Better support for customising context lines in --patch commands
#1915
Conversation
Welcome to GitGitGadgetHi @NinjaInShade, and welcome to GitGitGadget, the GitHub App to send patch series to the Git mailing list from GitHub Pull Requests. Please make sure that either:
You can CC potential reviewers by adding a footer to the PR description with the following syntax: NOTE: DO NOT copy/paste your CC list from a previous GGG PR's description, Also, it is a good idea to review the commit messages one last time, as the Git project expects them in a quite specific form:
It is in general a good idea to await the automated test ("Checks") in this Pull Request before contributing the patches, e.g. to avoid trivial issues such as unportable code. Contributing the patchesBefore you can contribute the patches, your GitHub username needs to be added to the list of permitted users. Any already-permitted user can do that, by adding a comment to your PR of the form Both the person who commented An alternative is the channel Once on the list of permitted usernames, you can contribute the patches to the Git mailing list by adding a PR comment If you want to see what email(s) would be sent for a After you submit, GitGitGadget will respond with another comment that contains the link to the cover letter mail in the Git mailing list archive. Please make sure to monitor the discussion in that thread and to address comments and suggestions (while the comments and suggestions will be mirrored into the PR by GitGitGadget, you will still want to reply via mail). If you do not want to subscribe to the Git mailing list just to be able to respond to a mail, you can download the mbox from the Git mailing list archive (click the curl -g --user "<EMailAddress>:<Password>" \
--url "imaps://imap.gmail.com/INBOX" -T /path/to/raw.txtTo iterate on your change, i.e. send a revised patch or patch series, you will first want to (force-)push to the same branch. You probably also want to modify your Pull Request description (or title). It is a good idea to summarize the revision by adding something like this to the cover letter (read: by editing the first comment on the PR, i.e. the PR description): To send a new iteration, just add another PR comment with the contents: Need help?New contributors who want advice are encouraged to join [email protected], where volunteers who regularly contribute to Git are willing to answer newbie questions, give advice, or otherwise provide mentoring to interested contributors. You must join in order to post or view messages, but anyone can join. You may also be able to find help in real time in the developer IRC channel, |
NinjaInShade
force-pushed
the
interactive-patch-context
branch
2 times, most recently
from
0484407 to
ff64c39
Compare
|
/allow |
|
User NinjaInShade is now allowed to use GitGitGadget. WARNING: NinjaInShade has no public email address set on GitHub; GitGitGadget needs an email address to Cc: you on your contribution, so that you receive any feedback on the Git mailing list. Go to https://github.com/settings/profile to make your preferred email public to let GitGitGadget know which email address to use. |
NinjaInShade
changed the title
--patch commands--patch commands
NinjaInShade
force-pushed
the
interactive-patch-context
branch
2 times, most recently
from
9b90dc6 to
b4b7854
Compare
|
/submit |
|
Submitted as [email protected] To fetch this version into To fetch this version to local tag |
![gitgitgadget[bot]](https://avatars.githubusercontent.com/in/12836?s=60&v=4){kind=small}
|
User |
|
User |
|
Thanks for letting me know, I'll make sure to amend that in v2 :)
…On Mon, 5 May 2025 at 21:33, gitgitgadget[bot] ***@***.***> wrote:
*gitgitgadget[bot]* left a comment (gitgitgadget/git#1915)
<#1915 (comment)>
User Eric Sunshine ***@***.***> has been added to the cc:
list.
—
Reply to this email directly, view it on GitHub
<#1915 (comment)>,
or unsubscribe
<https://github.com/notifications/unsubscribe-auth/APFXCEW367PGDBXHMJDZTHT247DJ7AVCNFSM6AAAAAB4M4W3LOVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDQNJSGI3DIOBUGQ>
.
You are receiving this because you were mentioned.Message ID:
***@***.***>
|
|
User |
|
User |
NinjaInShade
force-pushed
the
interactive-patch-context
branch
2 times, most recently
from
cf726da to
973dfad
Compare
|
/submit |
|
Submitted as [email protected] To fetch this version into To fetch this version to local tag |
| @@ -63,7 +63,7 @@ test_expect_success 'setup (initial)' ' | |||
| ' | |||
There was a problem hiding this comment.
On the Git mailing list, Junio C Hamano wrote (reply to this):
"Leon Michalak via GitGitGadget" <[email protected]> writes:
> From: Leon Michalak <[email protected]>
>
> Refactor to use the modern "test_grep" test utility instead of regular
> "grep" which provides better debug information if tests fail.
