| From: | Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> |
|---|---|
| To: | "Dennis Noordsij" <dennis(dot)noordsij(at)helsinki(dot)fi> |
| Cc: | pgsql-bugs(at)postgresql(dot)org, Oleg Bartunov <oleg(at)sai(dot)msu(dot)su>, Teodor Sigaev <teodor(at)sigaev(dot)ru> |
| Subject: | Re: BUG #4814: Segmentation fault when using indexed prefix FT search |
| Date: | 2009-05-18 01:59:17 |
| Message-ID: | 7385.1242611957@sss.pgh.pa.us |
| Views: | Raw Message | Whole Thread | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-bugs |
"Dennis Noordsij" <dennis(dot)noordsij(at)helsinki(dot)fi> writes:
> create table foo ( ts tsvector not null );
> create index foo_idx on foo using gin(ts);
> insert into foo values (to_tsvector('Lore ipsum'));
> select * from foo where 'ipsu:*'::tsquery @@ ts;
> [crash]
Hmm. It rather looks to me like this
cmp = DatumGetInt32(FunctionCall4(&ginstate->comparePartialFn[attrnum],
should be this
cmp = DatumGetInt32(FunctionCall4(&ginstate->comparePartialFn[attrnum-1],
which leaves me with even less of a warm and fuzzy feeling about
matchPartialInPendingList() than I had before. How much has that code
been tested at all? It seems quite clear that it's not exercised by
any regression test ...
regards, tom lane
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Heikki Linnakangas | 2009-05-18 11:46:41 | Re: BUG #4812: statistics not available on tables containing only hstore column |
| Previous Message | Dennis Noordsij | 2009-05-17 18:25:41 | BUG #4814: Segmentation fault when using indexed prefix FT search |