Searching TEXT String in a document stored in BLOB Column

Hi,
I fairly new to this concept of Oracle iText. I would like to get the atmost procedural approach to meet the following scenario.
I have a table with 3 columns (viz. id,name, content) of which content is a BLOB datatype. I have stored a simple .doc file in this column. (Just one row has been stored into this table ). I am aware that this document contains my search word say 'TOPIC'.
These are the following steps i did to get my soln but didnt meet my expected result.
1. Inserted the row into the table ..... successful
2. Tried to do a simple select (using contains)after creating context index....no rows selected
3. Tried to create preference (DIRECT_PREFERENCE) and create an index using this PREFERENCE and do the select for the search string......no rows selected
It would b great if u would let me know, how can i retrive the doc info ie. atleast the id,name from the database if i get a matched search string in the table.
Regards
Kevin

When you index BLOBs (or RAW, or file) the product assumes that you have some kind of
binary data -- like Microsoft Word or something like that. It then automatically
engages the INSO filter.
In your case, if I understand correctly, you have text data stored in a blob.
So INSO filtering is not appropriate. So you will want to tell the product not to
use INSO:
create index ...
parameters ('filter ctxsys.null_filter');
try that...

Similar Messages

Problem displaying PDF stored in BLOB column

Hello everyone.
I've been trying to follow this tutorial http://asktom.oracle.com/pls/asktom/f?p=100:11:0::::p11_question_id:232814159006 on displaying PDFs stored in BLOB columns. This is being performed on Apex 4.2, with my DB running 11g. I've got my procedure, which I'll post below:
create or replace procedure "PDF" (p_id IN gvo_documents.doc_id%type)
is
    l_lob    blob;
    l_amt    number default 30;
    l_off   number default 1;
    l_raw   raw(4096);
begin
    select contents into l_lob
        from gvo_documents
         where doc_id = p_id;
-- make sure to change this for your type!
    owa_util.mime_header( 'application/pdf' );
    begin
        loop
          dbms_lob.read( l_lob, l_amt, l_off, l_raw );
          htp.prn( utl_raw.cast_to_varchar2( l_raw ) );
          l_off := l_off+l_amt;
          l_amt := 4096;
    end loop;
        exception
           when no_data_found then
              NULL;
        end;
end;
I am trying to run this through a PL/SQL dynamic region and while I don't receive any error's, the content displayed is a huge mess of garbled text and odd characters. I've tried to run this procedure on numerous other document types, including word files and jpeg images, all with the necessary changes in my procedure, and regardless of what I use, I still get a large mess of strange characters. Does anyone have any information or ideas about why this is happening?

If I understand correctly, your requirements needs to be broken down into two problems:
1) click link that pops up a window displaying a new APEX page
2) an APEX page the displays the document, not downloads it.
I haven't done #1 (yet).
However, you may be able to generate a URL that points to the new page as part of the SELECT statement for the Report.
This has a related question, but no answer yet:
open pdf in popup browser window
The key is target="_blank" for the anchor tag.
To generate the URL, you should use the APEX_UTIL.prepare_URL() function.
If that doesn't work, a Dynamic Action that does some magical JavaScript stuff may be needed.
For #2, I lost the URL that showed how to display a PDF as part of a "form" page.
From what I remember:
Start with a blank page with one blank HTML region (all the Items go in the HTML region)
Add an Item for the PK/Doc_ID
part I forgot Create a Data Manipulation Process
- Automated Row Fetch
- On Load - After Header
- (stuff for your table/view)
part I forgot Create an (I believe) "File Browser" item type. For Settings:
- Storage Type "BLOB column specified in Item Source" (and place the column name of the BLOB there)
- MIME Type Column: (column name) -- since you have multiple types, this is a MUST HAVE
- Filename Column: (column name) -- I highly recommend you have this.
- Content Disposition == INLINE <-- this is the kicker
Also, you will need a Browser Plugin for each of the MIME Types (otherwise, the browser may try to 'download' the file)
Browsers can handle Image types internally. Adobe's plugin can handle PDFs. I don't know about Word/Excel.
Again, I don't remember the exact details, but that should cover most of it.
MK

