At 08:15 AM 3/20/01 -0500, Tom Briggs wrote:
> If you'll forgive me for asking what feels like a stupid question: if
>all these hash files are now known to work, why don't we change our
>default setup to use these types of hash files?
Frpom what I can tell, all this does is change which people are happy.
Everybody who has working dictionaries now gets busted. Some subset of
folks with currently-broken dictionaries become happy.
>Are there as many or more
>hash files that are already known to work?
There's a partial list in the source tree:
abi/test/wp/Spelling.abw
For more information, see the threads from Henrik Berg's original work to
get ispell working for various languages.
I still submit that there's a feasible technical solution to allow us to
support a wide variety of ispell variants. For details, see Vlad's thread
on the topic this month.
bottom line
-----------
As I've mentioned before, it would be *very* useful if someone would take
the time to do a *complete* inventory of *all* known ispell dictionaries to
see which ones do and don't work.
We now believe that our alternatives are to:
- insist on 8-bit dictionaries (and perhaps distribute them)
- add support for 7-bit dictionaries, too
However, there are a *lot* of possible ways to compile ispell dictionaries,
so it would be very helpful to know which of those ways are most common.
Paul
