413358 - cairo: wrong type of struct passed to cairo_surface_get_extents()

Status: RESOLVED FIXED

(Core :: Graphics, defect, P1)

Description

[Mats Palmgren (inactive)]

We pass a cairo_rectangle_int16_t to cairo_surface_get_extents() where a 
cairo_rectangle_int_t is expected:

cairo-pdf-surface.c: In function '_cairo_pdf_surface_emit_meta_surface':
cairo-pdf-surface.c:1569: warning: passing argument 2 of '_cairo_surface_get_extents' from incompatible pointer type
cairo-pdf-surface.c: In function '_cairo_pdf_surface_emit_surface_pattern':
cairo-pdf-surface.c:1650: warning: passing argument 2 of '_cairo_surface_get_extents' from incompatible pointer type
cairo-pdf-surface.c:1666: warning: passing argument 2 of '_cairo_surface_get_extents' from incompatible pointer type

There's also another issue that I want to address that I should have
seen in bug 413169:
http://bonsai.mozilla.org/cvsblame.cgi?file=/mozilla/gfx/cairo/cairo/src/cairo-types-private.h&rev=1.8&root=/cvsroot&mark=201,223-226#186
I believe line 201 should use 'int32_t' for x and y, like for example
'cairo_rectangle_int32' does, otherwise CAIRO_RECT_INT_MAX will not
actually fit:
cairo-scaled-font.c:1116: warning: overflow in implicit constant conversion

Patch coming up to fix both problems...
(and I'm confident there are no more of these because gfx/cairo/
now compiles without warnings in my tree...  A lot more is needed
for that of course, I'll open separate bugs on those soon.)

Comment 1

[Mats Palmgren (inactive)]

Attachment: Patch rev. 1

Comment 2

[Mike Schroepfer]

Do we need to upstream this?

Comment 3

[Vladimir Vukicevic [:vlad] [:vladv] (needinfo me, slow to respond)]

Already done; I didn't see this r? request, I ended up recreating it.  But I did miss the fix to point32_t, whoops!  We can either check this in, or I'll sync up our cairo and upstream cairo tomorrow.