously
proposed.
Make sure that we don't test lines that start or end inside the box,
since our algorithm will find intersections on the wrong ends on the
line, and not count them.
The PS Language Reference requires the PS miter to be >= 1.
The PDF Reference does not specify miter limits however acroread fails
to display pages when the miter is < 1. Older versions of ghostscript
crash when the PDF miter is < 1.
The image/meta-pattern is written once. A PS pattern twice the size of
the cairo pattern is created and the surface drawn four times in a
reflect pattern inside the PS pattern.
This fixes the extend-reflect and extend-reflect-similar PS test
failures.
Store the surface size in each pattern and smask group and use this
size when the pattern/group is emitted. This is required as a pattern
or group may be used from a meta surface with a different size to the
PDF surface. However the pattern or group is emitted after the the meta
surface content has been emitted and the surface size has been
restored to the the page size.
This fixes the following pdf test failures:
extend-reflect-similar
extend-repeat-similar
mask
Before there was just an assert statement here that the
determinant of the matrix was not infinite. That was bogus
since a user-provided can end up here. So instead, do the
correct error propagation of any CAIRO_STATUS_INVALID_MATRIX
error as necessary.
This eliminates the current failure of the invalid-matrix
test case.
Previously we left the return values alone, which set the
user up for a nasty trap, (using potentially uninitialized
values with no indication that there was an error). So now
we initialize these values to 0.0 if the cairo_t is in error.
The fixed functions include:
cairo_path_extents
cairo_stroke_extents
cairo_fill_extents
cairo_clip_extents
cairo_font_extents
Main goal was to be able to pass list of all cairo sources, enabled or
not, to check-doc-syntax.sh such that it doesn't check *.h, *.c, *.cpp
because that can be annoying when bisecting.
The NULL return value will only happen if the X Render extension
is not available. We've already got that NULL return value
documented, so it's not an error if the user asks for it. In
particular, it's definitely not a surface-type mismatch.
Thanks for Behdad for the suggested name improvement. Also, make
it more clear that the stride passed to cairo_image_surface_create_for_data
should come from calling cairo_format_stride_for_width with the
same width.
Document this function as a required call to get the correct
stride value before calling cairo_image_surface_create_for_data.
This means that previously-failing calls with non-multiple-of-4
stride values are now documented as errors. Also, we now have
the possibility of moving to more stringent alignment constraints,
(one can imagine doing 64-bit or 128-bit boundaries for example).
This Quartz API seems to only tile at integer coordinates; if the source image is
scaled to anything less than integer-aligned, seams appear between tiles. Detect
this and fall back to slower but more general CGPattern path.
