announce: nnmaildir 2001.12.19

announce: nnmaildir 2001.12.19

[Paul Jarc]

<URL:http://multivac.cwru.edu./nnmaildir/>
One new function, nnmaildir-article-number-to-file-name, and two
bugfixes: -request-retrieve-headers was assuming the article range was
sorted in ascending order, which is not the case for [/ o]; and
-open-server was assumed to be called before anything else, which is
not always the case when defining a server in the *Server* buffer
(thanks to Harry Putnam for helping to find this one).


paul

Re: announce: nnmaildir 2001.12.19

[Harry Putnam]

prj@po.cwru.edu (Paul Jarc) writes:

> <URL:http://multivac.cwru.edu./nnmaildir/>
> One new function, nnmaildir-article-number-to-file-name, and two
> bugfixes: -request-retrieve-headers was assuming the article range was
> sorted in ascending order, which is not the case for [/ o]; and
> -open-server was assumed to be called before anything else, which is
> not always the case when defining a server in the *Server* buffer
> (thanks to Harry Putnam for helping to find this one).

Paul, I noticed  a warning on our pages about skipping versions and
what it might cause, but I see no handy way to tell which version is
currently in use.  Nothing in the *.el file to indicate version.

I already know my version was the previous release but seems there
should be some way to tell.

Also maybe a condensed version of the info on our pages as a
`comments' section.  Findable with M-x finder-commentary.

I'm going to take a stab at that, and will send you the result.

Also, I've considered moving our discussion about nnmaildir from
gnu.emacs.gnus to here.  It has gotten somewhat detailed.

There should be a private message on you incoming spool some where
with a detailed report, see:

From: Harry Putnam <reader@newsguy.com>
In-Reply-To: <m3g069t8ka.fsf@multivac.cwru.edu> (prj@po.cwru.edu's message
 of "Tue, 18 Dec 2001 00:12:47 -0500")
Date: Tue, 18 Dec 2001 11:15:22 -0800
Message-ID: <x7vgf4tk51.fsf@reader.newsguy.com>

Re: announce: nnmaildir 2001.12.19

[Paul Jarc]

Harry Putnam <reader@newsguy.com> wrote:
> prj@po.cwru.edu (Paul Jarc) writes:
>> <URL:http://multivac.cwru.edu./nnmaildir/>
>
> Paul, I noticed  a warning on our pages about skipping versions and
> what it might cause,

That hasn't really been an issue lately.  A few versions ago, there
was a big change in the on-disk format, and I put in some code to
migrate the old information so nothing would be lost, and then removed
that code later.  So skipping the version in the middle could be
harmful.  I doubt there will be any more such changes, and if there
are, we can keep the migration code around longer.

> but I see no handy way to tell which version is currently in use.
> Nothing in the *.el file to indicate version.

See the variable nnmaildir-version (which could probably be removed
now, since nnmaildir is now in CVS).

> Also maybe a condensed version of the info on our pages as a
> `comments' section.  Findable with M-x finder-commentary.

Ah - is that simply pulled from the comment at the top of the file?  I
think I can take care of that.  I'm working on an addition to the
manual, too.

> There should be a private message on you incoming spool some where
> with a detailed report, see:

I haven't seen it.  Resend?


paul

Re: announce: nnmaildir 2001.12.19

[Harry Putnam]

[-- Attachment #1: Type: text/plain, Size: 2526 bytes --]

prj@po.cwru.edu (Paul Jarc) writes:

> Harry Putnam <reader@newsguy.com> wrote:
>> prj@po.cwru.edu (Paul Jarc) writes:
>>> <URL:http://multivac.cwru.edu./nnmaildir/>
>>
>> Paul, I noticed  a warning on our pages about skipping versions and
>> what it might cause,          ^^^

He he... In case you were wondering how your pages got to be our
pages, it was thru the magic of typo wizardry.

[...]

>> but I see no handy way to tell which version is currently in use.
>> Nothing in the *.el file to indicate version.
>
> See the variable nnmaildir-version (which could probably be removed
> now, since nnmaildir is now in CVS).

Why such a complicated unlikely name... :-)

>
>> Also maybe a condensed version of the info on our pages as a
>> `comments' section.  Findable with M-x finder-commentary.
>
> Ah - is that simply pulled from the comment at the top of the file?  I

Yes, but there isn't currently a comment section. At least not in the
download I got.

>> There should be a private message on you incoming spool some where
>> with a detailed report, see:
>
> I haven't seen it.  Resend?

I guess I'll go ahead and air it here.  Maybe someone else will be
interested.

Excerpted from previously sent message:

prj@po.cwru.edu (Paul Jarc) writes in mesg-id:
Message-ID: <x7vgf4tk51.fsf@reader.newsguy.com>

[...]
> 
> > But now that we know that nnmaildir-save-mail is being called, I'd say
> > we probably don't need to step through it.  Instead, debug-on-entry
> > nnmaildir-request-accept-article and step through that.  (But if the
> > debugger never pops up, then that means that nnmaildir-save-mail isn't
> > calling nnmaildir-request-accept-article, so we'd need to step through
> > nnmaildir-save-mail to find out why.)  As you step through, make note
> > of the results of the controlling tests of all the "if" and "when"
> > calls.
> 
> I collected quite a bunch of data, but thought maybe better to take
> this off list.  Probably getting a little boring, and too detailed.
 
 <Maybe more appropriate here on ding>
 
> Attached is a report from debug-on-entry
> nnmaildir-request-accept-article

This was compiled before your new version was released, not sure how
much may be different now.  Just now getting the new version installed.
 
> I recorded the steps (about 45) and `c'ed thru some stuff that looked
> repetitious. Hard to know where to stop, seems like it could go on
> into infinity.  I hope this is enough to figure out something.
> 
> The report starts with a brief description of my setup.


[-- Attachment #2: debug_data --]
[-- Type: application/octet-stream, Size: 18864 bytes --]