On Mon, Sep 26, 2011 at 8:04 PM, Benji York <benji.y...@canonical.com> wrote: > On Mon, Sep 26, 2011 at 2:14 PM, Gary Poster <gary.pos...@canonical.com> > wrote: >> I think this constraint comes from httplib2. > > After looking at the lazr.restfulclient and launchpadlib code, I can > confirm this. The problem is that httplib2 doesn't make the effort to > mutate cache files atomically. >
Thanks. >> Maybe httplib2 can be subclassed [...] > > Another workaround would be to provide a unique value for > launchpadlib_dir (perhaps the result of a call to tempfile.mkdtemp) to > each invocation of login_anonymously. > > You will loose some of the benefits of the cache, but if each worker is > relatively long-lived, the overall performance should be reasonable. > The workers are actually quite short-lived, but maybe that will have to do. Thanks again, jml _______________________________________________ Mailing list: https://launchpad.net/~launchpad-dev Post to : launchpad-dev@lists.launchpad.net Unsubscribe : https://launchpad.net/~launchpad-dev More help : https://help.launchpad.net/ListHelp
