We need to occasionally force fallbacks whilst testing the PDF
output, so export a debug interface to do so in order to avoid poking
around inside cairo internals.
References: https://bugs.freedesktop.org/show_bug.cgi?id=48577
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
We need to occasionally force fallbacks whilst testing the PostScript
output, so export a debug interface to do so in order to avoid poking
around inside cairo internals.
References: https://bugs.freedesktop.org/show_bug.cgi?id=48577
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
A PostScript surface embeds a CreationDate comment into its document
description pre-amble. Normally this is set to the time the surface is
written out, except we set this to a constant value in the boilerplate
for the purposes of mimicking a reference file. It may also be useful
for external applications, so make it a public export.
References: https://bugs.freedesktop.org/show_bug.cgi?id=48577
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
There is already a cairo-xcb section and there are no symbols that
should be documented in an extra -xrender section. It is not mentioned
in cairo-sections.txt either. So simply delete it.
Signed-off-by: Uli Schlachter <psychon@znc.in>
When xlib-xcb is enabled, this is just dead code which is never used. Thus, this
shouldn't hurt. However, this does include cairo-xlib-private.h. Shouldn't be a
problem? Well, that header contains static inline functions which some compiler
on Solaris will emit even when they are unused.
This brings us to the real problem: That static inline function refers to a
function which isn't compiled with xlib-xcb and thus linking fails with
undefined symbols.
This can be reproduced with GCC by adding a call to
_cairo_xlib_screen_put_gc(NULL, NULL, 0, 0); to
_cairo_xlib_fallback_compositor_get.
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=48489
Signed-off-by: Uli Schlachter <psychon@znc.in>
The bbox is used to compute the ink extents (and so the pattern extents
of a recording surface) and if given an integer translation we failed to
transform the bbox into the target space.
Fixes mask (pdf).
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Just in case the clip polygon turns out to be much larger than the
operation extents (silly us for not reducing it correctly) and lead to
catastrophe, such as:
Program received signal SIGSEGV, Segmentation fault.
pixman_fill_sse2 (bits=<optimized out>, stride=4, bpp=8, x=0, y=0,
width=3, height=-34811, data=0) at pixman-sse2.c:3369
3369 *(uint16_t *)d = data;
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
A typo using half_line_x instead of half_line_y when emitting dashed
segments of the rectilinear stroke.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Commit 2283ab9 introduced a logic error. Instead of falling back to the non-SHM
path when getting the image via SHM failed, we now did the fallback when getting
the image via SHM worked (which means that the SHM operation was a waste of
time).
Signed-off-by: Uli Schlachter <psychon@znc.in>
We need to fix y-axis origin when map a GL surface to image surface for
non-texture GL surface.
Test cases: extended-blend-alpha-mask, extended-blend-mask.
Although the image outputs is not right, but the image on the first grid
(upper-left corner) is correct comparing to image output.
A 'Since' field on non-public functions is not very meaningful,
because their signature, implementation and semantic is subject to
change without any visibility outside of cairo.
The following Python script was used to compute "Since: 1.X" tags,
based on the first version where a symbol became officially supported.
This script requires a concatenation of the the cairo public headers
for the officially supported beckends to be available as
"../../includes/1.X.0.h".
from sys import argv
import re
syms = {}
def stripcomments(text):
def replacer(match):
s = match.group(0)
if s.startswith('/'):
return ""
else:
return s
pattern = re.compile(
r'//.*?$|/\*.*?\*/|\'(?:\\.|[^\\\'])*\'|"(?:\\.|[^\\"])*"',
re.DOTALL | re.MULTILINE
)
return re.sub(pattern, replacer, text)
for minor in range(12,-2,-2):
version = "1.%d" % minor
names = re.split('([A-Za-z0-9_]+)', stripcomments(open("../../includes/%s.0.h" % version).read()))
for s in names: syms[s] = version
for filename in argv[1:]:
is_public = False
lines = open(filename, "r").read().split("\n")
newlines = []
for i in range(len(lines)):
if lines[i] == "/**":
last_sym = lines[i+1][2:].strip().replace(":", "")
is_public = last_sym.lower().startswith("cairo")
elif is_public and lines[i] == " **/":
if last_sym in syms:
v = syms[last_sym]
if re.search("Since", newlines[-1]): newlines = newlines[:-1]
if newlines[-1].strip() != "*": newlines.append(" *")
newlines.append(" * Since: %s" % v)
else:
print "%s (%d): Cannot determine the version in which '%s' was introduced" % (filename, i, last_sym)
newlines.append(lines[i])
out = open(filename, "w")
out.write("\n".join(newlines))
out.close()
The Python script is unable to fix all of the warnings, because some
point to comments that are not actually documentation.
Fixes the remaining 'documentation comment not closed with **/'
warnings.
Documentation comments should always start with "/**" and end with
"**/". This is not required by gtk-doc, but it makes the
documentations formatting more consistent and simplifies the checking
of documentation comments.
The following Python script tries to enforce this.
from sys import argv
from sre import search
for filename in argv[1:]:
in_doc = False
lines = open(filename, "r").read().split("\n")
for i in range(len(lines)):
ls = lines[i].strip()
if ls == "/**":
in_doc = True
elif in_doc and ls == "*/":
lines[i] = " **/"
if ls.endswith("*/"):
in_doc = False
out = open(filename, "w")
out.write("\n".join(lines))
out.close()
This fixes most 'documentation comment not closed with **/' warnings
by check-doc-syntax.awk.
The SOURCE and CLEAR are the odd pair in Cairo's range of operators that
are bound by the shape/mask, but are unbound by the source. This
regularly leads to bugs as only track the bound/unbound rectangles and
confuse the meaning when bound only by the mask.
What is required is that the unbound extents in this case is only
trimmed by the mask (the bounded extents are still the intersection of
all).
Fixes bug-source-cu
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
There is need to loop over number of glyphs to check wether the glyph
image is a ARGB32 as the font's antialias option can be used for checking.
If antialias is SUBPIXEL or BEST, the glyph surface will be ARGB32,
otherwise it will be A8 format. Therefore we will only be using
component-alpha at SUBPIXEL (or better) font quality and only then need
a mask for multiple pass glyph composition.
Flushing only releases the fallback if we flush twice with no
intervening damage (the theory is to try and reduce readbacks). So it is
possible for a correctly behaving application to call mark-dirty and there
still be a fallback.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
The gradient is relatively small and any differences between upload
methods should be hidden by the caching. As it stands, using pbo with
the gradient fails with fglrx. As the workaround to use a simple
TexImage2D is inconsequential (may even be a minute win) and simplifies
the code, just do it.
A couple of mistakes, such as inverting the logic as to when to flush
damage back from the shadow, meant that nothing happened when drawing to
the directfb surface.
Again still only compile tested.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
The damage wasn't being created on the right surface, so the damage to
the fallback image surface was not being tracked. Perform a little bit
of juggling so that we track dirty regions on the fallback surface itself.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
