[cairo] [RFC] cairo_path_extents()

[Bill Spitzak]

Kalle Vahlman wrote:
> 2008/1/21, Carl Worth <cworth at cworth.org>:
>> So the semantic question for the API is:
>>
>>         Should cairo_path_extents consider a lone move_to when
>>         computing its extents?

I would say no.

The original problem was with a path derived from a font including the 
moveto at the end. If the moveto counted it would be impossible to get 
the actual extents of the characters converted to the path. However if 
the moveto did not count, you can get both results, by default you get 
the extents of the characters, but by adding an rlineto(0,0) to the end 
of the path you get the result including the moveto. Since the ability 
to get both results is better then the moveto should not count.