[Gs-devel] Re: integration of FT2 and Ghostscript

[Ralph Giles]

On 13 Oct 2000, Francesco Zappa Nardelli wrote:

> the screenshots I've taken show that FreeType rasterizers work better
> than the original gs_eofill rule when filling charpaths (minor
> concerns in presence of anti-aliasing).

agreed.

> After the next, chaotic week, I plan to convert my hack in a
> fully-working patch, which could be used for wide-range testing --- I
> promise I'll write clean, good code.

Is there a particular reason you're patching in the freetype rasterizer?
I would rather have (as a configuration option) compiling in the gs
rasterizer *or* linking to the freetype library, either static or dynamic.
But I'm hardly authoritative on policy. Raph? Peter?

> About replacing the whole Type1 interpreter code with FT2, I prefer to
> wait for the new FT Type1 hinting code: David Turner told me once that
> he was going to rewrite it.  Currently native hinting is ignored in
> favour of David's auto-hinter, and I don't think this is a good
> approach with all fonts.

Are there patent issues with the native Type1 hints like there are with
TrueType? And how does the ghostscript hint code compare?

> A last little question: I am always addressing directly to you.  I am
> sure this is not a good habit: thus I am wondering about the existence
> of a ghostscript devel mailing list.

Yes, we just got a public development list this week and it's not heavily
advertised yet...understandable you missed it. =)

gs-devel at ghostscript.com

Thanks for you work, this is really exciting!

 -ralph