Background

In some situations, you may wish to completely remove records from the Symphony OCR database.  One example of this is when the user changes the selected PGs in the Worldox configuration screen.  Changes to the Worldox configuration screen impact the *finder* component of Symphony OCR only.  Any documents that are already in the database will continue to be processed, even if they reside in profile groups that have been de-selected.

Procedure

Important: this procedure uses the Advanced/Debug interface for Symphony OCR.  Be careful!

Option 1 - remove many explicit documents

1. Create a text file with the full path of the document records you wish to remove.  Place the file in a location that is available to Symphony OCR (generally a file on the disk of the workstation that runs Symphony OCR)
   Tip: You can use the Export CSV button in the Processing list to obtain a list of all documents in the processing queue.  This CSV can then be edited to include only the paths that you wish to delete.  Be sure to remove the other columns!
2. Click Advanced
3. Type the full path of the text file created in step 1 in the 'Or files in the following file' field:
   
4. Click the Delete Database Records for Specified Files button

Option 2 - Copy and paste from Worldox

This option works well if you wish to delete records for a handful of documents (up to about 150)

1. In Worlodox, search for the documents
2. Select the documents and type Ctrl+C - this will place the paths of all of the documents onto the clipboard
3. In Symphony OCR, click Advanced
4. Paste the paths into the 'List files to perform debug operations on' field
5. Click the Delete Database Records for Specified Files button

Option 3 - Remove all document records from Symphony OCR

If you wish to remove all document records from Symphony OCR, you can reset the Symphony OCR internal database.  When Symphony OCR is restarted, it will rebuild the database based on the documents that it finds.  This will result in a loss of all processing history, but any documents that have already been OCRed will not be affected, and records for those documents will be recreated and placed in the Processed list without consuming additional processing capacity.

1. Close Symphony OCR using the Quit side-bar link
2. Navigate to Program Files\Trumpet\Symphony OCR\Data
3. Rename the following files:
   
   • documents.db  -> documents.db_old
   • documents.lg  -> documents.lg_old
   • ProcessorPerformance.dat  ->  ProcessorPerformance.dat_old
4. Restart Symphony OCR