* [PATCH 0/1] Improve automatic setup of tracking for new branches @ 2021-07-28 13:50 Ben Boeckel 2021-07-28 13:50 ` [PATCH 1/1] config: support setting up a remote tracking branch upon creation Ben Boeckel 2021-07-29 2:01 ` [PATCH 0/1] Improve automatic setup of tracking for new branches Ben Boeckel 0 siblings, 2 replies; 20+ messages in thread From: Ben Boeckel @ 2021-07-28 13:50 UTC (permalink / raw) To: git Cc: Ben Boeckel, Martin Ågren, Junio C Hamano, Jeff King, Jeff Hostetler, Nguyễn Thái Ngọc Duy, Taylor Blau Hi, I recently discovered some of the nicer behaviors of the `--track` option for local branches (courtesy of `vim-fugitive`'s status rendering). However, after a few weeks of slowly working `-t origin/master` into my workflow, I figured that Git could help me out here. This patch adds three new configuration variables to initialize the `branch.<name>.*` settings for tracking remote branches. I suspect there are more tests that should be added. Thanks, --Ben Ben Boeckel (1): config: support setting up a remote tracking branch upon creation Documentation/config/branch.txt | 15 +++++++++ branch.c | 28 ++++++++++------ branch.h | 3 ++ config.c | 15 +++++++++ environment.c | 3 ++ t/t3200-branch.sh | 57 +++++++++++++++++++++++++++++++++ 6 files changed, 111 insertions(+), 10 deletions(-) base-commit: eb27b338a3e71c7c4079fbac8aeae3f8fbb5c687 -- 2.31.1 ^ permalink raw reply [flat|nested] 20+ messages in thread
* [PATCH 1/1] config: support setting up a remote tracking branch upon creation 2021-07-28 13:50 [PATCH 0/1] Improve automatic setup of tracking for new branches Ben Boeckel @ 2021-07-28 13:50 ` Ben Boeckel 2021-07-28 17:48 ` Junio C Hamano 2021-07-29 2:01 ` [PATCH 0/1] Improve automatic setup of tracking for new branches Ben Boeckel 1 sibling, 1 reply; 20+ messages in thread From: Ben Boeckel @ 2021-07-28 13:50 UTC (permalink / raw) To: git Cc: Ben Boeckel, Martin Ågren, Junio C Hamano, Jeff King, Jeff Hostetler, Nguyễn Thái Ngọc Duy, Taylor Blau The `branch.autoSetupMerge` works well for setting up tracking a local branch, but there is no mechanism to automatically set up a remote tracking situation. This patch adds the following configuration values: - `branch.defaultRemote`: initializes `branch.<name>.remote` if not otherwise given; - `branch.defaultPushRemote`: initializes `branch.<name>.pushRemote` (currently falls back on `remote.pushDefault` when pushing); and - `branch.defaultMerge`: initializes `branch.<name>.merge` if not otherwise given. Signed-off-by: Ben Boeckel <mathstuf@gmail.com> --- Documentation/config/branch.txt | 15 +++++++++ branch.c | 28 ++++++++++------ branch.h | 3 ++ config.c | 15 +++++++++ environment.c | 3 ++ t/t3200-branch.sh | 57 +++++++++++++++++++++++++++++++++ 6 files changed, 111 insertions(+), 10 deletions(-) diff --git a/Documentation/config/branch.txt b/Documentation/config/branch.txt index cc5f3249fc..6e9d446066 100644 --- a/Documentation/config/branch.txt +++ b/Documentation/config/branch.txt @@ -1,3 +1,18 @@ +branch.defaultRemote:: + When a new branch is created with 'git branch', 'git switch' or 'git + checkout', this value will be used to initialize the + "branch.<name>.remote" setting. + +branch.defaultPushRemote:: + When a new branch is created with 'git branch', 'git switch' or 'git + checkout', this value will be used to initialize the + "branch.<name>.pushRemote" setting. + +branch.defaultMerge:: + When a new branch is created with 'git branch', 'git switch' or 'git + checkout', this value will be used to initialize the + "branch.<name>.merge" setting. + branch.autoSetupMerge:: Tells 'git branch', 'git switch' and 'git checkout' to set up new branches so that linkgit:git-pull[1] will appropriately merge from the diff --git a/branch.c b/branch.c index 7a88a4861e..0e8ece8259 100644 --- a/branch.c +++ b/branch.c @@ -60,6 +60,9 @@ int install_branch_config(int flag, const char *local, const char *origin, const const char *shortname = NULL; struct strbuf key = STRBUF_INIT; int rebasing = should_setup_rebase(origin); + const char *actual_origin = origin ? origin : git_branch_remote; + const char *actual_push_origin = git_branch_push_remote; + const char *actual_remote = remote ? remote : git_branch_merge; if (skip_prefix(remote, "refs/heads/", &shortname) && !strcmp(local, shortname) @@ -70,12 +73,17 @@ int install_branch_config(int flag, const char *local, const char *origin, const } strbuf_addf(&key, "branch.%s.remote", local); - if (git_config_set_gently(key.buf, origin ? origin : ".") < 0) + if (git_config_set_gently(key.buf, actual_origin ? actual_origin : ".") < 0) + goto out_err; + + strbuf_reset(&key); + strbuf_addf(&key, "branch.%s.pushremote", local); + if (git_config_set_gently(key.buf, actual_push_origin) < 0) goto out_err; strbuf_reset(&key); strbuf_addf(&key, "branch.%s.merge", local); - if (git_config_set_gently(key.buf, remote) < 0) + if (git_config_set_gently(key.buf, actual_remote) < 0) goto out_err; if (rebasing) { @@ -88,27 +96,27 @@ int install_branch_config(int flag, const char *local, const char *origin, const if (flag & BRANCH_CONFIG_VERBOSE) { if (shortname) { - if (origin) + if (actual_origin) printf_ln(rebasing ? _("Branch '%s' set up to track remote branch '%s' from '%s' by rebasing.") : _("Branch '%s' set up to track remote branch '%s' from '%s'."), - local, shortname, origin); + local, shortname, actual_origin); else printf_ln(rebasing ? _("Branch '%s' set up to track local branch '%s' by rebasing.") : _("Branch '%s' set up to track local branch '%s'."), local, shortname); } else { - if (origin) + if (actual_origin) printf_ln(rebasing ? _("Branch '%s' set up to track remote ref '%s' by rebasing.") : _("Branch '%s' set up to track remote ref '%s'."), - local, remote); + local, actual_remote); else printf_ln(rebasing ? _("Branch '%s' set up to track local ref '%s' by rebasing.") : _("Branch '%s' set up to track local ref '%s'."), - local, remote); + local, actual_remote); } } @@ -119,9 +127,9 @@ int install_branch_config(int flag, const char *local, const char *origin, const error(_("Unable to write upstream branch configuration")); advise(_(tracking_advice), - origin ? origin : "", - origin ? "/" : "", - shortname ? shortname : remote); + actual_origin ? actual_origin : "", + actual_origin ? "/" : "", + shortname ? shortname : actual_remote); return -1; } diff --git a/branch.h b/branch.h index df0be61506..6ce978731d 100644 --- a/branch.h +++ b/branch.h @@ -14,6 +14,9 @@ enum branch_track { }; extern enum branch_track git_branch_track; +extern const char* git_branch_remote; +extern const char* git_branch_push_remote; +extern const char* git_branch_merge; /* Functions for acting on the information about branches. */ diff --git a/config.c b/config.c index f33abeab85..42fc510a9f 100644 --- a/config.c +++ b/config.c @@ -1599,6 +1599,21 @@ static int git_default_branch_config(const char *var, const char *value) return error(_("malformed value for %s"), var); return 0; } + if (!strcmp(var, "branch.defaultremote")) { + if (!value) + return config_error_nonbool(var); + return git_config_string(&git_branch_remote, var, value); + } + if (!strcmp(var, "branch.defaultpushremote")) { + if (!value) + return config_error_nonbool(var); + return git_config_string(&git_branch_push_remote, var, value); + } + if (!strcmp(var, "branch.defaultmerge")) { + if (!value) + return config_error_nonbool(var); + return git_config_string(&git_branch_merge, var, value); + } /* Add other config variables here and to Documentation/config.txt. */ return 0; diff --git a/environment.c b/environment.c index 2f27008424..3b4d54e7dc 100644 --- a/environment.c +++ b/environment.c @@ -60,6 +60,9 @@ int global_conv_flags_eol = CONV_EOL_RNDTRP_WARN; char *check_roundtrip_encoding = "SHIFT-JIS"; unsigned whitespace_rule_cfg = WS_DEFAULT_RULE; enum branch_track git_branch_track = BRANCH_TRACK_REMOTE; +const char* git_branch_remote = NULL; +const char* git_branch_push_remote = NULL; +const char* git_branch_merge = NULL; enum rebase_setup_type autorebase = AUTOREBASE_NEVER; enum push_default_type push_default = PUSH_DEFAULT_UNSPECIFIED; #ifndef OBJECT_CREATION_MODE diff --git a/t/t3200-branch.sh b/t/t3200-branch.sh index cc4b10236e..2edfb50872 100755 --- a/t/t3200-branch.sh +++ b/t/t3200-branch.sh @@ -797,6 +797,63 @@ test_expect_success 'test tracking setup via --track but deeper' ' test "$(git config branch.my7.merge)" = refs/heads/o/o ' +test_expect_success 'test tracking setup via branch.default* and --track' ' + git config branch.autosetupmerge always && + git config branch.defaultremote local && + git config branch.defaultmerge main && + git config remote.local.url . && + git config remote.local.fetch refs/heads/*:refs/remotes/local/* && + (git show-ref -q refs/remotes/local/main || git fetch local) && + git branch --track other/foo my2 && + git config branch.autosetupmerge false && + test "$(git config branch.my2.remote)" = other && + ! test "$(git config branch.my2.pushremote)" = other && + test "$(git config branch.my2.merge)" = refs/heads/foo +' + +test_expect_success 'test tracking setup via branch.default* and --no-track' ' + git config branch.autosetupmerge always && + git config branch.defaultremote local && + git config branch.defaultmerge main && + git config remote.local.url . && + git config remote.local.fetch refs/heads/*:refs/remotes/local/* && + (git show-ref -q refs/remotes/local/main || git