Physical Structure Details

Top  Previous  Next

A library is a disk directory on the machine where the UnForm server runs.  This can be a full path, or it can be a simple name, in which case its path becomes "serverpath/arc/name", a directory in the "arc" subdirectory of the UnForm server location.

 

Within this directory are several data files and subdirectories, described below.

 

cats.dat

Category indexes

control.dat

Control file used for internal management

docdata.dat

Document user-defined data, for future use

docs.dat

Indexed document property file

docs2.dat

Additional document property data

keywords.dat

Keyword indexes

subdocs.dat

Indexed sub-document properties

images/

Directory structure that contains encrypted/compressed image data files

recover/

Directory that contains transaction data used for library recovery

 

 

Differences from 7.x/8.0

All file names are different, so the same library directory can host both new and old formats without conflict

Keyword indexing has been added

The design allows for adding indexed, user-defined data in the future

The image store does not contain shadow records, so will be smaller

The recover directory stores transaction data by day, and provides a faster recovery facility should the library ever need to be rebuilt.  It is also possible to copy the recover directory to a new system and generate a duplicate library.  
 
If disk space becomes an issue, you can backup old recover files and remove them.  In the event a rebuild is necessary, you will have to restore the old files before running the rebuild.  If you don't need all the records rebuilt, such as records you'd like purged anyway, you only need to restore date files required.