RE: [SQL] Re: Matching and Scoring with multiple fields

Thanks to all of you that replied. I think Oliver's idea (which is pretty close to Stephan's) will probably do the trick I think. I will maybe look in the future to add the ability to allow users to weight fields with more priority. So customers could number the top five most important fields and

RE: [SQL] join if there, blank if not

Jacques, The problem with using the union in this way is that you get NULLs for a number weather or not it has an associated record in calls. To do a pure outer join, it would be something like this: select c.cdate, c.ctime, c.cextn, c.cnumber, n.ndesc from calls c, numbers n where c.cnumber=n.

Re: [SQL] importing in sql

karim> The file has the following format karim> karim> field2|field3|field4|field5| karim> karim> Each field is separated by a | (pipe), How do I insert the sequence karim> number before field2? I have like 30 files in this format and I need to karim> import them in one table with each line havi

[SQL] importing in sql

Hi, I am trying to import loads of data into postgres, I am having trouble with a field which is a sequence. I know how to import data using a delimiter with the copy command. But I need to know how I increase the sequence on each line of import. My Table has the following format. field1ser

Re: [SQL] importing in sql

On Mon, Jul 10, 2000 at 05:03:08PM -0500, Abdul Karim wrote: > Hi, I am trying to import loads of data into postgres, I am having > trouble with a field which is a sequence. I know how to import data > using a delimiter with the copy command. But I need to know how I > increase the sequence on ea

[SQL] Bug in to_char()

Greetings, Working with PostGreSQL 7.02, I found the following problem: The AM/PM designator in the to_char function does not work proper for 13:00 and 12:00. See the following: test=> select to_char('3-12-2000 14:00'::timestamp, 'Dy, HH12:MI PM'); to_char --- Sun, 02:00 P

Re: [SQL] join if there, blank if not

Gary, What you want here is an outer join. The syntax would look something like this: select c.cdate, c.ctime, c.cextn, c.cnumber, n.ndesc from calls c, numbers n where c.cnumber=n.nnumber union all select null as cdate, null as cextn, null as cnumber, n.nnumber, n.ndesc from numbers; (I haven'

[SQL] SQL

Hi, I'd like an information about the history of SQL. The similarities and differences between SQL1 and SQL2 and SQL3. Thank you, Frederico Papatella Guerino Belo Horizonte-MG/ Brasil [EMAIL PROTECTED]

Re: [SQL] Opposite of LOCK

Antti Linno <[EMAIL PROTECTED]> writes: > Is there any possibility to unlock tables after they've been locked? The > manual part of lock was fuzzy, so I ask from experts instead. MySQL > uses lock and unlock. Locks are released at transaction commit/abort. It has to be that way to preserve trans

Re: [SQL] date comparision ???

On Mon, Jul 10, 2000 at 11:52:16AM +0300, sandis wrote: > Thanks for your input. Unfortunately, it doesn't helped.. > > Here is the samples. > > This query works fine: > SELECT datums_ FROM jaunumi WHERE flag = 'a' AND date_part('year',datetime > '2000-06-02 06:11:01-07') = '2000' LIMIT 1; > dat

Re: [SQL] date comparision ???

Thanks for your input. Unfortunately, it doesn't helped.. Here is the samples. This query works fine: SELECT datums_ FROM jaunumi WHERE flag = 'a' AND date_part('year',datetime '2000-06-02 06:11:01-07') = '2000' LIMIT 1; datums_ -- 2000-07-06 18:51:27+03 (1 row) But this fai

[SQL] join if there, blank if not

Hi all, I just can't get my head round this one so I hope one of you can. I've got two tables, one holding phone calls, and another holding phone numbers. I want to do a select where if the number in the calls table exists in the numbers table, the description is included otherwise the descri

Re: [SQL] SERIAL type does not generate new ID ?

On Sat, 10 Jun 2000, Jean-Marc Libs wrote: Hi all, > I don't really understand what happens, so I put context, then problem: > > 1/ Context > -- > I have this table: > > CREATE TABLE film ( > film_id SERIAL PRIMARY KEY, snip > ) > ; > SELECT setval ('film_film_id_seq', 6); > > 2/ Prob

Re: [SQL] Opposite of LOCK

Antti Linno wrote: > Is there any possibility to unlock tables after they've been locked? The > manual part of lock was fuzzy, so I ask from experts instead. MySQL > uses lock and unlock. Yes, COMMIT/ROLLBACK. The transactional concept implies that you hold each lock you accquired si

[SQL] Opposite of LOCK

Is there any possibility to unlock tables after they've been locked? The manual part of lock was fuzzy, so I ask from experts instead. MySQL uses lock and unlock. Antti.