fetch local) && + git branch --no-track my2 && + git config branch.autosetupmerge false && + ! test "$(git config branch.my2.remote)" = local && + ! test "$(git config branch.my2.pushremote)" = other && + ! test "$(git config branch.my2.merge)" = refs/heads/main +' + +test_expect_success 'test tracking setup via branch.default*' ' + git config branch.autosetupmerge always && + git config branch.defaultremote local && + git config branch.defaultmerge main && + git config remote.local.url . && + git config remote.local.fetch refs/heads/*:refs/remotes/local/* && + (git show-ref -q refs/remotes/local/main || git fetch local) && + git branch my2 && + git config branch.autosetupmerge false && + test "$(git config branch.my2.remote)" = local && + ! test "$(git config branch.my2.pushremote)" = other && + test "$(git config branch.my2.merge)" = refs/heads/main +' + +test_expect_success 'test tracking setup via branch.default* with pushremote' ' + git config branch.autosetupmerge always && + git config branch.defaultremote local && + git config branch.defaultpushremote other && + git config branch.defaultmerge main && + git config remote.local.url . && + git config remote.local.fetch refs/heads/*:refs/remotes/local/* && + (git show-ref -q refs/remotes/local/main || git fetch local) && + git branch my2 && + git config branch.autosetupmerge false && + test "$(git config branch.my2.remote)" = local && + test "$(git config branch.my2.pushremote)" = other && + test "$(git config branch.my2.merge)" = refs/heads/main +' + test_expect_success 'test deleting branch deletes branch config' ' git branch -d my7 && test -z "$(git config branch.my7.remote)" && -- 2.31.1 ^ permalink raw reply related [flat|nested] 20+ messages in thread
* Re: [PATCH 1/1] config: support setting up a remote tracking branch upon creation 2021-07-28 13:50 ` [PATCH 1/1] config: support setting up a remote tracking branch upon creation Ben Boeckel @ 2021-07-28 17:48 ` Junio C Hamano 2021-07-28 18:26 ` Ben Boeckel 0 siblings, 1 reply; 20+ messages in thread From: Junio C Hamano @ 2021-07-28 17:48 UTC (permalink / raw) To: Ben Boeckel Cc: git, Martin Ågren, Jeff King, Jeff Hostetler, Nguyễn Thái Ngọc Duy, Taylor Blau Ben Boeckel <mathstuf@gmail.com> writes: > The `branch.autoSetupMerge` works well for setting up tracking a local > branch, but there is no mechanism to automatically set up a remote > tracking situation. This description is probably insufficient to explain what's missing, probably because "set up a remote tracking situation" is a bit fuzzy. Without this patch, I can do this already: $ git checkout -t -b topic origin/topic And after the above, we have [branch "topic"] remote = origin merge = refs/heads/topic Of course, you can instead use a short-hand DWIM, e.g. $ git checkout topic ;# assuming origin/topic exists gives us the same thing. In either case, topic knows to integrate with the 'topic' branch from remote 'origin' and push back there. So instead of saying "there is no mechanism to ...", be a bit more specific what you can and cannot do in this first paragraph. Then we can describe the solution we'd propose in the second and subsequent paragraphs. Thanks. > +branch.defaultRemote:: > + When a new branch is created with 'git branch', 'git switch' or 'git > + checkout', this value will be used to initialize the > + "branch.<name>.remote" setting. You mean without "-t"? I offhand do not think of a reason why this is a good idea. How would one create local topic branches that you plan to merge locally into your own larger topic branch to be shared with others? Shouldn't there be an easy way to countermand the setting by this configuration? > +branch.defaultPushRemote:: > + When a new branch is created with 'git branch', 'git switch' or 'git > + checkout', this value will be used to initialize the > + "branch.<name>.pushRemote" setting. Ditto. > +branch.defaultMerge:: > + When a new branch is created with 'git branch', 'git switch' or 'git > + checkout', this value will be used to initialize the > + "branch.<name>.merge" setting. So the expected use case is to fork multiple local branches, to integrate with the same branch from the remote? I think we can already do $ git checkout -t -b xyzzy origin/master $ git checkout -t -b frotz origin/master and you can lose -t with branch.autosetupmerge setting. As the "git branch" command needs to get the name of your branch (e.g. 'xyzzy' or 'frotz') and which remote tracking branch you start from (e.g. 'origin/master'), even with branch.{defaultRemote,defaultMerge} configuration, you wouldn't lose that many keystrokes, I suspect. Or do you plan to make $ git branch xyzzy a short-hand for $ git branch -t xyzzy origin/master when defaultRemote and defaultMerge are set to 'origin' and 'refs/heads/master'? It would be way too confusing to start the new branch from origin/master in such a case, as everybody learned that "git branch xyzzy" that does not say where the branch initially points at uses HEAD. ^ permalink raw reply [flat|nested] 20+ messages in thread
* Re: [PATCH 1/1] config: support setting up a remote tracking branch upon creation 2021-07-28 17:48 ` Junio C Hamano @ 2021-07-28 18:26 ` Ben Boeckel 2021-07-28 18:39 ` Junio C Hamano 0 siblings, 1 reply; 20+ messages in thread From: Ben Boeckel @ 2021-07-28 18:26 UTC (permalink / raw) To: Junio C Hamano Cc: git, Martin Ågren, Jeff King, Jeff Hostetler, Nguyễn Thái Ngọc Duy, Taylor Blau On Wed, Jul 28, 2021 at 10:48:56 -0700, Junio C Hamano wrote: > Ben Boeckel <mathstuf@gmail.com> writes: > > > The `branch.autoSetupMerge` works well for setting up tracking a local > > branch, but there is no mechanism to automatically set up a remote > > tracking situation. > > This description is probably insufficient to explain what's missing, > probably because "set up a remote tracking situation" is a bit > fuzzy. Fair enough. I finally understood what was going on with "tracking" only recently. Usually I would track the remote's branch of the same name, but I found that "tracking" `origin/master` instead of `myfork/topic` makes `fugitive` render the following: - commits on my topic that aren't integrated (this makes it easy to tell it to "amend this commit" using its keybinding; - commits on `master` that aren't on my topic (so I can see if there's anything relevant to rebase on top of); and - diffs against my fork's branch (as last fetched) so that I can see the status. Maybe this is a tooling issue, but I saw this as a potential way to avoid having to specify this information on every topic I create. > Without this patch, I can do this already: > > $ git checkout -t -b topic origin/topic I should note that the *vast* majority of my development is done using the fork-based workflow. I have `remote.pushDefault` set to my fork and I use `push.default = simple` (there are only a handful of repositories where "my fork" is also `origin` and I have, on multiple occasions, wanted to make forks even there). My normal pattern is that I'm on the target branch already, so I have: $ git checkout -b topic which doesn't do any tracking. Just `-t` makes it track my *local* `master` when I really want it to track `origin/master`. AFAIK, there's no shortcut for this other than to give the full `-t origin/master` at branch creation time (and as something I do all the time). > And after the above, we have > > [branch "topic"] > remote = origin > merge = refs/heads/topic > > Of course, you can instead use a short-hand DWIM, e.g. > > $ git checkout topic ;# assuming origin/topic exists > > gives us the same thing. In either case, topic knows to integrate > with the 'topic' branch from remote 'origin' and push back there. This doesn't match with my usual experience in fork-based workflows. I want the topic to track `master` and then my `remote.pushDefault` makes sure it goes to "the right place" when pushing. > So instead of saying "there is no mechanism to ...", be a bit more > specific what you can and cannot do in this first paragraph. > > Then we can describe the solution we'd propose in the second and > subsequent paragraphs. > > Thanks. I'll work on an improved cover letter to give more background. > > +branch.defaultRemote:: > > + When a new branch is created with 'git branch', 'git switch' or 'git > > + checkout', this value will be used to initialize the > > + "branch.<name>.remote" setting. > > You mean without "-t"? I offhand do not think of a reason why this > is a good idea. How would one create local topic branches that you > plan to merge locally into your own larger topic branch to be shared > with others? Shouldn't there be an easy way to countermand the > setting by this configuration? Everything goes through merge requests for the projects I work on day-to-day (even contributions to "my own" projects due to CI workflows). I added a test that `--no-track` works for this case (which given its rarity for me is the right tradeoff at least). > > +branch.defaultPushRemote:: > > + When a new branch is created with 'git branch', 'git switch' or 'git > > + checkout', this value will be used to initialize the > > + "branch.<name>.pushRemote" setting. > > Ditto. > > > +branch.defaultMerge:: > > + When a new branch is created with 'git branch', 'git switch' or 'git > > + checkout', this value will be used to initialize the > > + "branch.