bug#64908: 29.1; svg parse failure

[Alan Third]

On Sat, Aug 05, 2023 at 07:37:01PM +0200, David Ponce wrote:
> On 05/08/2023 18:31, Alan Third wrote:
> > On Sat, Aug 05, 2023 at 06:24:43PM +0200, David Ponce wrote:
> > > 
> > > Background is white by default.
> > 
> > Thanks.
> > 
> > > Just before call rsvg_handle_get_geometry_for_layer (line 11412 in 
> > > image.c):
> > > 
> > > (gdb) print zero_rect
> > > $1 = {x = 6.9533474147268633e-310, y = 2.4608421688060808e-319, width = 
> > > 5.8283471538673388e-315, height = 9.8813129168249309e-323}
> > > (gdb) print viewbox
> > > $2 = {x = 6.9533558057749482e-310, y = 6.9533473517748961e-310, width = 
> > > 1.4821969375237396e-323, height = 3.0586981611317814e-317}
> > > (gdb) print out_logical_rect
> > > $3 = {x = 1.2476920383715837e-316, y = 0, width = 0, height = 
> > > 3.3675983782902914e-317}
> > > 
> > > Just after:
> > > 
> > > (gdb) print zero_rect
> > > $4 = {x = 6.9533474147268633e-310, y = 2.4608421688060808e-319, width = 
> > > 5.8283471538673388e-315, height = 9.8813129168249309e-323}
> > > (gdb) print viewbox
> > > $5 = {x = 0, y = 0, width = 0, height = 0}
> > > (gdb) print out_logical_rect
> > > $6 = {x = 0, y = 0, width = 0, height = 0}
> > > (gdb) print $eax
> > > $7 = 1
> > 
> > Yeah, I see what's going on.
> > 
> > I was under the impression a new struct should always be set to all
> > zeroes, but evidently not.
> > 
> > Are you getting any build warnings about it being uninitialised?
> > 
> > Anyway, 3rd go at the patch attached.
> 
> Thanks
> 
> I just tried patch V3, and the result is the same:
> insert-image produces the same black square of 22x22 pixels.

Can you try changing the initialisation of zero_rect to {0, 0, 10,
10}?

I remember a lot of trial and error with this originally, and it looks
like they may have changed how it works.
-- 
Alan Third