Re: [BUG] LaTeX export in non-English language [9.6-pre (release_9.5.5-1087-g620a96.dirty @ /home/yantar92/.emacs.d/straight/build/org/)]

["Juan Manuel Macías" <maciaschain@posteo.net>]

Hi, Ihor,

I start by comment the error that you get when compiling. Is it possible
that you have an incomplete TeX Live installation? Distro repositories
usually offer TeX live in a modular way. On Arch Linux and its
derivatives, the texlive-lang and texlive-lang-extra packages should be
installed for non-Latin language support (this would include the luabidi
package). In the Ubuntu/Debian family I think the required package is
called texlive-lang-arabic. Anyway, I recommend installing TeX live in
its entirety (the drawback is that it takes up more space). I think that
in Ubuntu and family the meta-package is called texlive-full. Another
possibility that I recommend, especially on distros that package older
versions of TeX live, is to install the official TeX live distribution:
https://tug.org/texlive/acquire.html (this has the advantage of being
able to update everything more regularly via TeX live's own package
manager). If you want to keep the official packages of your distro, but
also want to try new things, it is also possible to create a portable
installation of TeX live on a pen drive:
https://tug.org/texlive/archive/portable-anchor.html

Ihor Radchenko writes:

> Aha! I missed that part in the newer manual.
>
> Note that the manual only talks about setting
> `org-latex-packages-alist'. Not the keyword.
>
> Maybe we can add the #+LATEX_HEADER example to the manual.

There are examples included in 'LaTeX header and sectioning structure'.
Perhaps it would be nice to add some more examples in the case of Bidi
languages, Chinese, Japanese, etc.? In the case of CJK languages, I
think additional LaTeX packages are required.

>> BTW I've noticed that the value of #+language is not case-agnostic. You
>> should put "he". Should it be case-agnostic? In the previous
>> implementation, when there were two language lists for babel and
>> polyglossia, neither was it.
>
> I think that it is a good idea. I do not recall common language
> abbreviations to be case-sensitive in other software.
>

I totally agree: that's something I missed when I made the previous
patch...

Best regards,

Juan Manuel