ConText Cartridge

i have created my own oracle database. i want to use context searching in it. but the user=ctxsys is not working in my database. how do i create this user in my database and what privileges should be given to this user.

<BLOCKQUOTE><font size="1" face="Verdana, Arial, Helvetica">quote:</font><HR>Originally posted by Meetu Srivastava ([email protected]):
i have created my own oracle database. i want to use context searching in it. but the user=ctxsys is not working in my database. how do i create this user in my database and what privileges should be given to this user.<HR></BLOCKQUOTE>
Assuming you're using 8.0.x on NT:
0. Startup your Context NT service: OracleConTextService80
1. Startup your database
2. Launch the Database Assistant utility.
Choose your database name
Choose Modify
Choose Context
Click Finish.
User CTXSYS with CTXSYS passwd created.
Sokun
null

Similar Messages

Free text retrieval with Context Cartridge

I am wondering whether it would a problem to create a Java
application doing free text retrieval via Context Cartridge.
I mean by that :
1. JDBC call via thin Oracle driver with CONTAINS in WHERE
clause. Is it supposed to work ?
2. Assuming we have several hits in a BLOB can we provide
functionality ( next hit prev hit ) in textarea field for Java
frame.
null

This one was forwarded to JDBC experts. Two experts (A and B)
replied. Their answers are below.
Michael Mitiaguin (guest) wrote:
I am wondering whether it would a problem to create a Java
application doing free text retrieval via Context Cartridge.
I mean by that :
1. JDBC call via thin Oracle driver with CONTAINS in WHERE
clause. Is it supposed to work ?
A. YES.
B. There is no limitation in JDBC of using CONTAINS in WHERE
clause.
2. Assuming we have several hits in a BLOB can we provide
functionality ( next hit prev hit ) in textarea field for Java
frame.
A. I don't really get what you mean. If you mean that you want to
map a Java Frame to a BLOB column and depending upon the hit in
the Java frame, you want to read the data from a certain position
(offset), then yes. If you are using 8.0.X drivers, you can use
the dbms_lob.read call. If you are using 8.1.X drivers, you can
use Blob.getBytes call.
B. It can be achieved by creating a wrapper of using the
following method in oracle.sql.BLOB in 8.1 driver.
public long position(byte[] pattern,
long start)
throws SQLException
Determines the byte position at which the specified byte pattern
begins within the BLOB value that this Blob object represents.
Begin search at position start. Parameters:
pattern - the byte array for which to search
start - the position at which to begin searching; the first
position is 1
Returns: the position at which the pattern appears, else -1.
Throws: SQLException - if there is an error accessing the BLOB
null

Context Cartridge for Oracle 8.0

Would someone please recommend a good source for design and
development with the Oracle Context Cartridge for Oracle 8. Our
pharamaceutical company requires access to text based documents
and key word in context searches through a web enabled
application suite. Does anyone know of any Oracle courses in
this area?
Thanks for the effort...
Henri Tuthill
null

The Oracle8 Context Cartridge Quickstart guide is the best source
to get acquainted with the Context option. You can search and
download this doc from the support site or email me if you have
any questions.
Vivek
HENRI TUTHILL (guest) wrote:
: Would someone please point me to a good source for developing
: applications that must utilize the context cartridge on oracle
: 8.0.5 RDBMS. Extensive text searches required in this
: application based on key word in context.
: regards and thanks...
: Henri T.
null

Context Cartridge for Oracle 8.0.6 for Solaris

6/21/00
I need some help with tuning Context Text queries a SQL query.
My query is of the format:
SELECT /*+ ORDERED USE_NL(T1) */ distinct T1.column4,sum(SCORE(10))
FROM Table1 T1, Table2 T2
WHERE T1.column_with_bit_map_index = 1 AND -- column is either 1 or 0
T1.column2_with_regular_index=2 AND
T2.column3_with_regular_index= T1.column3
AND CONTAINS(T2.LONG_Column1,'{a_frequently_occurring_word}',10)>0
GROUP BY T1.column4
ORDER BY 1 DESC
T1 has 600,000 rows, T2 has 100,000 rows.
My query takes 14 seconds if the CONTAINS clause is looking for a_frequently_occurring_word. If the CONTAINS cluase looks instead for a_rarely_occurring_word, the query response time is 4 seconds!
If I remove the CONTAINS clause, the query comes back in 4 seconds, which leads me to believ the best performance gain wil come from tuning the Context part of the query, not the regular SQL part of the query.
I followed the Context doc and attempted to tune Context as follows:
1. Created 20 QDML servers using ctxsys
2. Created a policy and index (CONTEXT_index1) for T2.LONG_Column1
3. Used exec. ctx_ddl.optimize_index('CONTEXT_index1',ctx_ddl.defragment_to_new_table);
4. Used exec ctx_adm.set_query_buffer_size(1000000);
There was no improvement in the query response time, i.e. it was still 14 seconds.
Any help would be greatly appreciated.
Arjuna Jayasinha
[email protected]

