[tex-live] new babel in TL

[Robin Fairbairns]

Mojca Miklavec <mojca.miklavec.lists at gmail.com> wrote:

> thank you very much for this import.
> 
> A few points that I would like to mention:
> 
> - I suggest to put serbian and serbianc into the same babel package if
> possible (both is Serbian, but one is in cyrillic script)

unless someone complains by tonight (when i'll next work on it, i'll do that

> - I know that you mentioned it, but I would still suggest to review if
> names match between babel-lang and hyph-lang. Just as an example,
> there is hyph-slovenian and babel-slovene. (I have deliberately chosen
> "slovenian" for hyph and even the document "babel-slovene/slovene.pdf"
> says "The Slovenian language". The name "slovene" as filename probably
> remained just for "compatibility" reasons.) Other examples are
> babel-portuges, norsk, ...

this is not new: we are generating a new setup from existing files.  in
other words, you could have complained ages ago.

i shall add these things to the list of "to do".
 
> - In the same spirit: there are some languages like "babel-frenchb" or
> "ukraineb" just because there is "frenchb.ldf". However, there is
> "germanb.ldf" and still "babel-german" and not "babel-germanb". My
> suggestion would be to call the package "babel-french",
> "babel-ukrainian" and not "babel-frenchb". If ever four other packages
> for French pop up, they can all go to babel-french, no matter how they
> will be called. The list is not extensive, I just listed the first few
> languages that I noticed.

i've never been quite sure why these "language-name"b versions exist.
(except in the case of languages like french, where the are lots of
alternative babel packages.)

using the existing name was

> - I miss at least Lithuanian
> (http://www.ctan.org/tex-archive/language/lithuanian/tex/generic/babel/)

please use mirror.ctan.org/language/...  you know it makes sense (and it
saves on dante's paid-for bandwidth)

> and a few other languages. Here's a list of some ldf files in TeX Live
> that aren't in babel-language (buy probably not all of them make
> sense):
> 
> tex/generic/thailatex/thai.ldf
> tex/generic/velthuis/hindi.ldf
> tex/latex/arabi/arabic.ldf
> tex/latex/arabi/farsi.ldf
> tex/latex/bgreek/bgreek.ldf
> tex/latex/bgreek/ibygreek.ldf
> tex/latex/cjk/texinput/thai/thaicjk.ldf
> tex/latex/cjk/utils/pyhyphen/pinyin.ldf
> tex/latex/ethiop/ethiop.ldf
> tex/latex/frenchle/french.ldf
> tex/latex/frenchle/frenchle.ldf
> tex/latex/ibycus-babel/ibycus.ldf
> tex/latex/lithuanian/lithuanian.ldf
> tex/latex/mongolian-babel/mongolian.ldf
> tex/latex/spanglish/spanglish.ldf
> tex/latex/spanish-mx/spanishmx.ldf
> tex/latex/turkmen/turkmen.ldf
> tex/latex/vntex/vietnam.ldf
> tex/latex/vntex/vietnamese.ldf
> tex/platex/japanese/japanese.ldf
> 
> - I'm almost sure that there was supposed to be Latvian somewhere. I
> was talking to
> Ilmars Poikans about it in 2008, but maybe it was only included in
> Polyglossia and not in babel. I'm CC-ing him.
> 
> - I suggest to delete furlan.ldf (or rather: ask Claudio Beccari to
> give permission to delete it). I removed the language synonym for
> hyphenation patterns soon after adding it and this is a relatively new
> file anyway. The author didn't know what he was supposed to do with
> synonyms.

won't delete without claudio's approval.

> - I just realized that "turkmen" contains babel-turkish, but I have
> put hyph-turkish under langother. That probably needs a change (moving
> hyph-turkish to langturkmen)?

dunno

> - I'm just curious: why are these ldf files in TEXMF/tex/generic
> rather than TEXMF/tex/latex?

presumably because they'll work under plain.  babel was designed to work
that way, from the very start (before context became a public entity,
for sure).

> Given a high chance that something could go wrong at our side, I would
> prefer to release the promised hyph-utf8 based on the old babel and a
> new version based on the new babel a few days later.

"new babel" as modified from your comments above?  please note that the
ctan work progresses slowly, and is unlikely to be complete this month.
to delay hyphenation stuff on that basis would seem to me a woeful
mistake.

robin
the modern internet-connected snail