Hi Behdad,
for my embedded project your proposal might be over the top :-)

Werner, may I suggest to enhance the synopsis section of FTC documentation.
cited from FTC documentation:

"Second, the cache calls, only when needed, a client-provided function to
convert a FTC_FaceID into a new FT_Face object. The latter is then
completely managed by the cache, including its termination through
FT_Done_Face."

Maybe this can be extended by:

To monitor termination of face objects, the finalizer callback in FTC_Face
generic can be used. In addition you might use generic data to store the
FTC_FaceID of the face.

br
Gernot


2010/9/28 Behdad Esfahbod <[email protected]>

> On 09/28/10 01:09, Werner LEMBERG wrote:
> >>> >> You can register a callback by setting face->generic.
> >> >
> >> > perfekt solution for my problem. I was not aware of the finalizer
> >> > callback.
> > Could you suggest an addition to the documentation to improve it?
>
> Actually let me hijack the thread.  While the face->generic is the perfect
> solution for an application's needs, it's inadequate for use by other
> libraries.  The problem is that there is only one generic slot per face.
> What's needed instead is something like the cairo_set_user_data() or
> g_object_set_data_full() APIs that allow multiple generic slots.  This can
> be
> implemented as a hash-table internally.  I think it would be enough to just
> add such API to FT_Face and perhaps FT_Library, but not other objects.
>
> Anyone feel like cooking such API?  It probably needs a new type other than
> FT_Generic since the callback to FT_Generic doesn't get passed the generic
> itself, just the containing object.
>
> Thanks,
>
> behdad
>
> _______________________________________________
> Freetype-devel mailing list
> [email protected]
> http://lists.nongnu.org/mailman/listinfo/freetype-devel
>
_______________________________________________
Freetype-devel mailing list
[email protected]
http://lists.nongnu.org/mailman/listinfo/freetype-devel

Reply via email to