Modifying MS WORD documents stored in BLOBS. (Oracle 9i Release 9.2.0.4.0)

I have a need to replace a string of characters in an MS Word document that is stored in a BLOB field in an Oracle 9i database.
Background: We have a large number of MS Word documents (~50,000) and many of them contain links to other Word documents stored on various servers throughout our network. We are moving a number of these servers, thus invalidating these links. I need to be able to update these links without having to open up every one of these documents and doing it manually.
Is it possible, using PL/SQL, to update these links in these MS Word documents that are stored in Oracle? I have currently written a quick PL/SQL routine that will find the links and provide their location within the document, but I am not sure how I can update them. The problem, as I see it, is that if the characters that I am replacing differ in length than those that are currently in the existing document, then I would have to shift the remaining bytes of the document either left or right to preserve the integrity of the document.
Is this possible, or should I pursue another option?

It does not have to manual, but I think you will need to use word to modify and save the documents.
The most automated process I can think of would be to write a client program to fetch a blob, run word, and have a word macro or add in to make the edits save, and then the client program updates with the new blob.
With that number of documents, I would probably try to fetch a bunch of blobs and have word process them all before updating them for performance reasons.
Unfortunately I would have no clue how to do the word automation, but I wouldn't touch those blobs using pl/sql

Printing Word Document stored in BLOB from Reports

I have stored images and MS-Word documents in the DB in a BLOB column. Using reports 6, I am able to print the images by setting the File Format property of the item to Image. What I need to know is how can I print the Word document that has been stored in the DB using reports? Any alternative suggestions welcome...
Thanks for your help.
Aparna

Hi Aparna,
I have tried the same with CLOB instead BLOB.
(MS Word)
Only problem is (as expected) any graphics (BOLD, Table etc) will not work when printing. It also replaces the TABS with A single space.
If you have found an answer, please let me know.
Tissa
([email protected])
null

How to classify the type of the file which stored in Blob column

I have a table which will stored various kinds of documents (e.g. PDF, MS word, MS Excel etc...) into blob column.
How can I read them from the database through web and display them into the correct software? i.e. when the file is PDF, it can be opened by Acrobat reader and .doc can be opened by MS word
thanks.

A bit off the topic of Oracle Text, but the client typically determines the application that will be used (you have your apps associated with file extensions on your machine and I have mine). Are you having problems related to this, or is this a question in advance of trying it out?
Thanks,
Ron

Image stored in BLOB column crashes report's engine

Hi,
I have a report where I display logos stored in a BLOB column on the DB.
The report is generated as RTF.
When images have sizes of 624x168 or 423x324 everything is OK. One image, 1221x224, is crashig the report with a dr. Watson error.
Any idea why?
The size of the "bad" image file is 38K, and the "good" ones are about 300K.
Monica

Sajjad wrote:
I altered student table to store photoes of students. For this I added STIMG as blob, MIMETYPE,FILENAME,IMAGE_LAST_UPDATE COLUMNS to studentadmitted table. Image stored and displayed in report very nicely. With the following query
select "ROWID",
"ST_ID",
"FIRST_NAME",
"LAST_NAME",
"FFIRST_NAME",
"FLAST_NAME",
"COURSE_ID",
"SESSION_ID",
"CONTACT#" "CONTACT_H",nvl(dbms_lob.getlength(STIMG),0) IMAGE
from "#OWNER#"."STUDENTADMITTED"
Images are displayed in column but they have different sizes.Stick to declarative BLOB images as above. They can be resized in reports using CSS. Add a style sheet in the page Inline CSS property:
td[headers="IMAGE"] img {
display: block;
width: 75px;
border: 1px solid #999;
padding: 6px;
background: #f6f6f6;
}where the <tt>IMAGE</tt> value in the attribute selector is the table header ID of the image column. Setting only one dimension (in this case the width) scales the image with the correct aspect ratio. (The border, padding and background properties are just eye candy...)
However, if the original images are large then scaling them in the browser this way is a huge waste of bandwidth and produces poorer quality images than creating proper scaled down versions using image tools. Downloading megapixel images and then reducing them to thumbnail size results in pointlessly pushing millions of bytes that will never be seen.
For improved performance and image quality, and where you require image-specific scaling the best approach is to use the database ORDImage object to produce thumbnail and preview versions automatically. Note that this is not possible in Oracle XE as Multimedia is not included