First, export should not be a primary method of backing up a database. At most, exports should compliment a physical backup process. There are a variety of ways to handle incremental backups when you are doing physical backups.
Second, there is no useful way to do an incremental export. Later versions of export (I don't think 8.0.6 had this) have a "incremental" option, but that will export all the rows of a table if any row changes or is added to the table since the last incremental export. For 99.9% of the people out there, this renders the option practically useless.
Justin
Distributed Database Consulting, Inc.
http://www.ddbcinc.com/askDDBC

ConText Cartridge not working correctly

im using Context to search a substring within a String . the String is column of type varchar2(5000). these Strings don't have any space between characters. Each string is some thing of type -- AVJHGFUEWBFJSGISGFJSDCJSHCLIHCNCCHDYWDJDHEUGFUEHUHGDSUNCDSHCUSHSNCKSHIFINCKCISCHCN...
if i want to search say a substring HCN , the hitlist should contain all the strings which have this substring. but the hitlist returned by the ConText is not correct. it returns only those strings which have this substring in first 32 characters.
i have checked the index created by ConText, for each string it has stored first 32 characters in the WORDLIST. im using Oracle Enterprise 8.03 on NT.

Context is really designed for word searching in a document and therefore you need to have a 'definition' of a word which you do not have. I suspect that there may be a limit of 32chs at the relaese you are working on. The Inter Media table (inter media is Context's successor at 8.1) which contains tokens only defines the length as 64 bytes and I cannot see any documented way of exceeding this.
Could you not just use an ordinary index?

ConText Viewer

Hy!
I'm trying to get the ctxv32.ocx control for VB6 from the ConText Cartridge. It does not appear to be in the installation options of Oracle 8i enterprise edition. Is there another way to install it. It would be greatly appreaciated if any one could submit his knowledge on this isue.
Thanks,

Hi Ashok,
It might be too late for your specific task but maybe the following information could help others: Meanwhile, there is an official How-to Guide that describes how to integrate BCV in your own Web Dynpro application. It can be found here:
[http://www.sdn.sap.com/irj/scn/index?rid=/library/uuid/b05d79aa-a0b3-2e10-b0be-dade61266534]
Regards, Sven

Context Indexing creation error...

I cannot use context cartridge...
When I try to create index, i receive following error message:
SQL> create index myindex on x(name) indextype is ctxsys.context
2 parameters ('lexer mylexer wordlist mywordlist stoplist mystoplist');
create index myindex on x(name) indextype is ctxsys.context
ERROR at line 1:
ORA-29855: error occurred in the execution of ODCIINDEXCREATE routine
ORA-20000: ConText error:
ORA-06520: PL/SQL: Error loading external library
ORA-06522: z: cannot open shared object file: No such file or directory
ORA-06512: at "CTXSYS.DRUE", line 122
ORA-06512: at "CTXSYS.TEXTINDEXMETHODS", line 34
ORA-06512: at line 1
I listed below the libraries Oracle Context have installed:
SQL> select library_name, file_spec from dba_libraries where library_name
like 'DR%'
LIBRARY_NAME FILE_SPEC
DR$LIB
DR$LIBX z
As you can see these libraries have strange file_specs, but i dont know what right file_specs i must use...
Please help...
Best regards
Oleg

But the status of the context index is VALID on the view USER_INDEXES...
Besides lsnrctl says following about extproc:
LSNRCTL> status e
Connecting to (ADDRESS=(PROTOCOL=IPC)(KEY=EXTPROCKEY))
STATUS of the LISTENER
Alias e
Version TNSLSNR for Linux: Version 8.1.5.0.0 - Production
Start Date 23-OCT-00 13:19:31
Uptime 1 days 2 hr. 21 min. 49 sec
Trace Level off
Security OFF
SNMP OFF
Listener Parameter File /pt01/app/oracle/product/8.1.5/network/admin/listener.ora
Listener Log File /pt01/app/oracle/product/8.1.5/network/log/e.log
Services Summary...
extproc has 1 service handler(s)
The command completed successfully
So, all is right with extproc...

Has Anyone gotten the Intelligent Agent to work???

