Re: [VM] Are there plans to release 8.2?

[Uday Reddy]

Göran Uddeborg writes:

> I've been waiting to upgrade the Fedora package of VM until 8.2 is
> declared released. After all, 8.2.0b is officially a beta, right?
> But there are some issues with waiting, most recently illustrated by
> the bug report https://bugzilla.redhat.com/show_bug.cgi?id=960295
>
> Do you have plans to make an official 8.2.0. Or should I package
> 8.2.0b anyway. True, that "officially released" stamp doesn't make
> that much difference in reality. But it would feel a bit better. :-)

Unfortunately, it does make quite a bit of difference. Once it is an
official release, it goes into Linux distributions, and people get it by
doing routine updates. If things break for them, it would be quite
annoying.

Things were quite different when people made conscious decision to upgrade.
But Linux has changed all that.

The earliest date I can think of for the official release of 8.2.0 is the
end of summer. I lost a window of opportunity last summer and haven't had
much time to devote to VM afterwards.

We will have another beta release before the official release.

Cheers,
Uday

[John Stoffel]

>>>>> "Uday" == Uday Reddy <usr.vm...@gmail.com> writes:

Uday> Göran Uddeborg writes:
>> I've been waiting to upgrade the Fedora package of VM until 8.2 is
>> declared released. After all, 8.2.0b is officially a beta, right?
>> But there are some issues with waiting, most recently illustrated by
>> the bug report https://bugzilla.redhat.com/show_bug.cgi?id=960295
>>
>> Do you have plans to make an official 8.2.0. Or should I package
>> 8.2.0b anyway. True, that "officially released" stamp doesn't make
>> that much difference in reality. But it would feel a bit better. :-)

Uday> Unfortunately, it does make quite a bit of difference. Once it is an
Uday> official release, it goes into Linux distributions, and people get it by
Uday> doing routine updates. If things break for them, it would be quite
Uday> annoying.

Uday> Things were quite different when people made conscious decision to upgrade.
Uday> But Linux has changed all that.

Uday> The earliest date I can think of for the official release of 8.2.0 is the
Uday> end of summer. I lost a window of opportunity last summer and haven't had
Uday> much time to devote to VM afterwards.

Uday> We will have another beta release before the official release.

Can you push out a Beta release now, with all the patches you have
accumulated? Release early and release often is a good mantra to
have. We've all been stuck in a holding pattern lately, and it's not
fun.

Personally, I've been thinking that it's time to finally drop VM and
move to some other mail reader with good IMAP support, probably mutt.
But the hassle of either A) learning new key bindings or B) changing
mutt to have all the emacs/vm bindings has been holding me back.

As far as I'm concerned, the problems I see lately are:

- on startup, vm doesn't auto-load vm-reply properly. It's probably a
stupidly simple bug, but it's annoying. Basically I have to 'r'eply
to a mesage before 'm' to create a new message will work.

- IMAP support so that vm-visit-imap-folder will ask for a password
properly. I figure using the code that vm-list-imap-folders uses
would be enough, but my elisp skills are non-existent.

- Fix IMAP support when run against exchange mail servers. If I read
a message, and file it locally, it seems that Exchange doesn't like
that command, and VM can't handle it either. I have to go through,
find all the filed emails and delete them before I can get VM to save
properly.


Uday, please just send out a patch to bring us to 8.2.0c so people can
at least *test* proposed changes, and maybe comment on them and give
you some feedback.

Thanks,
John

[Uday Reddy]

John Stoffel writes:

> As far as I'm concerned, the problems I see lately are:
>
> - on startup, vm doesn't auto-load vm-reply properly. It's probably a
> stupidly simple bug, but it's annoying. Basically I have to 'r'eply
> to a mesage before 'm' to create a new message will work.

If you start up by doing M-x vm, then everything starts up just fine. The
only time I know where this doesn't work is if you start VM by running
mail-user-agent, perhaps by clicking on a URL somewhere.

> - IMAP support so that vm-visit-imap-folder will ask for a password
> properly. I figure using the code that vm-list-imap-folders uses
> would be enough, but my elisp skills are non-existent.

What you are looking for is automatic password management. Emacs does it by
using EasyPG (a gnu encryption library). I am attaching an extract from the
VM manual that describes how to set it up.

> - Fix IMAP support when run against exchange mail servers. If I read
> a message, and file it locally, it seems that Exchange doesn't like
> that command, and VM can't handle it either. I have to go through,
> find all the filed emails and delete them before I can get VM to save
> properly.

Looking back at that discussion, I find that we discovered that your
Exchange server was misbehaving. I asked you for information on how VM
breaks as a result of it, and haven't heard from you since.

