SES, TREX ... and search in indexed table

Hi
I have a task, to use SES to create intarface (in ABAP) for search in one specyfic table. This table contains field with long description text and I would like to use SES to search over it. I read some documents and i checked SAP Help ... but i still don't know how to start. Is there any step by step tutorial for using SES? Do I have to create Business Object Type for this table? If i have to ... how to create it for to enable SES for it?
Regards
Marcin

Hi Marcin
I think you should create this message in the Enterprise Search forum, if you haven't done already.
Kind regards,
Martin Søgaard

Similar Messages

Flash and Search Engine Indexing

Hi all,
I am sure this is a newbie question that has been answered before but does using
flash on a website affect how the site is indexed by the main search engines?
thanks
TermPat

Well in theory, search engines like Google are suppose to be able to read the text used in Flash Website, improving SEO somewhat. But in my experience, it has not developed to a point that is equal to a straight HTML site. In particular, if you are using xml files to provide text or other data to the Flash pages, I don't see Google indexing that text.
So as a backup, I would suggest using swfobject to place your Flash on the Web page. Then add plenty of regular HTML as "alt content" to descibe in great detail the graphic elements. Since most viewers will not see this "alt content" you can actually over do the descriptions and get in MORE unique phrases, descriptions, and text content, than if you were just using HTML.
So in effect, you can get your graphic elements displayed with Flash but the text data described in HTML. Is it more work? Yes, But if SEO is important to you, Flash of and by itself will not index as well as just HTML.
Here's an example of using "alt content" to descibe the products in much greater detail than what you actually see on the Flash page. First view the Web page, its just a simple .swf with a little text on it. Then view the source code and see the "alt content" that lies behind the Flash.:
http://www.ksowetsuits.com/
For more info on swfobject:
http://code.google.com/p/swfobject/
Here's a little tip to measure exactly when and what gets indexed on your site. Create a unique word that cannot be found ANYWHERE on the Internet, something like "myuniquewordisdellamagestic"...can't be found anywhere on Google.
So include your special words in both your Flash file as well as the swfobject "alt content", using 2 unique words, to see which is indexed first. As soon as that word is indexed, it will appear top position, first page. So Google for it every so often, once it shows up you know that content has been indexed. See which gets indexed first, the text in Flash or the text in the "alt content".
I had one site where I used a complete text narration of a Flash video as the "alt content" and had the VIDEO (actually it's alt content) indexed within 4 hours. Now that's a case where using Flash PLUS alt content give a better WEb page than just the Flash alone or just the text narration alone. Best of both worlds I would say.
Best wishes,
Adninjastrator

Bookmarks and Search Engine Indexing

How do you allow a user to bookmark a page or a search engine to index a page?
I am new to JSF and I am trying to decide whether to go forward with it. The company I am doing work for is requiring that their site can be bookmarked and search engines can index it. I think that most every site would want this. I understand that there are certain pages you do not want bookmarked, like a shopping cart or a muti page form, but item pages, article pages and so on should be allowed to be bookmarked. I think I understand how JSF handles URLs and understand how the first request of a page is handled. With this, I do not see anyway a particular page can be bookmarked or indexed.
For instance, this page in JSF would not be able to be bookmarked or indexed. I have seen many other posts regarding this, but I have not seen any solutions.
I did see an idea about placing query strings in a URL:
� /article.jsp?article=383276
would be something like
� /article-383276/article.jsf
but there looks like there would be issues creating an ArticleBean on the first request and then accessing it.
I am also not sure how you would map /article-383276/article.jsf to /article.jsp, being that you would have many different URLs and then getting the URL in the address bar to reflect which article you were at.
/article-383276/article.jsf
/article-245621/article.jsf
/article-145845/article.jsf
and so on
I have also seen there are issues regarding the use of redirect.
I would sure like to use JSF on this project, but this limitation may make it impossible.
Thanks,
Warren

JSF pages accept arguments passed as part of URL via param.foo
http://forum.java.sun.com/thread.jspa?forumID=427&threadID=596838
Not sure if that servers your purpose

