freetype
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[ft] Glyph/Outline Caching on Linux non-FT_GLYPH_FORMAT_OUTLINE outlines


From: Jeff Finger
Subject: [ft] Glyph/Outline Caching on Linux non-FT_GLYPH_FORMAT_OUTLINE outlines
Date: Tue, 04 Dec 2007 12:56:07 +0200

I would like to integrate some font rasterization technology into FreeType. Any thoughts on the issues below would be greatly appreciated.

 

1. I am considering adding FreeType modules which create outlines in a format other than FT_GLYPH_FORMAT_OUTLINE and will rasterize the outlines with its own renderer. Has anyone out there tried anything similar, that is, using a non-FT_GLYPH_FORMAT_OUTLINE format? Am I about to jump off of a very tall cliff? One of my concerns is breaking glyph/outline caching, hence the question below on caching.

 

I would indeed prefer not to use another format. Doing so loses all the benefit of the existing FT_Outline technology. I do not see, though, a way to read my own file format, say format Q, create a FT_GLYPH_FORMAT_OUTLINE outline, and then render the Q-derived outlines with my own Q renderer without rendering *all* outlines in FT_GLYPH_FORMAT_OUTLINE format with the Q renderer. I do not want to render outlines from standard TTF, CFF, etc., files with the Q Rasterizer in order to have no impact on how existing fonts look.

 

One could add additional fields in the FT structures to specify which renderer to use, but that would break binary compatibility, thus breaking already compiled apps, as I see it.

 

2. On Linux (RHEL4, for now), where is glyph and/or outline caching taking place? Is the FreeType caching mechanism used, that is, the FTC_* methods? Is there a way to find this out short of reading the Linux source code? So far, I have not found anything on this topic in reading about fontconfig, Render, and XFt.

 

Thank you very much.

 

 

 


Checked by AVG Free Edition.
Version: 7.5.503 / Virus Database: 269.16.13/1167 - Release Date: 12/3/2007 12:20 PM


reply via email to

[Prev in Thread] Current Thread [Next in Thread]