Displaying the image stored in Blob column

Hi all,
I'm trying to print the report with an image. I've stored the image in a Blob column and the format of the image is Jpg.
I'm using the Reports 10g(10.1.2.0.2). When I'm trying to print the report, I'm getting an error Rep-62203. I want to know whether there are any settings which i need to set to display the image item. I mean to say that are there any registry settings.
Regards,
Alok Dubey
Edited by: Alok Dubey on Dec 1, 2008 12:55 PM

Hi all,
I'm trying to print the report with an image. I've stored the image in a Blob column and the format of the image is Jpg.
I'm using the Reports 10g(10.1.2.0.2). When I'm trying to print the report, I'm getting an error Rep-62203. I want to know whether there are any settings which i need to set to display the image item. I mean to say that are there any registry settings.
Regards,
Alok Dubey
Edited by: Alok Dubey on Dec 1, 2008 12:55 PM

How to search text content of scanned Documents in DMS

hello,
i need to scan the legacy document and put it in to DMS.
After that i want to search the scanned document using the Trex search.
I want to know how to achive it as scaaed document are as like images and we can not search the text in images.
Any helpfull ideas are welcome..
regards
shiv

Hi Alfredo,
Thanks for reply..Will you please explain in brief what are the OCR program and workflow.
Will you please send me any docs on [email protected]
Thanks
shiv.

Image not displayed programmtically stored in blob column

Hi,
I altered student table to store photoes of students. For this I added STIMG as blob, MIMETYPE,FILENAME,IMAGE_LAST_UPDATE COLUMNS to studentadmitted table. Image stored and displayed in report very nicely. With the following query
select "ROWID",
"ST_ID",
"FIRST_NAME",
"LAST_NAME",
"FFIRST_NAME",
"FLAST_NAME",
"COURSE_ID",
"SESSION_ID",
"CONTACT#" "CONTACT_H",nvl(dbms_lob.getlength(STIMG),0) IMAGE
from "#OWNER#"."STUDENTADMITTED"
Images are displayed in column but they have different sizes.
To control this I modified the query like this
select "ROWID",
"ST_ID",
"FIRST_NAME",
"LAST_NAME",
"FFIRST_NAME",
"FLAST_NAME",
"COURSE_ID",
"SESSION_ID",
"CONTACT#" "CONTACT_H",decode(nvl(dbms_lob.getlength(stimg),0),0,null,
'<img style="border: 4px solid #CCC; -moz-border-radius: 4px; -webkit-border-radius: 4px;" '||
'src="'||
apex_util.get_blob_file_src('P16_stIMG',st_id)||
'" height="75" width="75" alt="Product Image" title="Product Image" />')
detail_img
from "#OWNER#"."STUDENTADMITTED"
Report did'nt run with following error,
report error:ORA-06502: PL/SQL: numeric or value error: character to number conversion error
Win Server 2008
Oracle Database :11G R2
Apex Ver :4.2
Sajjad