Webinar: Understanding TREX Indexing and Search Options

SAP NetWeaver Know-How Network Webinar:
Understanding TREX Indexing and Search Options
Wednesday 25 August 2004
11 a.m. EDT
On Wednesday 25 August, Larry Brambrut, an EP RIG Consultant, hosts the webinar titled Understanding TREX Indexing and Search Options as part of the ongoing SAP NetWeaver Know-How Network Webinar Series.
Heres how Larry describes his webinar presentation:
This session will describe the enhancements to "Search and Classification"(TREX) in NetWeaver '04 and EP 6.0 SP2 Patch 6. The session will include a discussion of the CM enhancements such as new crawlers, new search UI options and plug-ins, and TREX enhancements such as the new TREX architecture, delta indexing, and new TREX Admin Tool.
SDN invites you to post your questions to the presenter prior to the webinar and continue the online discussion afterward.
How to Participate
(Please go to the SDN webinar schedule page to find more information)
Dial-in Information:
Date: Wednesday 25 August 2004
Time: 11 a.m. EDT
Within the U.S., call: +1.888.428.4473
Outside the U.S., call: +1.651.291.0618
Password: NetWeaver04
WebEx Information:
Topic: SAP NetWeaver Know-How Network
Date: Wednesday 25 August 2004
Time: 11 a.m. EDT
Meeting Number: 742391500
Meeting Password: netweaver04 (lowercase)
WebEx Link: sap.webex.com
Replay Information:
A recorded replay of this call will be available for approximately three months after the webinar. Access this recording by dialing the appropriate number and using the replay access code 720155.
Toll-free: +1.800.475.6701
International: +1.320.365.3844

About the SAP NetWeaver Know-How Webinar Series
The SAP NetWeaver Know-How Webinar Series is driven by the SAP NetWeaver Regional Implementation Group (RIG), part of the SAP Development organization. The mission of the SAP NetWeaver RIG is to enable customers, employees, and partners to successfully implement the SAP NetWeaver solution. This SAP RIG has expertise in BI, EP, XI, and WebAS. They contribute their implementation expertise to the SDN implementation forums as well as to the SAP NetWeaver Know-How Webinar Series.
Disclaimer
SDN is not responsible for any changes to the webinar schedule. The webinar schedule may be changed or cancelled without prior notice.

Hi there,
I just read this thread, and maybe someone here can answer my current trex question:
I have created an ordinary CM repository, and created an index with this repository as source. Now the problem: I would like to exclude files in the repository with specific mimetypes from the TREX indexing process.
I have verified that the TrexValidMimetypes.ini does not contain any reference to the Mimetypes I'm creating, but never the less, the document titles are searchable and are returned when searching.
How do I get around this issue?
Is it possible in NW04 or EP6.0 SP3 PXXX??
Regards,
Hco

TREX - ECC 6.0 searches in different tables

Hi, have some search related questions:
- Is it possible to use TREX for combined searches from non-indexed tables in ECC 6.0 (where data is stored in different areas such as material, classification, BOM, customers, vendors, address data and so on)?
- Is it possible to mirror tables in ECC 6.0 to TREX via change pointers so the search data is "fresh" (the returned search result will be used in combination with other data and then stored back to ECC 6.0 from the NW portal)?
- Is it possible to have cross related search attributes (fields from material master, classification, bom, dms, address and so on) on the same tab in the sap standard search help (sap search help will be called from ABAP WebDynpro's)?
The main issue is to do combined searches with different combinations (number of input fields (10-20) related to different modules/areas) in ABAP WebDynpro. To do searches with outer table joins in tables without
keys and indexes seems not be doable.
/Brg Patrick Olsson

Hi lorenzo,
I have successfully installed ECC6.0 IDES SR2, i have found ECC-DIMP 6.0 LEVEL 6 in Support packs, please could you tell me that how can i activate IS-MILL in my server and what are the preconditions.
Thanks and Regards
ChandraB.

How to store pdf into blob, index it and search it?

