From: Junio C Hamano Date: Fri, 14 Aug 2015 23:14:30 GMT Subject: Re: [PATCH 5/7] transport: Remove git_transport_options.push_cert Message-ID: In-Reply-To: <1439492451-11233-6-git-send-email-dborowitz@google.com> Dave Borowitz writes: > This field was set in transport_set_option, but never read in the push > code. The push code basically ignores the smart_options field > entirely, and derives its options from the flags arguments to the > push* callbacks. Note that in git_transport_push there are already > several args set from flags that have no corresponding field in > git_transport_options; after this change, push_cert is just like > those. > > Signed-off-by: Dave Borowitz > --- Thanks for cleaning up my mess. Honestly, to me, the smart transport is always a second-class citizen (and http walkers are not even citizens ;-)) and any support of new feature is added as an after-thought once the feature starts working with the native transport, and that development pattern clearly shows in a place like this. > transport.c | 3 --- > transport.h | 1 - > 2 files changed, 4 deletions(-) > > diff --git a/transport.c b/transport.c > index 40692f8..3dd6e30 100644 > --- a/transport.c > +++ b/transport.c > @@ -476,9 +476,6 @@ static int set_git_option(struct git_transport_options *opts, > die("transport: invalid depth option '%s'", value); > } > return 0; > - } else if (!strcmp(name, TRANS_OPT_PUSH_CERT)) { > - opts->push_cert = !!value; > - return 0; > } > return 1; > } > diff --git a/transport.h b/transport.h > index 18d2cf8..79190df 100644 > --- a/transport.h > +++ b/transport.h > @@ -12,7 +12,6 @@ struct git_transport_options { > unsigned check_self_contained_and_connected : 1; > unsigned self_contained_and_connected : 1; > unsigned update_shallow : 1; > - unsigned push_cert : 1; > int depth; > const char *uploadpack; > const char *receivepack;