# [PATCH 0/1] completion: send-email: don't complete revs when --no-format-patch 11 messages from 2024-01-08 to 2024-04-28. Participants: Britton Leo Kerin, Dragan Simic, Britton Kerin, Junio C Hamano, Patrick Steinhardt. Thread: https://gitlist.dev/t/60699 ## Britton Leo Kerin, 2024-01-08 09:36 Subject: [PATCH 0/1] completion: send-email: don't complete revs when --no-format-patch Message-ID: <9627364b-c0c9-4b85-a81a-ba1ef0735c9a@smtp-relay.sendinblue.com> URL: https://gitlist.dev/e/9627364b-c0c9-4b85-a81a-ba1ef0735c9a%40smtp-relay.sendinblue.com ``` Along the way I taught __git_find_last_on_cmdline to understand '--', which isn't stricly necessary but I think reads more clearly at the call sites. __git_find_on_cmdline could be changed to work the same, or this part dropped if people don't like it. Britton Leo Kerin (1): completion: don't comp revs when --no-format-patch contrib/completion/git-completion.bash | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-) base-commit: e79552d19784ee7f4bbce278fe25f93fbda196fa -- 2.43.0 ``` ## Dragan Simic, 2024-01-08 09:40 Subject: Re: [PATCH 0/1] completion: send-email: don't complete revs when --no-format-patch Message-ID: <0e8a1572100faae72db54becefe19f6b@manjaro.org> URL: https://gitlist.dev/e/0e8a1572100faae72db54becefe19f6b%40manjaro.org In-Reply-To: <9627364b-c0c9-4b85-a81a-ba1ef0735c9a@smtp-relay.sendinblue.com> ``` On 2024-01-08 10:36, Britton Leo Kerin wrote: > Along the way I taught __git_find_last_on_cmdline to understand '--', > which > isn't stricly necessary but I think reads more clearly at the call > sites. > __git_find_on_cmdline could be changed to work the same, or this part > dropped > if people don't like it. If I may suggest, there's no need for a cover letter for a single patch. If you want to include some notes in the patch submission, which aren't supposed to be part of the commit summary, you can do that in the patch itself. > Britton Leo Kerin (1): > completion: don't comp revs when --no-format-patch > > contrib/completion/git-completion.bash | 6 +++++- > 1 file changed, 5 insertions(+), 1 deletion(-) > > > base-commit: e79552d19784ee7f4bbce278fe25f93fbda196fa > -- > 2.43.0 ``` ## Britton Kerin, 2024-01-08 19:34 Subject: Re: [PATCH 0/1] completion: send-email: don't complete revs when --no-format-patch Message-ID: URL: https://gitlist.dev/e/CAC4O8c8Evt1XEggVWoVCFCQ-mUkS-%3DdBYEKM8uuq9WEi-sMdwg%40mail.gmail.com In-Reply-To: <0e8a1572100faae72db54becefe19f6b@manjaro.org> ``` On Mon, Jan 8, 2024 at 12:40 AM Dragan Simic wrote: > > On 2024-01-08 10:36, Britton Leo Kerin wrote: > > Along the way I taught __git_find_last_on_cmdline to understand '--', > > which > > isn't stricly necessary but I think reads more clearly at the call > > sites. > > __git_find_on_cmdline could be changed to work the same, or this part > > dropped > > if people don't like it. > > If I may suggest, there's no need for a cover letter for a single patch. > If you want to include some notes in the patch submission, which aren't > supposed to be part of the commit summary, you can do that in the patch > itself. Ok thanks, I'll do it that way in future. Britton ``` ## Britton Leo Kerin, 2024-01-09 01:08 Subject: [PATCH v2 0/1] completion: don't complete revs when --no-format-patch Message-ID: URL: https://gitlist.dev/e/c9656f53-daa1-4e4e-852c-31997cef6ba5%40smtp-relay.sendinblue.com In-Reply-To: <9627364b-c0c9-4b85-a81a-ba1ef0735c9a@smtp-relay.sendinblue.com> ``` Improve commit message Britton Leo Kerin (1): completion: don't complete revs when --no-format-patch contrib/completion/git-completion.bash | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-) Range-diff against v1: 1: ff4d2e55e3 ! 1: e56dbbacd9 completion: don't comp revs when --no-format-patch @@ Metadata Author: Britton Leo Kerin ## Commit message ## - completion: don't comp revs when --no-format-patch + completion: don't complete revs when --no-format-patch In this case the user has specifically said they don't want send-email to run format-patch so revs aren't valid argument completions (and it's -- 2.43.0 ``` ## Britton Leo Kerin, 2024-01-09 01:08 Subject: [PATCH v2 1/1] completion: don't complete revs when --no-format-patch Message-ID: URL: https://gitlist.dev/e/a718b5ee-afb0-44bd-a299-3208fac43506%40smtp-relay.sendinblue.com In-Reply-To: <20240109010830.458775-1-britton.kerin@gmail.com> ``` In this case the user has specifically said they don't want send-email to run format-patch so revs aren't valid argument completions (and it's likely revs and dirs do have some same names or prefixes as in