# [PATCH] gitremote-helpers(1): clarify refspec behaviour 3 messages from 2013-04-06 to 2013-04-17. Participants: John Keeping, Sverre Rabbelier, Felipe Contreras. Thread: https://gitlist.dev/t/33407 ## John Keeping, 2013-04-06 18:13 Subject: [PATCH] gitremote-helpers(1): clarify refspec behaviour Message-ID: <505914600ac7f4afc26f8b29fbece26156c87519.1365272021.git.john@keeping.me.uk> URL: https://gitlist.dev/e/505914600ac7f4afc26f8b29fbece26156c87519.1365272021.git.john%40keeping.me.uk ``` The documentation says that "If no 'refspec' capability is advertised, there is an implied `refspec *:*`" but this is only the case for the "import" command. Since there is a comment in transport-helper.c indicating that this default is for historical reasons, change the documentation to clarify that a refspec should always be specified. Signed-off-by: John Keeping --- Documentation/gitremote-helpers.txt | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/Documentation/gitremote-helpers.txt b/Documentation/gitremote-helpers.txt index 0c91aba..f506031 100644 --- a/Documentation/gitremote-helpers.txt +++ b/Documentation/gitremote-helpers.txt @@ -174,8 +174,8 @@ ref. This capability can be advertised multiple times. The first applicable refspec takes precedence. The left-hand of refspecs advertised with this capability must cover all refs reported by -the list command. If no 'refspec' capability is advertised, -there is an implied `refspec *:*`. +the list command. If a helper does not need a specific 'refspec' +capability then it should advertise `refspec *:*`. 'bidi-import':: This modifies the 'import' capability. -- 1.8.2.692.g17a9715 ``` ## Sverre Rabbelier, 2013-04-06 18:15 Subject: Re: [PATCH] gitremote-helpers(1): clarify refspec behaviour Message-ID: URL: https://gitlist.dev/e/CAGdFq_jrYu23yyQyDEDD%3D72f7UrhWBL0bnmZ7XLtBM%3DF4Hwzug%40mail.gmail.com In-Reply-To: <505914600ac7f4afc26f8b29fbece26156c87519.1365272021.git.john@keeping.me.uk> ``` In Sat, Apr 6, 2013 at 11:13 AM, John Keeping wrote: > The documentation says that "If no 'refspec' capability is advertised, > there is an implied `refspec *:*`" but this is only the case for the > "import" command. > > Since there is a comment in transport-helper.c indicating that this > default is for historical reasons, change the documentation to clarify > that a refspec should always be specified. > > Signed-off-by: John Keeping Acked-by: Sverre Rabbelier -- Cheers, Sverre Rabbelier ``` ## Felipe Contreras, 2013-04-17 22:30 Subject: Re: [PATCH] gitremote-helpers(1): clarify refspec behaviour Message-ID: URL: https://gitlist.dev/e/CAMP44s38Cqcr4mwDBexW4tHzv3BKffCYEn5oN0U6Wji10nmJMw%40mail.gmail.com In-Reply-To: <505914600ac7f4afc26f8b29fbece26156c87519.1365272021.git.john@keeping.me.uk> ``` On Sat, Apr 6, 2013 at 12:13 PM, John Keeping wrote: > --- a/Documentation/gitremote-helpers.txt > +++ b/Documentation/gitremote-helpers.txt > @@ -174,8 +174,8 @@ ref. > This capability can be advertised multiple times. The first > applicable refspec takes precedence. The left-hand of refspecs > advertised with this capability must cover all refs reported by > -the list command. If no 'refspec' capability is advertised, > -there is an implied `refspec *:*`. > +the list command. If a helper does not need a specific 'refspec' > +capability then it should advertise `refspec *:*`. But if it advertises a straight 'refspec *:*', nothing would work. If anything, it should be 'refs/heads/*:refs/remotes/$alias/*', but then tags would fail. Why not just tell the remote helpers to do the right thing and avoid this comment altogether? -- Felipe Contreras ```