On Thursday, October 10, 2002, at 09:17 AM, Jordi Mas wrote:

> En/na Dom Lachowicz ha escrit:
>> On Thursday, October 10, 2002, at 06:11 AM, Jordi Mas wrote:
>>> En/na Hubert Figuiere ha escrit:
>>>
>>>> On jeu, 2002-10-10 at 11:42, Hubert Figuiere wrote:
>>>>
>>>>> I'm getting the exact same problem.
>>>>> I suspect it is related to the fact that we don't link against 
>>>>> ispell by
>>>>> default anymore. Enabling ispell should solve the problem
>>>>
>>>> Here is a fix. I don't know if I'm doing the right thing, but it do
>>>> works and is quite clean. The idea is to link the source file were 
>>>> those
>>>> *ispell* function are implemented.
>>>> Attached is a patch that I haven't committed.
>>>
>>>
>>> The barbarism class uses the ispell internationalisation functions, 
>>> I should look into using the Pspell ones when we use Pspell, but I 
>>> have not done that bit because I have not look into Pspell yet.
>> Why are we using the ispell ones? If this is just an XML file, why 
>> not use the UTF-8 to UCS-4 string class, or iconv? After all, we just 
>> do append those words to the UCS4 buffer of strings returned from 
>> ispell/pspell. What am I missing?
>
> Yes, you are right, but I need mylower/mytolower and other stuff to 
> make the search of words with first letter upper case and return the 
> suggestion in upper case, etc. Are any local sensitive USC-4 available 
> functions to do this?

#include "ut_string.h"

UT_UCS4_isLower
UT_UCS4_isUpper
UT_UCS4_toLower
UT_UCS4_toUpper

...

Cheers,
Dom

Reply via email to