[cairo] problem in _cairo_gstate_glyph_extent today?

[Carl Worth]

On Fri, 14 Jan 2005 13:14:12 -0500, John Ellson wrote:
> The problem seems to be an invalid double for dy, but I can't tell how 
> it got there.

Are you sure?

> Has there been a change to the cairo_text_extents API that I need to 
> catch up on?

No. The implementation has been through a fair amount of shuffling,
but the API should be the same.

> Program received signal SIGFPE, Arithmetic exception.
> *INT_cairo_matrix_transform_distance (matrix=0x9aaac7c, dx=0xbff586c0, 
> dy=0xbff586b8)
>     at cairo_matrix.c:237
> 237         new_y = (matrix->m[0][1] * *dx

Here's the complete line of code as I have it:

    new_y = (matrix->m[0][1] * *dx
             + matrix->m[1][1] * *dy);

Which doesn't make the source of the SIGFPE very obvious. There's
certainly no divide-by-zero here.

> (gdb) p *dy
> $2 = 1.5251824510634041e-314

That just looks like 0 to me. And multiplying by 0 had better not give
a SIGFPE.

Can you send a minimal test case to exercise this bug?

Thanks.

-Carl
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 189 bytes
Desc: not available
Url : http://lists.freedesktop.org/archives/cairo/attachments/20050114/f0a7ce59/attachment.pgp