Re: [HACKERS] another locale problem - Mailing list pgsql-hackers

From Daniel Kalchev
Subject Re: [HACKERS] another locale problem
Date
Msg-id 199906110932.MAA25757@dcave.digsys.bg
Whole thread Raw
In response to Re: [HACKERS] another locale problem  (Tatsuo Ishii <t-ishii@sra.co.jp>)
Responses Re: [HACKERS] another locale problem
List pgsql-hackers
This worked, however it made all selects of ~* '^sometext' sequential.

Why can't we use index searches in this case? I believe it is :-)

But this

SELECT key FROM t WHERE key ~* '^ sometext';

(note the space after the carret) still does not work! It says index scan will 
be used and finds nothng. If 'sometext' is all uppercase the proper result is 
returned. The problem seems to be in the regex handling anyway. I tried 
compiling regcomp.c with -funsigned-char, but that was not enough. Further 
hacking...

Daniel

>>>Tatsuo Ishii said:> Ok. Now I almost certain that:> >                         (strcmp(opname,"~*")> == 0 &&
isalpha((unsignedchar)n->val.val.str[pos])))> > should work for you. Can you confirm this?> --> Tatsuo Ishii
 




pgsql-hackers by date:

Previous
From: Angelos Karageorgiou
Date:
Subject: Re: [HACKERS] another locale problem
Next
From: Daniel Kalchev
Date:
Subject: Re: [HACKERS] another locale problem