Re: PDF outline not capturing Cyrillic text

[Oliver Corff]

Dear Peter, Dear All,

this problem is presumably not limited to groff. I remember the same
issue when I was building LaTeX texts with foreign language elements (in
my case, among others: Chinese) with the package to create internal
links from (hyperref, iirc) table of contents to chapters,etc. As soon
as there were non-ASCII characters in the anchor, the result was
(halfway working) garbage.

My memory of this issue should only serve as a point of reference, I
think modern versions of hyperref can produce non-ASCII links and output.

Best regards,

Oliver.


On 17/09/2022 23:35, Peter Schaffter wrote:
> Greetings, all.
>
> Source documents written in Cyrillic and processed with mom/pdfmom
> break the PDF outline:
>
> 1. The text of titles and headings is not displayed in the outline.
>
> 2. If there are items where English is automatically part of the
>     outline label (cover, title page), the label prints (minus the
>     text), but the linking hierarchy fails with some viewers if there
>     are intervening titles or headings, e.g.
>       doc cover (Cyrillic)
>       copyright notice (in English)
>       table of contents (auto relocated)
>       title page (Cyrillic)
>       chapter title (Cyrillic)
>       body text with Cyrillic headings
>     With the above arrangement, the outline in okular and evince shows
>     only
>       Cover: (no text)
>       Copyright:
>       Title Page: (no text)
>     In okular, clicking on Copyright takes you to the correct page,
>     however in evince, clicking on Copyright takes you to the table
>     of contents page.
>
> At a guess, it looks as if gropdf or pdfmark isn't recognizing Cyrillic
> characters as valid input for creating pdf bookmarks.  I'm at a
> loss as to how to overcome this.  Ideas?