[cairo] About cairo_show_text and UTF-8...

[Behdad Esfahbod]

On Fri, 6 Jan 2006, Stuart Parmenter wrote:

> On 1/5/06, Behdad Esfahbod <behdad at cs.toronto.edu> wrote:
> > On Thu, 5 Jan 2006, Carl Worth wrote:
> >
> > > I'd be glad to see cairo_show_text improved to get glyphs from
> > > multiple fonts in order to provide better glyph coverage. But this
> > > work has not been done yet.
> > >
> > > Any volunteers?
> >
> > How about making cairo use Pango? :)
>
> I was hoping to avoid entering this thread, but making cairo depend on
> Pango is a bad idea.  As nice as Pango is, it isn't the right solution
> for all situations.  While yes, Pango works on Windows, there is a
> good chance that you are going to want to use Uniscribe and friends
> directly instead.  Last I checked, Pango doesn't work at all on Mac
> where ATSUI rules the land.  Pango also isn't covered by the same
> licenses as cairo which would probably make it harder as well.

Next version of Pango has support for ATSUI.  On Windows Pango
uses Uniscribe for most of the stuff.

> In Mozilla, for fonts we're going to be using Pango on unix platforms,
> Uniscribe directly on Windows and ATSUI on Mac.  We will be using
> cairo on all platforms.
>
> I think cairo's font and text APIs are more-or-less at the right level
> now and I don't think adding more text handling to cairo is the right
> solution.  I think that the current Pango calls are easy enough for
> people using Pango... maybe pango could add a pango_cairo_cairo_text()
> function that just does the 4 lines of code that Owen posted earlier.

pango_cairo_cairo_text would be quite simplistic and unusable
unless you don't want to set any font attributes whatsoever...
So I don't see enough reasons to add it.

--behdad
http://behdad.org/

"Commandment Three says Do Not Kill, Amendment Two says Blood Will Spill"
	-- Dan Bern, "New American Language"