[VM] bug on running vm-visit-imap-folder?

[Robert Marshall]

running 8.2.0b

Previously (not sure when the behaviour changed) when I ran
vm-visit-imap-folder I used to get prompted for the short name of the
imap account (and tab completion worked), then the password for the
account and then the account:folder - and tab completion also worked

Now the initial prompt expects account:folder as the first input and the
password only when you attempt to get the new email (I see this follows
the doc string). However when I attempt to tab complete (at the initial
prompt) I get the message:

read-passwd: Command attempted to use minibuffer while in minibuffer
vm-read-imap-folder-name: IMAP folder required in the format account-name:folder-name

I assume this is unintended, anyone else seeing this? Is it attempting
to read the password before it knows which imap folder?

emacs 24.1.50.1

Robert

[Uday Reddy]

Post by Robert Marshall
I assume this is unintended, anyone else seeing this? Is it attempting
to read the password before it knows which imap folder?

If you attempt to tab-complete the folder name, then VM has to ask the
server for the list of folders. So, it needs the password.

However, you are saying that the password reader is giving an error. Please
file a bug report with the exact key strokes that cause this to happen.

(The two separate prompts have been fused into one, quite some time ago I
think. The original separate prompts were before I had figured out how to
merge them into one.)

(If you want the password handling to be really smooth, you can use an
.authinfo.gpg file. Check the index in the manual under "passwords".)

Cheers,
Uday

[Robert Marshall]

Post by Uday Reddy

Post by Robert Marshall
I assume this is unintended, anyone else seeing this? Is it
attempting to read the password before it knows which imap folder?

If you attempt to tab-complete the folder name, then VM has to ask the
server for the list of folders. So, it needs the password.
However, you are saying that the password reader is giving an error.
Please file a bug report with the exact key strokes that cause this to
happen.

I think the problem is that I'm attempting to complete on the *server*
name so it doesn't know which server to send/ask the password to/for!

I'll file a bug

Post by Uday Reddy
(The two separate prompts have been fused into one, quite some time
ago I think. The original separate prompts were before I had figured
out how to merge them into one.)
(If you want the password handling to be really smooth, you can use an
.authinfo.gpg file. Check the index in the manual under "passwords".)

Will look into it, thanks

Robert

[Uday S Reddy]

Post by Robert Marshall

Post by Uday Reddy
If you attempt to tab-complete the folder name, then VM has to ask the
server for the list of folders. So, it needs the password.

I think the problem is that I'm attempting to complete on the *server*
name so it doesn't know which server to send/ask the password to/for!

If the server name completion works, VM then goes on to look for folder
names. So, I think this is fine.

Normally, VM asks for the password at this point. However, your Emacs
version is giving an error. It is likely that the problem is with the
Emacs 24 development version. If you try it on Emacs 23.3 or even the
Emacs 24 prerelease versions, it should work.

Cheers,
Uday

[Robert Marshall]

Post by Uday S Reddy

Post by Robert Marshall

Post by Uday Reddy
If you attempt to tab-complete the folder name, then VM has to ask
the server for the list of folders. So, it needs the password.

I think the problem is that I'm attempting to complete on the
*server* name so it doesn't know which server to send/ask the
password to/for!

If the server name completion works, VM then goes on to look for
folder names. So, I think this is fine.
Normally, VM asks for the password at this point. However, your Emacs
version is giving an error. It is likely that the problem is with the
Emacs 24 development version. If you try it on Emacs 23.3 or even the
Emacs 24 prerelease versions, it should work.

Yes I can confirm that this works with 23.3, I will try updating my bzr
pull and see if it then works and if not consider filing a bug against
emacs

Thanks

Robert