Hi

indeed ,
I want to get rid of those hacks
mainly for LastChangeNumber, The others I find less interesting, but
someone added them for a reason (I hope)

but it's a bit breaking,option 4 breaks the layout of the logs :-(
If it was not for that, I just went along


with kind regards
Ruben Willems

On 27 August 2012 10:09, Leszek Ciesielski <[email protected]> wrote:

> I'd vote for 4. Also - weren't there some hacks floating around to get
> them into the build, particularly for LastChangeNumber? Those seem
> quite useful for things like tagging the assembly versions with custom
> information.
>
> On Sun, Aug 26, 2012 at 5:58 AM, Ruben Willems <[email protected]>
> wrote:
> > Hi all
> >
> > I noticed that there is a discrepancy between the integration parameters
> > sent to the build log (xml file)
> > and the ones to the tasks / publishers :-(
> >
> > The build log has the following extra :
> >
> > LastIntegrationStatus
> > LastSuccessfulIntegrationLabel
> > LastChangeNumber
> > LastModificationDate
> > CCNetForceBuildReason
> >
> >
> > This could lead to wrong interpretations :
> > in the build log you see the CCNetForceBuildReason, but you do not have
> it
> > via an environmentvariable nor property nant/msbuild
> >
> >
> > Now the question : how to make these the same?
> >
> > 1) remove them also from the build log?
> > 2) add them to the integration properties with the current names
> > 3) add them to the integration properties with a CCNet prefix like the
> other
> > integration properties?
> >
> >
> http://www.cruisecontrolnet.org/projects/ccnet/wiki/Integration_Properties
> > 4) add them to the integration properties with a CCNet prefix like the
> other
> > integration properties and update the names also in the build log file
> >      so you can see in a build whats available in the tasks to work with
> > 5) only add a few to the integration properties
> > 5) ....
> >
> >
> >
> > For example LastChangeNumber holds the revision number/changeset of
> > sourcecontrol, like the git hash
> >
> >
> > with kind regards
> > Ruben Willems
>

Reply via email to