RE: on-the-fly html encoding

Daniel Stenberg wrote:
> Why not just convert the mailbox to HTML and remove the original
> mailbox file? It seems like a waste of resources to do on-the-fly
> converting of static contents.

The problem I have is that the mbox -> hypermail process is not reversible, and hypermail doesn't provide archive maintenance functions.

The simplest installation creates a single archive which grows over time, with degrading performance (chiefly in large index files). There have been a variety of ad hoc solutions to creating monthly archives, but they all depend on having source messages to send through hypermail.

Having failed to save mbox source, or to establish a segmenter up front, I have no way to delete unwanted messages, segment a large archive into smaller ones on month, year, or other boundaries, etc.

The missing enhancement, imho, is a tool or tools to perform general archive maintenance. Having a hypermail -> mbox process seems like a short path to that end, although it leaves a potentially brute force task of manipulating mbox files. Shane's idea (I think) is to keep the unadulterated source, and generate HTML as needed. Hypermail's current status is to generate HTML and discard the source.

Maybe I'm overstating the difficulty? Say I have an archive with 0000.html...1000.html, some messages of which have pointers to attached documents which have been decoded, filed and referenced by hypermail. How do I:

   Delete messages (and any attachments that came with) older    than 12 months?
   Move messages older than 12 months to another archive?    Delete message 0164.html and its attachments?

and so on.
_____________ Hewlett-Packard Personal Storage Business Tom von Alten mailto:Tom_vonAlten_at_boi.hp.com

          This posting is for informational purposes only.
          It is not a statement of the Hewlett-Packard Co.