Subject: Re: rendering - how bad is AW font displaying desing?
From: Leonard Rosenthol (leonardr@lazerware.com)
Date: Wed Jan 10 2001 - 09:34:10 CST
At 4:51 AM +0100 1/10/01, Mike Nordell wrote:
>So what we *really* need, is some kind of "context sensitive
>glyph-replacer"? Would that suffice, or am I yet again misunderstanding
>something.
That's just one part of it.
What you need to layout and display text is (in a sense) what
AbiWord already has (though in a more limited fashion), which is a
set of routines that can do (at least) the following:
* Given a chunk of text (series of "code points") in a given "style"
(font, size, lang, etc.) convert that to a set of glyphs.
* Using the "set of glyphs", be able to measure the width (and
height) of the glyphs as well as actually render them to the
destination media.
That's really all that is at the core of a text engine. When
dealing with simple Roman ASCII text, the above routines are pretty
trivial. However, when dealing with languages like Arabic or Thai
that are contextual, or CJKV where you have composite glyphs, it can
get pretty hairy!
>If I *did* get you right, are there any "engines" available that does this
>kind of stuff, or are we breaking new ground?
This is what Pango is all about, for example! It provides
all the routines for doing the above stuff.
>Wait a minute. Is "line-breaking" *also* a renderer issue?
_I_ consider line breaking to be outside the core renderer,
since how the glyphs are placed on a page is up to higher level
logic. The renderer should ONLY be concerned with glyph
composition.
Leonard
--
----------------------------------------------------------------------------
You've got a SmartFriend in Pennsylvania
----------------------------------------------------------------------------
Leonard Rosenthol Internet: leonardr@lazerware.com
America Online: MACgician
Web Site: <http://www.lazerware.com/>
FTP Site: <ftp://ftp.lazerware.com/>
PGP Fingerprint: C76E 0497 C459 182D 0C6B AB6B CA10 B4DF 8067 5E65
This archive was generated by hypermail 2b25 : Wed Jan 10 2001 - 12:06:03 CST