Re: Seeking for the fore-part of the key

Havasvölgyi Ottó <havasvolgyi(dot)otto(at)gmail(dot)com> schrieb:

> Hi,
>
> Let's assume I have a table (called tbl) with a column 'name' and an index on
> it.
> How to optimally find records whose name column begins with - say - "Pre"?
> I know it can be done with LIKE (WHERE name LIKE 'Pre%'), but will this use the
> index mentioned above to optimize the search?

for instance:

test=*# \d words
Tabelle »public.words«
Spalte | Typ | Attribute
--------+------+-----------
w | text |
Indexe:
»idx_words_w« btree (lower(w) text_pattern_ops)

test=*# explain analyse select * from words where lower(w) like lower('foo%');
QUERY PLAN
------------------------------------------------------------------------------------------------------------------------
Bitmap Heap Scan on words (cost=40.65..1779.81 rows=1595 width=13) (actual time=0.021..0.021 rows=0 loops=1)
Filter: (lower(w) ~~ 'foo%'::text)
-> Bitmap Index Scan on idx_words_w (cost=0.00..40.25 rows=1595 width=0) (actual time=0.019..0.019 rows=0 loops=1)
Index Cond: ((lower(w) ~>=~ 'foo'::text) AND (lower(w) ~<~ 'fop'::text))
Total runtime: 0.047 ms
(5 Zeilen)

Andreas
--
Really, I'm not out to destroy Microsoft. That will just be a completely
unintentional side effect. (Linus Torvalds)
"If I was god, I would recompile penguin with --enable-fly." (unknown)
Kaufbach, Saxony, Germany, Europe. N 51.05082°, E 13.56889°