[cl-pdf-devel] Re: [cl-typesetting-devel] Unicode support feedbackneeded

[Dmitriy Ivanov]

"Marc Battyani" <marc.battyani at fractalconcept.com> wrote:

| Dmitriy Ivanov <divanov at aha.ru> wrote:
|>
|> I am pleased to propose quite a big rewrite of CL-PDF. The purpose is
|> two-fold:
|>
|> 1) Integrate Type0 (CID-keyed) and Type1 fonts based on
|> custom-encoding. 2) Approach binary mode of output file.
|>
|> The attachment contains the two files:
|>
|> 1) di-encodings.lisp
|> Slightly changed encoding class hierarchy.
|>
|> 2) di-contrib.lisp
|> Patches for several source files.
|
| Thanks Dmitriy, I will test this as soon I can find some time to do it.
|
| As for the binary mode of the pdf file, I'm starting to wonder if it's
| a good idea after all. In any case the pdf format is a mixed char/byte
| one and there are advantages and drawback to using each mode in Lisp.

I think, the binary mode can be reasonably used when opening an external
file to write the document. In other places, we take encoding into account
and convert everything to strings that actually store ub8 codes. Page stream
content is always a string and ready for compressing.

That is why my latest code used your version of compress-string from
zlib.lisp - the subtle version of
     string-to-octets
 is unneeded.
--
Sincerely,
Dmitriy Ivanov
lisp.ystok.ru