Hans Hagen wrote:
> Oleg Kolosov wrote:
>
>> Hans Hagen wrote:
>>
>>> Oleg Kolosov wrote:
>>>
>>>
>>>> Hello!
>>>>
>>>> I'm trying to generate searchable pdf with cyrrillic glyphs with the
>>>> following:
>>>>
>>>> \enableregime[utf]
>>>> \mainlanguage[ru]
>>>> \setupencoding[default=t2a]
>>>> \useencoding[pfr]
>>>> \usepdffontresource t2a
>>>> \usetypescript[pscyr][\defaultencoding] % type-pscyr is my own
>>>> typescript file
>>>> \setupbodyfont[pscyr,14pt]
>>>>
>>>> also tried with:
>>>>
>>>> \startencoding[t2a]
>>>> \usepdffontresource t2a
>>>> \stopencoding
>>>>
>>>> It seems that \usepdffontresource does nothing. I see pdfr-def loaded
>>>> in log, but not pdfr-t2a. \input pdfr-t2a (or pdfr-ec) says that
>>>> \startpdffontresource is undefined command. I've created pdfr-t2a.tex
>>>> by replacing definitions in pdfr-ec with ones from cmap latex package
>>>> (found in file t2a.cmap). I'm using ConTeXt mkII since mkIV is in
>>>> active development. Tried also with ec as default encoding with the
>>>> same result (pdfr-ec.tex is not loaded).
>>>>
>>>> Please help me create header for minimal file which will generate
>>>> searchable PDF.
>>>>
>>>>
>>> pdftex does it itself (i.e. create the vectors) using pdfr-def.tex
>>> (unless i did something wrong)
>>>
>>> Hans
>>>
>>>
>>>
>> It's unlikely. I've tested it with minimal file and english text is
>> indeed searchable, but cyrillic is not, with copy-paste I get some
>> strange symbols. I'm using type1 fonts from PSCyr package with my own
>> typescript, does this matter? I've attached typescript file just in
>> case (it's still incomplete but works fine for me). Maybe I miss some
>> definition or option? BTW cyrillic in PDF TOC works fine (with inclusion
>> of spec-tst.tex).
>>
>
> can you check if the file has the right entris for your font?
>
> pdfr-def.tex
>
> the old mechanism is obsolete so pdfr-t2a will not do anything
>
> Hans
>
>
Codes seem to be in place but doesn't match actual font in T2A encoding.
For ex. I have cyrillic capital a in font on 00C1 where in pdfr-def this
is Aacute. According to enco-utf my 00C1 in font should map to 0410
position (I hope this is understandable description). Maybe there is
some switch to enable such mapping? I don't understand these
encoding/mapping issues enough to create necessary table myself. Maybe
you can provide some example, so I will be able to help?
P. S. Sorry for late response.
--
Best Regards,
Oleg Kolosov