>
> This is a prerequisite to the commits that follow which add to both test
> files.
>
> Signed-off-by: Leon Michalak <[email protected]>
> ---
These mostly look sensible, but I would title & phrase the commit
description to 'use "test_grep"', not 'refactor to &'. It's shorter
and more direct ;-)
Thanks.There was a problem hiding this comment.
On the Git mailing list, Leon Michalak wrote (reply to this):
On Mon, 12 May 2025 at 14:42, Junio C Hamano <[email protected]> wrote:
> These mostly look sensible, but I would title & phrase the commit
> description to 'use "test_grep"', not 'refactor to &'. It's shorter
> and more direct ;-)
Thanks - will make sure to change that in v3 :)
| @@ -0,0 +1,10 @@ | |||
| `-U<n>`:: | |||
There was a problem hiding this comment.
On the Git mailing list, Junio C Hamano wrote (reply to this):
"Leon Michalak via GitGitGadget" <[email protected]> writes:
> diff --git a/Documentation/diff-context-options.adoc b/Documentation/diff-context-options.adoc
> new file mode 100644
> index 000000000000..e161260358ff
> --- /dev/null
> +++ b/Documentation/diff-context-options.adoc
> @@ -0,0 +1,10 @@
> +`-U<n>`::
> +`--unified=<n>`::
> + Generate diffs with _<n>_ lines of context. Defaults to `diff.context`
> + or 3 if the config option is unset.
> +
> +`--inter-hunk-context=<n>`::
> + Show the context between diff hunks, up to the specified _<number>_
> + of lines, thereby fusing hunks that are close to each other.
> + Defaults to `diff.interHunkContext` or 0 if the config option
> + is unset.
It might not be trivial to do but I wonder if we cannot do better
than this to share more of the same text across manual pages. These
two being options understood by `git diff`, we certainly have an
existing description for them, no?
Other than that, looking good to me.
Thanks.There was a problem hiding this comment.
On the Git mailing list, Leon Michalak wrote (reply to this):
On Mon, 12 May 2025 at 17:45, Junio C Hamano <[email protected]> wrote:
> It might not be trivial to do but I wonder if we cannot do better
> than this to share more of the same text across manual pages. These
> two being options understood by `git diff`, we certainly have an
> existing description for them, no?
Yes, I did of course notice documentation for `git diff` also has
these; ultimately my justification for not changing that to use this
new .adoc include as well was for a couple reasons:
- these two options are not together in the `git diff` documentation
(not *so* important, and they probably should actually be together?)
- there is an extra if def which adds on "implies --patch" text in the
`git diff` documentation which isn't the behaviour the add-patch
commands are going for, so that makes the intent a little different
here
But would be good to hear if anyone else has any thoughts.
|
This patch series was integrated into seen via git@0945e76. |
|
There was a status update in the "Cooking" section about the branch "git add/etc -p" now honors diff.context configuration variable, and learns to honor -U<n> option. Expecting a reroll. cf. <[email protected]> source: <[email protected]> |
|
This patch series was integrated into seen via git@2ef95cf. |
@NinjaInShade please follow the advice in https://gitgitgadget.github.io/reply-to-this to reply on the Git mailing list, otherwise your reply will not reach the intended recipient. |
Hey! So far I have been replying via gmail and it has made it's on way on to the mailing list (and on this github PR too). Am I missing something? |
|
This patch series was integrated into seen via git@2aced5b. |
@NinjaInShade yes: #1915 (comment) was sent as a reply to this PR (via email, but not to the Git mailing list). |
|
This patch series was integrated into seen via git@392000c. |
|
There was a status update in the "Cooking" section about the branch "git add/etc -p" now honors diff.context configuration variable, and learns to honor -U<n> option. Expecting a reroll. cf. <[email protected]> source: <[email protected]> |
This patch compliments the previous commit, where builtins that use add-patch infrastructure now respect diff.context and diff.interHunkContext file configurations. In particular, this patch helps users who don't want to set persistent context configurations or just want a way to override them on a one-time basis, by allowing the relevant builtins to accept corresponding command line options that override the file configurations. This mimics commands such as diff and log, which allow for both context file configuration and command line overrides. Signed-off-by: Leon Michalak <[email protected]>
NinjaInShade
force-pushed
the
interactive-patch-context
branch
from
2774b93 to
9731e5b
Compare
|
/submit |
Interesting - I have been responding with gmail this whole time and somehow (most) my replies have gotten to the mailing list. I just assumed this was intentional |
|
Submitted as [email protected] To fetch this version into To fetch this version to local tag |
|
On the Git mailing list, Phillip Wood wrote (reply to this): Hi Leon
The range-diff below looks good to me. Thanks for working on this
Phillip
On 29/07/2025 08:01, Leon Michalak via GitGitGadget wrote:
> This series of patches attempt to give --interactive/--patch compatible
> builtins ("add", "commit", "checkout", "reset", "restore" and "stash")
> better support and nicer experience for configuring how many context lines
> are shown in diffs through a variety of ways.
