On Sun, 28 Feb 2010 23:02:39 -0500
Robert Haas <robertmhaas(at)gmail(dot)com> wrote:
> On Sun, Feb 28, 2010 at 8:43 AM, Ivan Sergio Borgonovo
> <mail(at)webthatworks(dot)it> wrote:
> > I've no idea what to do next other than asking if someone can
> > give a look to the code and check my comprehension of what a
> > tsquery should be in memory.
> Have you tried valgrind?
It was something related to my misinterpretation on how a tsquery is
operand.length is the length of the cstring *excluding* the \0.
I was mislead by the fact that in tsvector there are no ending \0
while in tsquery there are.
It still need more extensive testing but I think I got it working.
Once I'm sure I got it I'll make the polished source available and
write some docs for all the poor guys like me willing to start to
write extensions ;)
thanks to everybody for the patience.
Ivan Sergio Borgonovo
In response to
pgsql-hackers by date
|Next:||From: Peter Eisentraut||Date: 2010-03-02 10:26:20|
|Subject: Re: function side effects|
|Previous:||From: Heikki Linnakangas||Date: 2010-03-02 09:58:20|
|Subject: Re: WAL replay does not verify integrity|