On Wed, May 25, 2011 at 12:07:24PM +0200, Michael Hanselmann wrote: > Am 25. Mai 2011 11:09 schrieb Iustin Pop <[email protected]>: > > On Mon, May 23, 2011 at 02:50:14PM +0200, Michael Hanselmann wrote: > >> --- a/lib/cli.py > >> +++ b/lib/cli.py > >> + def AddJobId(self, name, status, job_id): > >> + """Adds a job ID to the internal queue. > >> + > >> + """ > >> + self.jobs.append((0, status, job_id, name)) > > > > In what context will this be used? > > For LU-generated jobs. Here's an example from the client code of > “gnt-node migrate”: > > for (status, job_id) in result[constants.JOB_IDS_KEY]: > jex.AddJobId(None, status, job_id) > > > And why are you using duplicate indices, do we simply not care? > > The only place where they're used is for sorting the results. If > “SubmitPending” and “QueueJob” were to be called multiple times, there > can also be duplicate indices (the former uses enumerate() on the > internal queue). Should I change JobExecutor, maybe in a separate > patch, to use a counter for all job additions?
That sounds good, thanks! iustin
