freetype
[Top][All Lists]
Advanced

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

Fw: Re: [ft] opentype problems


From: mpsuzuki
Subject: Fw: Re: [ft] opentype problems
Date: Tue, 26 Apr 2005 21:15:25 +0900

Excuse me
I attached 2 big pictures and refused by list manager.
Here I forward the text only.

Regards,
mpsuzuki

Begin forwarded message:

Date: Tue, 26 Apr 2005 20:51:23 +0900
From: address@hidden
To: address@hidden
Cc: address@hidden, Werner LEMBERG <address@hidden>
Subject: Re: [ft] opentype problems


Hi

On Tue, 26 Apr 2005 20:12:49 +0900
address@hidden wrote:
>>> +    current_font.flags |= FT_LOAD_IGNORE_GLOBAL_ADVANCE_WIDTH;
>>
>>Hmm, perhaps the "global advance width" in the font should be adjusted
>>to zero or turned off?
>>
>>address@hidden:~$ ftinfo -a yngn.otf | grep -i width
>>yngn.otf: 0: horizontal: advance_Width_Max: 65535
>
>Your suggestion sounds worthful to test. However, TrueType
>specification says advanceWidthMax must be consistent with
>horizontal metrics (of hmtx?), I suppose "turn-off" is
>impossible.

Just I've checked if fixing advanceWidthMax of yngn.otf
can be counter bug-fix for old freetype. I modified the
value by binary editor. (It's lucky that freetype can
load the font file with wrong checksum :-))

00000000  00 01 00 00 00 14 01 00 00 04 00 40 47 44 45 46 address@hidden
          [snip]
000000D0  9C 84 04 5E 00 00 01 4C 00 00 00 36 68 68 65 61 ...^...L...6hhea
000000E0  06 D3 FD 6C 00 00 01 84 00 00 00 24 68 6D 74 78 ...l.......$hmtx
                      ^^^^^^^^^^^ (hhea starts 0x00000184)
          [snip]
00000180  00 00 00 00|00 01 00 00|02 F8|FF 37|00 3B|FF FF ...........7.;..
                     |  version  | xxx | xxx | xxx |<HERE>

When I decrease 0xFFFF to smaller values, the glyph width gets
smaller, but the rendering result of old ftview (including
wrong glyph width bug) is still different from current ftview.
See attached picture, I suppose Burmese ligature functionalities
are broken.

I have to check adding the flag FT_LOAD_IGNORE_GLOBAL_ADVANCE_WIDTH
is more than fixing advanceWidthMax.

Regards,
mpsuzuki





reply via email to

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