Developers of Sympa

[David Verdin <address@concealed>]

Hi all,

Back from sick days, sorry for my absence.

Le 30/06/14 06:21, IKEDA Soji a écrit :

On Thu, 26 Jun 2014 09:35:04 +0200
David Verdin <address@concealed> wrote:

While waiting, I wish to go on refactoring about spool.
In sympa-6.2-branch-old, spool classes were implemented based on
both filesystem and RDBMS then latter was withdrawn later.
Thanks to our experience in older branch, works on current
sympa-6.2-branch will not need tracing such roundabout.

Repeatable integration tests make bold refactoring easier without
concerning unstability.  That's why I think I may start this work
right now.

Frankly, I'm not sure. I remember the hard work it represented. Though I 
tested it thoroughly, I know that it was rather unstabilizing. I wished 
we concentrate on getting rid of the bugs and - finally - get the 6.2 
out of the table so we can have time this summer for the true refactoring.

All the work on the spools is actually refactoring work. As it does not 
adds functionnalities or improves the overall software behaviour, I 
prefer to keep it for 7.0.
Additionnally, it is already in this branch, so all we have to do is 
check for its completeness.

<<Here was my complaint with over ten lines which was removed before posting>>

The fact that refactoring increases unstabiliy is curious, isn't
it?

I think it is partly due to poor testing environment (it does not
always mean hardware).  If the test suit with sufficient coverage
was prepared, and it was able to run frequently, sign of unstability
might be found relatively earlier.

Sure, but if we want the 6.2 to get off the table, we should focus on releasing it.
And make the real unit tests and refactorings on the 7.0.

We discussed with Marc and he proposed that we have a development server 
avaialble for all developpers. This way, we could set up a fake PKI and 
DNS records related to DKIM.
Everyone would have his private instance on this server, so you could 
test anything you like.

Would you like to have such a server?

If there was such host, I probably use it (sorry, I write this not
thinking of who maintains it).

However, fake CA and key pairs may be prepared (and distributed)
easily.  Fake DNS is a bit difficult to prepare, but if stub (mock)
module of Net::DNS is provided, it may not be neccessary to run
actual name service.

The DNS does not have to be fake. There will be real DNS entries published in the RENATER domain.

So no need to handle this.

Cheers,

David

# As I wrote, why I requested them is according to view of
# interoperability.

BTW I would like to make another proposal: How about prepare any
space for integration test in the repostory?

Regards,

--- Soji

Cheers,

David

Regards,

--- Soji

   * 7.0 version expected in beta by the end of May, stable in July
   * 7.1 version expected in beta by the end of December.

7.1 should contain SaaS functionnalities (config edition through the
web, refactored configuration management, eased virtual host deplyment).
You can add whatever you like in this version as soon as it fits within
the end-of-the-year deadline.
That gives us indicators for our work on Sympa.
A word about this schedule?

Best regards,

David

-- 
A bug in Sympa? Quick! To the bug tracker!
<https://sourcesup.renater.fr/tracker/?group_id=23>
RENATER logo
*David Verdin*
Études et projets applicatifs

Tél : +33 2 23 23 69 71
Fax : +33 2 23 23 71 21

www.renater.fr <http;//www.renater.fr>
	RENATER
263 Avenue du Gal Leclerc
35042 Rennes Cedex

-- 
A bug in Sympa? Quick! To the bug tracker! 
<https://sourcesup.renater.fr/tracker/?group_id=23>
RENATER logo
*David Verdin*
Études et projets applicatifs

Tél : +33 2 23 23 69 71
Fax : +33 2 23 23 71 21

www.renater.fr <http;//www.renater.fr>
	RENATER
263 Avenue du Gal Leclerc
35042 Rennes Cedex

--
A bug in Sympa? Quick! To the bug tracker!

David Verdin Études et projets applicatifs
Tél : +33 2 23 23 69 71 Fax : +33 2 23 23 71 21   www.renater.fr	RENATER 263 Avenue du Gal Leclerc 35042 Rennes Cedex

Attachment: smime.p7s
Description: Signature cryptographique S/MIME