Sajjad wrote:
I altered student table to store photoes of students. For this I added STIMG as blob, MIMETYPE,FILENAME,IMAGE_LAST_UPDATE COLUMNS to studentadmitted table. Image stored and displayed in report very nicely. With the following query
select "ROWID",
"ST_ID",
"FIRST_NAME",
"LAST_NAME",
"FFIRST_NAME",
"FLAST_NAME",
"COURSE_ID",
"SESSION_ID",
"CONTACT#" "CONTACT_H",nvl(dbms_lob.getlength(STIMG),0) IMAGE
from "#OWNER#"."STUDENTADMITTED"
Images are displayed in column but they have different sizes.Stick to declarative BLOB images as above. They can be resized in reports using CSS. Add a style sheet in the page Inline CSS property:
td[headers="IMAGE"] img {
display: block;
width: 75px;
border: 1px solid #999;
padding: 6px;
background: #f6f6f6;
}where the <tt>IMAGE</tt> value in the attribute selector is the table header ID of the image column. Setting only one dimension (in this case the width) scales the image with the correct aspect ratio. (The border, padding and background properties are just eye candy...)
However, if the original images are large then scaling them in the browser this way is a huge waste of bandwidth and produces poorer quality images than creating proper scaled down versions using image tools. Downloading megapixel images and then reducing them to thumbnail size results in pointlessly pushing millions of bytes that will never be seen.
For improved performance and image quality, and where you require image-specific scaling the best approach is to use the database ORDImage object to produce thumbnail and preview versions automatically. Note that this is not possible in Oracle XE as Multimedia is not included

Displaying images stored in blob columns

I've seen discussions on how to upload and download blob columns but I want to store a blob in the database and be able to display the blob (which is a jpg) as I would any other <img> on my web page. Instead of having to store the jpg on a file system somewhere, and use the data in the database to formulate a link to the image on the file share, I want to pull it right out of the blob. Does someone have a sample of how to do this? Can I do this?
Thanks

Laurence,
If you uploaded the image via the Application Builder, you can use an image tag like so:
<img src="#WORKSPACE_IMAGES#my_image.jpg" />If you followed the upload/download how to, you should be able to reference an image like this:
<img src="download_my_file?p_file=1232897645" />Where 1232897645 is the unique ID of the uploaded image. To make referencing images easier, you could write a version of the download_my_file procedure that takes in the name of the image as a parameter.
Sergio

Full text search in BLOB columns NOT Working

Hi, I am using Oracle 9i.
Presently I use following code to create index.
begin
Ctx_Ddl.Drop_Preference ('text_only');
end;
begin
Ctx_Ddl.Create_Preference ( 'text_only', 'basic_lexer');
Ctx_Ddl.Set_Attribute ( 'text_only', 'index_text', 'true' );
Ctx_Ddl.Set_Attribute ( 'text_only', 'index_themes', 'false' );
end;
CREATE TABLE MY_XML_DOCS(
XML_RES_ID NUMBER (16) NOT NULL,
XML_RES BLOB NOT NULL,
XML_RES_PATH VARCHAR2 (128),
XML_DOCS_CREATED_BY VARCHAR2 (16) NOT NULL,
XML_DOCS_CREATED_DT DATE NOT NULL,
XML_DOCS_MODIFIED_BY VARCHAR2 (16),
XML_DOCS_MODIFIED_DT DATE ) ;
drop index FULL_TEXT_INDEX ;
create INDEX FULL_TEXT_INDEX ON MY_XML_DOCS(XML_RES)
INDEXTYPE IS ctxsys.CONTEXT
parameters ('lexer text_only stoplist ctxsys.empty_stoplist');
BEGIN
CTX_DDL.sync_index('FULL_TEXT_INDEX ','10M');
END;
And I've some XML documents stored in blob columns in the above table. But my query with 'contains' param does not return expected results.(it presently returns no results). Do I miss anything? Do I need to specify any thing else while I create the Index? Kindly advice.

I would agree the CLOB storage would be preferable for XML. If you are using BLOB storage, by default the documents are going to be FILTERed by the INSO_FILTERS. A BLOB column will give you reduced storage in a unicode environment though.
If you wish to store your data in a BLOB though, you should specify the NULL_FILTER in your create index statement to avoid having the INSO_FILTERS invoked.
Further, you may also require a CHARSET column in your table specifying the character set of your data for each row if it can vary. The Column should then also be specified in the 'PARAMETERS' clause of the create index statement, this way Oracle knows the correct character set to use when converting the binary data to character data.
My guess that you are not getting any query results because none of the documents are actually being indexed. Be sure to query the ctx_user_index_errors view to determine if you are getting errors at indexing time.

