amavisd-new regression suite

[C.J. Collier]

Hello folks,

I'm doing a quick review of the amavisd source here:

http://anonscm.debian.org/cgit/amavisd-new/pkg-amavisd-new.git/tree/amavisd

First of all, is there an online repository upstream of this one that I
should be using instead? The only thing I could find was the static
release .tar.xz files referenced here:

http://anonscm.debian.org/cgit/amavisd-new/pkg-amavisd-new.git/tree/debian/watch

I would like to make some changes to address some or all of the issues
raised by perlcritic:

***@hype0:/usr/src/git/alioth/pkg-amavisd-new$ perlcritic amavisd | wc -l
288


But I'm afraid that I will break things if I just go hacking around in
there. I would expect there to be a rather extensive set of tests in the
t/ directory for a project of this size. I'm happy to contribute some if
they don't exist. I just want to have a bit of a safety net to fall back
on.

Cheers,

C.J.

[Benning, Markus]

Post by C.J. Collier
http://anonscm.debian.org/cgit/amavisd-new/pkg-amavisd-new.git/tree/amavisd
First of all, is there an online repository upstream of this one that
I should be using instead? The only thing I could find was the static
http://anonscm.debian.org/cgit/amavisd-new/pkg-amavisd-new.git/tree/debian/watch
I would like to make some changes to address some or all of the issues

wc -l
288

But I'm afraid that I will break things if I just go hacking around in
there. I would expect there to be a rather extensive set of tests in
the t/ directory for a project of this size. I'm happy to contribute
some if they don't exist. I just want to have a bit of a safety net
to fall back on.

Unfortunately there is no public source code repository which could be
used
to directly contribute to amavisd-new.

I recently created a repository from the tar balls at:

https://github.com/benningm/amavisd-new

It could be used to collect patches from this mailling list an others.



Markus

[C.J. Collier]

Thank you Markus,

I was just minutes ago looking on the CPAN for amavis-related modules and
came across your name. I'd never seen a collectd plugin before :-)

I saw your unofficial repository on github as well when I was looking for
online repos but moved on when I saw that it is unofficial. I will plan to
submit pull requests against it once I get to the point of submitting
patches.

Cheers,

C.J.

Post by C.J. Collier
http://anonscm.debian.org/cgit/amavisd-new/pkg-amavisd-new.git/tree/amavisd
First of all, is there an online repository upstream of this one that
I should be using instead? The only thing I could find was the static
http://anonscm.debian.org/cgit/amavisd-new/pkg-amavisd-new.git/tree/debian/watch
I would like to make some changes to address some or all of the issues

wc -l
288

But I'm afraid that I will break things if I just go hacking around in
there. I would expect there to be a rather extensive set of tests in
the t/ directory for a project of this size. I'm happy to contribute
some if they don't exist. I just want to have a bit of a safety net
to fall back on.

Unfortunately there is no public source code repository which could be used
to directly contribute to amavisd-new.
https://github.com/benningm/amavisd-new
It could be used to collect patches from this mailling list an others.
Markus
--
https://markusbenning.de/

[Quanah Gibson-Mount]

--On Monday, March 14, 2016 9:32 AM +0100 "Benning, Markus"

Post by Benning, Markus
Unfortunately there is no public source code repository which could be
used
to directly contribute to amavisd-new.

It bothers me significantly that Mark refuses to address this in any way
what so ever. Direct questions about it are utterly ignored (zero
response).

--Quanah

--

Quanah Gibson-Mount
Platform Architect
Zimbra, Inc.
--------------------
Zimbra :: the leader in open source messaging and collaboration
A division of Synacor, Inc

[C.J. Collier]

Sounds like time to fork.

I call a vote.

Aye.

Sent from my PDP-11

--On Monday, March 14, 2016 9:32 AM +0100 "Benning, Markus" <
Unfortunately there is no public source code repository which could be

Post by Benning, Markus
used
to directly contribute to amavisd-new.

It bothers me significantly that Mark refuses to address this in any way
what so ever. Direct questions about it are utterly ignored (zero
response).
--Quanah
--
Quanah Gibson-Mount
Platform Architect
Zimbra, Inc.
--------------------
Zimbra :: the leader in open source messaging and collaboration
A division of Synacor, Inc

[Phil Regnauld]

Post by C.J. Collier
Sounds like time to fork.
I call a vote.

Hmm - seems that Mark has been rather reactive in integrating
suggestions and patches over the years.

It's legitimate to ask what the plans are for the future of amavis -
but maybe a bit premature to suggest forking ?

What would be the benefits ? Who would maintain it ?

If the only issue is that of versioning, it's easy to maintain vendor
imports in a repo, and have regression testing on the side.

Cheers,
P.

[Benning, Markus]

Post by C.J. Collier
Sounds like time to fork.
I call a vote.
Aye.

I think the best way would be to re-setup/open the current project
together with Mark to make it open for others to directly contribute
and integrate changes.

It would be good to know Mark plans for amavis.

Markus

[Quanah Gibson-Mount]

--On Wednesday, March 16, 2016 11:36 AM +0100 "Benning, Markus"

Post by Benning, Markus

Post by C.J. Collier
Sounds like time to fork.
I call a vote.
Aye.

I think the best way would be to re-setup/open the current project
together with Mark to make it open for others to directly contribute
and integrate changes.
It would be good to know Mark plans for amavis.

Yes, I'm definitely not calling for a fork. But I am frustrated by the
lack of visibility into the code or the ability to easily contribute back
fixes, etc. In particular being able to get a log of changes and the
related diffs would be exceptionally helpful.

--Quanah


--

Quanah Gibson-Mount
Platform Architect
Zimbra, Inc.
--------------------
Zimbra :: the leader in open source messaging and collaboration
A division of Synacor, Inc

[C.J. Collier]

On Tue, Mar 15, 2016 at 10:38 PM, C.J. Collier <

Post by C.J. Collier
Sounds like time to fork.

After discussion with management, I need to make it exceedingly clear that
this is my personal opinion and that my employer in no way condones or
approves of this message. I am very sorry if it sounded as my personal
feelings on the matter were influenced in any way by my employer. I was
following a trend of foolish mistakes such as supporting the re-naming of
Zebra to SysCOW:

http://blog.gmane.org/gmane.network.zebra/month=20030801/page=5

I'm still interested in seeing development happen in a more collaborative,
open & transparent, iterative way.

And I retract my vote of Aye.

[Quanah Gibson-Mount]

--On Friday, March 18, 2016 12:31 PM -0700 "C.J. Collier"

Post by C.J. Collier

Post by C.J. Collier
On Tue, Mar 15, 2016 at 10:38 PM, C.J. Collier
Sounds like time to fork.

After discussion with management, I need to make it exceedingly clear
that this is my personal opinion and that my employer in no way condones
or approves of this message.  I am very sorry if it sounded as my
personal feelings on the matter were influenced in any way by my
employer.  I was following a trend of foolish mistakes such as

Thanks for the clarification. ;)

Post by C.J. Collier
I'm still interested in seeing development happen in a more
collaborative, open & transparent, iterative way.

Yes, I think this is what needs to happen.

Post by C.J. Collier
And I retract my vote of Aye.

Thanks.

What bothers me the most atm is the completely and utter silence on the
matter from Mark. Is there an issue with moving to a repository that's
accessible to everyone? I.e., does the Institut "Jožef Stefan" have a
policy against being transparent? etc.

--Quanah

--

Quanah Gibson-Mount
Platform Architect
Zimbra, Inc.
--------------------
Zimbra :: the leader in open source messaging and collaboration
A division of Synacor, Inc