Re: brltty: does this really need to depend on libicu?
Samuel Thibault <email@example.com> writes:
> Frans Pop, le Tue 14 Oct 2008 14:54:35 +0200, a écrit :
>> a bit over 5MB of that was due to libicu38 now being included on the
>> CD, which seems to be a new dependency from brltty.
>> Does brltty really need such a huge library for unicode support?
> It is not really for unicode support, it is for character description:
> when a blind person encounters a character which she doesn't know (and
> thus can not read in braille), she can at least get a description of it.
> That feature can not be compiled as a module (it's deep inside the core
> of brltty), so there is no easy way to make it optional other than
> building a brltty-noicu package.
> Mario, what do you think?
If ICU is really just used for character descriptions (I am not 100% sure)
we could write some code to parse UnicodeData.txt (which is about 1MB).
However, this is very likely much to late in the release process.
We'd need to coordinate this change with upstream first.
A brltty-noicu package doesn't make much sense IMO, it will
only make the user experience more complicated.
⡍⠁⠗⠊⠕ | Debian Developer <URL:http://debian.org/>
.''`. | Get my public key via finger firstname.lastname@example.org
: :' : | 1024D/7FC1A0854909BCCDBE6C102DDFFC022A6B113E44
`- <URL:http://delysid.org/> <URL:http://www.staff.tugraz.at/mlang/>