> > Prior to these patches, the user could not choose how many context lines
> they saw in --patch commands (apart from one workaround by using
> GIT_DIFF_OPTS=-u<number> ..., however this isn't a good user experience or a
> persistent solution). Additionally, the behaviour around reading from the
> diff.context and diff.interHunkContext configs was also inconsistent with
> other diff generating commands such as "log -p".
> > The summarised changes below hopefully make this experience better and fix
> some inconsistencies:
> > * diff.context and diff.interHunkContext configs are now respected by
> --patch compatible commands
> * --unified and --inter-hunk-context command line options have been added
> to --patch compatible commands (which take prescendence over file
> configs)
> * "add" and "commit" in --interactive mode now expose a new "context"
> subcommand which configures the amount of context lines you wish to see
> in subsequent diffs generated from other subcommands such as "patch" or
> "diff"
> > The original discussion for this can be read at:
> > * https://lore.kernel.org/git/CAP9jKjGb-Rcr=RLJEzeFdtrekYM+qmHy+1T1fykU3n9cV4GhGw@mail.gmail.com/
> > Changes since v1:
> > * Update commit descriptions
> * Update tests to use the more modern and robust test_grep and test_config
> utils
> * Reword some documentation / user messages
> * Ensure each commit is atomic and builds/passes tests on it's own
> * Make new command line options DRY
> * Add tests for interhunk context interaction
> * Error if context config/command line options are negative
> * Drop previous last commit to do with new subcommand for --interactive
> add/commit. My motivations behind this patch series originally where
> quite simple, just for add-patch commands to respect context configs.
> This subcommand, after the discussion in v1, will require more thought
> and a larger implementation that what I had anticipated. I would prefer
> to leave this for another time as it's the least impactful but the most
> time intensive and complicated idea.
> > Changes since v2:
> > * Update tests to only test single command (following Philip's suggestion)
> * Add negative option checks
> * Minor commit re-wording
> > Changes since v3:
> > * Update commit descriptions
> * Read struct properties directly instead of assigning to variables first
> * Simplify config setting / error checking
> * Remove redundant tests in later commit as they were replaced with better
> test(s)
> * Change tests to use single quotes (this messes with the grep so was
> unable to explicitly test single quotes in the error messages, so decided
> to use regex . instead, which is what some other tests that have this
> problem seem to use as well)
> > Changes since v4:
> > * Add back tests to maintain good coverage and remove redundant tests
> > Leon Michalak (4):
> t: use test_grep in t3701 and t4055
> t: use test_config in t4055
> add-patch: respect diff.context configuration
> add-patch: add diff.context command line overrides
> > Documentation/diff-context-options.adoc | 10 ++
> Documentation/git-add.adoc | 2 +
> Documentation/git-checkout.adoc | 2 +
> Documentation/git-commit.adoc | 2 +
> Documentation/git-reset.adoc | 2 +
> Documentation/git-restore.adoc | 2 +
> Documentation/git-stash.adoc | 2 +
> add-interactive.c | 45 +++++++--
> add-interactive.h | 17 +++-
> add-patch.c | 14 ++-
> builtin/add.c | 21 ++++-
> builtin/checkout.c | 31 +++++-
> builtin/commit.c | 16 +++-
> builtin/reset.c | 17 +++-
> builtin/stash.c | 56 ++++++++---
