On Tue, Nov 11, 2008 at 08:22:38PM -0800, Junio C Hamano <gitster@xxxxxxxxx> wrote: > The function migrate_file() introduced by [2/4] is called for any > remote definition that did not come from config (by definition, it > either came from remotes/foo or branches/foo). The function adds the > entries for the given remote definition to the config file, and then > removes remotes/foo file if the remote definition came from it. So it > is a logically consistent change if you only called this function only > for remote definitions that came from remotes/foo. > > But the function is called for a remote definition that originally > came from branches/foo as well. It happily adds the definition to the > config, even though it *fails to remove* branches/foo file. > > Do you still think 2/4 is a logically contained good change? OK, here is an updated series, which is supposed to fix this issue. 1/4 and 4/4 is unchanged. Miklos Vajna (4): remote: add a new 'origin' variable to the struct git-remote rename: support remotes->config migration git-remote rename: support branches->config migration git-remote: document the migration feature of the rename subcommand Documentation/git-remote.txt | 4 ++++ builtin-remote.c | 37 +++++++++++++++++++++++++++++++++++++ remote.c | 3 +++ remote.h | 7 +++++++ t/t5505-remote.sh | 33 +++++++++++++++++++++++++++++++++ 5 files changed, 84 insertions(+), 0 deletions(-) -- 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