<name>.merge" setting. > > So the expected use case is to fork multiple local branches, to > integrate with the same branch from the remote? I think we can > already do > > $ git checkout -t -b xyzzy origin/master > $ git checkout -t -b frotz origin/master > > and you can lose -t with branch.autosetupmerge setting. As the "git > branch" command needs to get the name of your branch (e.g. 'xyzzy' > or 'frotz') and which remote tracking branch you start from > (e.g. 'origin/master'), even with branch.{defaultRemote,defaultMerge} > configuration, you wouldn't lose that many keystrokes, I suspect. There are times that I want to branch from HEAD, but track `master` (for example, a branch destined for backporting to an older branch). The equivalent of: $ git checkout release $ git checkout -b backported-branch $ git branch --set-upstream-to=origin/master > Or do you plan to make > > $ git branch xyzzy > > a short-hand for > > $ git branch -t xyzzy origin/master > > when defaultRemote and defaultMerge are set to 'origin' and > 'refs/heads/master'? It would be way too confusing to start the > new branch from origin/master in such a case, as everybody learned > that "git branch xyzzy" that does not say where the branch initially > points at uses HEAD. No, it would just *track* `origin/master`, not branch from it. It should be shorthand for: $ git branch xyzzy $ git branch --set-upstream-to=origin/master xyzzy Though I personally use `git checkout -b` far more often to create branches. And since "every" branch I make would have `-t origin/master`, I wanted to have configuration to do this part for me. Hopefully this gives a clearer picture of where I'm coming from. Thanks, --Ben ^ permalink raw reply [flat|nested] 20+ messages in thread
* Re: [PATCH 1/1] config: support setting up a remote tracking branch upon creation 2021-07-28 18:26 ` Ben Boeckel @ 2021-07-28 18:39 ` Junio C Hamano 0 siblings, 0 replies; 20+ messages in thread From: Junio C Hamano @ 2021-07-28 18:39 UTC (permalink / raw) To: Ben Boeckel Cc: git, Martin Ågren, Jeff King, Jeff Hostetler, Nguyễn Thái Ngọc Duy, Taylor Blau Ben Boeckel <mathstuf@gmail.com> writes: > Hopefully this gives a clearer picture of where I'm coming from. I think you gave descriptions to a reasonable level of detail to explain what your expected workflow is, to what extent the current tools support the workflow already, and what are still missing (and you want to extend the system to help). The reason I asked these questions is to make sure that future readers of "git log" after your contribution becomes part of our history will not have to ask them (as you won't be as readily available to answer their questions). So please don't be content just because you answered and helped _me_ understand where you're coming from. Make sure an updated documentation and proposed log message explains them well to our future readers without your assistance. Thanks. ^ permalink raw reply [flat|nested] 20+ messages in thread
* [PATCH 0/1] Improve automatic setup of tracking for new branches 2021-07-28 13:50 [PATCH 0/1] Improve automatic setup of tracking for new branches Ben Boeckel 2021-07-28 13:50 ` [PATCH 1/1] config: support setting up a remote tracking branch upon creation Ben Boeckel @ 2021-07-29 2:01 ` Ben Boeckel 2021-07-29 2:01 ` [PATCH 1/1] config: support a default remote tracking setup upon branch creation Ben Boeckel ` (2 more replies) 1 sibling, 3 replies; 20+ messages in thread From: Ben Boeckel @ 2021-07-29 2:01 UTC (permalink / raw) To: git Cc: Ben Boeckel, Martin Ågren, Junio C Hamano, Jeff King, Jeff Hostetler, Nguyễn Thái Ngọc Duy, Taylor Blau Hi, I recently discovered some of the nicer behaviors of the `--track` option for local branches (courtesy of `vim-fugitive`'s status rendering). However, after a few weeks of slowly working `-t origin/master` into my workflow, I figured that Git could help me out here. This patch adds two new configuration variables to initialize the `branch.<name>.*` settings for tracking remote branches. I searched the docs (including `Documentation/gitworkflows.txt`), but didn't see anywhere to describe the fork-based workflow common on forges (such as GitHub and GitLab) where this felt "at home". I suppose I could describe the workflow I'm used to, but I think I'll hammer that out as blog posts before submitting it as documentation. I suspect there are more tests that should be added. Thanks, --Ben --- v1 -> v2: - removed `branch.defaultPushRemote` (`remote.pushDefault` covers this case already) - improved the commit message with more background and an expanation of the intended uses Ben Boeckel (1): config: support a default remote tracking setup upon branch creation Documentation/config/branch.txt | 14 ++++++++++++ branch.c | 22 ++++++++++--------- branch.h | 2 ++ config.c | 10 +++++++++ environment.c | 2 ++ t/t3200-branch.sh | 39 +++++++++++++++++++++++++++++++++ 6 files changed, 79 insertions(+), 10 deletions(-) base-commit: eb27b338a3e71c7c4079fbac8aeae3f8fbb5c687 -- 2.31.1 ^ permalink raw reply [flat|nested] 20+ messages in thread
* [PATCH 1/1] config: support a default remote tracking setup upon branch creation 2021-07-29 2:01 ` [PATCH 0/1] Improve automatic setup of tracking for new branches Ben Boeckel @ 2021-07-29 2:01 ` Ben Boeckel 2021-07-30 13:35 ` Philippe Blain 2021-08-02 13:02 ` Ævar Arnfjörð Bjarmason 2021-07-30 1:04 ` [PATCH 0/1] Improve automatic setup of tracking for new branches Junio C Hamano 2021-07-30 13:35 ` Philippe Blain 2 siblings, 2 replies; 20+ messages in thread From: Ben Boeckel @ 2021-07-29 2:01 UTC (permalink / raw) To: git Cc: Ben Boeckel, Martin Ågren, Junio C Hamano, Jeff King, Jeff Hostetler, Nguyễn Thái Ngọc Duy, Taylor Blau The `branch.autoSetupMerge` works well today for setting up tracking a local branch, but there is no easy mechanism to automatically set up a remote tracking situation for workflows which use a single set of branches for integration without specifying `--track` to every branch creation command or branching directly from the remote ref. This patch adds the following configuration values: - `branch.defaultRemote`: initializes `branch.<name>.remote` if not otherwise given; - `branch.defaultMerge`: initializes `branch.<name>.merge` if not otherwise given. These effectively make branch creation commands such as `git branch`, `git switch`, and `git checkout -b` have an implicit `-t ${branch.defaultRemote}/${branch.defaultMerge}` argument and is equivalent to doing: $ git branch new-topic # or another branch creation command $ git branch --set-upstream-to=${branch.defaultRemote}/${branch.defaultMerge} new-topic In a fork-based workflow where contributions to the main project flow in through forks of the main repository rather than pushing directly to it (not uncommon of forge-hosted projects), the following setup may be used: $ git config remote.pushDefault myfork # push to `myfork` by default $ git config push.default simple # the default $ git config branch.autoSetupMerge always # always setup tracking $ git config branch.defaultRemote origin # track from `origin` $ git config branch.defaultMerge main # track the `main` branch With this setup, branches will automatically be set up to track `origin/main` while being pushed to `myfork` by default. Some tools (at least Vim's fugitive plugin) will show the commit differences between both the tracking branch and the default push target. This allows such tools to easily show "what needs to be merged?", "what has happened on the target branch since this topic has been created?", and "what is my status against my fork's current status?" at a glance. Additionally, with the `extensions.worktreeConfig` setting, when a separate work tree which is used for changes against a different branch (e.g., a branch tracking a prior release), the `branch.defaultMerge` setting may be changed independently, e.g., to `maint`. Signed-off-by: Ben Boeckel <mathstuf@gmail.com> --- Documentation/config/branch.txt | 14 ++++++++++++ branch.c | 22 ++++++++++--------- branch.h | 2 ++ config.c | 10 +++++++++ environment.c | 2 ++ t/t3200-branch.sh | 39 +++++++++++++++++++++++++++++++++ 6 files changed, 79 insertions(+), 10 deletions(-) diff --git a/Documentation/config/branch.txt b/Documentation/config/branch.txt index cc5f3249fc..ab5cd2c1ed 100644 --- a/Documentation/config/branch.txt +++ b/Documentation/config/branch.txt @@ -1,3 +1,17 @@ +branch.defaultRemote:: + Tells 'git branch', 'git switch' and 'git checkout' to set up new branches + so that it will track a branch on the specified remote. This can be + used, in conjunction with `branch.defaultMerge`, in projects where + branches tend to target a single branch. This will be used to + initialize the "branch.<name>.remote" setting. + +branch.defaultMerge:: + Tells 'git branch', 'git switch' and 'git checkout' to set up new branches + so that it will track a branch with this name. This can be used, in + conjunction with `branch.defaultRemote` in projects where branches tend + to target a single branch. This will be used to initialize the + "branch.