Jeff King <peff@xxxxxxxx> writes: > On Thu, Apr 26, 2012 at 08:18:58AM -0700, Junio C Hamano wrote: > >> Jeff King <peff@xxxxxxxx> writes: >> >> > Don't we explicitly not want to do this when the protocol is http? Curl >> > doesn't respect HTTP_PROXY. >> >> Yes. Here is what I'll queue. > > I had raised several other points in my (admittedly belated) review[1]. > So I was kind of expecting a re-roll rather than a patch on top. It was > mostly readability issues, but there is also a potential segfault when > the url has no protocol. I'm not sure it can happen in normal git use, > but: > > git remote-https url-without-protocol > > may dereference null (it's via snprintf, so on my system glibc magically > rewrites this as "(null)_proxy", but other systems will segfault). > > -Peff > > [1] http://article.gmane.org/gmane.comp.version-control.git/195443 > http://article.gmane.org/gmane.comp.version-control.git/195445 Indeed. Thanks for reminding me not to advance this beyond 'pu'. -- To unsubscribe from this list: send the line "unsubscribe git" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html