I am trying to store pdf into a column, index the column and search it.
I store pdf file into blob column with either the following code or with an interface in apex
declare
Dest_loc BLOB;
Src_loc BFILE := BFILENAME('DIR_TESTCASE', 'intermedia.pdf');
BEGIN
INSERT INTO pdffiles(id, description)
VALUES(1, 'InterMedia Concepts')
RETURN text INTO Dest_loc;
DBMS_LOB.FILEOPEN(Src_loc, DBMS_LOB.LOB_READONLY);
DBMS_LOB.LOADFROMFILE(Dest_loc, Src_loc, dbms_lob.getlength(Src_loc));
DBMS_LOB.FILECLOSE(Src_loc);
COMMIT;
END;
then i need to create an index
begin
ctx_ddl.create_preference('pdf
lexer', 'BASICLEXER');
ctx_ddl.set_attribute('pdf_lexer', 'printjoins', '_-');
end;
create index PDFFILES_TEXT on PDFFILES(TEXT) indextype is ctxsys.context
parameters ('lexer pdf_lexer');
sqlplus says index created but windows gives an error and closes ctxhx.exe
and of cource when I tried to make a search like
select id, description
from pdffiles
where contains(text,'Oracle') > 0;
it doesn't return any results.
My database is 10g XE, I also tried it on Oracle 10.2.0.3 EE

It is possible to store pdf and doc files in a blob column, create a text index, and search them, and display them. However, you cannot expect to just select doc and pdf data directly from a blob column and expect to be able to read it, due to all of the special formatting characters. You need to use ctx_doc.snippet or ctx_doc.markup or some such thing to format it and make it readable. Please see the demonstration below. You can search the online documentation for complete syntax, additional information, and examples.
SCOTT@orcl_11g> CREATE TABLE my_files
2 (id NUMBER,
3 doc BLOB)
4 /
Table created.
SCOTT@orcl_11g> CREATE OR REPLACE DIRECTORY DIR_TESTCASE AS 'c:\oracle11g'
2 /
Directory created.
SCOTT@orcl_11g> declare
2 Dest_loc BLOB;
3 Src_loc BFILE;
4 BEGIN
5 INSERT INTO my_files (id, doc) VALUES (1, EMPTY_BLOB())
6 RETURNING doc INTO Dest_loc;
7 Src_loc := BFILENAME ('DIR_TESTCASE', 'banana.pdf');
8 DBMS_LOB.FILEOPEN (Src_loc, DBMS_LOB.LOB_READONLY);
9 DBMS_LOB.LOADFROMFILE (Dest_loc, Src_loc, dbms_lob.getlength (Src_loc));
10 DBMS_LOB.FILECLOSE (Src_loc);
11 INSERT INTO my_files (id, doc) VALUES (2, EMPTY_BLOB())
12 RETURNING doc INTO Dest_loc;
13 Src_loc := BFILENAME ('DIR_TESTCASE', 'test1.doc');
14 DBMS_LOB.FILEOPEN (Src_loc, DBMS_LOB.LOB_READONLY);
15 DBMS_LOB.LOADFROMFILE (Dest_loc, Src_loc, dbms_lob.getlength (Src_loc));
16 DBMS_LOB.FILECLOSE (Src_loc);
17 END;
18 /
PL/SQL procedure successfully completed.
SCOTT@orcl_11g> INSERT INTO my_files (id, doc)
2 VALUES (3, UTL_RAW.CAST_TO_RAW ('blob data for id 3'))
3 /
1 row created.
SCOTT@orcl_11g> COMMIT
2 /
Commit complete.
SCOTT@orcl_11g> SELECT id, DBMS_LOB.GETLENGTH (doc) FROM my_files
2 /
 ID DBMS_LOB.GETLENGTH(DOC)
 1 222824
 2 19968
 3 18
SCOTT@orcl_11g> CREATE INDEX my_files_idx ON my_files(doc)
2 INDEXTYPE IS CTXSYS.CONTEXT
3 /
Index created.
SCOTT@orcl_11g> SELECT COUNT (*) FROM dr$my_files_idx$i
2 /
COUNT(*)
 308
SCOTT@orcl_11g> COLUMN first_45 FORMAT A45
SCOTT@orcl_11g> SELECT id,
2 UTL_RAW.CAST_TO_VARCHAR2 (DBMS_LOB.SUBSTR (doc, 18, 1)) AS first_45
3 FROM my_files
4 /
 ID FIRST_45
 1 %PDF-1.5
%âãÏÓ
 1
 2 ÐÏà¡±á
 3 blob data for id 3
SCOTT@orcl_11g> EXEC CTX_DOC.SET_KEY_TYPE ('ROWID')
PL/SQL procedure successfully completed.
SCOTT@orcl_11g> COLUMN keywords_in_context FORMAT A45 WORD_WRAPPED
SCOTT@orcl_11g> SELECT id,
2 CTX_DOC.SNIPPET
3 ('MY_FILES_IDX',
4 ROWID,
5 'contents OR (fruit of the month) OR data',
6 '<<',
7 '>>')
8 AS keywords_in_context
9 FROM my_files
10 WHERE CONTAINS (doc, 'contents OR (fruit of the month) OR data') > 0
11 /
 ID KEYWORDS_IN_CONTEXT
 1 <<Fruit of the Month>>
 Banana
 Bananas are the most popular
 2 This is the original <<contents>> of
 test1.doc.
 3 blob <<data>> for id 3
SCOTT@orcl_11g>