<name>.merge" setting. + branch.autoSetupMerge:: Tells 'git branch', 'git switch' and 'git checkout' to set up new branches so that linkgit:git-pull[1] will appropriately merge from the diff --git a/branch.c b/branch.c index 7a88a4861e..097d5af647 100644 --- a/branch.c +++ b/branch.c @@ -60,6 +60,8 @@ int install_branch_config(int flag, const char *local, const char *origin, const const char *shortname = NULL; struct strbuf key = STRBUF_INIT; int rebasing = should_setup_rebase(origin); + const char *actual_origin = origin ? origin : git_branch_remote; + const char *actual_remote = remote ? remote : git_branch_merge; if (skip_prefix(remote, "refs/heads/", &shortname) && !strcmp(local, shortname) @@ -70,12 +72,12 @@ int install_branch_config(int flag, const char *local, const char *origin, const } strbuf_addf(&key, "branch.%s.remote", local); - if (git_config_set_gently(key.buf, origin ? origin : ".") < 0) + if (git_config_set_gently(key.buf, actual_origin ? actual_origin : ".") < 0) goto out_err; strbuf_reset(&key); strbuf_addf(&key, "branch.%s.merge", local); - if (git_config_set_gently(key.buf, remote) < 0) + if (git_config_set_gently(key.buf, actual_remote) < 0) goto out_err; if (rebasing) { @@ -88,27 +90,27 @@ int install_branch_config(int flag, const char *local, const char *origin, const if (flag & BRANCH_CONFIG_VERBOSE) { if (shortname) { - if (origin) + if (actual_origin) printf_ln(rebasing ? _("Branch '%s' set up to track remote branch '%s' from '%s' by rebasing.") : _("Branch '%s' set up to track remote branch '%s' from '%s'."), - local, shortname, origin); + local, shortname, actual_origin); else printf_ln(rebasing ? _("Branch '%s' set up to track local branch '%s' by rebasing.") : _("Branch '%s' set up to track local branch '%s'."), local, shortname); } else { - if (origin) + if (actual_origin) printf_ln(rebasing ? _("Branch '%s' set up to track remote ref '%s' by rebasing.") : _("Branch '%s' set up to track remote ref '%s'."), - local, remote); + local, actual_remote); else printf_ln(rebasing ? _("Branch '%s' set up to track local ref '%s' by rebasing.") : _("Branch '%s' set up to track local ref '%s'."), - local, remote); + local, actual_remote); } } @@ -119,9 +121,9 @@ int install_branch_config(int flag, const char *local, const char *origin, const error(_("Unable to write upstream branch configuration")); advise(_(tracking_advice), - origin ? origin : "", - origin ? "/" : "", - shortname ? shortname : remote); + actual_origin ? actual_origin : "", + actual_origin ? "/" : "", + shortname ? shortname : actual_remote); return -1; } diff --git a/branch.h b/branch.h index df0be61506..7d7990dda7 100644 --- a/branch.h +++ b/branch.h @@ -14,6 +14,8 @@ enum branch_track { }; extern enum branch_track git_branch_track; +extern const char* git_branch_remote; +extern const char* git_branch_merge; /* Functions for acting on the information about branches. */ diff --git a/config.c b/config.c index f33abeab85..a46c5a43a1 100644 --- a/config.c +++ b/config.c @@ -1599,6 +1599,16 @@ static int git_default_branch_config(const char *var, const char *value) return error(_("malformed value for %s"), var); return 0; } + if (!strcmp(var, "branch.defaultremote")) { + if (!value) + return config_error_nonbool(var); + return git_config_string(&git_branch_remote, var, value); + } + if (!strcmp(var, "branch.defaultmerge")) { + if (!value) + return config_error_nonbool(var); + return git_config_string(&git_branch_merge, var, value); + } /* Add other config variables here and to Documentation/config.txt. */ return 0; diff --git a/environment.c b/environment.c index 2f27008424..d550deabbd 100644 --- a/environment.c +++ b/environment.c @@ -60,6 +60,8 @@ int global_conv_flags_eol = CONV_EOL_RNDTRP_WARN; char *check_roundtrip_encoding = "SHIFT-JIS"; unsigned whitespace_rule_cfg = WS_DEFAULT_RULE; enum branch_track git_branch_track = BRANCH_TRACK_REMOTE; +const char* git_branch_remote = NULL; +const char* git_branch_merge = NULL; enum rebase_setup_type autorebase = AUTOREBASE_NEVER; enum push_default_type push_default = PUSH_DEFAULT_UNSPECIFIED; #ifndef OBJECT_CREATION_MODE diff --git a/t/t3200-branch.sh b/t/t3200-branch.sh index cc4b10236e..82137e8451 100755 --- a/t/t3200-branch.sh +++ b/t/t3200-branch.sh @@ -797,6 +797,45 @@ test_expect_success 'test tracking setup via --track but deeper' ' test "$(git config branch.my7.merge)" = refs/heads/o/o ' +test_expect_success 'test tracking setup via branch.default* and --track' ' + git config branch.autosetupmerge always && + git config branch.defaultremote local && + git config branch.defaultmerge main && + git config remote.local.url . && + git config remote.local.fetch refs/heads/*:refs/remotes/local/* && + (git show-ref -q refs/remotes/local/main || git fetch local) && + git branch --track other/foo my2 && + git config branch.autosetupmerge false && + test "$(git config branch.my2.remote)" = other && + test "$(git config branch.my2.merge)" = refs/heads/foo +' + +test_expect_success 'test tracking setup via branch.default* and --no-track' ' + git config branch.autosetupmerge always && + git config branch.defaultremote local && + git config branch.defaultmerge main && + git config remote.local.url . && + git config remote.local.fetch refs/heads/*:refs/remotes/local/* && + (git show-ref -q refs/remotes/local/main || git fetch local) && + git branch --no-track my2 && + git config branch.autosetupmerge false && + ! test "$(git config branch.my2.remote)" = local && + ! test "$(git config branch.my2.merge)" = refs/heads/main +' + +test_expect_success 'test tracking setup via branch.default*' ' + git config branch.autosetupmerge always && + git config branch.defaultremote local && + git config branch.defaultmerge main && + git config remote.local.url . && + git config remote.local.fetch refs/heads/*:refs/remotes/local/* && + (git show-ref -q refs/remotes/local/main || git fetch local) && + git branch my2 && + git config branch.autosetupmerge false && + test "$(git config branch.my2.remote)" = local && + test "$(git config branch.my2.merge)" = refs/heads/main +' + test_expect_success 'test deleting branch deletes branch config' ' git branch -d my7 && test -z "$(git config branch.my7.remote)" && -- 2.31.1 ^ permalink raw reply related [flat|nested] 20+ messages in thread
* Re: [PATCH 1/1] config: support a default remote tracking setup upon branch creation 2021-07-29 2:01 ` [PATCH 1/1] config: support a default remote tracking setup upon branch creation Ben Boeckel @ 2021-07-30 13:35 ` Philippe Blain 2021-07-30 14:07 ` Ben Boeckel 2021-07-30 17:32 ` Junio C Hamano 2021-08-02 13:02 ` Ævar Arnfjörð Bjarmason 1 sibling, 2 replies; 20+ messages in thread From: Philippe Blain @ 2021-07-30 13:35 UTC (permalink / raw) To: Ben Boeckel, git Cc: Martin Ågren, Junio C Hamano, Jeff King, Jeff Hostetler, Nguyễn Thái Ngọc Duy, Taylor Blau Le 2021-07-28 à 22:01, Ben Boeckel a écrit : > The `branch.autoSetupMerge` works well today for setting up tracking a > local branch, but there is no easy mechanism to automatically set up a > remote tracking situation for workflows which use a single set of > branches for integration without specifying `--track` to every branch > creation command or branching directly from the remote ref. This patch > adds the following configuration values: > > - `branch.defaultRemote`: initializes `branch.<name>.remote` if not > otherwise given; > - `branch.defaultMerge`: initializes `branch.<name>.merge` if not > otherwise given. > > These effectively make branch creation commands such as `git branch`, > `git switch`, and `git checkout -b` have an implicit `-t > ${branch.defaultRemote}/${branch.defaultMerge}` argument and is > equivalent to doing: > > $ git branch new-topic # or another branch creation command > $ git branch --set-upstream-to=${branch.defaultRemote}/${branch.defaultMerge} new-topic > > In a fork-based workflow where contributions to the main project flow in > through forks of the main repository rather than pushing directly to it > (not uncommon of forge-hosted projects), the following setup may be > used: > > $ git config remote.pushDefault myfork # push to `myfork` by default > $ git config push.default simple # the default > $ git config branch.autoSetupMerge always # always setup tracking OK, so if I understand correctly this exisiting setting has to be changed to 'always' for the new settings you are adding to take effect, right ? I think this does make sense, reading the description of 'branch.autoSetupMerge', but maybe it should be spelled explicitely in the text of the commit message, and not just mentioned here in this terminal session excerpt. > $ git config branch.defaultRemote origin # track from `origin` > $ git config branch.defaultMerge main # track the `main` branch Small nit: maybe I would invert these two, so it can read: $ git config branch.defaultMerge main # track the `main` branch ... $ git config branch.defaultRemote origin # ... from `origin` > > With this setup, branches will automatically be set up to track > `origin/main` while being pushed to `myfork` by default. Some tools > (at least Vim's fugitive plugin) will show the commit differences > between both the tracking branch and the default push target. This > allows such tools to easily show "what needs to be merged?", "what has > happened on the target branch since this topic has been created?", and > "what is my status against my fork's current status?" at a glance. > > Additionally, with the `extensions.worktreeConfig` setting, when a > separate work tree which is used for changes against a different branch > (e.g., a branch tracking a prior release), the `branch.defaultMerge` > setting may be changed independently, e.g., to `maint`. This last paragraph is not explicitely needed, as nothing relating to 'extensions.worktreeConfig' is changed here right ? It's just the normal way that this setting works. > > Signed-off-by: Ben Boeckel <mathstuf@gmail.com> > --- > Documentation/config/branch.txt | 14 ++++++++++++ > branch.c | 22 ++++++++++--------- > branch.h | 2 ++ > config.c | 10 +++++++++ > environment.c | 2 ++ > t/t3200-branch.sh | 39 +++++++++++++++++++++++++++++++++ > 6 files changed, 79 insertions(+), 10 deletions(-) > > diff --git a/Documentation/config/branch.txt b/Documentation/config/branch.txt > index cc5f3249fc..ab5cd2c1ed 100644 > --- a/Documentation/config/branch.txt > +++ b/Documentation/config/branch.txt > @@ -1,3 +1,17 @@ > +branch.defaultRemote:: > + Tells 'git branch', 'git switch' and 'git checkout' to set up new branches > + so that it will track a branch on the specified remote. This can be > + used, in conjunction with `branch.defaultMerge`, in projects where > + branches tend to target a single branch. This will be used to > + initialize the "branch.