At 1237305418 time_t, Uli Schlachter wrote: > + if (MIN(geometry.width, geometry.height) <= 2 * border_width)
You should store the MIN() value since you use it later. :-)
> + {
> + /* The border_width is so big that there would be no space left in
> the
> + * window. This now adjusts border_width so that there is at least
> 1px
> + * space inside the window, except when we were already called with a
> + * width/height of 0, in which case this will likely lead to a X
> + * protocol error.
> + */
> + border_width = MIN(geometry.width, geometry.height) / 2;
> + if (border_width > 0)
> + border_width--;
> + fprintf(stderr, "error: Attempted to create a window with too large
> border_width / too small size\n");
I don't like that, and you should use warn().
I'd sugget to rather change the behaviour of the geometry:
- sw->geometries.internal.width = geometry.width - 2*border_width;
- sw->geometries.internal.height = geometry.height - 2*border_width;
+ sw->geometries.internal.width = geometry.width;
+ sw->geometries.internal.height = geometry.height;
Which should avoid any error and probably clarify things.
What do you think guys?
Cheers,
--
Julien Danjou
// ᐰ <[email protected]> http://julien.danjou.info
// 9A0D 5FD9 EB42 22F6 8974 C95C A462 B51E C2FE E5CD
// My root password is
signature.asc
Description: Digital signature
