Re: Bug in ILIKE?

Tom Lane wrote:
> Good:
>
> regression=# select 'T' ilike 't';
> ?column?
> ----------
> t
> (1 row)
>
> Not so good:
>
> regression=# select 'T' ilike E'\\t';
> ?column?
> ----------
> f
> (1 row)
>
> ISTM backslash is only supposed to turn off the pattern-language
> specialness of characters, not render them case sensitive. The reason
> this happens is that the backslash case in MatchText() checks for exact
> equality. I think it should be checking for TCHAR() equality, same as
> when it is just checking two ordinary characters. Comments? Is this a
> backpatchable bug fix, or should we only change the behavior for 8.4 and
> beyond?
>
>

Hmm. A quick test shows this working the other way in 8.2, so I assume
this is a byproduct of the work we did to improve the efficiency of
LIKE/ILIKE in 8.3 ;-(

The docs actually don't state what are the semantics of escape followed
by something that is not escape or a metachar. Does the spec say
anything about that?

cheers

andrew