<name>.remote" setting. > + > +branch.defaultMerge:: > + Tells 'git branch', 'git switch' and 'git checkout' to set up new branches > + so that it will track a branch with this name. This can be used, in > + conjunction with `branch.defaultRemote` in projects where branches tend > + to target a single branch. This will be used to initialize the > + "branch.<name>.merge" setting. For the two setting above, if 'branch.autoSetupMerge' must be set to 'always' for the settings to work, I think it should be explicitely mentioned. > + > branch.autoSetupMerge:: > Tells 'git branch', 'git switch' and 'git checkout' to set up new branches > so that linkgit:git-pull[1] will appropriately merge from the > diff --git a/branch.c b/branch.c > index 7a88a4861e..097d5af647 100644 > --- a/branch.c > +++ b/branch.c > @@ -60,6 +60,8 @@ int install_branch_config(int flag, const char *local, const char *origin, const > const char *shortname = NULL; > struct strbuf key = STRBUF_INIT; > int rebasing = should_setup_rebase(origin); > + const char *actual_origin = origin ? origin : git_branch_remote; > + const char *actual_remote = remote ? remote : git_branch_merge; > > if (skip_prefix(remote, "refs/heads/", &shortname) > && !strcmp(local, shortname) > @@ -70,12 +72,12 @@ int install_branch_config(int flag, const char *local, const char *origin, const > } > > strbuf_addf(&key, "branch.%s.remote", local); > - if (git_config_set_gently(key.buf, origin ? origin : ".") < 0) > + if (git_config_set_gently(key.buf, actual_origin ? actual_origin : ".") < 0) > goto out_err; > > strbuf_reset(&key); > strbuf_addf(&key, "branch.%s.merge", local); > - if (git_config_set_gently(key.buf, remote) < 0) > + if (git_config_set_gently(key.buf, actual_remote) < 0) > goto out_err; > > if (rebasing) { > @@ -88,27 +90,27 @@ int install_branch_config(int flag, const char *local, const char *origin, const > > if (flag & BRANCH_CONFIG_VERBOSE) { > if (shortname) { > - if (origin) > + if (actual_origin) > printf_ln(rebasing ? > _("Branch '%s' set up to track remote branch '%s' from '%s' by rebasing.") : > _("Branch '%s' set up to track remote branch '%s' from '%s'."), > - local, shortname, origin); > + local, shortname, actual_origin); > else > printf_ln(rebasing ? > _("Branch '%s' set up to track local branch '%s' by rebasing.") : > _("Branch '%s' set up to track local branch '%s'."), > local, shortname); > } else { > - if (origin) > + if (actual_origin) > printf_ln(rebasing ? > _("Branch '%s' set up to track remote ref '%s' by rebasing.") : > _("Branch '%s' set up to track remote ref '%s'."), > - local, remote); > + local, actual_remote); > else > printf_ln(rebasing ? > _("Branch '%s' set up to track local ref '%s' by rebasing.") : > _("Branch '%s' set up to track local ref '%s'."), > - local, remote); > + local, actual_remote); > } > } > > @@ -119,9 +121,9 @@ int install_branch_config(int flag, const char *local, const char *origin, const > error(_("Unable to write upstream branch configuration")); > > advise(_(tracking_advice), > - origin ? origin : "", > - origin ? "/" : "", > - shortname ? shortname : remote); > + actual_origin ? actual_origin : "", > + actual_origin ? "/" : "", > + shortname ? shortname : actual_remote); > > return -1; > } > diff --git a/branch.h b/branch.h > index df0be61506..7d7990dda7 100644 > --- a/branch.h > +++ b/branch.h > @@ -14,6 +14,8 @@ enum branch_track { > }; > > extern enum branch_track git_branch_track; > +extern const char* git_branch_remote; > +extern const char* git_branch_merge; > > /* Functions for acting on the information about branches. */ > > diff --git a/config.c b/config.c > index f33abeab85..a46c5a43a1 100644 > --- a/config.c > +++ b/config.c > @@ -1599,6 +1599,16 @@ static int git_default_branch_config(const char *var, const char *value) > return error(_("malformed value for %s"), var); > return 0; > } > + if (!strcmp(var, "branch.defaultremote")) { > + if (!value) > + return config_error_nonbool(var); > + return git_config_string(&git_branch_remote, var, value); > + } > + if (!strcmp(var, "branch.defaultmerge")) { > + if (!value) > + return config_error_nonbool(var); > + return git_config_string(&git_branch_merge, var, value); > + } > > /* Add other config variables here and to Documentation/config.txt. */ > return 0; > diff --git a/environment.c b/environment.c > index 2f27008424..d550deabbd 100644 > --- a/environment.c > +++ b/environment.c > @@ -60,6 +60,8 @@ int global_conv_flags_eol = CONV_EOL_RNDTRP_WARN; > char *check_roundtrip_encoding = "SHIFT-JIS"; > unsigned whitespace_rule_cfg = WS_DEFAULT_RULE; > enum branch_track git_branch_track = BRANCH_TRACK_REMOTE; > +const char* git_branch_remote = NULL; > +const char* git_branch_merge = NULL; Can the new settings be implemented without adding more global variables ? I think we are trying to move away from these. Apart from that the code looks OK to me. > enum rebase_setup_type autorebase = AUTOREBASE_NEVER; > enum push_default_type push_default = PUSH_DEFAULT_UNSPECIFIED; > #ifndef OBJECT_CREATION_MODE > diff --git a/t/t3200-branch.sh b/t/t3200-branch.sh > index cc4b10236e..82137e8451 100755 > --- a/t/t3200-branch.sh > +++ b/t/t3200-branch.sh > @@ -797,6 +797,45 @@ test_expect_success 'test tracking setup via --track but deeper' ' > test "$(git config branch.my7.merge)" = refs/heads/o/o > ' > > +test_expect_success 'test tracking setup via branch.default* and --track' ' > + git config branch.autosetupmerge always && You can use 'test_config branch.autosetupmerge always' so that the config is only active for the duration of the 'test_expect_success' block (see t/test-lib-functions.sh). > + git config branch.defaultremote local && > + git config branch.defaultmerge main && > + git config remote.local.url . && > + git config remote.local.fetch refs/heads/*:refs/remotes/local/* && > + (git show-ref -q refs/remotes/local/main || git fetch local) && > + git branch --track other/foo my2 && > + git config branch.autosetupmerge false && > + test "$(git config branch.my2.remote)" = other && Here and for the following line you can use 'test_cmp_config'. > + test "$(git config branch.my2.merge)" = refs/heads/foo > +' This tests checks that an explicit '--track' argument overrides the new configs. I would name it something like "'--track overrides 'branch.default{merge,remote}'" or something like this. I would also add another test before this one that just checks that the new settings by themselves work as expected; i.e. a simple 'git checkout -b' and verifying that the tracking info is correctly configured according to 'branch.default{merge,remote}'. > + > +test_expect_success 'test tracking setup via branch.default* and --no-track' ' > + git config branch.autosetupmerge always && > + git config branch.defaultremote local && > + git config branch.defaultmerge main && > + git config remote.local.url . && > + git config remote.local.fetch refs/heads/*:refs/remotes/local/* && > + (git show-ref -q refs/remotes/local/main || git fetch local) && > + git branch --no-track my2 && > + git config branch.autosetupmerge false && > + ! test "$(git config branch.my2.remote)" = local && > + ! test "$(git config branch.my2.merge)" = refs/heads/main Here you expect the configs to be absent, so for more clarity you could do git config branch.my2.remote >remote && test_must_be_empty remote and the same for merge. > +' > + > +test_expect_success 'test tracking setup via branch.default*' ' > + git config branch.autosetupmerge always && > + git config branch.defaultremote local && > + git config branch.defaultmerge main && > + git config remote.local.url . && > + git config remote.local.fetch refs/heads/*:refs/remotes/local/* && > + (git show-ref -q refs/remotes/local/main || git fetch local) && > + git branch my2 && > + git config branch.autosetupmerge false && > + test "$(git config branch.my2.remote)" = local && > + test "$(git config branch.my2.merge)" = refs/heads/main > +' > + > test_expect_success 'test deleting branch deletes branch config' ' > git branch -d my7 && > test -z "$(git config branch.my7.remote)" && > Oh, so here is the 'just test the new settings' test. Let's move that one to be before the two others. Another test that could be added is one that does not change 'branch.autosetupmerge' but sets the new configs, and checks that the tracking info is *not* set. Cheers, Philippe. ^ permalink raw reply [flat|nested] 20+ messages in thread