I have Oracle installed and all is well except for a few minor
problems that I think the Intelligent Agent will solve. Has
anyone gotten this to work on RedHat 6.0 with Oracle
Enterprise Edition 8.0.5.1??? I was able to get it to work
under Oracle8 SE, but I need EE for the ConText Cartridge.
I have gotten to the point where I can get the dbsnmp process
running, but OEM cannot talk to it.
Please help
--Bryan
null

I'm having the same problem. Does anyone out there have a
solution?
I am running RedHat Linux 6.0 straight out of the box. The only
change I made was to set the kernel parameters according to the
release notes and recompile the kernel.
P.S. Oracle deserves our thanks for providing free developer
licenses to so many of its products, but this is the buggiest
install I've ever seen in my 12 years of working with their
products. Didn't anyone from Oracle test the install? This
problem and the zero length libnjni8.so file are the real
doozies.
I've followed the release notes and installation guide to the
letter and have found several documentation / distribution bugs
that should have leapt out at the first person who tried it.
Thorsten Khnemann (guest) wrote:
: Bryan Batchelder (guest) wrote:
: : anyone else who could not get it to compile with TCL 7.5 to
: : please at least let me know other people are having the same
: : problem.
: Same problem here, the
: "gmake -f ins_oemagent.mk agent" gives the message
: "gcc: unrecognized option
: `-R/home/oracle/run/product/8.1.5/lib/'" and hundreds
: of unresolved references.
: Tested on RedHat 6.0, egcs 1.1.2
null

XML support in Oracle 8i

Hi,
I have a requirement of storing XML documents in the Oracle
database. As Oracle claims support for XML in Oracle 8i, I would
be thankful if the following questions could be answered.
There will be around 15000 or more XML documents generated
annually, each of around 200 pages . Various views of the same
XML instance will be shown to different customers depending on
their preferences. All views will be generated dynamically and
the results will be sent as HTML pages to the client or XML
instances which will then be manipulated and shown as required
on the clients machine.
Please bear with me in case a few questions are naive
Questions :
1) Does Oracle 8i propose to have XML as a new datatype
2) Does a document descriptor mean the mapping of the XML
elements to certain columns in an Oracle table? Does it have
anything to do with the DTD? Will an XML instance be mapped to a
single table only or across multiple tables in a master-detail
relationship as the heirarchy in the instance gets too complex ?
3) How will the ConText cartridge be able to identify XML
documents? How will the queries be written to enable 'element-
level' searching?
4)What is iFS , especially in relation to XML storage?
Regards,
Kedar
null

Kedar Sadekar (guest) wrote:
: 1) Does Oracle 8i propose to have XML as a new datatype
Not at this time.
: 2) Does a document descriptor mean the mapping of the XML
: elements to certain columns in an Oracle table? Does it have
: anything to do with the DTD? Will an XML instance be mapped to
:a single table only or across multiple tables in a master-detail
: relationship as the heirarchy in the instance gets too complex?
A Document Descriptor is basicly a mapping between a DTD and the
tables and columns in the RDBMS. More that one table can be
mapped to.
: 3) How will the ConText cartridge be able to identify XML
: documents? How will the queries be written to enable 'element-
: level' searching?
Context, now interMedia Text, does not 'identify' XML documents
per se, but can index colums with imbedded XML and search the
contents between specified tags.
: 4)What is iFS , especially in relation to XML storage?
Check out http://www.oracle.com/xml/documents/xml_twp/ for a
complete description.
Oracle XML Team
http://technet.oracle.com
Oracle Technology Network
null

Wich is correct way to handle Workflow of documents ?

We are starting a project to handle Workflow of documents.
We have an Enterprise with some Users that edit and publish documents;
and related subcompanies that have Users that read only the documents.
We have to handle the revisions of documents, the process phases and the validation
of a revisor for each phase and also control the document access.
Our backend database is Oracle 8 for Wordgroup (in the Enterprise) and IIS for the client access (from subcompanies).
We want to use this architecture : client (HTML) and server (servlet, JSP).
These documents are written with Word97 and are stored in Oracle 8.
We plan in the future on updating Oracle 8 FWG > Oracle8 Enterprise > Oracle8i and the migration from IIS (NT Web Server) to (UNIX Solaris).
My questions are:
1) How control the open and save of documents with Oracle connection ?
2) Is better to store the documents inside Oracle or just insert the URL in tables ?
3) If I want to use ConText cartridge for searching mechanism where I have to store
this documents ?
4) For the servlets I need an Application Server ? Wich release ?
Could you help me to get the correct solution. I would appreciate any suggestions.
Thanks
Lorenzo Baldovini.
[email protected]