Need to create link to Word Doc in blob column in search results report

I got the Oracle Text boolean search of word documents in a blob column of a table working.
Now I need to be able to create a link in the results report.
I know that Oracle creates a link for each document in the column when you attach a file, I just need to know what the link should be to allow users to open the document in the search results report.
Right now the application is set to do a search in the attached word documents, and in the results window it shows the names of the people who match the search, but it won't allow me to create a link to their resume in the results report.
the name of the table is CONTRACTOR_LIST and the blob column is RESUME.
How would you go about creating a simple link to the word file in the results window?
Here is the code for the query
select score(1) relevance, Name, Resume
from contractor_list
where CONTAINS (resume, :P1_SEARCH, 1) > 0
order by 1 desc
That works, but I can't get it to link to the resume file.
Here is the code I'm using for the link
javascript:popupURL("#RESUMEL#")
This just give me an error page and the the link is
http://server-namer:8080/apex/[datatype]
the error is
Bad Request
The HTTP client sent a request that this server could not understand.
Thanks again!
Edited by: gjones77 on Dec 2, 2008 6:14 AM
Edited by: gjones77 on Dec 2, 2008 7:08 AM

It is within the table I believe (I'm not a DBA or a developer) since I created a BLOB column and then used the file browse feature to allow users to attach a resume to the table in order to be able to perform a search of the attached documents.
I'm just having a hard time pointing the link in the search results report to the document in the blob column.
The information on that page is great if you're trying to create a link to the document on the initial report.
But I created a query using Oracle Text to run a report that does a boolean search of the attached word documents in the table.
When it displays the search results, it doesn't create a link to the document and I can't figure out how to do it.
Here's a link the the instructions I used to create the initial search report with Oracle Text, mind you I only created the index and query, I didn't add in all the link data since they're using documents on websites and I'm using documents in a table.
http://www.oracle.com/technology/products/database/application_express/pdf/apex_text_application_v1.6.pdf
If you can help me with this I'd really appreciate it.
Thanks again.
Greg
Edited by: gjones77 on Dec 2, 2008 8:14 AM

How to search text in pdf file?

Hi all
I have to store the cover of a newspaper that include images and text and then should be able to search keywords in the cover.
I've read about to store in pdf format and use intermedia text.
I am just wondering the way to store and to do the search .
Thanks all

Hi,
You need store the PDF document in a BLOB column and create a CTXSYS index type.
e.g.: (.doc files)
CREATE INDEX I_DOC ON DOC_TABLE (DOC_COLUMN) INDEXTYPE IS CTXSYS.CONTEXT PARAMETERS ('SYNC (ON COMMIT)');Then you can test typing this SQL below:
select score(1) from DOC_TABLE where contains(DOC_COLUMN, 'My text', 1) > 0;In my case, i use this index for purpose to search on Word Documents (.doc)
Maybe this link help you to create an index type using FILTERS, in order to search on PDF files:
http://www.oracle.com/technology/products/text/htdocs/altfilters.htm
Cheers

APEX- How to open world document (and other files) stored in a blob column?