* Re: [PATCH 1/1] config: support a default remote tracking setup upon branch creation 2021-07-30 13:35 ` Philippe Blain @ 2021-07-30 14:07 ` Ben Boeckel 2021-07-30 17:32 ` Junio C Hamano 1 sibling, 0 replies; 20+ messages in thread From: Ben Boeckel @ 2021-07-30 14:07 UTC (permalink / raw) To: Philippe Blain Cc: git, Martin Ågren, Junio C Hamano, Jeff King, Jeff Hostetler, Nguyễn Thái Ngọc Duy, Taylor Blau On Fri, Jul 30, 2021 at 09:35:49 -0400, Philippe Blain wrote: > Le 2021-07-28 à 22:01, Ben Boeckel a écrit : > > $ git config remote.pushDefault myfork # push to `myfork` by default > > $ git config push.default simple # the default > > $ git config branch.autoSetupMerge always # always setup tracking > > OK, so if I understand correctly this exisiting setting has to be changed > to 'always' for the new settings you are adding to take effect, right ? > I think this does make sense, reading the description of 'branch.autoSetupMerge', > but maybe it should be spelled explicitely in the text of the commit message, > and not just mentioned here in this terminal session excerpt. Good point. I'll add it. > > $ git config branch.defaultRemote origin # track from `origin` > > $ git config branch.defaultMerge main # track the `main` branch > > Small nit: maybe I would invert these two, so it can read: > > $ git config branch.defaultMerge main # track the `main` branch ... > $ git config branch.defaultRemote origin # ... from `origin` Agreed. > > Additionally, with the `extensions.worktreeConfig` setting, when a > > separate work tree which is used for changes against a different branch > > (e.g., a branch tracking a prior release), the `branch.defaultMerge` > > setting may be changed independently, e.g., to `maint`. > > This last paragraph is not explicitely needed, as nothing relating to > 'extensions.worktreeConfig' is changed here right ? It's just the normal > way that this setting works. Yes. I'll mention more explicitly that this is the reason for preserving split settings (rather than a single `branch.defaultTrack = origin/main` setting that I had thought about until I saw the `--worktree` flag to `git config` and was intrigued). > > +branch.defaultRemote:: > > + Tells 'git branch', 'git switch' and 'git checkout' to set up new branches > > + so that it will track a branch on the specified remote. This can be > > + used, in conjunction with `branch.defaultMerge`, in projects where > > + branches tend to target a single branch. This will be used to > > + initialize the "branch.<name>.remote" setting. > > + > > +branch.defaultMerge:: > > + Tells 'git branch', 'git switch' and 'git checkout' to set up new branches > > + so that it will track a branch with this name. This can be used, in > > + conjunction with `branch.defaultRemote` in projects where branches tend > > + to target a single branch. This will be used to initialize the > > + "branch.<name>.merge" setting. > > For the two setting above, if 'branch.autoSetupMerge' must be set to 'always' for > the settings to work, I think it should be explicitely mentioned. Will update. > > diff --git a/environment.c b/environment.c > > index 2f27008424..d550deabbd 100644 > > --- a/environment.c > > +++ b/environment.c > > @@ -60,6 +60,8 @@ int global_conv_flags_eol = CONV_EOL_RNDTRP_WARN; > > char *check_roundtrip_encoding = "SHIFT-JIS"; > > unsigned whitespace_rule_cfg = WS_DEFAULT_RULE; > > enum branch_track git_branch_track = BRANCH_TRACK_REMOTE; > > +const char* git_branch_remote = NULL; > > +const char* git_branch_merge = NULL; > > Can the new settings be implemented without adding more global variables ? > I think we are trying to move away from these. Apart from that the code > looks OK to me. I'm all for that, but didn't see guidance on where such things should be stored. There's not a "context" object passed around, but I guess stuffing it into `repository` is fine? This also gives a nice place to free them rather than just leaking them too. Alas, after some cursory investigation, `config.c@@git_default_branch_config` does not have access to "the repository" and the `cb` in `git_default_config` seems to be ~always be `NULL`. So maybe that will have to wait for further refactoring of the configuration tracking logic? > > diff --git a/t/t3200-branch.sh b/t/t3200-branch.sh > > index cc4b10236e..82137e8451 100755 > > --- a/t/t3200-branch.sh > > +++ b/t/t3200-branch.sh > > @@ -797,6 +797,45 @@ test_expect_success 'test tracking setup via --track but deeper' ' > > test "$(git config branch.my7.merge)" = refs/heads/o/o > > ' > > > > +test_expect_success 'test tracking setup via branch.default* and --track' ' > > + git config branch.autosetupmerge always && > > You can use 'test_config branch.autosetupmerge always' so that the > config is only active for the duration of the 'test_expect_success' block > (see t/test-lib-functions.sh). Nifty. > > + git config branch.defaultremote local && > > + git config branch.defaultmerge main && > > + git config remote.local.url . && > > + git config remote.local.fetch refs/heads/*:refs/remotes/local/* && > > + (git show-ref -q refs/remotes/local/main || git fetch local) && > > + git branch --track other/foo my2 && > > + git config branch.autosetupmerge false && > > + test "$(git config branch.my2.remote)" = other && > > Here and for the following line you can use 'test_cmp_config'. > > > + test "$(git config branch.my2.merge)" = refs/heads/foo > > +' > > This tests checks that an explicit '--track' argument overrides the new configs. > I would name it something like "'--track overrides 'branch.default{merge,remote}'" > or something like this. I would also add another test before this one that just > checks that the new settings by themselves work as expected; > i.e. a simple 'git checkout -b' and verifying that the > tracking info is correctly configured according to 'branch.default{merge,remote}'. > > > + > > +test_expect_success 'test tracking setup via branch.default* and --no-track' ' > > + git config branch.autosetupmerge always && > > + git config branch.defaultremote local && > > + git config branch.defaultmerge main && > > + git config remote.local.url . && > > + git config remote.local.fetch refs/heads/*:refs/remotes/local/* && > > + (git show-ref -q refs/remotes/local/main || git fetch local) && > > + git branch --no-track my2 && > > + git config branch.autosetupmerge false && > > + ! test "$(git config branch.my2.remote)" = local && > > + ! test "$(git config branch.my2.merge)" = refs/heads/main > > Here you expect the configs to be absent, so for more clarity you could > do > > git config branch.my2.remote >remote && > test_must_be_empty remote > > and the same for merge. > > > +' > > + > > +test_expect_success 'test tracking setup via branch.default*' ' > > + git config branch.autosetupmerge always && > > + git config branch.defaultremote local && > > + git config branch.defaultmerge main && > > + git config remote.local.url . && > > + git config remote.local.fetch refs/heads/*:refs/remotes/local/* && > > + (git show-ref -q refs/remotes/local/main || git fetch local) && > > + git branch my2 && > > + git config branch.autosetupmerge false && > > + test "$(git config branch.my2.remote)" = local && > > + test "$(git config branch.my2.merge)" = refs/heads/main > > +' > > + > > test_expect_success 'test deleting branch deletes branch config' ' > > git branch -d my7 && > > test -z "$(git config branch.my7.remote)" && > > > > Oh, so here is the 'just test the new settings' test. Let's move that > one to be before the two others. > > Another test that could be added is one that does not change > 'branch.autosetupmerge' but sets the new configs, and checks that the > tracking info is *not* set. I'll make the test suite updates as well. Thanks for the review, --Ben ^ permalink raw reply [flat|nested] 20+ messages in thread
* Re: [PATCH 1/1] config: support a default remote tracking setup upon branch creation 2021-07-30 13:35 ` Philippe Blain 2021-07-30 14:07 ` Ben Boeckel @ 2021-07-30 17:32 ` Junio C Hamano 1 sibling, 0 replies; 20+ messages in thread From: Junio C Hamano @ 2021-07-30 17:32 UTC (permalink / raw) To: Philippe Blain Cc: Ben Boeckel, git, Martin Ågren, Jeff King, Jeff Hostetler, Nguyễn Thái Ngọc Duy, Taylor Blau Philippe Blain <levraiphilippeblain@gmail.com> writes: >> diff --git a/environment.c b/environment.c >> index 2f27008424..d550deabbd 100644 >> --- a/environment.c >> +++ b/environment.c >> @@ -60,6 +60,8 @@ int global_conv_flags_eol = CONV_EOL_RNDTRP_WARN; >> char *check_roundtrip_encoding = "SHIFT-JIS"; >> unsigned whitespace_rule_cfg = WS_DEFAULT_RULE; >> enum branch_track git_branch_track = BRANCH_TRACK_REMOTE; >> +const char* git_branch_remote = NULL; >> +const char* git_branch_merge = NULL; Style: (1) asterisk sticks to the identifier, not type, in our codebase. (2) do not initialize globals and statics to 0 or NULL. > Can the new settings be implemented without adding more global variables ? This is worth considering in the longer term. For things like these new configuration items and existign git_branch_track, we already have reasonably made abstraction that branch.c is where interesting actions happen (like setting up remote tracking, etc), so there is no reason for them to be in environment.c or *.h to be visible to anywhere outside branch.c file. I wonder if it is a matter of moving git_default_branch_config() to branch.c from config.c and make it global, while moving these global variables also to branch.c and make them file-local? I am still unsure without the expected use-case well documented, if it is clear enough for users to learn how and when these new configurations should be used (as opposed to following the traditional triangular workflow) with only the documentation updates in this patch, but at least I can trust others like you to give input to polish this into a reasonable shape. Thanks for a review (and thanks for starting the effort, Ben). ^ permalink raw reply [flat|nested] 20+ messages in thread
