Re: [SQL] where col1 not ilike ('str1%', 'str2%'... 'strN%') support?

2011-08-31 Thread Emre Hasegeli
2011/8/30 Emi Lu em...@encs.concordia.ca: First, where not (col1 ~~* any(array['str1%', 'str2%'... 'strN%'])) will work for me. But I feel ilike ('str1', ... 'strN') is more intuitive, isn't it? It is not. It is like where id = (3, 5, 7). -- Emre Hasegeli Veri Tabanı Yöneticisi Tart

Re: [SQL] where col1 not ilike ('str1%', 'str2%'... 'strN%') support?

2011-08-31 Thread Jasen Betts
On 2011-08-30, Emi Lu em...@encs.concordia.ca wrote: Hi Tom, select * from tablename where not (col1 ~~* any(array['str1%', 'str2%'... 'strN%'])); If next version could have not ilike ('', '') added into window functions, that's will be great! Why? And what's this got to do with window

Re: [SQL] where col1 not ilike ('str1%', 'str2%'... 'strN%') support?

2011-08-31 Thread Emi Lu
On 08/31/2011 03:16 AM, Emre Hasegeli wrote: 2011/8/30 Emi Luem...@encs.concordia.ca: First, where not (col1 ~~* any(array['str1%', 'str2%'... 'strN%'])) will work for me. But I feel ilike ('str1', ... 'strN') is more intuitive, isn't it? It is not. It is like where id = (3, 5, 7). What

[SQL] ERROR: cannot subscript type text because it is not an array

2011-08-31 Thread lxnf98mm
I have table with field l type lseg Column | Type | Modifiers +-+--- cnt| integer | used | boolean | l | lseg| c | text| select l from t where used = false group by l[0]; ERROR: cannot subscript type text because it is not an array

Re: [SQL] where col1 not ilike ('str1%', 'str2%'... 'strN%') support?

2011-08-31 Thread Pavel Stehule
2011/8/31 Emi Lu em...@encs.concordia.ca: On 08/31/2011 03:16 AM, Emre Hasegeli wrote: 2011/8/30 Emi Luem...@encs.concordia.ca: First, where not (col1 ~~* any(array['str1%', 'str2%'... 'strN%'])) will work for me. But I feel ilike ('str1', ... 'strN') is more intuitive, isn't it? It is

[SQL] function based index problem

2011-08-31 Thread Viktor Bojović
Hi, on table entry (17M records) there is one index: CREATE INDEX ndxlen ON uniprot_frekvencije.entry USING btree (length(sequence::text)); When using = in search which returns only two records, query runs much (hundred times) slower. i don't know why it doesn't use index scan. I just

Re: [SQL] function based index problem

2011-08-31 Thread David Johnston
From: pgsql-sql-ow...@postgresql.org [mailto:pgsql-sql-ow...@postgresql.org] On Behalf Of Viktor Bojovic Sent: Wednesday, August 31, 2011 5:27 PM To: pgsql-sql@postgresql.org; pgsql-ad...@postgresql.org Subject: [SQL] function based index problem Hi, on table entry (17M records) there is

Re: [SQL] function based index problem

2011-08-31 Thread Viktor Bojović
On Thu, Sep 1, 2011 at 12:09 AM, David Johnston pol...@yahoo.com wrote: ** ** *From:* pgsql-sql-ow...@postgresql.org [mailto: pgsql-sql-ow...@postgresql.org] *On Behalf Of *Viktor Bojovic *Sent:* Wednesday, August 31, 2011 5:27 PM *To:* pgsql-sql@postgresql.org; pgsql-ad...@postgresql.org