Indexing and Searching PDF Files

Hi All,
I am trying to store and search PDF files in the oracle database.
I can insert and index the PDF files just fine but cannot get any result. I always get No Rows.
Here's what I am doing and the issues I am facing.
I created a Table with fields
ID (VARCHAR)
NAME (VARCHAR)
DOC (BLOB)
I inserted the PDF file in the BLOB field through a Java program and insert worked fine as I verified by retreiving the PDF and writing to file.
I created index using following SQL:
create index my_index on PDF_TABLE(PDF_FLD) indextype is ctxsys.context
parameters ('datastore ctxsys.default_datastore
filter ctxsys.inso_filter');
The index was created successfully without any problems.
I ran query as follows and got no rows although the searched text is in PDF
SELECT SCORE(1), PDF_FLD from PDF_TABLE WHERE CONTAINS (PDF_FLD, 'Table of Cotents',
1) > 0;
I tried alternate queries as well with no luck.
Any ideas ??
Thanks

After creation the index you need execute next operations.
first, check that your index tables conatins indexed terms. Execute
select token_text from dr$YOUR_INDEX$i;
Second, you will need to check the index errors table CTX_INDEX_ERRORS. This is owned by the user CTXSYS, and most users do NOT have # SELECT privilege to it by default.
If it's OK, then check that your PDF documents is supported by INSO filter.
Citation:
"PDF - Portable Document Format
Acrobat Versions 2.1, 3.0, 4.0, and 5.0 including Japanese PDF"
(Appendix B. Supported Document Formats in Oracle Text Reference 9.2)
For Oracle 9i you could install 9.2.0.4 patchset (it included INSO FILTER 7.5)
P.S.
for the beginning, you could find answers for your question about Oracle Text here
http://otn.oracle.com/products/text
Sorry for my English.
Best regards, Victor Zogin.

Talent Management TREX and Enterprise Search

Hi All,
in the area of talent management I try to use the TREX search functionality. I have two problems:
1) For Connector HRTMC_AES_DOCUMETS no indexes are available although I started indexing many times. Other Connectors have active indexes. Appraisal documents exist in the system. The indexing and search for hrtmc_aes_templates are successful.
2) How does the connector hrtmc structural authority work. If I do a search for my username on this connector I get a result. If I do a search for a colleagues username on the same connector I get no result. If this colleague wants to search for e.g. a central person he gets no result. How can I fill up the index for struct authorization with the colleagues username?
many thanks for answers and have a nice day
Martin

Hi RD,
the TREX you will need for this will be specific to HCM and called via HCM's own strategy in ECC 6.0. This - for all I know - should be through AS ABAP for indexing and AS Java for searching.
This should have nothing to do with the Portal's access to TREX, as the functionality is HCM's, even if provided to Portal in an iView. For the specifics, please check your HCM Master Guide and other documentation and/or discuss with your technical consultant for ECC 6.0.
Best, Karsten

Multilinual indexing and searching!

