Further thought on the subject come up with a possible no-scripting,
no-programming solution which might of use to you.

You might care to use MHonArc [1] or hypermail [1a] to convert those
emails which are not already in HTML format to it and then index the
lot using htdig:// [2] as necessary, which may not be much as MHonArc
has built-in indexing. You will have to experiment.

[1] http://www.mhonarc.org/
or
[1a] http://www.hypermail.org/docs.html

[2] http://www.htdig.org/

This would provide a full word index to all the mails. Take care,
htdig is a disk hog, and needs about as much, if not more, disk space
for its indices as the original data. The htdig site has a demo, so
you can see what you are going to get.

All care, but not one jot of responsibility. It should work though.


On 7/31/08, Andrew Errington <[EMAIL PROTECTED]> wrote:
> On Wed, July 30, 2008 21:33, Paul Swafford wrote:
>> I have a M$ solution for this issue but looking for a Linux alternative.
> <snip>
>> Prize for the most elegant solution!
>
> The most elegant solution, in my view, is the one that requires least
> work.  So, leave the M$ solution in place.
>
> Prize, please.
>
> A
>
> PS Joking aside, you have given no context for this application, so there
> are too many unknowns.  It might be that the application could be
> streamlined by integrating other parts of the system instead of just
> replacing this chunk.
>
> That notwithstanding, you already *have* a database containing the
> requisite information- it is the mailstore itself.  There are already
> tools for trawling the mailstore, so why copy the data to a 'database'?  A
> 'database' is just a filestore with a process sitting on top of it to
> manage it.
>
>


-- 
Sincerely etc.
Christopher Sawtell

Reply via email to