Hi,
You really need to take a look at the XMLDB Developers guide.
For updating XML with SQL/XML see UPDATEXML and for XQuery see [Using XQuery with Oracle XMLDB|http://download.oracle.com/docs/cd/B19306_01/appdev.102/b14259/xdb_xquery.htm#sthref1673]
HTH,
Chris

Intermedia and Oracle8i

Hi,
is InterMedia Context package included in oracle 8i ?
What is the name of this package ?
There is some documentations about (I have just see the URL
given same Thread before...)?
The problem always insert and search xml documents into the dbms.
TIA,
pietro
null

The previous Context Cartridge is now called interMedia-Text and
is part of the Oracle interMedia option for 8i. Details are at
http://www.oracle.com/database/options/intermedia.html.
The interMedia-Text cartridge will not help you insert XML
documents into the database, only search them.
Oracle XML Team
http://technet.oracle.com
Oracle Technology Network
pietro (guest) wrote:
: Hi,
: is InterMedia Context package included in oracle 8i ?
: What is the name of this package ?
: There is some documentations about (I have just see the URL
: given same Thread before...)?
: The problem always insert and search xml documents into the
dbms.
: TIA,
: pietro
null

Case statement support in CPL XML

Hello,
I'm trying to find a way of implementing a case statement based on H.323_ID of the destination in the already existing and working CPL script on VCS X7.1. The problem is, I can't find any guidance anywhere in the Web. Is the case statement supported at all?
Thanks!

Kedar Sadekar (guest) wrote:
: 1) Does Oracle 8i propose to have XML as a new datatype
Not at this time.
: 2) Does a document descriptor mean the mapping of the XML
: elements to certain columns in an Oracle table? Does it have
: anything to do with the DTD? Will an XML instance be mapped to
:a single table only or across multiple tables in a master-detail
: relationship as the heirarchy in the instance gets too complex?
A Document Descriptor is basicly a mapping between a DTD and the
tables and columns in the RDBMS. More that one table can be
mapped to.
: 3) How will the ConText cartridge be able to identify XML
: documents? How will the queries be written to enable 'element-
: level' searching?
Context, now interMedia Text, does not 'identify' XML documents
per se, but can index colums with imbedded XML and search the
contents between specified tags.
: 4)What is iFS , especially in relation to XML storage?
Check out http://www.oracle.com/xml/documents/xml_twp/ for a
complete description.
Oracle XML Team
http://technet.oracle.com
Oracle Technology Network
null

