[scribus] Stalking the wild 'RTL language support' in Scribus

[Hans Schmidt]

Am 13.06.2014 16:56, schrieb Kunda Loves Scribus:
> Ok proposal for group,
> first the representatives:
>   RTL: Mossab
>   CJK: Hans
>   Scribus code code refactoring: Andreas
>
>
> Needed:
> * Code refactoring on PDF Extraction is needed before moving forward with
> Mossab's RTL patch improvements
> * CJK scripter that can implement Hans's consultations before
> implementation of RTL
> * Any other interested parties that have skills that can aid this process

I can start by writing a small document about the general issues with CJK.

Some keywords:

- Text flow either horizontally (Western style) or vertically/RTL
- Therefore text flow of pages the same as Arabic (document starts at
“last” page)
- Text is generally laid out on predefined grid (e.g. 80x20 characters)
instead of only base line.
- No spacing between words (except Korean) and different line breaking rules
- Phonetic annotations (“Ruby” or “Furigana”) between the lines
- In vertical mode still sometimes in-line horizontal writing
- Support for input methods (Linux: Fcitx, Ibus, UIM etc; Windows: MS IME)

Well, I guess a Wiki page would be best to collect all the information
(for CJK, RTL, also other non-latin scripts and even some latin
scripts). The page should then be understandable by someone who cannot
read the language.

We can create a general wiki page called “Language support” with
subpages for every script group:

Roman alphabet (there is still much work for them as well)
Arabic/Hebrew
Chinese/Japanese/Korean
Indian scripts
Thai
etc...

On the subpage, every expert will write as much as he can about the
general script (well, can largely be referenced to Wikipedia, I guess)
which is easily understood even if the reader does not speak the
language. Then there should be typographic details explained. Maybe in
order of importance (it cannot all be implemented at first, I guess. Top
priority would maybe be to write single words correctly and after that a
complete paragraph. Eventually a complete document/book). This should at
best be referenced by typographic books.

This should then also, with a help of a developer, be technically
written: How could this be implemented in Scribus?

Especially important is that there are connections between the different
groups: CJK books are printed right to left, which is the same as
Arabic. Indian scripts are connected (as far as I remember...), which is
similar to Arabic. Mongol is written vertically, but from left-to-right
(still some similarities with CJK).
This is better as if every group does its own thing and we afterwards
have 3 different methods to set the pages RTL.  Also, if some
implementation is done in advance (RTL does some progress now), it
should easily be extendable to other languages.

I myself would love to take part in the development, but unfortunately
my programming skills are relatively basic and mainly limited to Python.
Maybe as a training I will try to implement the general text layout in a
PyQt demo, which can then be ported to C++.... But unfortunately I
wouldn’t count on me able to finish it.


One other thing: I think a _huge_ problem for Scribus is the general
English image of Scribus. The web page is only in English, the ML is
completely in English, the bug tracker is in English. Of course English
is the lingua franca especially in software development, but this has
the general effect that Scribus is generally a western project. I know
that many Japanese do not speak English at all. Even if they are good
programmers, they will be too scared to take part in such a English
project. I think this is really a huge problem when wanting to have
non-western participation.
For starters, I think making the website multilingual would be a
necessary step. Nobody without English knowledge even gets to the
download page and therefore misses Scribus altogether (even though the
software itself is translated!).
Also, if possible, some people should act as intermediaries between
English and other languages. For example, if a Japanese likes to
contribute, he could write a message in Japanese on the web page and I
could tell the others what he wants. We should actively promote
communication from other languages. Writing “Your contribution is
welcome! You can write in your own language!” could be one step.