VM

8.2.0b - virtual folders not working

Bug #924016 reported by Uday Reddy
8
This bug affects 1 person
Affects Status Importance Assigned to Milestone
VM
Fix Committed
High
Uday Reddy

Bug Description

Julian Bradfield reports (viewmail-info, 2012-01-22):

My standard virtual folder (which is just the concatenation of three
files) comes up with no messages in it.
The three actual folder files have been loaded in to XEmacs, and are in
VM mode, but have "Unrecognized folder type".

If I visit the real files directly, VM recognizes them (they're mmdf -
well, two are mmdf and one is empty, but I tried removing the empty
one, just in case).

Tags: 8.2 virtual

Related branches

Revision history for this message
Uday Reddy (reddyuday) wrote :

Hi Julian,

Does

     M-: (vm-visit-folder FOLDER nil t)

 work for these folders?

Cheers,
Uday

Revision history for this message
Uday Reddy (reddyuday) wrote : Re: [Bug 924016] [NEW] 8.2.0b - virtual folders not working

I couldn't reproduce this problem with 8.2.0b using a couple of folders that
I freshly converted into MMDF format. You would also need to watch and see
if something is going wrong when your component folders are being visited.

Cheers,
Uday

Revision history for this message
Julian Bradfield (jcb+vm) wrote : [Bug 924016] Re: 8.2.0b - virtual folders not working

> M-: (vm-visit-folder FOLDER nil t)

No, it doesn't work. Unrecognized folder type.

It's not mmdf-specific, happens with vanilla mbox folders too.

Revision history for this message
Uday Reddy (reddyuday) wrote :

Julian Bradfield writes:

> > M-: (vm-visit-folder FOLDER nil t)
>
> No, it doesn't work. Unrecognized folder type.
>
> It's not mmdf-specific, happens with vanilla mbox folders too.

So, this call is breaking for you for all mail folders?

Does it work ok if the last argument is nil?

I still can't reproduce the problem. Checked it on XEmas too. So, you will
need to narrow down the problem. I suppose you know what to do? What
exactly is vm-folder-type after the folder is visited?

Cheers,
Uday

Revision history for this message
Julian Bradfield (jcb+vm) wrote :

>Does it work ok if the last argument is nil?

Yes, it does.

>I still can't reproduce the problem. Checked it on XEmas too. So, you will

How do you get vm into XEmacs? I can't even load new VM unless I first
load vm-macro and vm-vars .
(With 21.4 and 21.5, vanilla, no .vm)

>need to narrow down the problem. I suppose you know what to do? What
>exactly is vm-folder-type after the folder is visited?

'mmdf

As far as I can see, vm-build-message-list is not being called.

Revision history for this message
Uday Reddy (reddyuday) wrote :

Ok, I found the problem. I was probably not using the exact released
version when I tested at home.

The fix is in revision 1310 of the trunk, the day after the release. The
log doesn't say anything about it. I guess I somehow missed putting this in
the release and forgot all about it.

> How do you get vm into XEmacs? I can't even load new VM unless I first
> load vm-macro and vm-vars .
> (With 21.4 and 21.5, vanilla, no .vm)

I have this in my init file.

(if running-fsf
    (require 'vm-autoloads)
  ;; Use this only if there is no site-package installed
  (load-library (substitute-in-file-name "$VMDIR/lisp/auto-autoloads"))
  )

VMDIR is the directory where I built VM. I don't use a "package" normally,
but if the package is installed, XEmacs is supposed to load its
auto-autoloads file automatcially.

Cheers,
Uday

Revision history for this message
John Hein (xpqheqdvq4) wrote :

Is the fix for this the following:

@@ -285,7 +286,7 @@
       ;; builds message list, reads attributes if they weren't
       ;; read from an index file.
       ;; but that is not what the code is doing! - USR, 2011-04-24
- (unless revisit
+ (unless revisiting
        (vm-assimilate-new-messages :read-attributes t
                                    :gobble-order (not did-read-index-file)
                                    :run-hooks nil))

If so, this is already in the 8.2.x branch. I just wanted to make sure it'll be in 8.2.0

Revision history for this message
Uday Reddy (reddyuday) wrote :

Yup. Whatever is in 8.2.x branch will get included in 8.2.0.

Julian, do you want to download the tarball from

  http://bazaar.launchpad.net/~vm/vm/8.2.x/revision/1323

and see if it works?

Cheers,
Uday

Revision history for this message
Julian Bradfield (jcb+vm) wrote :

I've already tried 1310, and it was ok.

Uday Reddy (reddyuday)
Changed in vm:
status: Triaged → Fix Committed
Uday Reddy (reddyuday)
Changed in vm:
milestone: 8.2.90a → 8.2.89a
Revision history for this message
Uday Reddy (reddyuday) wrote :

Notes:

The change seems to have been made in rev. 1194, to handle Bug 614413.

Revision history for this message
Uday Reddy (reddyuday) wrote :

Actually the original change was made in rev. 653.1.107, to handle Bug 561718.

Revision history for this message
Uday Reddy (reddyuday) wrote :

Discussion on viewmail-info (2012-02-02):

>Hi Julian, this might be a duplicate of
> https://bugs.launchpad.net/vm/+bug/909501
>Can you try revision 1311 please?

Yes, that fixes the folder changing.
However, it's still creating Summary and Presentation buffers for all
the component folders, which seems highly undesirable. (Especially
since I sometimes use virtual folders with a hundred component
folders!)

Revision history for this message
Uday Reddy (reddyuday) wrote :

Rev. 1351 suppresses the summary generation and retrieval of new mail etc. Presentation buffer is still created. (This was present in VM 8.1.1 as well.)

We might consider getting rid of the Presentation buffer sometime down the road.

Julian, can you test to see if this works ok?

Cheers,
Uday

Uday Reddy (reddyuday)
no longer affects: vm/8.2.x
Changed in vm:
milestone: 8.2.89a → 8.2.0b1
Uday Reddy (reddyuday)
Changed in vm:
milestone: 8.2.1a → 8.2.0
Uday Reddy (reddyuday)
Changed in vm:
milestone: 8.2.1a → 8.2.90a
To post a comment you must log in.
This report contains Public information  
Everyone can see this information.

Other bug subscribers

Remote bug watches

Bug watches keep track of this bug in other bug trackers.