Thread: locale and spanish acute
I'm doing some searches with LIKE and have a problem with words with acute
vocals. I would like más to be treated the same way as mas, and vice-verse.
Is this because I don't have the locale set correctly?
Saludos... ;-)
--
Porqué usar una base de datos relacional cualquiera,
si podés usar PostgreSQL?
-----------------------------------------------------------------
Martín Marqués | mmarques@unl.edu.ar
Programador, Administrador, DBA | Centro de Telematica
Universidad Nacional
del Litoral
-----------------------------------------------------------------
Martín Marqués writes: > I'm doing some searches with LIKE and have a problem with words with acute > vocals. I would like más to be treated the same way as mas, and vice-verse. > > Is this because I don't have the locale set correctly? No, it's because it doesn't work that way. What you probably want to do is "accent-fold" your strings before sending them through pattern matching. -- Peter Eisentraut peter_e@gmx.net http://funkturm.homeip.net/~peter
On Mié 12 Sep 2001 20:10, you wrote:
> Martín Marqués writes:
> > I'm doing some searches with LIKE and have a problem with words with
> > acute vocals. I would like más to be treated the same way as mas, and
> > vice-verse.
> >
> > Is this because I don't have the locale set correctly?
>
> No, it's because it doesn't work that way. What you probably want to do
> is "accent-fold" your strings before sending them through pattern
> matching.
What does "accent-fold" mean?
--
Porqué usar una base de datos relacional cualquiera,
si podés usar PostgreSQL?
-----------------------------------------------------------------
Martín Marqués | mmarques@unl.edu.ar
Programador, Administrador, DBA | Centro de Telematica
Universidad Nacional
del Litoral
-----------------------------------------------------------------
Martín Marqués writes: > On Mié 12 Sep 2001 20:10, you wrote: > > Martín Marqués writes: > > > I'm doing some searches with LIKE and have a problem with words with > > > acute vocals. I would like más to be treated the same way as mas, and > > > vice-verse. > > > > > > Is this because I don't have the locale set correctly? > > > > No, it's because it doesn't work that way. What you probably want to do > > is "accent-fold" your strings before sending them through pattern > > matching. > > What does "accent-fold" mean? Convert all accented characters to some other characters in a way that is meaningful to you. -- Peter Eisentraut peter_e@gmx.net http://funkturm.homeip.net/~peter
On Mié 12 Sep 2001 21:52, Peter Eisentraut wrote:
> Martín Marqués writes:
> >
> > What does "accent-fold" mean?
>
> Convert all accented characters to some other characters in a way that is
> meaningful to you.
OK, I think I got what you told me.
lets see... I have this query:
select count(*) from tab1 where col1 LIKE '%mas%'
but I would like it to catch rows with "más" also.
The only thing I can think of is makeing a function that applied to col1
would give me col1 but without accents. Say the function is called
no_accents, so:
no_accents(más)=mas
Now, the problem is that I have little idea on making such a function. Is it
very difficult?
Saludos... :-)
--
Porqué usar una base de datos relacional cualquiera,
si podés usar PostgreSQL?
-----------------------------------------------------------------
Martín Marqués | mmarques@unl.edu.ar
Programador, Administrador, DBA | Centro de Telematica
Universidad Nacional
del Litoral
-----------------------------------------------------------------
Martín Marqués writes: > select count(*) from tab1 where col1 LIKE '%mas%' > > but I would like it to catch rows with "más" also. > > The only thing I can think of is makeing a function that applied to col1 > would give me col1 but without accents. Say the function is called > no_accents, so: > > no_accents(más)=mas > > Now, the problem is that I have little idea on making such a function. Is it > very difficult? CREATE FUNCTION no_accents(text) RETURNS text AS ' my $arg = $_[0]; $arg =~ tr/áéëíñóú/aeeinou/; return $arg; ' LANGUAGE 'plperl'; Something similar should be possible in PL/Tcl if you prefer that. Or you bother with C and run a simple loop over the text string. -- Peter Eisentraut peter_e@gmx.net http://funkturm.homeip.net/~peter
On Fri, 2001-09-14 at 15:49, Peter Eisentraut wrote: > Martín Marqués writes: > > > select count(*) from tab1 where col1 LIKE '%mas%' > > > > but I would like it to catch rows with "más" also. select count(*) from tab1 where col1 ILIKE to_ascii($1) works for me. The variable is passed via .jsp Cheers Tony Grant -- RedHat Linux on Sony Vaio C1XD/S http://www.animaproductions.com/linux2.html Macromedia UltraDev with PostgreSQL http://www.animaproductions.com/ultra.html
On Lun 17 Sep 2001 04:31, Tony Grant wrote:
> On Fri, 2001-09-14 at 15:49, Peter Eisentraut wrote:
> > Martín Marqués writes:
> > > select count(*) from tab1 where col1 LIKE '%mas%'
> > >
> > > but I would like it to catch rows with "más" also.
>
> select count(*) from tab1 where col1 ILIKE to_ascii($1)
>
> works for me.
But I want it to work with más too. For that I would need to change that
query for this one:
select count(*) from tab1 where to_ascii(col1) ILIKE to_ascii($1)
Right?
Saludos... :-)
--
Porqué usar una base de datos relacional cualquiera,
si podés usar PostgreSQL?
-----------------------------------------------------------------
Martín Marqués | mmarques@unl.edu.ar
Programador, Administrador, DBA | Centro de Telematica
Universidad Nacional
del Litoral
-----------------------------------------------------------------
On Mon, 2001-09-17 at 22:49, Martín Marqués wrote: > But I want it to work with más too. For that I would need to change that > query for this one: > > select count(*) from tab1 where to_ascii(col1) ILIKE to_ascii($1) > > Right? Right that will work because the content and the request is changed. Cheers Tony -- RedHat Linux on Sony Vaio C1XD/S http://www.animaproductions.com/linux2.html Macromedia UltraDev with PostgreSQL http://www.animaproductions.com/ultra.html