Re: [Devel] Bug Report TypeCIDCFF

[Ivan Nincic]

Werner,

This bug report was related to dev. build dated 5/7/2003.
The error that occured during face load was fixed since then.

However I think there is another bug related to this font (plus hundereds of 
others I have).
In this case the problem is that there is only a single code in the 
encoding, when in fact there should be more entries (i.e. encoding built by 
FT is wrong).

The following PDF contains this font (as an embedded font) and various PDF 
viewers dispaly text properly:

 http://www3.telus.net/public/ab114139/sample.pdf

See the following related messages:

http://www.freetype.org/pipermail/devel/2003-May/009332.html
http://www.freetype.org/pipermail/devel/2003-May/009335.html

In general the encoding problem seems to be related to sub-setted, CFF 
fonts.

Regards,

Ivan



> From: Werner LEMBERG <address@hidden>
> Reply-To: address@hidden
> To: address@hidden, address@hidden
> Subject: Re: [Devel] Bug Report TypeCIDCFF
> Date: Sun, 01 Jun 2003 18:27:05 +0200 (CEST)
>
> > I have a a bug report related to latest freetype2-current.tar.gz
> > build 5/7/2003 (the same problem occurs for 2.1.4 release build).
> >
> > FT_Error error = FT_New_Memory_Face( gl_ft_lib,
> >         font_buf,
> >         font_buf_sz,
> >         0,
> >         (FT_Face*) &m_font_face );
> >
> > returns error code 0x55 while attempting to load the attached font
> > (same is for FT_New_Face(..)).
> > When font_buf_sz is extended beyond the actual size of the font
> > buffer the face is sometimes loaded and sometimes not.
> >
> > The font is a subsetted Type1 CID / CFF font.
>
> I've investigated this now.  The result of FreeType is correct!  The
> subsetted font T1CFF.cid is buggy (or truncated): It has a CIDcount
> value of 94 in its top dictionary while providing just a single glyph.
> If you look at the data starting at offset 0x28 (the beginning of the
> object data of the top DICT INDEX) you get
>
>   byte    value
>   -------------
>
>   248
>    28  => 392
>
>   248
>    29  => 393
>
>   139  =>   0
>
>    12
>    30  => ros   => registry = 392
>                    ordering = 393
>                    supplement = 0
>
>   233  =>  94
>
>    12
>    34  =>          CIDcount = 94
>
> [see 5176.CFF.pdf for details]
>
>
>     Werner
>
> _______________________________________________
> Devel mailing list
> address@hidden
> http://www.freetype.org/mailman/listinfo/devel

_________________________________________________________________
STOP MORE SPAM with the new MSN 8 and get 2 months FREE* 
http://join.msn.com/?page=features/junkmail