Searching Files and Workspaces

[jgo6d]

I have been asked to create a tool that would allow a user to search the
files in the Groove workspaces on his or her computer. The user could search
by file name, creation date, file type, etc. I am having difficulty finding
information on how this might be done and how the Groove file system is set
up. I thought that I might be able to make an application that would simply
search the file information for the Groove files stored on the hard drive.
If anyone has any suggestions on how this might be done or information on
accessing the Groove file system, please let me know. Thank you.

 

[Hugh Pyle [MSFT]]

Groove Files tool contents are stored on disk in an encrypted binary form,
which is not accessible directly.

You could write an application which uses Groove Web Services APIs to read
the contents of each files tool into your index. If you plan to do this,
you should be careful about performance (since opening every workspace in
quick succession will increase Groove's memory usage; and extracting the
full content of every file to a temporary indexing location will be slow and
also use plenty of RAM), and about security (because Groove users can
typically assume that the content of their workspaces is encrypted on-disk).

Do you need to index only the metadata (file names, types etc) or the full
content of files?

 

[jgo6d]

Mr. Pyle,

Thank you for your response. I will be working with the metadata and not
the actual contents of files. I don't know if this makes a difference in
terms of having to open each workspace and file. If there is no way for me
to get the metadata without opening every workspace or file, I may decide to
do a batch-type update instead where running the application stores workspace
and file information which can later be searched by the user. The obvious
problem is searching potentiall out-dated information.

Any other input you may have would be greatly appreciated. Also, is there a
good source for Groove Web Services API documentation? Thank you

 

[Hugh Pyle [MSFT]]

Indexing only the file metadata will be substantially easier (& faster) than
indexing file contents. You will still need to open each workspace,
enumerate its tools, and for each Files tool call Read() with the
"recursive" flag. To avoid excessive memory growth, try not to open lots of
workspaces in quick succession (i.e. build your index slowly).

You could also subscribe for Groove events, to be notified when files are
added/updated/removed, so the full scan would only be necessary one time.
See the doc on GrooveSubscriptions::CreateAggregated2() - the "handling web
services events" topic in the help file.

SDK is here:
http://www.microsoft.com/downloads/...E9-E1B9-4A10-BEEA-1FD906B77F92&displaylang=en



--
Hugh Pyle / Program Manager / Microsoft Office Groove
http://blogs.msdn.com/hughpyle/

This posting is provided "AS IS" with no warranties, and confers no rights.


in message
news:[email protected]...