Hi!
Craig Small wrote:
>
> G'day,
> I don't understand why indexer needs to have each charset put into the
> source when libc already provides many charsets for all sorts of
> languages. Won't isalnum() do the job? Someone showed me some test
> code it it gave the same results.
This will work only for current LOCALE. UdmSearch allows to use any
charset
independant from current locale. Moreover, we have to add charsets table
to
make recoding possibility. For example, Russian language web sites use 5
different charsets, where 2 of them are extremelly popular.
Note that it is usefull to implement support for several LocalCharsets.
The
example is Debian.org. So, libc library will be useless in this case.
BTW. MySQL has it's own charset library, it does not use libc for locale
related stuff.
> It would mean that indexer would understand any language that libc does,
> which is quite a lot.
______________
If you want to unsubscribe send "unsubscribe udmsearch"
to [EMAIL PROTECTED]