INTERMEDIA TEXT - INSO FILTER의 사용방법과 한글 설정

제품 : ORACLE SERVER
작성날짜 : 2003-02-05
INTERMEDIA TEXT - INSO FILTER의 사용방법과 한글 설정
====================================================
Purpose
Intermedia text에서 INSO filter를 사용할 때 한글을 설정하여 이용하는
방법을 알아보자.
Explanation
INSO Filter 에는 자체적인 속성이 없기 때문에, 한글 환경을 설정하기
위해서는 ‘ctxhx’에 적절한 입력 값을 제공하는 새로운 script 를 만
들어야 한다. 그 방법은 다음 두 과정을 거쳐야 한다.
첫번째, 새로운 shell script 작성 : ctxhx 에 원하는 매개 변수 설정
두번째, INSO_FILTER 대신 USER_FILTER 를 사용하여 작성한 script 를
command로 설정
다음은 INSO filter 를 이용하여 한글 문서를 검색하는 과정을 보여준다 .
데모 환경은 다음과 같다 .
* KOREAN_LEXER 사용
* Acrobab PDF, Word97, PPT97, Excel, Html 문서 검색
* 한글 검색을 위해 , INSO Filter 를 변형한 USER_FILTER 사용
1. 한글 검색 용 INSO Filter 생성
다음은 INSO Filter‘ctxhx’를 이용하여 한글 문서를 필터링하도록
'autofilt'의 내용이다. 이 파일은 실행 권한이 주어져야 한다.
$ cd $ORACLE_HOME/ctx/bin
$ vi autofilt
#!/bin/sh
$ORACLE_HOME/ctx/bin/ctxhx $1 $2 hangeul koreanhangul T
$ chmod 777 autofilt
* NT의 경우
메모장 등으로 oracle_home\ctx\bin directory에 autofilt.bat 라는
이름의 화일을 다음과 같이 생성한다.
oracle_home 이 E:\ora815 인 경우
set oracle_home=E:\ORA815
E:\ora815\ctx\bin\ctxhx.exe %1 %2 hangeul koreanhangul T
2. 테이블 생성
SQL> create table mydocs (
2 docid number primary key,
3 title varchar2(256),
4 fname varchar2(30),
5 document blob );
3. Pro*c 나 sql*loader등을 이용하여 long raw 또는 blob type에
file을 입력한다.
[예제] sql*loader를 이용한 방법
sql*loader를 이용하여 , ‘mydocs’ 테이블에 있는 BLOB 에 5개의 문서를
입력하기 위한 과정을 위해 쓰이는 file들은 다음과 같다.
* controlfile : mydocs.ctl
INFILE 'mydocs.dat'
FIELDS TERMINATED BY ";"
optionally enclosed by '"'
(docid, title, fname,
ext_fname filler char(40),
document lobfile(ext_fname)
terminated by eof )
* datafile : mydocs.dat
0; ConText Cartridge Datasheet ; context.pdf ; context.pdf
1; Image/Vir Cartridge 소개 ; img_vir.ppt ; img_vir.ppt
2; Oracle Video Server 소개 ; ovs 소개 .doc ; ovs 소개 .doc
관리 시스템 ; kms.html ; kms.html
4; Oracle8i New Feature ; oracle8i.xls ; oracle8i.xls
* sql*loader의 실행
$ sqlldr userid=ctxdemo/ctxdemo control=mydocs.ctl data=mydocs.dat direct=no
4. Indexing Object 의 생성
* 한글 검색을 위해 KOREAN_LEXER preference 를 사용
* autofilt 를 필터링에 쓰이는 명령어로 하는 USER_FILTER preference 사용
SQL> begin
2 ctx_ddl.create_preference('MYDOCS_LEXER', 'KOREAN_LEXER');
3 ctx_ddl.create_preference('MYDOCS_FILTER', 'USER_FILTER');
4 ctx_ddl.set_attribute('MYDOCS_FILTER', 'command', 'autofilt');
5 end;
6 /
PL/SQL procedure successfully completed.
* NT인 경우 아래와 같이 attribute를 지정해야 한다.
ctx_ddl.set_attribute('MYDOCS_FILTER', 'command', 'autofilt.bat');
이제 위에서 정의된 indexing object 들을 이용하여 ,
mydocs 의 document field 에 Text index를 생성하는 과정을 보여준다 .
SQL> create index mydocs_document_index
2 on mydocs(document) indextype is ctxsys.context
3 parameters('lexer mydocs_lexer
4 filter mydocs_filter');
Index created.
5. 데이타 조회
* '비디오'와 '서버'를 포함한 문서의 검색 결과를 보여 준다 .
SQL> select ctxsys.score(0) scr, docid, title, fname
2 from mydocs
3 where ctxsys.contains(mydocs.document, '비디오 and 서버 ',0) > 0 ;
SCR DOCID TITLE FNAME
38 2 Oracle Video Server ovs소개.doc
1 row selected.
* BLOB 에 들어 있는 Word97 문서를 필터링한 결과를 출력하는 과정을 보여준다.
SQL> create table
2 query_id number,
3 document clob);
Table created.
SQL> exec CTX_DOC.FILTER('mydocs_document_index', 2, 'filtertab', 0);
PL/SQL procedure successfully completed.
SQL> select document from filtertab;
DOCUMENT
Oracle Video Server 소개
.....

Indixes and like statement

hi folx,
hmm i need to know of there is any chance
to create an index or a kind of hash table
for a column (varchar) wich needs to be compared to '%string%'. No index is usefull for me
cause to many differences for bitmap index.
normal and reverse indexes would imho
not be used by the parser.
I thought about context cartridge but varchar(80) is too small i think to get any performance back from the query.
thanks in advance
mfg
markus

Statement should be quicker than Prepared Statement, because
Statment do only one thing: send that sql to server or run that sql directly.
Prepared Statement should do two (or more than two)things:
1. parse your sql first, prepare a store procedure, then call that store procedure.
Or
2. prase your sql first, then use your value to replace "?" for getting a new sql, then work like Statement.
Prepared Statement is quiker when you use it repeatedly.

Clone Repository Database

Hi team,
I want to copy the EM Grid Database to another server and use it for test purpose.
We have 11g OMS and 11gR2 database.
Whats the best method to copy the EM database to another server?
I have two options
1. Create new empty database, take full export of current EM DB and import to new DB.
2. rman clone the DB to new server.
Thanks,
Rane

There were major changes between 8.0 and 8.1. You certainly couldn't just install the 8.0 Context cartridge onto 8.l.

ConText Cartridge

Similar Messages

Maybe you are looking for