Re: Patch: remove hard-coded indicator size (revised)

["Daniel Richard G." <straker fox mit edu>]

On 7 Jul 2000, Owen Taylor wrote:

> Well one reason for them to be there is for subclassing. But mostly
> they are there just as a way of representing a fixed global parameter
> in a fairly encapsulated way, to allow for the addition of 
> later configuration mechanisms. But I wouldn't say they have much
> point beyond a #define.

Okay, fair enough. So then, don't look at the patch as allowing theme
engines to do the Wrong Thing in yet another way-- look at it as fixing
GtkCheckMenuItem to use that very convention for its indicator size. I
mean, right now, in gtkcheckmenuitem.c:
gtk_real_check_menu_item_draw_indicator(), you have: 

      width = 8;
      height = 8;
[...]
	  gtk_paint_check (widget->style, widget->window,
			   state_type, shadow_type,
			   area, widget, "check",
			   x, y, width, height);

which is, IMHO, bad. Unfriendly vibes from hard-coded values aside,
GtkCheckButton already draws its check using an indicator_size class
variable. So the problem is not only one of flexiblity, but also one of
consistency. 


--Straker


                       //,,    //,,
//=================// //////  ////// //==================================\\
 Straker Skunk     / //////  ////// /      Skunks are such wonderful
 <skunk@mit.edu>  / //////  ////// /     creatures... soft, and cuddly,
 -- -- -- -- -- --\ \\\\\\  \\\\\\ \      and if you annoy them they
 Daniel Richard G. \ \\\\\\  \\\\\\ \      make you stink like hell
 ========--====--==-\ \\\\\\  \\\\\\ \---------------=--==--====--========
 //mit.edu/straker  / //////  ////// / Furry|Course VI-3|MIT Class of 2001
 //skunk.mit.edu   / //////  ////// / 80% Beaver 90% Penguin 100% SKUNK!!!
\\===============// //////  ////// //====================================//
                     ''//    ''//