[Gs-devel] Report of Oct. 3 quarterly Artifex/artofcode/Aladdinmeeting

[mpsuzuki at hiroshima-u.ac.jp]

Dear Vlad,

>Files produced by ps2pdf from GS 6.5 are incorrect in this respect
>- they don't export encoding vector of the font (mapping glyph's
>index to unicode) - so ANY text (even ascii) in pdf files produced
>by ps2pdf can't be copied to clipboard in proper encoding (what gets
>to clipboard is glyphindexes - e.g. you get characters "\x1\x2\x3"
>when copy string "abc" from pdf file produced by ps2pdf).

Hmm? Even ASCII? When I make PDF from pure US-ASCII text with
standard basefonts like Courier, Helvetica etc, ps2pdf does not
embed any fonts, so I can copy the text from xpdf window (or
Acrobat Reader) to xclipboard. You can't?

If the specified font is encoded in Latin, Greek, Cyrillic or etc,
ps2pdf makes Type3 bitmap font after subsetting and embeddes it
into PDF. And, as you experienced, no text can be copied to
other applications.

As Peter explained, it's almost impossible to make correct
encoding vector for bitmap Type3. gs7.0x is greatly developped
in font embedding system, it can embeddes non US-ASCII encoded
characters in Type1. By this development, gs7.0x can provide
text as text, instead of a serie of unknown graphic objects.

# gs6.5x can convert PDF without embedding Latin/Greek/Cyrillic fonts?
# The simple and best way is it... I suppose.

This development is too big and indepth to bring into gs6.5x.
And there's license issue (as Peter told). Please do not
expect it for gs6.5x branch.

Regards,

mpsuzuki