[MERGE] Fix bug #229076 by fixing header names before sending the request.

[Vincent Ladeuil]

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

    Andrew> Matt Nordhoff wrote:
    Andrew> [...]
    >> Not that it matters, but...
    >> 
    >> > +        headers = dict((name.title(), val) for name, val in headers.items())
    >> 
    >> Why not iteritems()?

    Andrew> Well, you could just as easily ask “Why iteritems()?” ;)

    Andrew> I highly doubt there will be a measurable performance
    Andrew> benefit to bzr either way,

Sure. 

On the other hand a few lines above (not shown by the patch),
headers is build as a copy of other attributes, so doing yet
another copy with items() doesn't make sense either and is a
waste.

    Andrew> so we may as well use the simpler, shorter method
    Andrew> name.  In fact, I would not be at all surprised to
    Andrew> find that items() is faster than iteritems() for
    Andrew> dicts of that size...

    Andrew> Basically, it doesn't matter either way.

True. English people talk about bike-shedding, French people talk
about sodomizing flies (with an emphasis on the uselessness and
the difficulties involved ;-)

A particular fly specie is called drosophila
(http://en.wikipedia.org/wiki/Drosophila) mostly known in genomic
research and especially small.

Here we are drosodomizing (coined while rebuilding our kitchen
regarding our tendency to perfectionism).

      Vincent