From: Steve Fabian
| 2/ For portable drives (and pseudodrives) it would be useful for their
| segment of the database to reside in their own root directory, WITHOUT
| driveletter, because even on the same system they may be plugged into
| a different USB port at different times, thus responding to different
| driveletters.
From: Kachupp
| 2/ Each volume has its own serial number It doesn't get anymore
| specific than that.
|
| I have a script that does that specifically for USB plug and play It
| detects that very "difference"
| It isn't tricky at all with a btm, can't be too hard for the OS/shell
| either. Hello plugin Vince!
When used on a single system using different ports, this would be perfect. But it would require that my memory stick be cataloged on each system on which I may use it, leaving behind a junk file when I leave. Furthermore, since the serial number is just a 32b integer, there is no assurance that you don't have a disk with the same serial number as one of mine.
Another point about my scheme: when you rebuild your system, adding a NEW drive as C:, you need to recatalog everything. My scheme (each drive has its own index file) would not require that. Considering the availability of memory, I would use for the directory database a method similar to the one for global aliases, etc. - when a drive becomes accessible (by TCC start-up OR by plugging in OR by network mapping) TCC should read its index file (WITH drive letters added as applicable at the moment). Any updates to the table (explicit via CDD /S or implicit by directory creation or deletion using a TCC command which is allowed to update the datase) are written to the device's index file immediately. When a device becomes inaccessible (unmapping or turning off power of a network drive, unplugging a USB device, etc.) its entries would be removed from the live database table. BTW, this method would retain the ability to temporarily add read-only devices (CD-ROM, DVD-ROM) to the database, with the additional benefit that when a different CD is inserted into the drive, the previous one's directory database will not be used to access it.
--
Steve
