🚀 go-pugleaf

In article <830821635snz@wbriscoe.demon.co.uk>
           walter@wbriscoe.demon.co.uk "walter briscoe" writes:

[snip]

> 
> I expect to submit the work to Malcolm within the week unless I find a
> catastrophic hole I cannot putty within that time.

I am afraid I found a catastrophe: an agency found me a contract!
I shall try to find time to finish renumbering articles from the front
at the weekend.

Input             Current output     Revised Output
01 Marked unread  01 Marked Unread   27 Marked Unread
02 Old A
...
27 Old Z
28 New A          02 New A           28 New A
29 New B          03 New B           29 New B

With my current implementation, articles are renumbered to fit with the
unread articles; with the new scheme, the unread articles are renumbered
to fit with the articles which are newer than the threshold.  In
consequence, the history information is only wrong - if there - for the
articles kept because they are marked as unread.  This requires a second
pass through the index file for each affectd group.
--
walter briscoe