Home » Asset Suite & PassPort » Object embedding & linking (OLE) » Miscellaneous object embedding and linking (OLE) topics

Reading the URL Address of a linked document

I'm trying to figure out if it's possible to read/report the URL address for documents linked in PassPort. I use the raw_to_long to display the text from embedded documents, but the returned ole_object_id field is null when the document is linked.

‹ Fixing very slow OLE response times Portal/J working with OLE via ActiveX ›
By Chris36695 | | add new comment

RE: Reading the URL Address of a linked document

Thanks for the response. I'm a Business/Data Analyst, so I forwarded your comments to our PassPort DBA who was unable to decode the BLOB to view the links in previous attempts. His comment was; "COOL!...The last missing piece of the puzzle was the two missing bytes on the front of the file to get it uncompressed…"

I'll let you know if it works out and what had to be done to get the URL info.

By Chris36695 | reply

Java program to do uncompress

There's a Java program to do the Unix uncompress at: http://www.unidata.ucar.edu/support/help/MailArchives/netcdf/msg03482.html

The code needs to be modified to remove the check for the magic number (those first 2 bytes).

By writing a Java Servlet and invoking OleRead, the uncompress, and a search program, it is possible to implement a dynamic "strip data out of TIDBLOB" and have it be user triggered via the "URL alias" data dictionary element.

By webmaster | reply

Reading OLE(?) containing link from TIDBLOB.

Hi,

It is possible to read and report link definitions from OLE's stored within PassPort. Oracle's UTL_RAW.CAST_TO_VARCHAR2 function will work with text data stored within TIDBLOB where TIDBLOB.OLE_OBJECT_TYPE = 'T', but will not work when TIDBLOB.OLE_OBJECT_TYPE = 'b' (little b) because little b in this case indicates an OLE storing a link. If your data has TIDBLOB.OLE_OBJECT_TYPE = 'b', then to extract the OLE link location it is necessary to read the data into a file, uncompress it, and look for a pattern.

See if you can get the recently posted OleRead.java to work. It reads the data out of TIDBLOB and writes it to a file. Use a hex editor like UltraEdit to add two bytes (x1F x9D) to the beginning of the file. Put the file on a Unix box and rename the file so that it ends with ".Z". Use the Unix uncompress command to uncompress the file. Look at the file with a hex editor. Search for mapped drives by searching for x5C x5C, then search for logical drives by searching for x3A x5C.

Please let me know how this works out.

By webmaster | reply

Comment viewing options

Select your preferred way to display the comments and click "Save settings" to activate your changes.