It is a bug?

2002-12-24 Thread Walter Procopio
s.rstpg, ecoras.rsdat, ecoras.rsrub
FROM   ecoras,  ecotxt
LEFT JOIN ecotes ON ecoras.rstcd = ecotes.rstcd
LEFT JOIN ecorub ON ecoras.rsnum = ecorub.rsnum and ecoras.rsrub =
ecorub.rsrub
WHEREecoras.rsnum = 0
AND ecoras.rsori = ecotxt.rsori
AND ecoras.rscod = ecotxt.rscod
AND (rsdat >= '2002-12-01') and (rsdat <= '2002-12-23')
AND (match(ecotxt.rstxt) against('JAVA' IN BOOLEAN MODE))
ORDER BY ecoras.rspst, ecoras.rstda, ecotes.rstes, ecoras.rstpg,
ecoras.rsart


In this query the where-clause matches the primary key (index) of ecoras
table:


ecoras.rsnum = 0
AND ecoras.rsori = ecotxt.rsori
AND ecoras.rscod = ecotxt.rscod


If I do not use an index, I will obtain slow performances, but I must have
always a valid resultset like in the second query.

Is it a bug or a join-condition has to match an index?

Thanks and best regard
Walter Procopio


-
Before posting, please check:
   http://www.mysql.com/manual.php   (the manual)
   http://lists.mysql.com/   (the list archive)

To request this thread, e-mail <[EMAIL PROTECTED]>
To unsubscribe, e-mail <[EMAIL PROTECTED]>
Trouble unsubscribing? Try: http://lists.mysql.com/php/unsubscribe.php




Is it a bug?

2002-12-30 Thread Walter Procopio
s.rstpg, ecoras.rsdat, ecoras.rsrub
FROM   ecoras,  ecotxt
LEFT JOIN ecotes ON ecoras.rstcd = ecotes.rstcd
LEFT JOIN ecorub ON ecoras.rsnum = ecorub.rsnum and ecoras.rsrub =
ecorub.rsrub
WHEREecoras.rsnum = 0
AND ecoras.rsori = ecotxt.rsori
AND ecoras.rscod = ecotxt.rscod
AND (rsdat >= '2002-12-01') and (rsdat <= '2002-12-23')
AND (match(ecotxt.rstxt) against('JAVA' IN BOOLEAN MODE))
ORDER BY ecoras.rspst, ecoras.rstda, ecotes.rstes, ecoras.rstpg,
ecoras.rsart


In this query the where-clause matches the primary key (index) of ecoras
table:


ecoras.rsnum = 0
AND ecoras.rsori = ecotxt.rsori
AND ecoras.rscod = ecotxt.rscod


If I do not use an index, I will obtain slow performances, but I must have
always a
valid resultset like in the second query.

Is it a bug or a join-condition has to match an index?

Thanks and best regard
Walter Procopio

-
Before posting, please check:
   http://www.mysql.com/manual.php   (the manual)
   http://lists.mysql.com/   (the list archive)

To request this thread, e-mail <[EMAIL PROTECTED]>
To unsubscribe, e-mail <[EMAIL PROTECTED]>
Trouble unsubscribing? Try: http://lists.mysql.com/php/unsubscribe.php




Is it a bug

2002-12-31 Thread Walter Procopio
s.rstpg, ecoras.rsdat, ecoras.rsrub
FROM   ecoras,  ecotxt
LEFT JOIN ecotes ON ecoras.rstcd = ecotes.rstcd
LEFT JOIN ecorub ON ecoras.rsnum = ecorub.rsnum and ecoras.rsrub =
ecorub.rsrub
WHEREecoras.rsnum = 0
AND ecoras.rsori = ecotxt.rsori
AND ecoras.rscod = ecotxt.rscod
AND (rsdat >= '2002-12-01') and (rsdat <= '2002-12-23')
AND (match(ecotxt.rstxt) against('JAVA' IN BOOLEAN MODE))
ORDER BY ecoras.rspst, ecoras.rstda, ecotes.rstes, ecoras.rstpg,
ecoras.rsart


In this query the where-clause matches the primary key (index) of ecoras
table:


ecoras.rsnum = 0
AND ecoras.rsori = ecotxt.rsori
AND ecoras.rscod = ecotxt.rscod


If I do not use an index, I will obtain slow performances, but I must have
always a
valid resultset like in the second query.

