Hi Brett:
Its a moving target.
How can there possibly be any progress when the target keeps changing?
Stop the commits, fix the process by establishing automated functional
tests (unit testing doesn't help much in this situation) w/Selenium, and
then begin again.
Ruth
Brett Palmer wrote:
Ruth,
You make a good point. This has been a topic for a couple of years now. As
OFBiz continues to grow doing even a simple "smoke test" on the build will
be difficult. This is why I think the only scalable solution is to run a
series of automated unit and functional (selenium like) tests on the
application for each daily build.
If this was automated we could send the ofbiz-dev list an email with the the
broken tests and the information (stake trace, etc) about the failure.
There are a few of us working on this but getting valid user tests from the
community would be very helpful.
Brett
On Sun, Dec 6, 2009 at 4:57 PM, Ruth Hoffman <rhoff...@aesolves.com> wrote:
Hi David:
I wasn't going to say anything about this, but my most recent experience
with the nightly trunk download has me fuming again...Just because an
organization is made up of volunteers, doesn't mean there is no need for
rules, respect for others and most importantly leadership.
Here's how I would start to "fix things". I'd say: No more commits until
the community gets the one or more processes in place necessary to
coordinate testing, builds and releases. Anyone who violates the rule,
looses the right to commit. When all the processes are in place and agreed
upon, then you can give all the violators back the right to commit.
You, David, have the power to give developer's commit access to the source
code repository. You, David, can take it away. Or am I wrong about that?
Who, BTW, gave all these people commit access to the source code repository
initially?
Regards,
Ruth
David E Jones wrote:
To be clear, I'd like to hear from other people too. If anyone has any
ideas about how we can get people to do this, by all means let's discuss it!
This need and concern has come up in many places many times over the years
of the project. I have some thoughts on good ways to go about this (like the
UBPL stuff, automated testing which is going on now, etc, etc), but how to
get people to do things, especially in a volunteer organization like this,
is another question...
-David
On Dec 6, 2009, at 12:29 PM, Jacques Le Roux wrote:
Hi David,
I have no answers yet, I must says I have not even thought about it...
For the moment I only propose to concentrate on existing known bugs
repertoried in opened Jira issues.
Thanks
Jacques
() ascii ribbon campaign against HTML e-mail
/\ www.asciiribbon.org
----- Original Message ----- From: "David E Jones" <d...@me.com>
To: <dev@ofbiz.apache.org>
Sent: Sunday, December 06, 2009 6:36 PM
Subject: Re: Bugs and open Jira issues
Jacques,
I appreciate this feeling. In general OFBiz would benefit a lot from
more testing, more definition of what to test against (ie what does a pass
or fail look like, ie what is the design to test against), and in general
more care about respecting others by not breaking things that already exist.
The questions is, how do we get people to do this?
-David
On Dec 5, 2009, at 1:51 PM, Jacques Le Roux wrote:
Hi,
I'd like to express a feeling I have. Actually it's not only my own
feeling but also something some users have expressed recently.
I'm quite happy to see that these last times a lot of effort have been
made in order to fix OFBiz (yes to fix OFBiz!)
It's really great to see new features in OFBiz. But I really wonder if
we should not slow down the pace in integrating new features for a short
period of time and should not make and even greatest effort to have a more
stable OFBiz.
There are 180 bugs opened in Jira. Don't you think it's time for the
community to have a look at them and to fix the most important ones (109 are
considered as at least important) ?
Thanks
Jacques
