Re: new things in 1.5.18
Thus spake Rocco Rutte [05/23/08 @ 15.28.00 +0200]:
>> First, I see that when mutt is Evaluating Cache or Fetching Headers (or
>> other things too), there is now a percentage displayed to the right of
>> the fraction that tells how far along mutt is in the process. There
>> used to simply be a fraction. Can I get rid of this? Since I've set
>> the (wonderful) new "time_inc" to 10000, what happens now is that mutt
>> displays "0%" unless the operation takes more than 10 seconds. I would
>> rather it not display any percentage at all. "0%" makes it look like
>> mutt has frozen. Also, the less cluttered the message the better.
>
> You cannot only remove the percentage (in general the percentage doesn't
> reflect the fraction and vice versa, but I think for IMAP it does).
>
> However, when you set time_inc this large, you won't get other progress
> updates within 10s since the whole progress line is updated, not just
> the percentage. So why don't you disable them at all when you don't need
> the information?
OK, so am I right that the following are my options? On the one hand, I could
set read_inc and write_inc to zero, which means there will be no fractions or
percentages; however this would make it pointless to set time_inc to a large
value, since I would get no updates anyway. Well, not pointless, because
there's net_inc too, and maybe I want to see progress updates on the uploading
and downloading of files, but only for big files that take a long time.
On the other hand, I could leave read_inc and write_inc at 1000, and set
time_inc to something large (now it's 10000); this way, I'd see fractions and
percentages, but: (1) they won't be updated unless the process takes more than
10 seconds, and (2) if it takes more than ten seconds the only updates I see
will be things like (12000/15545) "on the thousands" rather than (12878/15545).
Is that right? If so, I suppose the best thing is to let read_inc and
write_inc be 1000 or even bigger, while letting time_inc be 10 seconds. That
way, I will get no updates unless the mailbox is big, but I will have to deal
with fractions and percentages. At least this way, I can avoid being fooled
into thinking that mutt has frozen, since all I have to do is pay attention for
the 10-second-mark. If I set read_inc, write_inc to zero, then a *long*
operation could make me think that mutt's frozen or the internet connection is
flaking.
>> Second, when I change IMAP folders, there seem to be more steps than
>> before that mutt goes through before showing me what's in my folders.
>> I used to see Evaluating Cache, and then if there were no new messages
>> since the last caching, mutt would go straight to "Sorting Mailbox" and
>> then show me the folder. If there were new messages since last
>> caching, then mutt would say "Fetching Headers" while it grabbed the
>> rest of them. Now, I notice that even when the folder has just been
>> cached and no new content is added, mutt will *still* take the time to
>> display "Fetching Headers". 1.5.17 did not do this, if I recall
>> correctly. Can I switch this off somehow?
>
> Does this always happen for the same folder with the same mutt version?
> Mutt 1.5.18 contains a fix for header caching with a structrure
> different to 1.5.17. Whenever the structure differs, mutt 1.5.18 will
> consider a message missing cached with 1.5.17 and vice-versa. So
> fetching headers again for already cached folders could well be a
> one-time symptom.
It's not a one-time symptom. I have 1.5.18 on one machine and 1.5.17 on
another with no overlap. I rebuilt the caches fresh for 1.5.18. 1.5.18
insists on Fetching Headers when there are none to be fetched since they are
already cached. It's true that mutt doesn't spend very *long* on the needless
fetching, since it realizes fairly quickly that there's nothing for it to be
doing. But I still think it should not do that at all once it's evaluated the
cache. That adds time to the opening of the mailbox.
That's the gist of my query about 1.5.18; mutt now takes longer to open
mailboxes, because it's now going through all these apparently needless
processes, where 1.5.17 got right to the point. And displaying messages to the
effect that "mutt is doing X" when mutt really isn't doing X counts as a
needless process that makes opening a mailbox longer than necessary.
Also, 1.5.18 has not solved my previously posted problem of the seemingly
arbitrary need to completely rebuild the hcache for some of my mailboxes, even
when they haven't been changed. Some of my boxes have > 15000 messages, so
building the cache takes a long time.
> Note that you could make it even worse when constantly switching between
> mutt 1.5.17 and 1.5.18 if they use the same hcache...
Right, I figured that might happen, so that's why I kept one mutt version per
machine.
-gmn