Using a database repository, it is easy enough to select a sum of the length of the message_body column to get a close estimate, however I have converted to a dbfile repository and have concerns with how the mail file names are produced. They appear in the form:
4D61696C313037373230383830373132332D33.Repository.FileStreamStore
I naturally assumed the first part to be hex encoded, so I decoded that and found something more useful:
Mail1077208807123-3.Repository.FileStreamStore
Yay, a message_name! So, given that discovery is it "okay" to select the message_names for a repository, encode them, list the files from the file system after appending ".Repository.FileStreamStore" to each, and sum up the file sizes?
Is this a total hack or what? Is there a better way? I realize this is a performance hit. I could keep a running tab and update it after deliveries and deletes or something, but that is for later determination. Am I missing something simple or is this the best approach? Is it reliable to use this methodology or will I be screwing myself when later versions of James/Avalon come out?
Any comments are welcome (both positive and negative).
--------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
