Re: more missing boxed types

[Owen Taylor <otaylor redhat com>]

James Henstridge <james daa com au> writes:

> On 1 Jul 2001, Owen Taylor wrote:
> 
> > I think language bindings will _probably_ want to use non-interned
> > strings for PangoLanguage. While its marginally less efficient, it's
> > more convenient to use "en-us", then Pango.Language ("en-us"),
> > lang.to_string ().
> >
> > The language property on GtkTextTag just uses a string, for
> > example. It's perhaps a bit like PangoFontDescription in this manner.
> >
> > I wouldn't object to a patch to add boxed types for PangoGlyphString
> > and PangoLanguage, though even they are a little marginal.
> 
> Okay.  I have put together the patch.  The details are available at:
>   http://bugzilla.gnome.org/show_bug.cgi?id=56957
> 
> As well as adding typecodes for PangoLanguage and PangoGlyphString, it
> moves the typecode definitions for PangoFontDescription and PangoTabArray
> from libgtk to libpango (to match the policies for the other boxed types).
> Patches for pango and gtk are included in the bug report.  I don't think
> this patch will affect anyone (it may possibly affect people who were
> using the GTK_TYPE_PANGO_* defines outside of gtk+ though) so I would
> consider it a fairly low risk patch.

The patch looks simple, correct, and like a good improvement.

Please go ahead and commit,
                                        Owen