Cheers,
Uday

--------

If your environment has the "EasyPG" utility and your version of
Emacs supports it, i.e., has the `epa-file' and `auth-source' libraries,
then you can store password information in a file such as
`.authinfo.gpg'. The `EasyPG' protocol allows you to store this
information in an encrypted form so that it cannot be read by third
parties. Each line in the `.authinfo.gpg' file should be of the form

machine HOST login USER password PASSWORD port PORT

where HOST, USER, PASSWORD and PORT are as detailed above. Ensure that
the variable `auth-sources' is customized to refer to your authinfo
file. *Note Help for users: (auth)Help for users. Then VM will read
passwords from the file and you don't need to type them in when
accessing mail servers.

If you have multiple login accounts on the same HOST then VM will
only use the first login listed in the authinfo file. To allow for
multiple login's, the HOST entry in the authinfo line can be replaced
by an account name as defined internally in VM. These account names are
defined via the variables `vm-pop-folder-alist' and
`vm-imap-account-alist', described below.

[Patrick P Murphy]

On Fri, 24 May 2013 07:26:10 +0100, Uday Reddy <usr.vm...@gmail.com>
said:

> John Stoffel writes:

>> - on startup, vm doesn't auto-load vm-reply properly. It's probably
>> a stupidly simple bug, but it's annoying. Basically I have to
>> 'r'eply to a mesage before 'm' to create a new message will work.

> If you start up by doing M-x vm, then everything starts up just fine.
> The only time I know where this doesn't work is if you start VM by
> running mail-user-agent, perhaps by clicking on a URL somewhere.

And running "emacs -f vm" from the command line, or an X session
startup.

- Pat

--
Patrick P. Murphy, Ph.D. Webmaster (East), Computing Security Manager
http://www.nrao.edu/~pmurphy/ http://chien-noir.com/maze.shtml

[John Stoffel]

>>>>> "Uday" == Uday Reddy <usr.vm...@gmail.com> writes:

Uday> John Stoffel writes:
>> As far as I'm concerned, the problems I see lately are:
>>

>> - on startup, vm doesn't auto-load vm-reply properly. It's probably a
>> stupidly simple bug, but it's annoying. Basically I have to 'r'eply
>> to a mesage before 'm' to create a new message will work.

Uday> If you start up by doing M-x vm, then everything starts up just
Uday> fine. The only time I know where this doesn't work is if you
Uday> start VM by running mail-user-agent, perhaps by clicking on a
Uday> URL somewhere.

All I know is that it doesn't work for me. I have:

(require 'vm-autoloads)

in my .emacs file. I guess I need to chase this down and see what's
happening here. Maybe I have a corrupted one, or an old one in my
emacs search path which is screwing things up.

>> - IMAP support so that vm-visit-imap-folder will ask for a password
>> properly. I figure using the code that vm-list-imap-folders uses
>> would be enough, but my elisp skills are non-existent.

Uday> What you are looking for is automatic password management.
Uday> Emacs does it by using EasyPG (a gnu encryption library). I am
Uday> attaching an extract from the VM manual that describes how to
Uday> set it up.

Umm... so how come 'vm-list-imap-folders' properly asks for my
password, but 'vm-visit-imap-folder' does not? I have to list my
folders first, then I can visit the folder.

>> - Fix IMAP support when run against exchange mail servers. If I read
>> a message, and file it locally, it seems that Exchange doesn't like
>> that command, and VM can't handle it either. I have to go through,
>> find all the filed emails and delete them before I can get VM to save
>> properly.

Uday> Looking back at that discussion, I find that we discovered that your
Uday> Exchange server was misbehaving. I asked you for information on how VM
Uday> breaks as a result of it, and haven't heard from you since.

Sorry, my fault. I was trying to come up with a repeatable test case
that was easy to show the problem. I *seems* to be because I do:

- vm-visit-imap-folder
- read a message
- file the message to a *local* folder.
- then expunge any changes with vm-expunge-folder followed by
vm-save-folder. It's the vm-save-folder which seems to try to
update the imap flags. Exchange fails the update command and
vm doesn't seem to properly flush all the changes.

But it's not quite clear to me what the root cause is. I'll try to
get a test case for you, to make it more clear what's going on. But I
wonder if vm is just seeing the errors from the IMAP connection, and
assumes that all the commands failed? And it's not clear when it's
going to update it, because I just read a new message and filed it to
a folder, but it worked.

And even with multiple seen/saved messages it works, even though the
trace log shows:

VM STORE 929 +FLAGS.SILENT (filed)
VM BAD Command Argument Error. 11
VM STORE 930 +FLAGS.SILENT (\seen filed)
VM BAD Command Argument Error. 11

So I wonder if maybe it's when more than X number of failures happens
that causes my issue? It seems to have when I have more than 10 or
save seen/filed messages, but that's just a guess. I'll see about
getting more details.

But this all just ignores my root question, why don't you release what
you have now as 8.2.0c so we can see what's coming down the pike?
Release early and often!

John

[Tim Cross]

Just a minor point, but which might throw another light on things. 

All Linux distos I know of will only include the new version of VM into their NEXT release, not upgrade existing versions (this would only happen when there is significant security fixes and even then, usually only the security patches are backported). 