> commit.h | 3 +-
> parse-options.h | 2 +
> t/t3701-add-interactive.sh | 119 +++++++++++++++++++-----
> t/t4055-diff-context.sh | 42 ++++-----
> t/t9902-completion.sh | 2 +
> 20 files changed, 324 insertions(+), 83 deletions(-)
> create mode 100644 Documentation/diff-context-options.adoc
> > > base-commit: cf6f63ea6bf35173e02e18bdc6a4ba41288acff9
> Published-As: https://github.com/gitgitgadget/git/releases/tag/pr-1915%2FNinjaInShade%2Finteractive-patch-context-v5
> Fetch-It-Via: git fetch https://github.com/gitgitgadget/git pr-1915/NinjaInShade/interactive-patch-context-v5
> Pull-Request: https://github.com/gitgitgadget/git/pull/1915
> > Range-diff vs v4:
> > 1: bbb2bc7082b = 1: bbb2bc7082b t: use test_grep in t3701 and t4055
> 2: feace2d3676 = 2: feace2d3676 t: use test_config in t4055
> 3: 994029d6602 = 3: 994029d6602 add-patch: respect diff.context configuration
> 4: 2774b930406 ! 4: 9731e5b76fb add-patch: add diff.context command line overrides
> @@ parse-options.h: int parse_opt_tracking_mode(const struct option *, const char *
> OPT_SET_INT_F('4', "ipv4", (v), N_("use IPv4 addresses only"), \
> > ## t/t3701-add-interactive.sh ##
> -@@ t/t3701-add-interactive.sh: test_expect_success 'hunk splitting works with diff.suppressBlankEmpty' '
> - test_cmp expect actual
> +@@ t/t3701-add-interactive.sh: test_expect_success 'add -p rejects negative diff.context' '
> + test_grep "diff.context cannot be negative" output
> '
> > --test_expect_success 'add -p respects diff.context' '
> -- test_write_lines a b c d e f g h i j k l m >file &&
> +for cmd in add checkout restore 'commit -m file'
> +do
> + test_expect_success "${cmd%% *} accepts -U and --inter-hunk-context" '
> @@ t/t3701-add-interactive.sh: test_expect_success 'hunk splitting works with diff.
> + test_write_lines a b c d e f g h i j k l m n o p q r s t u v >file &&
> + git commit -m file file &&
> + test_write_lines a b c d e F g h i j k l m n o p Q r s t u v >file &&
> - git add file &&
> -- test_write_lines a b c d e f G h i j k l m >file &&
> -- echo y | git -c diff.context=5 add -p >actual &&
> -- test_grep "@@ -2,11 +2,11 @@" actual
> ++ git add file &&
> + echo y | git -c diff.context=5 -c diff.interhunkcontext=1 \
> + reset -p -U 4 --inter-hunk-context 2 >actual &&
> + test_grep "@@ -2,20 +2,20 @@" actual
> - '
> -
> --test_expect_success 'add -p respects diff.interHunkContext' '
> -- test_write_lines a b c d e f g h i j k l m n o p q r s >file &&
> -- git add file &&
> -- test_write_lines a b c d E f g i i j k l m N o p q r s >file &&
> -- echo y | git -c diff.interhunkcontext=2 add -p >actual &&
> -- test_grep "@@ -2,16 +2,16 @@" actual
> ++'
> ++
> +test_expect_success 'stash accepts -U and --inter-hunk-context' '
> + test_write_lines a b c d e F g h i j k l m n o p Q r s t u v >file &&
> + git commit -m file file &&
> @@ t/t3701-add-interactive.sh: test_expect_success 'hunk splitting works with diff.
> + echo y | git -c diff.context=5 -c diff.interhunkcontext=1 \
> + stash -p -U 4 --inter-hunk-context 2 >actual &&
> + test_grep "@@ -2,20 +2,20 @@" actual
> - '
> -
> --test_expect_success 'add -p rejects negative diff.context' '
> -- test_config diff.context -1 &&
> -- test_must_fail git add -p 2>output &&
> -- test_grep "diff.context cannot be negative" output
> --'
> ++'
> ++
> +for cmd in add checkout commit reset restore "stash save" "stash push"
> +do
> + test_expect_success "$cmd rejects invalid context options" '
> @@ t/t3701-add-interactive.sh: test_expect_success 'hunk splitting works with diff.
