Re: U-umlaut and two-letter combination

[Philippe Lhoste <PhiLho@...>]

Florian Berger wrote:
> Asko Kauppi wrote:
> > I know that problem, and afik string.find() does not offer a means to  it.
>
> Ok, I have to think about using regexp as Shmuel suggested in his mail.
>
> > However, how about string.gsub(txt,"ue","ü") first, and then do the  
> > string.find?
>
> That might work for a simple case but I'm looking for a general search 
> and I don't want to modify the text to be searched. If I for example 
> first replace "ue" with "ü" and then search for "München" everything is 
> ok but after that I cannot search word "glue".

A more "natural" transformation is to do it the way around: ü => ue
In French, we would transform œ (œ in HTML) to oe.

It can be a way to do collation too.

Of course, such transformation may be unwanted, eg. if you want to keep 
a difference between both ways of writing.

If performance isn't an issue, you can also do two searches...

--
Philippe Lhoste
--  (near) Paris -- France
--  http://Phi.Lho.free.fr
--  --  --  --  --  --  --  --  --  --  --  --  --  --