[tex-live] Making texts externally replaceable in PDFs, e.g. with sed(1)

[Zdenek Wagner]

Hi,

the biggest problem is that fonts are usually embedded as a subset so
you can replace the text cintaining the embedded characters. The text
must have the same length bcause the adresses of streams and other
objects must remain unchanged. For modifications of a PDF file I would
suggest QPDF: http://qpdf.sourceforge.net/
I have not tried it myself but it seems capable of such transformatins.

Zdeněk Wagner
http://ttsm.icpf.cas.cz/team/wagner.shtml
http://icebearsoft.euweb.cz

pá 14. 12. 2018 v 20:21 odesílatel Ulrike Fischer <news3 at nililand.de> napsal:
>
> Am Fri, 14 Dec 2018 16:50:17 +0100 schrieb Osipov, Michael:
>
>
> > So for a sed(1)-based postprocessor it is virtually impossible to map
> > "<0008002700320026002c00270008>" to "%DOCID%" w/o analyzing the PDF objects.
>
> I don't understand this. Where is the difference between
> <0008002700320026002c00270008> and (%DOCID%)? Both is a quite unique
> string.
>
>
>
> >
> > Requesting XeLaTex to produce
> >> Td[(%DOCID%)]TJ ET
> >
> > will not work
>
> It is easy:
>
> \documentclass[12pt,a4paper]{article}
>  \special{dvipdfmx:config z 0} %uncompress pdf
>
>  \begin{document}
>
>  {\fontencoding{OT1}\ttfamily \%DOCID\%
>   \makebox[0pt]{%
>    \raisebox{5cm}[0pt][0pt]{1234567890ABCDEFGHIJKLMNOPQRSTUVWXYZ}}%
>  }
>
>  bllb
>
>  \end{document}
>
> This gives -68.742 Td[(%DOCID%)]TJ in the pdf.
>
> The chars in the raisebox are there so that the chars perhaps needed
> for the replacement are also in the pdf.
>
> Be aware that you must exchange the same number of chars. %DOCINFO%
> can be replaced by 7 other chars, but not less or more or the xref
> table will be faulty.
>
> --
> Ulrike Fischer
> https://www.troubleshooting-tex.de/
>