Showing posts with label contains. Show all posts
Showing posts with label contains. Show all posts

Thursday, March 22, 2012

A clause of the query contained only ignored words

Hi all,
I am getting this error when my query contains a word like "of":
"A clause of the query contained only ignored words."
The SQL statement uses the CONTAINS clause. This used to work fine on SQL
Server 2000, but I recently switched to SQL Server 2003 and can't get it to
work. I don't want to have any ignored words. I tried to rename all noise.*
files into _noise.* and then rebuilt the Full-Text Catalog, but that did not
fix it.
Anything else I need to do to get rid of all noise words?
Thanks.
Abe
http://www.indexserverfaq.com/noise.htm
Hilary Cotter
Looking for a SQL Server replication book?
http://www.nwsu.com/0974973602.html
Looking for a FAQ on Indexing Services/SQL FTS
http://www.indexserverfaq.com
"Abe Simpson" <abe@.simpson.com> wrote in message
news:%23QMJJHpGFHA.3916@.TK2MSFTNGP12.phx.gbl...
> Hi all,
> I am getting this error when my query contains a word like "of":
> "A clause of the query contained only ignored words."
> The SQL statement uses the CONTAINS clause. This used to work fine on SQL
> Server 2000, but I recently switched to SQL Server 2003 and can't get it
to
> work. I don't want to have any ignored words. I tried to rename all
noise.*
> files into _noise.* and then rebuilt the Full-Text Catalog, but that did
not
> fix it.
> Anything else I need to do to get rid of all noise words?
> Thanks.
> Abe
>
|||Able,
Could you post the full output of the following SQL code as it would be most
helpful information in helping you.
use <your_database_name_here>
go
SELECT @.@.language
SELECT @.@.version
EXEC sp_help_fulltext_columns
EXEC sp_help <your_FT-enable_table_name_here>
go
You are getting this error because "of" is considered a noise or stop word
that is maintained in the language-specific noise word files under
\FTDATA\SQLServer\Config\noise.* where * is a three-letter code for your
FT-enable column's "Language for Word Breaker". Specifically, noise.enu = US
English and you can edit this file with notepad.exe and remove "of" and
other words and single letters and digits, but leave at least a single
space, if you decide to remove all noise words. Note, you will have to stop
the "Microsoft Search" (MSSearch.exe) service before saving the file and
then run a Full Population on all of your FT Catalog for the change to be
effective.
Note, there is no such version as "SQL Server 2003". Are you perhaps
refereing to SQL Server 2005 (codename Yukon)?
If so, the noise word files are maintained in a different location
(\MSSQL.1\MSSQL\Binn\FTERef) and with different file names, such as
noiseENU.txt for US English. You will also need to stop a different service
MSFTESQL and then run a Full Population. If you are not using SQL Server
2005, could you post the full output of SELECT @.@.version ?
Thanks,
John
SQL Full Text Search Blog
http://spaces.msn.com/members/jtkane/
"Abe Simpson" <abe@.simpson.com> wrote in message
news:#QMJJHpGFHA.3916@.TK2MSFTNGP12.phx.gbl...
> Hi all,
> I am getting this error when my query contains a word like "of":
> "A clause of the query contained only ignored words."
> The SQL statement uses the CONTAINS clause. This used to work fine on SQL
> Server 2000, but I recently switched to SQL Server 2003 and can't get it
to
> work. I don't want to have any ignored words. I tried to rename all
noise.*
> files into _noise.* and then rebuilt the Full-Text Catalog, but that did
not
> fix it.
> Anything else I need to do to get rid of all noise words?
> Thanks.
> Abe
>
sql

Sunday, March 11, 2012

7619 error message when you run a full text query that contains certain Turkish character

I receive a 7619 error message when I run a full text query that
contains certain Turkish character strings on an instance of SQL Server
2000 that is running on a Windows 2000-based computer. How can I solve
this problem?
The parser is interpreting these characters as noise or noise words. Which
word breaker are you using? I am assuming you are using the neutral word
breaker.
If this is the case the problem could be in your code page setting in the
web page. You should set the session.codepage to match the lcid for Turkish.
Hilary Cotter
Looking for a SQL Server replication book?
http://www.nwsu.com/0974973602.html
Looking for a FAQ on Indexing Services/SQL FTS
http://www.indexserverfaq.com
"Ismail H. AYDIN" <dicisay@.hotmail.com> wrote in message
news:1117203863.089340.322710@.g49g2000cwa.googlegr oups.com...
>I receive a 7619 error message when I run a full text query that
> contains certain Turkish character strings on an instance of SQL Server
> 2000 that is running on a Windows 2000-based computer. How can I solve
> this problem?
>
|||I am running the query from SQL Query Analyzer and then error occurs.
Not from asp.net or asp page. The noise word is 'Ak'. And the other
noise charactes are '', '', ''.
*** Sent via Developersdex http://www.codecomments.com ***

Saturday, February 11, 2012

4 lookups against single large table

I need to do a 4 column lookup against a large table (1 Million rows) that contains 4 different record types. The first lookup will match on colums A, B, C, and D. If no match is found, I try again with colums A, B, C, and '99' in column D. If no match, try again with column A, B, D, and '99' in Column C. Finally, if no match in any of the above, use column A, '99' in B, '99' in C, '99' in D. I will retreive 2 columns from the lookup table.

My thought is that breaking this sequence out into 4 different tables/ lookups would be most efficient. The other option would be to write a script that handled this logic in a single transform with an in-memory table. My concern is that the size of the table would be too large to load into memory.

Any ideas/suggestions would be appreciated.

You only need to pull in the 4 columns from the table that you will be looking up against. Assuming they are normal integers (i.e. 4 bytes) you would need 4 * 4 * 1000000 =16MB per lookup. That's not really all that much. My advice would be to give it a go and if you're havig problems - see where the bottlenecks are.

More on memory per package: http://blogs.conchango.com/jamiethomson/archive/2005/05/29/1486.aspx

-Jamie