* Re: [PATCH 1/1] config: support a default remote tracking setup upon branch creation 2021-07-29 2:01 ` [PATCH 1/1] config: support a default remote tracking setup upon branch creation Ben Boeckel 2021-07-30 13:35 ` Philippe Blain @ 2021-08-02 13:02 ` Ævar Arnfjörð Bjarmason 2021-08-02 13:16 ` Ben Boeckel 1 sibling, 1 reply; 20+ messages in thread From: Ævar Arnfjörð Bjarmason @ 2021-08-02 13:02 UTC (permalink / raw) To: Ben Boeckel Cc: git, Martin Ågren, Junio C Hamano, Jeff King, Jeff Hostetler, Nguyễn Thái Ngọc Duy, Taylor Blau On Wed, Jul 28 2021, Ben Boeckel wrote: > The `branch.autoSetupMerge` works well today for setting up tracking a > local branch, but there is no easy mechanism to automatically set up a > remote tracking situation for workflows which use a single set of > branches for integration without specifying `--track` to every branch > creation command or branching directly from the remote ref. This patch > adds the following configuration values: > > - `branch.defaultRemote`: initializes `branch.<name>.remote` if not > otherwise given; > - `branch.defaultMerge`: initializes `branch.<name>.merge` if not > otherwise given. Not a new issue per-se, but what if you've got a branch called defaultRemote? It seems to me that any new branch.<name>.* config closes the door for a <name> we squat on. Given that we have checkout.defaultRemote and this also affects switch/checkout it seems better to continue in the checkout.* namespace even if it wasn't for that, but given the config squatting issue especially so.... For what it's worth I usually use the checkout.defaultRemote option (which I added) and: git checkout master && git branch -m master <topic-name> See 8d7b558baeb (checkout & worktree: introduce checkout.defaultRemote, 2018-06-05). It seems to me from that patch diff that you could modify some docs / tests for this, no? E.g. how it interacts with git-worktree. I like this direction, but just have a concern that this is a place where we need to consider all the UX in the area overall, and that any options/config don't overtly interact in a bad way. ^ permalink raw reply [flat|nested] 20+ messages in thread
* Re: [PATCH 1/1] config: support a default remote tracking setup upon branch creation 2021-08-02 13:02 ` Ævar Arnfjörð Bjarmason @ 2021-08-02 13:16 ` Ben Boeckel 2021-08-02 15:20 ` Ævar Arnfjörð Bjarmason 0 siblings, 1 reply; 20+ messages in thread From: Ben Boeckel @ 2021-08-02 13:16 UTC (permalink / raw) To: Ævar Arnfjörð Bjarmason Cc: git, Martin Ågren, Junio C Hamano, Jeff King, Jeff Hostetler, Nguyễn Thái Ngọc Duy, Taylor Blau On Mon, Aug 02, 2021 at 15:02:41 +0200, Ævar Arnfjörð Bjarmason wrote: > On Wed, Jul 28 2021, Ben Boeckel wrote: > > The `branch.autoSetupMerge` works well today for setting up tracking a > > local branch, but there is no easy mechanism to automatically set up a > > remote tracking situation for workflows which use a single set of > > branches for integration without specifying `--track` to every branch > > creation command or branching directly from the remote ref. This patch > > adds the following configuration values: > > > > - `branch.defaultRemote`: initializes `branch.<name>.remote` if not > > otherwise given; > > - `branch.defaultMerge`: initializes `branch.<name>.merge` if not > > otherwise given. > > Not a new issue per-se, but what if you've got a branch called > defaultRemote? It seems to me that any new branch.<name>.* config closes > the door for a <name> we squat on. It doesn't seem that shadowing is actually a thing: % git init Initialized empty Git repository in …/git-shadow/.git/ % git config foo.bar true % git config foo.bar.baz true % git config --get foo.bar true % git config --get foo.bar.baz true > Given that we have checkout.defaultRemote and this also affects > switch/checkout it seems better to continue in the checkout.* namespace > even if it wasn't for that, but given the config squatting issue > especially so.... > > For what it's worth I usually use the checkout.defaultRemote option > (which I added) and: > > git checkout master && > git branch -m master <topic-name> > > See 8d7b558baeb (checkout & worktree: introduce checkout.defaultRemote, > 2018-06-05). It seems to me from that patch diff that you could modify > some docs / tests for this, no? E.g. how it interacts with git-worktree. I think it would be weird for `checkout.*` to affect `git branch` which does no checkout at all. I want it to set up for simple branch creation as well, so this would be a hole in my use case. > I like this direction, but just have a concern that this is a place > where we need to consider all the UX in the area overall, and that any > options/config don't overtly interact in a bad way. I'll have to look at adding test cases as to how it interacts with `checkout.defaultRemote`. Thanks, --Ben ^ permalink raw reply [flat|nested] 20+ messages in thread
* Re: [PATCH 1/1] config: support a default remote tracking setup upon branch creation 2021-08-02 13:16 ` Ben Boeckel @ 2021-08-02 15:20 ` Ævar Arnfjörð Bjarmason 0 siblings, 0 replies; 20+ messages in thread From: Ævar Arnfjörð Bjarmason @ 2021-08-02 15:20 UTC (permalink / raw) To: Ben Boeckel Cc: git, Martin Ågren, Junio C Hamano, Jeff King, Jeff Hostetler, Nguyễn Thái Ngọc Duy, Taylor Blau On Mon, Aug 02 2021, Ben Boeckel wrote: > On Mon, Aug 02, 2021 at 15:02:41 +0200, Ævar Arnfjörð Bjarmason wrote: >> On Wed, Jul 28 2021, Ben Boeckel wrote: >> > The `branch.autoSetupMerge` works well today for setting up tracking a >> > local branch, but there is no easy mechanism to automatically set up a >> > remote tracking situation for workflows which use a single set of >> > branches for integration without specifying `--track` to every branch >> > creation command or branching directly from the remote ref. This patch >> > adds the following configuration values: >> > >> > - `branch.defaultRemote`: initializes `branch.<name>.remote` if not >> > otherwise given; >> > - `branch.defaultMerge`: initializes `branch.<name>.merge` if not >> > otherwise given. >> >> Not a new issue per-se, but what if you've got a branch called >> defaultRemote? It seems to me that any new branch.<name>.* config closes >> the door for a <name> we squat on. > > It doesn't seem that shadowing is actually a thing: > > % git init > Initialized empty Git repository in …/git-shadow/.git/ > % git config foo.bar true > % git config foo.bar.baz true > % git config --get foo.bar > true > % git config --get foo.bar.baz > true You're right, I was misrecalling (or mis-imagining) some edge case there that doesn't exist. I also tested setting branch.defaultRemote=true and moving a branch.defaultRemote.* branch with "git branch -m", but it also does the right thing. Nevermind. >> Given that we have checkout.defaultRemote and this also affects >> switch/checkout it seems better to continue in the checkout.* namespace >> even if it wasn't for that, but given the config squatting issue >> especially so.... >> >> For what it's worth I usually use the checkout.defaultRemote option >> (which I added) and: >> >> git checkout master && >> git branch -m master <topic-name> >> >> See 8d7b558baeb (checkout & worktree: introduce checkout.defaultRemote, >> 2018-06-05). It seems to me from that patch diff that you could modify >> some docs / tests for this, no? E.g. how it interacts with git-worktree. > > I think it would be weird for `checkout.*` to affect `git branch` which > does no checkout at all. I want it to set up for simple branch creation > as well, so this would be a hole in my use case. *nod*, although your approach has the opposite problem of making branch creation with "checkout" and "switch" (and presumably "worktree") impacted by "branch.*' config. In a way that's more sensible, in that we can imagine those commands calling "git branch" under the hood (which msotly doesn't actually happen, except I think in the worktree case, but it's the same underlying APIs). .. >> I like this direction, but just have a concern that this is a place >> where we need to consider all the UX in the area overall, and that any >> options/config don't overtly interact in a bad way. > > I'll have to look at adding test cases as to how it interacts with > `checkout.defaultRemote`. > > Thanks, ....right, none of that mess is a showstopper, I'm just prodding you to look at if any of those edge cases are made better/worse by these additions. Thanks! ^ permalink raw reply [flat|nested] 20+ messages in thread