Hi,
I've installed a Multi_Lexer for German, Swedish, French and English, how it is described in the interMedia documentation. Creating an index works and searching works, too.
But I still have a problem. Is there a way to combine ALTERNATE_SPELLING and BASE_LETTER in one way, so I can search after the word 'H|tte' with the expressions 'h|tte' or 'huette' or 'hutte'?
The way I installed the lexer makes only one possibility able to use. Must I create e table with three columns och three indexes with the same contents?
Who has got a solution?

Right now at query time "H|tte" can only be converted to one form - currently it's the
alternate spelling version.

Trex "synonym search" and taxonomy "search from here" issues

Hi Trex experts
I have two minor issues regarding Trex and Taxonomies
Taxonomy
I have created a Taxonomy and need to add the search from here function so its possible to create a more specific search from within one are of the taxonomy.
When I add a layout set where this command is enabled then its not shown.
I guess that its my repository manager for the Trex repository that I need to add a service to, but I cant figure out which one, or is this feature not possible for taxonomies?
Trex
Is it possible to define synonym search entries? Like I need to create a dependency on a bunch of terms, like when a user for example search for vacation then search for holiday should be included automatically in the search result as well. Is there some kind of configuration file where I can add a list of dependencies?
Thanks in advance
Regards
John

Hi John,
Related to synomym you should read this material:
"How to Enable Semantic Search or Search for Synonyms in TREX"
Related to taxonomy, the answer is yes it is possible.
If "search from here" botton is not shown, is due to you don't create an index (search option).
Patricio.

Same TREX server for KM indexing and e-Recruitment

Hi,
Is it possible to use the same TREX server for KM indexing on the portal and e-Recruiting searches on R/3?
Many thanks
Jane

Hi,
in general and from a technical view it's possible.
But it depends on the load of the applications, perhaps it's a better approach to separate them e.g. for a productive environment.
Regards, Thorsten

How to get list of block identifiers in a empty table and an empty index

We have an application that has issue with ITL waits: this application is running many INSERT statements on a table that has 2 NUMBER columns and one primary key index. The application is designed to run INSERT statements but they are never committed (this is a software package).
To check what are the really allocated ITL slots, I know that I can dump data block but I don't know how to get the block identifiers/numbers for an "empty" table and an "empty" index. Does someone knows how to do that ?
PS: I already had a look to the Metalink notes and I have a Metalink SR for that but maybe OTN forum is faster ?

You should be able to find the first data/index block with the following, even on an empty table/index.
select header_file, header_block +1
from dba_segments
where segment_name = '<index or table name>';

Spotlight searching no longer working - indexing and search disabled.

I've been searching the web and tried everything:
Server 10.5.8
In Server Admin - the attached drive is a SharePoint with Spotlight search on.
I've used mdutil to enable Spotlight.
I've checked permissions.
I can search the Boot Drive. I can't search the attached drive.
mdutil returns indexing and search disabled when used to turn it on.
very frustrating.
Anyone out there have a clue?
Thanks,
Mark

HI James,
Open System Preferences/Spotlight and click the Privacy tab. Where you see; Delete any locations listed, Quit System Preferences and restart your Mac and see if you can use Spotlight.
Spotlight Tips
Spotlight: How to re-index folders or volumes
Carolyn

Creating and Searching index files

Hi,
This is my first posting so apologiese in advance if it is difficult to understand.
Firstly, I am from a development background of some 6 years (c#, asp.net, vb.net, SQL) so i have a solid grounding and am happy to give any new development a try.
I have a client who has 10,000's of scanned pdf documents but no real way of retrieving and searching these documents. In fact another person in my team is involved in making sure these pdfs are searchable by doing some OCR and rescanning work. I have been asked to come up with a way to allow client to retrieve(open) and search the pdf files.
Here is what i am proposing based on what I have leared about Acrobat Pro 9 and Adobe Reader 8/9.
I actually want to do a lot more than above but think that is enough to get me started. In a nutshell the client would like a web based solution to search and retrieve (open) there scanned pdf documents. I have read on other posts in the forum an understand it can be quite difficult to search pdf's (pdx file) from Internet Explorer.
As a start i had the following tasks in mind as needing completed:
Task 1 - Create directory structure on file server for all the scanned pdfs. Create catalog(s) for these documents and then create an index which creates a .pdx file. Given the number of documents it looks like I will have X number of catalogs and X number of pdx files.
Task 2 - create a web application that allows a tree view to be displayed that replicates the directory structure above so a user can open the pdf from the browser.
Task 3 - I know the users can use Adobe Reader advanced find functionality and select the indexes i have created. However, I would like to create a plugin for Adobe Reader that will load all of the indexes into the selectable index list and select all by default. Given the volume of documents it is likely the user will not know exactly what they are looking for so will need to search across all indexes.
Thanks in advance for any help anyone can provide me in getting started with these tasks.

