[HarfBuzz] Problem with shaping Serif fonts

[Eduardo Castiñeyra]

Is there any way to avoid that? Despite ff being a single glyph, it represents two clusters, imagine I want to animate the tracking of a word, it would look like this

e ff i c i e n t 

Although glyph-wise it would be correct, it would be very hard to convince any customer about that.

In some cases it would be nice to force the shaper to output at least one glyph per cluster. 

Does it also depend on the language? If one script have one ligature it doesn't mean every language makes use of it.

Regards!

Sent from a mobile device

On 09/10/2013, at 06:23, Khaled Hosny <khaledhosny at eglug.org> wrote:

> On Wed, Oct 09, 2013 at 01:39:17AM +0200, Oliver Meier wrote:
>> Hello
>> 
>> I recently started playing with harfbuzz and encountered a strange problem.
>> I don't know what I'm doing wrong, but somehow multiple f's shaped
>> for serif fonts like DejaVuSerif or FreeSerif get grouped in pairs
>> or so..
>> 
>>        hb_buffer_set_direction( hb_buffer, HB_DIRECTION_LTR );
>>        hb_buffer_set_script( hb_buffer, HB_SCRIPT_LATIN );
>>        hb_buffer_set_language( hb_buffer,
>> hb_language_from_string("en", strlen("en")) );
>> 
>>        hb_buffer_add_utf8( hb_buffer, "ffff", strlen("ffff"), 0,
>> strlen("ffff") );
>>        hb_shape( hb_ft_font, hb_buffer, NULL, 0 );
>> 
>>        hb_glyph_info_t     *glyph_info =
>>            hb_buffer_get_glyph_infos(hb_buffer, &glyph_count);
>> 
>> At this point glyph_count=2 instead of 4. In the kerning i see that
>> f's are indeed "grouped".
>> 
>> Can someone point me in the right direction, or is this a bug?
> 
> https://en.wikipedia.org/wiki/Typographic_ligature#Latin_alphabet
> 
> BTW, if you have any code that assumes the number of output glyphs will
> be equal to the number of input characters, it is wrong.
> 
> Regards,
> Khaled
> _______________________________________________
> HarfBuzz mailing list
> HarfBuzz at lists.freedesktop.org
> http://lists.freedesktop.org/mailman/listinfo/harfbuzz