Maybe 2.5?
Marc, are changes to results.xml finished? This might
be reason enough to skip to 2.5 or 3.0. It might require sqlunit
to update their reporting to remain compatible for instance.
Paul.
Marc Guillemot wrote:
This was the question I wanted to ask (to you included ;-)).
Everybody agrees or nobody cares?
I will start changing target release of all fixed issues since 2.1 to 3.0.
Marc.
dna wrote:
Hi
Isn't there any interest?
On 16 mars 07, at 09:17, Marc Guillemot wrote:
releases are for WebTest a bit artificial as the latest build is
(nearly) always the best "release". Nevertheless many people pay high
attention to official releases and WebTest 2.1 is already 6 months old
meaning that many improvements have occurred since this time.
+1
I think that next release's name should be 3.0 rather than 2.2 due
particularly to WT-232 that bring very significant changes in WebTest
internals.
Shouldn't major version be used for changes that have major
(prositive or negative...) impacts for the end-user?
Such as a change that could break existing script, as the migration
to htmlunit was, or a change that brings new possibilities.
That way, I could as an end-user better decide if/when I want to
upgrade the webtest version.
A change to a minor version could be done pretty much anytime,
whereas refactoring and testing amay have to be planned along with a
change to a major version.
I would see the use of Ant 1.7, java 1.5, xpath 2.0 or successful
XmlHttpRequest as possible reasons for a webtest 3.
What do you think?
For my part I see currently 3 issues that I want to see fixed
before the
next official release:
- Provide a core structure and utility to facilitate WebTest projects
http://webtest-community.canoo.com/jira/browse/WT-287
- Upgrade to htmlunit 1.11
http://webtest-community.canoo.com/jira/browse/WT-288
and fix the remaining unwanted differences in generated report due to:
- Configure steps before execution rather that when added to <webtest>
http://webtest-community.canoo.com/jira/browse/WT-232
What do you think?
Maybe uPgrade all possible dependencies?
A quick look has shown at least log4j and oro could be upgraded.
Best
dna
--
You can sometimes write faster code in C, but
you can always write code faster in Perl.
-- Moral from perlembed
_______________________________________________
WebTest mailing list
[email protected]
http://lists.canoo.com/mailman/listinfo/webtest
_______________________________________________
WebTest mailing list
[email protected]
http://lists.canoo.com/mailman/listinfo/webtest
