I'd like to separate the grouping issue from the execution issue. Two different requirements - albeit connected.

I'd also resist making any commitment to a specific execution paradigm.

Let's start from the grouping and then move on to the execution? Agreed?

On 16/07/2013 09:11, Anthony Swan wrote:
Yep no worries. Are you generally happy with the approach we've taken or did you have some other architecture in mind?


Tony


On Tue, Jul 16, 2013 at 6:04 PM, Mauro Talevi <mauro.tal...@aquilonia.org <mailto:mauro.tal...@aquilonia.org>> wrote:

    No, work has not started yet.

    Contributions always welcome.
    http://jbehave.org/reference/stable/how-to-contribute.html

    Be sure to start from the 4.x branch.

    Cheers

    On 16/07/2013 08:56, Anthony Swan wrote:
    Funny I went through jira over and over and over but didn't see
that. I see it's still unassigned. Is the code already done? If not, we're happy to contribute and keep it going.


    Tony


    On Tue, Jul 16, 2013 at 5:49 PM, Mauro Talevi
    <mauro.tal...@aquilonia.org <mailto:mauro.tal...@aquilonia.org>>
    wrote:

        Hi Tony,

        it's already on the cards for 4.0:
        http://jira.codehaus.org/browse/JBEHAVE-873

        Cheers

        On 16/07/2013 08:37, Anthony Swan wrote:

            Hi all,

            We have an application that has lots of configuration
            toggles that get set in one or more databases or other
            systems.  In order to facilitate running in parallel as
            many tests as possible, we've created a separate
            execution path to change the running of stories into
            running of groups of stories as some stories are mutually
            exclusive of others.   All the stories within a single
            group are run in parallel and each group waits to finish
            before moving to the next.   We played around with
            multiple invocations of Jbehave and running filters but
            that made the report management unwieldy and made it
            harder for the people writing the tests to figure out
            what was going on.

            Is there any interest from the maintainers in bringing
            this into the code base?  If so, I will refactor our code
            to be more generic and do a little refactoring of the
            core code then submit a patch.

            What I've done is extend the Embedder (so that I could
            extend the StoryManager and StoryReporter) and extended
            the StoryReporter.


            If everyone agrees this is good, I will probably change
            it to work with the StoryMapping and get the Mapper to
            output multiple groups which the StoryManager could then run.


            Thoughts?

            Tony



        ---------------------------------------------------------------------
        To unsubscribe from this list, please visit:

        http://xircles.codehaus.org/manage_email






Reply via email to