jose.kahan_at_w3.org (Jose Kahan) writes:
>Hi,
>
>This patch removed the call to the hashed msgid num. The patch
>should have been applied to message_name().
Why? It looks to me like the patch I applied fixes behavior that, even when using hashed filenames, was broken and now works. The msgnum was not being used as a filename, although it was used in a similar way.
>If using the hashed msgid num is something you don't want, let's remove
>it completely!
The hashed filenames don't appear to do anything usefull, so removing
them sounds like a good idea.
As long as each message has a unique msgnum, that number works well for
identifying messages. The patch I applied is only a problem for the goal
of being able to distinguish messages that have the same msgnum. But that
goal appears harder than insuring that a msgnum uniquely identifies a
message.
The message_name function for controlling filenames is probably better
than what we had before (it should allow us to control the number of
leading zeroes in filenames), so I'm glad you implemented that.
-- ------------------------------------------------------------------------------ Peter McCluskey | http://www.rahul.net/pcm |Received on Wed 29 Jan 2003 07:25:51 PM GMT
This archive was generated by hypermail 2.2.0 : Thu 22 Feb 2007 07:33:54 PM GMT GMT