Documentation/MyFirstContribution.txt 'psuh'). Signed-off-by: Britton Leo Kerin --- contrib/completion/git-completion.bash | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-) diff --git a/contrib/completion/git-completion.bash b/contrib/completion/git-completion.bash index 185b47d802..c983f3b2ab 100644 --- a/contrib/completion/git-completion.bash +++ b/contrib/completion/git-completion.bash @@ -1242,10 +1242,12 @@ __git_find_last_on_cmdline () while test $# -gt 1; do case "$1" in --show-idx) show_idx=y ;; + --) shift && break ;; *) return 1 ;; esac shift done + [ $# -eq 1 ] || return 1 # return 1 if we got wrong # of non-opts local wordlist="$1" while [ $c -gt "$__git_cmd_idx" ]; do @@ -2429,7 +2431,9 @@ _git_send_email () return ;; esac - __git_complete_revlist + if [ "$(__git_find_last_on_cmdline -- "--format-patch --no-format-patch")" != "--no-format-patch" ]; then + __git_complete_revlist + fi } _git_stage () -- 2.43.0 ``` ## Junio C Hamano, 2024-02-07 22:12 Subject: Re: [PATCH v2 1/1] completion: don't complete revs when --no-format-patch Message-ID: URL: https://gitlist.dev/e/xmqq4jej6i1b.fsf%40gitster.g In-Reply-To: ``` "Britton Leo Kerin" writes: > In this case the user has specifically said they don't want send-email > to run format-patch so revs aren't valid argument completions (and it's > likely revs and dirs do have some same names or prefixes as in > Documentation/MyFirstContribution.txt 'psuh'). > > Signed-off-by: Britton Leo Kerin > --- > contrib/completion/git-completion.bash | 6 +++++- > 1 file changed, 5 insertions(+), 1 deletion(-) Any reviews people want to offer to this one? Thanks. > diff --git a/contrib/completion/git-completion.bash b/contrib/completion/git-completion.bash > index 185b47d802..c983f3b2ab 100644 > --- a/contrib/completion/git-completion.bash > +++ b/contrib/completion/git-completion.bash > @@ -1242,10 +1242,12 @@ __git_find_last_on_cmdline () > while test $# -gt 1; do > case "$1" in > --show-idx) show_idx=y ;; > + --) shift && break ;; > *) return 1 ;; > esac > shift > done > + [ $# -eq 1 ] || return 1 # return 1 if we got wrong # of non-opts > local wordlist="$1" > > while [ $c -gt "$__git_cmd_idx" ]; do > @@ -2429,7 +2431,9 @@ _git_send_email () > return > ;; > esac > - __git_complete_revlist > + if [ "$(__git_find_last_on_cmdline -- "--format-patch --no-format-patch")" != "--no-format-patch" ]; then > + __git_complete_revlist > + fi > } > > _git_stage () ``` ## Patrick Steinhardt, 2024-02-08 07:57 Subject: Re: [PATCH v2 1/1] completion: don't complete revs when --no-format-patch Message-ID: URL: https://gitlist.dev/e/ZcSJaRczdHApmnVi%40tanuki In-Reply-To: ``` On Mon, Jan 08, 2024 at 04:08:30PM -0900, Britton Leo Kerin wrote: > In this case the user has specifically said they don't want send-email > to run format-patch so revs aren't valid argument completions (and it's > likely revs and dirs do have some same names or prefixes as in > Documentation/MyFirstContribution.txt 'psuh'). > > Signed-off-by: Britton Leo Kerin > --- > contrib/completion/git-completion.bash | 6 +++++- > 1 file changed, 5 insertions(+), 1 deletion(-) > > diff --git a/contrib/completion/git-completion.bash b/contrib/completion/git-completion.bash > index 185b47d802..c983f3b2ab 100644 > --- a/contrib/completion/git-completion.bash > +++ b/contrib/completion/git-completion.bash > @@ -1242,10 +1242,12 @@ __git_find_last_on_cmdline () > while test $# -gt 1; do > case "$1" in > --show-idx) show_idx=y ;; > + --) shift && break ;; > *) return 1 ;; > esac > shift > done > + [ $# -eq 1 ] || return 1 # return 1 if we got wrong # of non-opts > local wordlist="$1" > > while [ $c -gt "$__git_cmd_idx" ]; do > @@ -2429,7 +2431,9 @@ _git_send_email () > return > ;; > esac > - __git_complete_revlist > + if [ "$(__git_find_last_on_cmdline -- "--format-patch --no-format-patch")" != "--no-format-patch" ]; then > + __git_complete_revlist > + fi > } While this second hunk here makes perfect sense to me, there is no explanation why we need to change `__git_find_last_on_cmdline ()`. It's already used with "--guess --no-guess" in another place, so I would think that it ought to work alright for this usecase, too. Or is it that the existing callsite of this function is buggy, too? If so, we should likely fix that in a separate patch together with a test. Also, adding a test for git-send-email that exercises this new behaviour would be very much welcome, too. Patrick ``` ## Britton Kerin, 2024-02-14 20:49 Subject: Re: [PATCH v2 1/1] completion: don't complete revs when --no-format-patch Message-ID: URL: