15 characters) and the same (even the spaces).Removing one
and
> the index on varchar was created. Any idea why?
>
> I will try to play with the indexes, if I understand well I'd better
index
> the three columns in once, because that will automatically index each of
> t
e columns in once, because that will automatically index each of
them?or am I wrong?
Anyway, thank you for all your advices, I'm really learning a lot of things
with that case!
Melanie
From: [EMAIL PROTECTED]
To: "mel list_php" <[EMAIL PROTECTED]>
CC: mysql@lists
"mel list_php" <[EMAIL PROTECTED]> wrote on 03/18/2005 10:35:30 AM:
> Hi again,
>
> Thanks for the explanation about the join and the group by.
>
> I wanted to test your query (almost a simple copy/paste :-)) ).
>
> The first 2 queries are ok, but the third one still is too long :
> mysql> CREA
n matches and the other tables is still too
heavy
Any idea?
From: [EMAIL PROTECTED]
To: "mel list_php" <[EMAIL PROTECTED]>
CC: mysql@lists.mysql.com
Subject: Re: help on query/group by
Date: Fri, 18 Mar 2005 09:14:02 -0500
Responses embedded below
"mel list_php"
Responses embedded below
"mel list_php" <[EMAIL PROTECTED]> wrote on 03/18/2005 05:57:29 AM:
> Hi Shawn,
>
> Thank you very much, I'm impressed by the time you took to answer me,
and
> the quality of the reply!!!
> I forwarded the answer to my friend.
> I'm wondering, I knew the mechanism of t
n thank you very much for your help, I will give temporary tables
an other chance!!!
Melanie
From: [EMAIL PROTECTED]
To: "mel list_php" <[EMAIL PROTECTED]>
CC: mysql@lists.mysql.com
Subject: Re: help on query/group by
Date: Wed, 16 Mar 2005 13:52:44 -0500
I have a favorite techn
I have a favorite technique for improving the results of queries , like
this, which involve fairly large JOINed tables. It's a form of
"divide-and-conquer" in that you pre-compute what you can then make the
JOINS you need to finish up the results.
I agree that the "AND s2.syn LIKE '%'" in the O
Hi,
A friend of mine asked me to have a look at one of his query, and I'm
stuck
Here was his query:
SELECT drugID, protID, COUNT(DISTINCT pmid),
MAX(s1.syn) AS o1, MAX(s2.syn) AS o2
FROM matches
INNER JOIN synonyms AS s1 ON drugID=s1.nameID AND s1.syn LIKE 'a%'
INNER JOIN synonyms AS s2