> + test_grep -E ".--inter-hunk-context. requires .(--interactive/)?--patch." actual
> + '
> +done
> -
> - test_done
> -
> - ## t/t4055-diff-context.sh ##
> -@@ t/t4055-diff-context.sh: test_expect_success 'The -U option overrides diff.context' '
> - test_grep ! "^ firstline" output
> - '
> -
> -+test_expect_success 'The -U option overrides diff.context for "add"' '
> -+ test_config diff.context 8 &&
> -+ git add -U4 -p >output &&
> -+ test_grep ! "^ firstline" output
> -+'
> -+
> -+test_expect_success 'The -U option overrides diff.context for "commit"' '
> -+ test_config diff.context 8 &&
> -+ ! git commit -U4 -p >output &&
> -+ test_grep ! "^ firstline" output
> -+'
> -+
> -+test_expect_success 'The -U option overrides diff.context for "checkout"' '
> -+ test_config diff.context 8 &&
> -+ git checkout -U4 -p >output &&
> -+ test_grep ! "^ firstline" output
> -+'
> +
> -+test_expect_success 'The -U option overrides diff.context for "stash"' '
> -+ test_config diff.context 8 &&
> -+ ! git stash -U4 -p >output &&
> -+ test_grep ! "^ firstline" output
> -+'
> -+
> -+test_expect_success 'The -U option overrides diff.context for "restore"' '
> -+ test_config diff.context 8 &&
> -+ git restore -U4 -p >output &&
> -+ test_grep ! "^ firstline" output
> -+'
> -+
> - test_expect_success 'diff.context honored by "diff"' '
> - test_config diff.context 8 &&
> - git diff >output &&
> + test_done
> > ## t/t9902-completion.sh ##
> @@ t/t9902-completion.sh: test_expect_success 'double dash "git checkout"' '
> |
|
On the Git mailing list, Junio C Hamano wrote (reply to this): Phillip Wood <[email protected]> writes:
> Hi Leon
>
> The range-diff below looks good to me. Thanks for working on this
>
> Phillip
Yeah, the new one matches the previous one plus your fixes to the
tests, which I agree with fully.
Will queue. Let's mark the topic for 'next'.
Thanks, both of you.
|
|
On the Git mailing list, Leon Michalak wrote (reply to this): Great! Thanks Philip, Junio and everyone else who helped me get this
finished and in.
Was a good experience :) |
|
This patch series was integrated into seen via git@7462283. |
|
This patch series was integrated into next via git@393fa4b. |
|
This patch series was integrated into seen via git@a2ab174. |
|
There was a status update in the "Cooking" section about the branch "git add/etc -p" now honors diff.context configuration variable, and learns to honor -U<n> option. Will merge to 'master'. source: <[email protected]> |
|
This patch series was integrated into seen via git@da00594. |
|
This patch series was integrated into seen via git@c1b859f. |
|
There was a status update in the "Cooking" section about the branch "git add/etc -p" now honors diff.context configuration variable, and learns to honor -U<n> option. Will merge to 'master'. source: <[email protected]> |
|
This patch series was integrated into seen via git@d63f334. |
|
This patch series was integrated into master via git@d63f334. |
|
This patch series was integrated into next via git@d63f334. |
|
Closed via d63f334. |
This series of patches attempt to give
--interactive/--patchcompatible builtins ("add", "commit", "checkout", "reset", "restore" and "stash") better support and nicer experience for configuring how many context lines are shown in diffs through a variety of ways.Prior to these patches, the user could not choose how many context lines they saw in
--patchcommands (apart from one workaround by usingGIT_DIFF_OPTS=-u<number> ..., however this isn't a good user experience or a persistent solution). Additionally, the behaviour around reading from thediff.contextanddiff.interHunkContextconfigs was also inconsistent with other diff generating commands such as "log -p".The summarised changes below hopefully make this experience better and fix some inconsistencies:
diff.contextanddiff.interHunkContextconfigs are now respected by--patchcompatible commands--unifiedand--inter-hunk-contextcommand line options have been added to--patchcompatible commands (which take prescendence over file configs)--interactivemode now expose a new "context" subcommand which configures the amount of context lines you wish to see in subsequent diffs generated from other subcommands such as "patch" or "diff"The original discussion for this can be read at:
Changes since v1:
test_grepandtest_configutilsChanges since v2:
Changes since v3:
.instead, which is what some other tests that have this problem seem to use as well)Changes since v4:
cc: "Kristoffer Haugsbakk" [email protected]
cc: Eric Sunshine [email protected]
cc: Christian Couder [email protected]
cc: Phillip Wood [email protected]