Re: [PATCH] Jumping to message by ID

[Andrea Monaco]

  > The name of the function doesn't seem to reflect what it does.  Can
  > you come up with a better name?

What about rmail-match-email-address-around-point?  Of course I'm
looking for message ids, but those have the same syntax as email
addresses and the latter are surely known to whoever reads my function.


  > If so, perhaps it would make sense to update the data each time we
  > update the summary, instead of punishing the first call after
  > getting new email?

You're right, that's the same issue as rmail-summary-by-thread.

In this case I only need the message id hash table so I could just
update that, at the risk of making it out of sync with the vectors of
parents and descendants.

The only solution is updating all the data structures each time new mail
is got or mail is expunged, instead of recreating them from scratch; I
can look into that.



Andrea Monaco