Some links that may be helpful.
Acrobat Developer Center:
http://www.adobe.com/devnet/acrobat.html
Forums:
Acrobat SDK
http://forums.adobe.com/community/acrobat/acrobat_sdk
Acrobat Scripting
http://forums.adobe.com/community/acrobat/acrobat_scripting
Various topic specific forums at the AUC
http://acrobatusers.com/forum
Be well...

Search in Nested Tables and Insert the result into new Nested Table!

How can I search in Nested Tables ex: (pr_travel_date_range,pr_bo_arr) using the SQL below and insert the result into a new Nested Table: ex:g_splited_range_arr.
Here are the DDL and DML SQLs;
Don't worry about the NUMBER( 8 )
CREATE OR REPLACE TYPE DATE_RANGE IS OBJECT ( start_date NUMBER( 8 ), end_date NUMBER( 8 ) );
CREATE OR REPLACE TYPE DATE_RANGE_ARR IS TABLE OF DATE_RANGE;
DECLARE
 g_splited_range_arr DATE_RANGE_ARR := DATE_RANGE_ARR( );
 g_travel_range DATE_RANGE := DATE_RANGE( '20110101', '99991231' );
 g_bo_arr DATE_RANGE_ARR := DATE_RANGE_ARR( DATE_RANGE( '20110312', '20110317' ), DATE_RANGE( '20110315', '20110329' ) );
 FUNCTION split_date_sql( pr_travel_date_range DATE_RANGE,
 pr_bo_arr DATE_RANGE_ARR )
 RETURN DATE_RANGE_ARR
 IS
 l_splited_range_arr DATE_RANGE_ARR;
 BEGIN
 SELECT start_date, end_date
 INTO l_splited_range_arr(start_date, end_date)
 FROM (WITH all_dates
 AS (SELECT tr_start_date AS a_date, 0 AS black_out_val FROM TABLE( pr_travel_date_range )
 UNION ALL
 SELECT tr_end_date, 0 FROM TABLE( pr_travel_date_range )
 UNION ALL
 SELECT bo_start_date - 1, 1 FROM TABLE( pr_bo_arr )
 UNION ALL
 SELECT bo_end_date + 1, -1 FROM TABLE( pr_bo_arr )),
 got_analytics
 AS (SELECT a_date AS start_date,
 LEAD( a_date ) OVER (ORDER BY a_date, black_out_val) AS end_date,
 SUM( black_out_val ) OVER (ORDER BY a_date, black_out_val) AS black_out_cnt
 FROM all_dates)
 SELECT start_date, end_date
 FROM got_analytics
 WHERE black_out_cnt = 0 AND start_date < end_date
 ORDER BY start_date);
 RETURN l_splited_range_arr;
 END;
BEGIN
 g_splited_range_arr := split_date_sql(g_travel_range,g_bo_arr);
 FOR index_g_splited_range_arr IN g_splited_range_arr .FIRST .. g_splited_range_arr .LAST LOOP
 DBMS_OUTPUT.PUT_LINE('g_splited_range_arr[' || index_g_splited_range_arr || ']: ' || g_splited_range_arr(index_g_splited_range_arr).start_date || '-' || g_splited_range_arr(index_g_splited_range_arr).end_date );
 END LOOP;
EXCEPTION
 WHEN NO_DATA_FOUND
 THEN
 NULL;
 WHEN OTHERS
 THEN
 NULL;
END;Or can I create a VIEW with parameters of Nested Tables in it so I can simply call
SELECT *
BULK COLLECT INTO g_splited_range_arr
FROM view_split_date(g_travel_range,g_bo_arr);