Is it a bug or a join-condition has to match an index?

Thanks and best regard
Walter Procopio 


-
Before posting, please check:
   http://www.mysql.com/manual.php   (the manual)
   http://lists.mysql.com/   (the list archive)

To request this thread, e-mail <[EMAIL PROTECTED]>
To unsubscribe, e-mail <[EMAIL PROTECTED]>
Trouble unsubscribing? Try: http://lists.mysql.com/php/unsubscribe.php




re: Is it a bug

2003-01-07 Thread Walter Procopio
> On Tuesday 31 December 2002 10:21, Walter Procopio wrote:
>
> > I have installed mysql max 4.0.4 beta on a Suse (ver. 8.0) linux server
and
> > I have created a database with this schema:
>
> [skip]
>
> >
> > And I try to execute this query:
> >
> >
> > SELECTecoras.rstda, ecotes.rstes, ecoras.rsart, ecoras.rsimm,
> > ecorub.rsdes, ecoras.rstpg, ecoras.rsdat, ecoras.rsrub
> > FROM   ecoras,  ecotxt
> > LEFT JOIN ecotes ON ecoras.rstcd = ecotes.rstcd
> > LEFT JOIN ecorub ON ecoras.rsnum = ecorub.rsnum and ecoras.rsrub =
> > ecorub.rsrub
> > WHEREecoras.rsori = ecotxt.rsori
> > AND ecoras.rscod = ecotxt.rscod
> > AND (rsdat >= '2002-12-01') and (rsdat <= '2002-12-23')
> > AND (match(ecotxt.rstxt) against('JAVA' IN BOOLEAN
MODE))
> > ORDER BY ecoras.rspst, ecoras.rstda, ecotes.rstes, ecoras.rstpg,
> > ecoras.rsart
> >
> >
> > I have no data in my resultset, but data-base contains data which match
> > conditions.
> > If I excute the following query I have a valid resultset.
> >
> >
> > SELECTecoras.rstda, ecotes.rstes, ecoras.rsart, ecoras.rsimm,
> > ecorub.rsdes, ecoras.rstpg, ecoras.rsdat, ecoras.rsrub
> > FROM   ecoras,  ecotxt
> > LEFT JOIN ecotes ON ecoras.rstcd = ecotes.rstcd
> > LEFT JOIN ecorub ON ecoras.rsnum = ecorub.rsnum and ecoras.rsrub =
> > ecorub.rsrub
> > WHEREecoras.rsnum = 0
> > AND ecoras.rsori = ecotxt.rsori
> > AND ecoras.rscod = ecotxt.rscod
> > AND (rsdat >= '2002-12-01') and (rsdat <= '2002-12-23')
> > AND (match(ecotxt.rstxt) against('JAVA' IN BOOLEAN
MODE))
> > ORDER BY ecoras.rspst, ecoras.rstda, ecotes.rstes, ecoras.rstpg,
> > ecoras.rsart
> >
> >
> > In this query the where-clause matches the primary key (index) of ecoras
> > table:
> >
> >
> > ecoras.rsnum = 0
> > AND ecoras.rsori = ecotxt.rsori
> > AND ecoras.rscod = ecotxt.rscod
> >
> >
> > If I do not use an index, I will obtain slow performances, but I must
have
> > always a
> > valid resultset like in the second query.
> >
> > Is it a bug or a join-condition has to match an index?
>
> Could you test it with latest version of MySQL server (4.0.7)?
> Could you provide some data for testing?
>
>
>

I have tried the release 4.0.7 and it is ok. It solves another problem I had
using full text query.

Thank you.