https://gitlist.dev/e/CAC4O8c88Z3ZqxH2VVaNPpEGB3moL5dJcg3cOWuLWwQ_hLrJMtA%40mail.gmail.com In-Reply-To: ``` On Wed, Feb 7, 2024 at 10:57 PM Patrick Steinhardt wrote: > > On Mon, Jan 08, 2024 at 04:08:30PM -0900, Britton Leo Kerin wrote: > > In this case the user has specifically said they don't want send-email > > to run format-patch so revs aren't valid argument completions (and it's > > likely revs and dirs do have some same names or prefixes as in > > Documentation/MyFirstContribution.txt 'psuh'). > > > > Signed-off-by: Britton Leo Kerin > > --- > > contrib/completion/git-completion.bash | 6 +++++- > > 1 file changed, 5 insertions(+), 1 deletion(-) > > > > diff --git a/contrib/completion/git-completion.bash b/contrib/completion/git-completion.bash > > index 185b47d802..c983f3b2ab 100644 > > --- a/contrib/completion/git-completion.bash > > +++ b/contrib/completion/git-completion.bash > > @@ -1242,10 +1242,12 @@ __git_find_last_on_cmdline () > > while test $# -gt 1; do > > case "$1" in > > --show-idx) show_idx=y ;; > > + --) shift && break ;; > > *) return 1 ;; > > esac > > shift > > done > > + [ $# -eq 1 ] || return 1 # return 1 if we got wrong # of non-opts > > local wordlist="$1" > > > > while [ $c -gt "$__git_cmd_idx" ]; do > > @@ -2429,7 +2431,9 @@ _git_send_email () > > return > > ;; > > esac > > - __git_complete_revlist > > + if [ "$(__git_find_last_on_cmdline -- "--format-patch --no-format-patch")" != "--no-format-patch" ]; then > > + __git_complete_revlist > > + fi > > } > > While this second hunk here makes perfect sense to me, there is no > explanation why we need to change `__git_find_last_on_cmdline ()`. It's > already used with "--guess --no-guess" in another place, so I would > think that it ought to work alright for this usecase, too. Or is it that > the existing callsite of this function is buggy, too? If so, we should > likely fix that in a separate patch together with a test. > > Also, adding a test for git-send-email that exercises this new behaviour > would be very much welcome, too. I'll look this one over again and add some tests eventually. Britton ``` ## Junio C Hamano, 2024-02-21 01:34 Subject: Re: [PATCH v2 1/1] completion: don't complete revs when --no-format-patch Message-ID: URL: https://gitlist.dev/e/xmqqa5nusiqv.fsf%40gitster.g In-Reply-To: ``` Britton Kerin writes: >> While this second hunk here makes perfect sense to me, there is no >> explanation why we need to change `__git_find_last_on_cmdline ()`. It's >> already used with "--guess --no-guess" in another place, so I would >> think that it ought to work alright for this usecase, too. Or is it that >> the existing callsite of this function is buggy, too? If so, we should >> likely fix that in a separate patch together with a test. >> >> Also, adding a test for git-send-email that exercises this new behaviour >> would be very much welcome, too. > > I'll look this one over again and add some tests eventually. Thank you, both. ``` ## Junio C Hamano, 2024-03-05 21:58 Subject: Re: [PATCH v2 1/1] completion: don't complete revs when --no-format-patch Message-ID: URL: https://gitlist.dev/e/xmqqfrx4z6gg.fsf%40gitster.g In-Reply-To: ``` Britton Kerin writes: > On Wed, Feb 7, 2024 at 10:57 PM Patrick Steinhardt wrote: >> ... >> Also, adding a test for git-send-email that exercises this new behaviour >> would be very much welcome, too. > > I'll look this one over again and add some tests eventually. Just pinging the thread to keep it visible in the list of recent topics I need to keep an eye on. No rush. Thanks. ``` ## Junio C Hamano, 2024-04-28 22:55 Subject: Re: [PATCH v2 1/1] completion: don't complete revs when --no-format-patch Message-ID: URL: https://gitlist.dev/e/xmqqplu92j3e.fsf%40gitster.g In-Reply-To: ``` Britton Kerin writes: > On Wed, Feb 7, 2024 at 10:57 PM Patrick Steinhardt wrote: >> >> On Mon, Jan 08, 2024 at 04:08:30PM -0900, Britton Leo Kerin wrote: >> ... >> While this second hunk here makes perfect sense to me, there is no >> explanation why we need to change `__git_find_last_on_cmdline ()`. It's >> already used with "--guess --no-guess" in another place, so I would >> think that it ought to work alright for this usecase, too. Or is it that >> the existing callsite of this function is buggy, too? If so, we should >> likely fix that in a separate patch together with a test. >> >> Also, adding a test for git-send-email that exercises this new behaviour >> would be very much welcome, too. > > I'll look this one over again and add some tests eventually. It has been a while, but is there any progress on this topic? Thanks. ```