[reviewing MERGE] Extend -Dhpss to emit a count of HPSS calls to stderr

[Vincent Ladeuil]

>>>>> "Andrew" == Andrew Bennetts <andrew at canonical.com> writes:

    Andrew> Vincent Ladeuil wrote:
    Andrew> [...]
    >> 
    >> Right, I realized that during dinner. But if that explained the
    >> huge difference in counters, that doesn't explain all the mediums
    >> with 0 calls. Or is it an artefact of the multiple inheritance ?
    >> But in that case why ?

    Andrew> Yes, it's because RemoteTransport inherits from
    Andrew> SmartClientMedium.  So each RemoteTransport that is
    Andrew> instantiated (as happens a lot, e.g. on every clone()
    Andrew> call) adds an entry to the counts dict.

    Andrew> Specifically, RemoteSSHTransport.__init__ calls
    Andrew> RemoteTransport.__init__ calls
    Andrew> SmartClientMedium.__init__, which adds “self” as an
    Andrew> object to track for the HPSS call reporting.  (In the
    Andrew> case of your patch, via the register_medium() call.)

Thanks.

And confirmed by trying against my python-2.6 branch with the
patch deleting the multiple inheritance: the spurious mediums
have disappeared :)

     Vincent