This is the mail archive of the libc-alpha@sources.redhat.com mailing list for the glibc project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

Re: Updating/adding locale for Ethiopia and Eritrea in GNU libc.


> It is probably a good idea to include the references as comments in the
> locales as well.

There simply are no standards in this area to refer to.  In the header
of each locale the language expert that I worked with is acknowledged
in comments.  This is the case for aa, om, sid, so and tig.  Generally
each locale represents two months of development work with the authority
cited.  For am, ti and gez, the work is my own, applying common knowledge
and formats used in sundry government and commercial publications.

 

> The package contains a simple build script, I would appreciate any
> feedback on problems encountered in building and using the locales.
 
> Looking at the sorted lists, I notice the following order:
 
>   a
>   A
>   b
>   B
 
> I believe I've asked before.  Are you sure capital letters (A, B)
> should come after small letters (a, b)?  Do any of the glibc
> maintainers know why iso14651_t1 uses this order?  In most languages I
> know it is the other way around, with A ordered before a.

This order comes from iso14641_t1, most locales defer to it.  Capital
before small is more familiar to me (from years of using the "C" locale),
but since I was uncertain I let iso14641_t1 handle the matter.  If for
no other reason than that the languages that introduced Roman script to
East African nations were doing the same.


 
> What is this Abegede Collation?  I suspect it will be easier to get
> _one_ collation order included.


Well...  if we could change history, much could be simplified ;-)
The order itself comes from the Hebrew sort order, it was used for
Ge'ez works for a period, but has now become academic.  Its not often
used outside of Ge'ez.  However, in the Amharic-English dictionary of
Dr Amsalu Aklilu, he still felt it important enough to acknowledge it
and explain why he _wasn't_ using it:

"In alphabetising Amharic entries, the conventional ha.le.ha.me. order
is followed rather than the traditional a.be.ge.de. one since the later
would create unnecessary confusion, and it is not a system that is very
well known among the population at large."

It is still important for people working in Ge'ez.  I never thought it
presented much in the way of complication.  Those he need it will use it,
those who don't won't.  Seems simple, for those who need the order I
think they will be very grateful it is available, its an excellent
utilization of glibc's collation capabilities.

/Daniel


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]