[HarfBuzz] Arabic shaping questions

[Jonathan Kew]

On 20 Jan 2011, at 16:30, Deron Kazmaier wrote:

> On 1/19/11 8:34 AM, Behdad Esfahbod wrote:
>> On 01/19/11 00:59, Deron Kazmaier wrote:
>>> So, is something missing from this Frutiger font that Harfbuzz requires? I
>>> can't say that it is a defective font (brand new from Linotype, must be good,
>>> right ;-), so is this a font that requires code not yet written?? Or something
>>> I am not handling on my end?
>> The Arabic shaper is complete as far as I'm concerned.  Can't tell without
>> having the font though.
>> 
>> This is a good time to say that I happily accept font donations from foundries
>> and individuals for testing purposes.
>> 
>> behdad
>> 
> 
> ok, nothing on my end seems to change what Harfbuzz does. Frutiger glyphs never get any substitutions, and any other font with Arabic glyphs that I have tested work just fine. Very odd.
> 
> Since Behdad confirms that this font works fine for him, I need to test with some other Harfbuzz application that works and run it through the debugger and see what the difference is. My only conclusion so far is that this font must use a different set of routines within the shaper. Don't ask me how that bit of code would fail and not the rest...
> 
> So, what applications are out there I can test with? Any simply test apps? There is the harfbuzz-ng branch of Pango, but that would not be easy to debug.

Firefox 4 uses harfbuzz-ng, so you could try that font with the latest FF4 beta (or nightly) and see if it shapes properly.

JK