While this does not regate your concern regarding breaking existing setups, I do think it modifies it slightly as the upgrade will only occur when the user explicitly asks for a distribution upgrade. Most users are expecting some breakage when doing a full distro upgrade compared to a normally update. 

The other point to consider is that this could actually be beneficial. Distros put their next version through considerable testing and new package versions are normally only included if they pass this testing. Releasing could assist VM by getting additional testing. If issues come in which cannot be resolved due to lack of time;resources, no big deal as the worst case scenario is the disto ships with the last known (current) good verison rather than the new one. 

Just a different perspective. As the bulk of maintenance falls on Uday's shoulders, I'm fine with whatever decision he makes as he is the only one in a position to judge what level of commitment he can afford at this time. 

Tim

On 21 May 2013 09:01, Uday Reddy <usr.vm...@gmail.com> wrote:

Göran Uddeborg writes:

> I've been waiting to upgrade the Fedora package of VM until 8.2 is
> declared released.  After all, 8.2.0b is officially a beta, right?
> But there are some issues with waiting, most recently illustrated by
> the bug report https://bugzilla.redhat.com/show_bug.cgi?id=960295
>
> Do you have plans to make an official 8.2.0.  Or should I package
> 8.2.0b anyway.  True, that "officially released" stamp doesn't make
> that much difference in reality.  But it would feel a bit better. :-)

Unfortunately, it does make quite a bit of difference.  Once it is an

official release, it goes into Linux distributions, and people get it by

doing routine updates.  If things break for them, it would be quite

annoying.

Things were quite different when people made conscious decision to upgrade.

But Linux has changed all that.

The earliest date I can think of for the official release of 8.2.0 is the

end of summer.  I lost a window of opportunity last summer and haven't had

much time to devote to VM afterwards.

We will have another beta release before the official release.

Cheers,
Uday

--
Tim Cross

[Uday Reddy]

John Stoffel writes:

> But this all just ignores my root question, why don't you release what
> you have now as 8.2.0c so we can see what's coming down the pike?
> Release early and often!

Because I have already mentioned that I didn't have spare time.

"Spare time" is what free software depends on, which is both its strength as
well as its weakness. To benefit from free software, users should be
willing to spend enough effort of their own, to understand how things work,
to read user manuals and to figure out how to get around problems.

There is no such thing as "8.2.0c". The "a" suffix (short for alpha-testing
release) is for asking people "down the hall" to test out new software. The
"b" suffix (short for beta-testing release) is for asking users to try out a
preview of a new release and report problems that should be fixed before the
eventual release. Recently people started doing another stage called
"release candidate" which follows beta releases.

http://en.wikipedia.org/wiki/Software_release

What happened with 8.2.0 line is that I thought I was ready to go to
beta-testing stage which, in retrospect, wasn't right. So, perhaps I should
go back to alpha releases so that we don't have to worry about the release
process.

The problems you reported with the 8.2.0b had easy work-arounds and, so,
weren't critical. They were not reason enough to rush things.

Cheers,
Uday

[Uday Reddy]

John Stoffel writes:

> Sorry, my fault. I was trying to come up with a repeatable test case
> that was easy to show the problem. I *seems* to be because I do:
>
> - vm-visit-imap-folder
> - read a message
> - file the message to a *local* folder.
> - then expunge any changes with vm-expunge-folder followed by
> vm-save-folder. It's the vm-save-folder which seems to try to
> update the imap flags. Exchange fails the update command and
> vm doesn't seem to properly flush all the changes.
>
> But it's not quite clear to me what the root cause is. I'll try to
> get a test case for you, to make it more clear what's going on.

We already know that the "root cause" is that your Exchange server is
violating the IMAP standard. I have quoted the standard to you in the last
exchange. (See below.) You should be complaining to the people that run
your Exchange server, not me. You should also go back and re-read the old
thread, and continue the discussion there, not in this thread.

What I have asked from you is a *precise description* of what goes wrong
with VM as a result of this problem. "doesn't seem to properly flush all
the changes" is too vague to tell me anything.

Cheers,
Uday

--------


The IMAP standard (RFC 3501) says:

59) It is not an error for the client to store a flag that is not in
the PERMANENTFLAGS list; however, the server will either ignore the
change or make the change in the session only.

So, unfortunately, your Exchange server is not adhering to the standard.
Please do complain to whoever is running your Exchange server.

All that I can do is to make VM resilient so that it carries on regardless
of the error messages. So, what happens after these error messages arise?
Does VM break in some way?

Cheers,
Uday

[Rene]

John Stoffel <john@...> writes:

> Personally, I've been thinking that it's time to finally drop VM and
> move to some other mail reader with good IMAP support, probably mutt.
> But the hassle of either A) learning new key bindings or B) changing
> mutt to have all the emacs/vm bindings has been holding me back.

Has anybody considered switching to mu4e?

http://www.djcbsoftware.nl/code/mu/mu4e.html

How soft/hard is the transition from vm?

--
Rene