Re: [ft-devel] Auto Embolding of Lohit Devanagari

[Alexei Podtelezhnikov]

On Wed, Feb 29, 2012 at 10:00 AM, Alexei Podtelezhnikov
<address@hidden> wrote:
> On Fri, Feb 24, 2012 at 9:54 PM, suzuki toshiya
> <address@hidden> wrote:
>> I think there are 2 issues in auto-embolden Lohit Devanagari;
>> a) disconnection of horizontal strokes (which should be connected with 
>> previous/next glyph)
>
> On line 357 in ftview.c, I think it is incorrect to double the
> strength before it is added to the advance on line 376. To confirm
> this I looked at the lines 895 and 959 in ftoutln.c. Indeed, each
> outline point is moved by half the strength of emboldening resulting
> in a total increase of the width by the strength value. So it easy to
> fix in ftview.c on line 357, but much harder to propagate this sacred
> knowledge to the higher level libraries.

I also think that emboldening should not touch bearing values.
The screenshots from ftview are attached. Here is the proposed patch.


diff --git a/src/ftview.c b/src/ftview.c
index 2cca5b0..ea4b4ca 100644
--- a/src/ftview.c
+++ b/src/ftview.c
@@ -354,7 +354,6 @@
           error = FT_Outline_Embolden( &slot->outline, xstr );
           /* ignore error */

-          xstr = xstr * 2;
           ystr = xstr;
         }
         else if ( slot->format == FT_GLYPH_FORMAT_BITMAP )
@@ -380,10 +379,7 @@

         slot->metrics.width        += xstr;
         slot->metrics.height       += ystr;
-        slot->metrics.horiBearingY += ystr;
         slot->metrics.horiAdvance  += xstr;
-        slot->metrics.vertBearingX -= xstr / 2;
-        slot->metrics.vertBearingY += ystr;
         slot->metrics.vertAdvance  += ystr;

         if ( slot->format == FT_GLYPH_FORMAT_BITMAP )

emboldened.png
Description: PNG image

emboldened1.png
Description: PNG image