Hi,
I have an application developed in Oracle Apex 3.2.1 and RDBMS Oracle 9.2.0.8 (SUN SOLARIS SPARC 64 BITS).
I have a table called BAC_ARQUIVO, with the following structure:
CREATE TABLE BAC_ARQUIVO
(NAME VARCHAR2(4000), /* Name of the file stored in BLOB column */
SUBJECT VARCHAR2(4000), /* A brief description about this file stored */
ID NUMBER, /* An identification of this file - It´s a PK column of this table*/
BLOB_CONTENT BLOB, /* Blob field that store diferent kind of files - XLS, DOC, PPT, PDF, EXE, ZIP, etc);
MIME_TYPE VARCHAR2(4000) /* Identification about which kind of file is stored on BLOB field */
The field BLOB_CONTENT of this table can store:
- Executable files;
- MS-World Documents (DOC);
- Zip Files (ZIP);
- Ms-Excel Files (XLS);
- Ms-Powerpoint files (PPT);
- Adobe documents (PDF);
Actually, in this APEX application, i have developed a interactive Report in which the user can download these files to desktop without problems.
But now, my user would like to know if it´s possible that APEX application can OPEN some of these files stored on this BLOB column WITHOUT download it.
My user would like to see a hiperlink (or button) that, if pressed/acessed, can open a new page (or a page inside application) with the document. He doesn´t need to edit this oppened file. He only wants to see it.
For example, if the file stored in the blob column is ms-word type , so the application will open a MS-WINDOWS application with these file.
Is it possible?
I saw in this forum that it´s possible to open a file stored on "/i/" directory, but i need to open the file stored in BLOB column. I don´t want to store the file in operational system of the server.
Could any one demonstrate to me how it could be done?
Best regards,
Sergio Coutinho

Hi Sergio,
This link might be helpful:
http://www.oracle.com/webfolder/technetwork/tutorials/obe/db/apex/r31/apex31nf/apex31blob.htm
Also refer the BLOB Download Format Mask :
{message:id=9716335}
Here I have given the format mask for APEX 4.0, which will slightly differ for 3.2.1 and proposed changes
in the format mask are:
1) Format Mask: DOWNLOAD
2) Content Disposition: Inline
This will be achieved with it:
>
But now, my user would like to know if it´s possible that APEX application can OPEN some of these files stored on this BLOB column WITHOUT download it.
My user would like to see a hiperlink (or button) that, if pressed/acessed, can open a new page (or a page inside application) with the document. He doesn´t need to edit this oppened file. He only wants to see it.
>
Hope it helps!
Regards,
Kiran

How to store a PDF in a BLOB column, then select/search it using SQL?

We have a simple requirement.
(1.) We have an Oracle 10g DB.
(2.) We want to store PDFs/MS Word docs inside BLOB column in a table.
(3.) We need to show PDF/MS Word content to the user, without opening the PDFs/MS World like we normally do (i.e. Using PDV reader/ MS Word).
(4.) i.e. We need to use SQL statements and
(a.) Get all contents of the PDF/MS WORD doc.
(b.) Search for specific strings inside these docs in the BLOB columns.How to do this pls? Code sample would be really great.
Edited by: user12240205 on Jul 18, 2012 9:35 PM
Edited by: user12240205 on Jul 18, 2012 9:37 PM

rp0428 wrote:
>
We have a simple requirement.
>
Really? You mean as opposed to a requirement that might actually be doable?Could be that simple is used as in stupid ? Like in, "+look at this simply idiotic requirement, blah.. blah...+"? ;-)
@user12240205:
BLOBs contains binary data. Not text data. How do you propose to search binary data for strings?
PDF and Word formats include actual text strings and identifiers as part of the formatting. How do you propose that this be identified and ignored when searching for specific strings in the document's content?
PL/SQL does not have the ability to take a page from a PDF/Word document, highlight the search string, and display that to the client. A format-parser is needed to parse the PDF/Word format and generate that as content (e.g. HTML or XML) that can be rendered by a client. How do you propose PL/SQL serves up search strings that it found, when it cannot interpret the document format? (cannot even say what page, section or paragraph, that search string occurs)
Documents can be loaded via the DBMS_LOB interface. Either from the db server's file system, or by the client supplying the document as chunks for "write appending" to a BLOB. See Oracle® Database PL/SQL Packages and Types Reference guide for details.
Text search and document indexing are supported by an Oracle feature called Content Management<a> (which includes interMedia, UltraSearch, Oracle Text, etc).
Oracle also supports (on Window server platforms), the Microsoft Component Object Model (COM). COM interfaces (such as COM for Word) can be used for specialised processing (e.g. searching Word documents). Oracle's COM feature allows one to interface directly with external s/w (e.g. MS Word) via its COM interface.

Searching TEXT String in a document stored in BLOB Column

Similar Messages

Maybe you are looking for