@riedelme
For your questions:
1) I don't want to store in the database as a nested table
2) I don't want to retrieve data from the database. Data will come from function split_date() parameter and data will be processed in the function and function will return it in nested table format. For more detail please look at the raw function SQL.
I have a SQL like:
WITH all_dates
 AS (SELECT tr_start_date AS a_date, 0 AS black_out_val FROM travel
 UNION ALL
 SELECT tr_end_date, 0 FROM travel
 UNION ALL
 SELECT bo_start_date - 1, 1 FROM black_out_dates
 UNION ALL
 SELECT bo_end_date + 1, -1 FROM black_out_dates),
 got_analytics
 AS (SELECT a_date AS start_date,
 LEAD( a_date ) OVER (ORDER BY a_date, black_out_val)
 AS end_date,
 SUM( black_out_val ) OVER (ORDER BY a_date, black_out_val)
 AS black_out_cnt
 FROM all_dates)
SELECT start_date, end_date
 FROM got_analytics
 WHERE black_out_cnt = 0 AND start_date < end_date
ORDER BY start_date;I want to change the tables black_out_dates and travel to Nested Array so I can use it in a function with Nested Array travel and Nested Array black_out_dates parameters and the function will return Nested Array of date ranges.
Here is what I want in raw SQL:
 DECLARE
 g_splited_range_arr DATE_RANGE_ARR := DATE_RANGE_ARR( );
 g_travel_range DATE_RANGE := DATE_RANGE( '20110101', '99991231' );
 g_bo_arr DATE_RANGE_ARR := DATE_RANGE_ARR( DATE_RANGE( '20110312', '20110317' ), DATE_RANGE( '20110315', '20110329' ) );
 FUNCTION split_date_sql( pr_travel_date_range DATE_RANGE,
 pr_bo_arr DATE_RANGE_ARR )
 RETURN DATE_RANGE_ARR
 IS
 l_splited_range_arr DATE_RANGE_ARR;
 BEGIN
 SELECT start_date, end_date
 INTO l_splited_range_arr(start_date, end_date)
 FROM (WITH all_dates
 AS (SELECT tr_start_date AS a_date, 0 AS black_out_val FROM TABLE( pr_travel_date_range )
 UNION ALL
 SELECT tr_end_date, 0 FROM TABLE( pr_travel_date_range )
 UNION ALL
 SELECT bo_start_date - 1, 1 FROM TABLE( pr_bo_arr )
 UNION ALL
 SELECT bo_end_date + 1, -1 FROM TABLE( pr_bo_arr )),
 got_analytics
 AS (SELECT a_date AS start_date,
 LEAD( a_date ) OVER (ORDER BY a_date, black_out_val) AS end_date,
 SUM( black_out_val ) OVER (ORDER BY a_date, black_out_val) AS black_out_cnt
 FROM all_dates)
 SELECT start_date, end_date
 FROM got_analytics
 WHERE black_out_cnt = 0 AND start_date < end_date
 ORDER BY start_date);
 RETURN l_splited_range_arr;
 END;
 BEGIN
 g_splited_range_arr := split_date_sql(g_travel_range,g_bo_arr);
 FOR index_g_splited_range_arr IN g_splited_range_arr .FIRST .. g_splited_range_arr .LAST LOOP
 DBMS_OUTPUT.PUT_LINE('g_splited_range_arr[' || index_g_splited_range_arr || ']: ' || g_splited_range_arr(index_g_splited_range_arr).start_date || '-' || g_splited_range_arr(index_g_splited_range_arr).end_date );
 END LOOP;
 EXCEPTION
 WHEN NO_DATA_FOUND
 THEN
 NULL;
 WHEN OTHERS
 THEN
 NULL;
 END;I must change the tables black_out_dates and travel in a way so it will be something like
FROM TABLE( pr_travel_date_range )to get the result into l_splited_range_arr so it will be something like
 SELECT start_date, end_date
 INTO l_splited_range_arr(start_date, end_date)
 FROM (

SES, TREX ... and search in indexed table

Similar Messages

Maybe you are looking for