-
Before posting, please check:
   http://www.mysql.com/manual.php   (the manual)
   http://lists.mysql.com/   (the list archive)

To request this thread, e-mail <[EMAIL PROTECTED]>
To unsubscribe, e-mail <[EMAIL PROTECTED]>
Trouble unsubscribing? Try: http://lists.mysql.com/php/unsubscribe.php




Full text limitation?

2003-01-08 Thread Walter Procopio
Hi,

I have installed mysql max 4.0.7 gamma on a Suse (ver. 8.0 professional)
linux server. I have tried many full text queries and I had only a problem
with a query like this:

SELECT count(a.rsori) pippo
FROM ecoras a, ecotxt b
WHERE MATCH ( b.rstxt ) AGAINST ( '-linux' IN BOOLEAN MODE )
AND a.rsori = b.rsori
AND a.rscod = b.rscod;

If I use only the minus operator I have not a valid resultset. If I use the
minus operator with another one I have a reultset.
For example:

SELECT count(a.rsori) pippo
FROM ecoras a, ecotxt b
WHERE MATCH ( b.rstxt ) AGAINST ( '+linux -windows' IN BOOLEAN MODE )
AND a.rsori = b.rsori
AND a.rscod = b.rscod;

Is it a full text limit?

Thanks
Walter Procopio


-
Before posting, please check:
   http://www.mysql.com/manual.php   (the manual)
   http://lists.mysql.com/   (the list archive)

To request this thread, e-mail <[EMAIL PROTECTED]>
To unsubscribe, e-mail <[EMAIL PROTECTED]>
Trouble unsubscribing? Try: http://lists.mysql.com/php/unsubscribe.php




Re: Full text limitation?

2003-01-09 Thread Walter Procopio
- Original Message -
From: "Paul DuBois" <[EMAIL PROTECTED]>
To: "Walter Procopio" <[EMAIL PROTECTED]>;
<[EMAIL PROTECTED]>
Sent: Wednesday, January 08, 2003 5:44 PM
Subject: Re: Full text limitation?


> At 14:03 +0100 1/8/03, Walter Procopio wrote:
> >Hi,
> >
> >I have installed mysql max 4.0.7 gamma on a Suse (ver. 8.0 professional)
> >linux server. I have tried many full text queries and I had only a
problem
> >with a query like this:
> >
> >SELECT count(a.rsori) pippo
> >FROM ecoras a, ecotxt b
> >WHERE MATCH ( b.rstxt ) AGAINST ( '-linux' IN BOOLEAN MODE )
> >AND a.rsori = b.rsori
> >AND a.rscod = b.rscod;
> >
> >If I use only the minus operator I have not a valid resultset.
>
> Isn't that the same as asking for records that don't match anything?
>


I think I was not exaustive, in mysql handbook I can read:


+A leading plus sign indicates that this word must be present in every
row returned.

- A leading minus sign indicates that this word must not be present in
any row returned.
  By default (when neither plus nor minus is specified) the word is
optional, but the rows that contain it will be rated higher.
  This mimicks the behaviour of MATCH() ... AGAINST() without the IN
BOOLEAN MODE modifier.

If I execute a query like in the example, I don't obtain all rows where this
word is not present. Instead if I use a leading plus sign I obtain the right
resultset.
I do not think it is a bug, maybe it is a limitation in full-text search and
I would know how can I obtain a valid resultset.


> >  If I use the
> >minus operator with another one I have a reultset.
> >For example:
> >
> >SELECT count(a.rsori) pippo
> >FROM ecoras a, ecotxt b
> >WHERE MATCH ( b.rstxt ) AGAINST ( '+linux -windows' IN BOOLEAN MODE )
> >AND a.rsori = b.rsori
> >AND a.rscod = b.rscod;
> >
> >Is it a full text limit?
> >
> >Thanks
> >Walter Procopio
>


-
Before posting, please check:
   http://www.mysql.com/manual.php   (the manual)
   http://lists.mysql.com/   (the list archive)

To request this thread, e-mail <[EMAIL PROTECTED]>
To unsubscribe, e-mail <[EMAIL PROTECTED]>
Trouble unsubscribing? Try: http://lists.mysql.com/php/unsubscribe.php