* Re: [PATCH 0/1] Improve automatic setup of tracking for new branches 2021-07-29 2:01 ` [PATCH 0/1] Improve automatic setup of tracking for new branches Ben Boeckel 2021-07-29 2:01 ` [PATCH 1/1] config: support a default remote tracking setup upon branch creation Ben Boeckel @ 2021-07-30 1:04 ` Junio C Hamano 2021-07-30 1:33 ` Ben Boeckel 2021-07-30 13:35 ` Philippe Blain 2 siblings, 1 reply; 20+ messages in thread From: Junio C Hamano @ 2021-07-30 1:04 UTC (permalink / raw) To: Ben Boeckel Cc: git, Martin Ågren, Jeff King, Jeff Hostetler, Nguyễn Thái Ngọc Duy, Taylor Blau Ben Boeckel <mathstuf@gmail.com> writes: > I searched the docs (including `Documentation/gitworkflows.txt`), but > didn't see anywhere to describe the fork-based workflow common on forges > (such as GitHub and GitLab) where this felt "at home". Before you came, many users have used Git with these forges, so it might be just the matter of correct terminology to use to find what to read about. Does the keyword "triangular workflow" find what may help your way of working? ^ permalink raw reply [flat|nested] 20+ messages in thread
* Re: [PATCH 0/1] Improve automatic setup of tracking for new branches 2021-07-30 1:04 ` [PATCH 0/1] Improve automatic setup of tracking for new branches Junio C Hamano @ 2021-07-30 1:33 ` Ben Boeckel 0 siblings, 0 replies; 20+ messages in thread From: Ben Boeckel @ 2021-07-30 1:33 UTC (permalink / raw) To: Junio C Hamano Cc: git, Martin Ågren, Jeff King, Jeff Hostetler, Nguyễn Thái Ngọc Duy, Taylor Blau On Thu, Jul 29, 2021 at 18:04:08 -0700, Junio C Hamano wrote: > Ben Boeckel <mathstuf@gmail.com> writes: > > I searched the docs (including `Documentation/gitworkflows.txt`), but > > didn't see anywhere to describe the fork-based workflow common on forges > > (such as GitHub and GitLab) where this felt "at home". > > Before you came, many users have used Git with these forges, so it > might be just the matter of correct terminology to use to find what > to read about. Does the keyword "triangular workflow" find what may > help your way of working? Oh, I'm familiar with it, I just didn't see a place which discussed it in the docs with respect to it as a *workflow*. The existing references are about the `@{push}`…selector (is there a name for these `@{}` suffixes as a collection?) and a note under "Future Work" in partial-clone. Neither of these seemed like relevant places to discuss how to set up remotes and tracking branches for the triangular workflow. I would expect it to be discussed in `gitworkflows.txt`, but I think such an addition is worthy of its own patchset rather than tacking it onto this. Thanks, --Ben ^ permalink raw reply [flat|nested] 20+ messages in thread
* Re: [PATCH 0/1] Improve automatic setup of tracking for new branches 2021-07-29 2:01 ` [PATCH 0/1] Improve automatic setup of tracking for new branches Ben Boeckel 2021-07-29 2:01 ` [PATCH 1/1] config: support a default remote tracking setup upon branch creation Ben Boeckel 2021-07-30 1:04 ` [PATCH 0/1] Improve automatic setup of tracking for new branches Junio C Hamano @ 2021-07-30 13:35 ` Philippe Blain 2021-07-30 13:57 ` Ben Boeckel 2 siblings, 1 reply; 20+ messages in thread From: Philippe Blain @ 2021-07-30 13:35 UTC (permalink / raw) To: Ben Boeckel, git Cc: Martin Ågren, Junio C Hamano, Jeff King, Jeff Hostetler, Nguyễn Thái Ngọc Duy, Taylor Blau Hi Ben, Le 2021-07-28 à 22:01, Ben Boeckel a écrit : > Hi, > > I recently discovered some of the nicer behaviors of the `--track` > option for local branches (courtesy of `vim-fugitive`'s status > rendering). However, after a few weeks of slowly working `-t > origin/master` into my workflow, I figured that Git could help me out > here. > > This patch adds two new configuration variables to initialize the > `branch.<name>.*` settings for tracking remote branches. > > I searched the docs (including `Documentation/gitworkflows.txt`), but > didn't see anywhere to describe the fork-based workflow common on forges > (such as GitHub and GitLab) where this felt "at home". I suppose I could > describe the workflow I'm used to, but I think I'll hammer that out as > blog posts before submitting it as documentation. I'm in favor of a change like the one you propose, thanks for working on this! The 'triangular' aka 'forking' workflow is, as you discovered, only explicitely mentioned in the description of '@{push}' [1]. 'gitworkflows(5)' is mostly about how the workflow used by the Git projet itself (from the description): This document attempts to write down and motivate some of the workflow elements used for git.git itself. > > I suspect there are more tests that should be added. > > Thanks, > > --Ben > > --- > v1 -> v2: > - removed `branch.defaultPushRemote` (`remote.pushDefault` covers this > case already) > - improved the commit message with more background and an expanation > of the intended uses > Small nit: usually when sending a second version of a patch, you would use the '-v2' argument to 'git format-patch' so that the patch and cover letter is prefixed [PATCH v2]. [1] https://git-scm.com/docs/gitrevisions#Documentation/gitrevisions.txt-emltbranchnamegtpushemegemmasterpushemempushem ^ permalink raw reply [flat|nested] 20+ messages in thread
* Re: [PATCH 0/1] Improve automatic setup of tracking for new branches 2021-07-30 13:35 ` Philippe Blain @ 2021-07-30 13:57 ` Ben Boeckel 2021-07-30 16:01 ` Philippe Blain 0 siblings, 1 reply; 20+ messages in thread From: Ben Boeckel @ 2021-07-30 13:57 UTC (permalink / raw) To: Philippe Blain Cc: git, Martin Ågren, Junio C Hamano, Jeff King, Jeff Hostetler, Nguyễn Thái Ngọc Duy, Taylor Blau On Fri, Jul 30, 2021 at 09:35:39 -0400, Philippe Blain wrote: > I'm in favor of a change like the one you propose, thanks for working on this! Thanks :) . Nice to know it'll be of use for others. > The 'triangular' aka 'forking' workflow is, as you discovered, only explicitely > mentioned in the description of '@{push}' [1]. 'gitworkflows(5)' is mostly about > how the workflow used by the Git projet itself (from the description): > > This document attempts to write down and motivate some of the workflow > elements used for git.git itself. Ah, I missed that. > Le 2021-07-28 à 22:01, Ben Boeckel a écrit : > > v1 -> v2: > > - removed `branch.defaultPushRemote` (`remote.pushDefault` covers this > > case already) > > - improved the commit message with more background and an expanation > > of the intended uses > > > > Small nit: usually when sending a second version of a patch, you would use > the '-v2' argument to 'git format-patch' so that the patch and cover letter > is prefixed [PATCH v2]. Yes, I realized that I had forgotten the `--reroll-count=` argument when making this patch (I suppose a way to store the Cc list for a topic somewhere would be nice so I didn't lean so heavily on shell history would help this). FWIW, my main gripe with the email-based workflow is the lack of coordinated metadata (LWN has numerous comments by me about my views if you're curious, but I should really formalize them into blog posts). But when in Rome :) . Thanks, --Ben ^ permalink raw reply [flat|nested] 20+ messages in thread
* Re: [PATCH 0/1] Improve automatic setup of tracking for new branches 2021-07-30 13:57 ` Ben Boeckel @ 2021-07-30 16:01 ` Philippe Blain 2021-07-30 17:45 ` Ben Boeckel 2021-08-02 21:17 ` Johannes Schindelin 0 siblings, 2 replies; 20+ messages in thread From: Philippe Blain @ 2021-07-30 16:01 UTC (permalink / raw) To: Ben Boeckel Cc: git, Martin Ågren, Junio C Hamano, Jeff King, Jeff Hostetler, Nguyễn Thái Ngọc Duy, Taylor Blau Hi again, Le 2021-07-30 à 09:57, Ben Boeckel a écrit : > On Fri, Jul 30, 2021 at 09:35:39 -0400, Philippe Blain wrote: >> Le 2021-07-28 à 22:01, Ben Boeckel a écrit : >> >> Small nit: usually when sending a second version of a patch, you would use >> the '-v2' argument to 'git format-patch' so that the patch and cover letter >> is prefixed [PATCH v2]. > > Yes, I realized that I had forgotten the `--reroll-count=` argument when > making this patch (I suppose a way to store the Cc list for a topic > somewhere would be nice so I didn't lean so heavily on shell history > would help this). There is 'format.cc', but it's not branch-specific, so you would have to use one worktree per branch with extension.worktreeConfig... > > FWIW, my main gripe with the email-based workflow is the lack of > coordinated metadata (LWN has numerous comments by me about my views if > you're curious, but I should really formalize them into blog posts). But > when in Rome :) . > I agree. I almost always use Gitgitgadget [1], which keeps track of the CC list for me, of updating the re-roll count, of adding the in-reply-to header such that subsequent versions of the series are sent as a response to the cover letter of the previous version, generating a range-diff against the previous version, commenting on the PR when the series is mentioned in "What's cooking", etc. Some things it does not support are: sending a patch as a response to some random mail on the list, which is sometimes useful, reading the commit notes to generate in-patch commentaries [2], customizing the diff generated by 'format-patch'. Other things are listed at [3]. Recently I've even been using only the terminal with Gitgitgadget: I use 'git branch --edit-description' to write my cover letter, and then use the 'gh' GitHub CLI to open my PR: $ git config --get-regexp alias.desc* alias.desc-title !git config branch.$(git branch --show-current).description | head -1 alias.desc-body !git config branch.$(git branch --show-current).description | tail -n+3 $ gh pr create --title "$(git desc-title)" --body "$(git desc-body)" --head phil-blain:$(git branch --show-current) Cheers, Philippe. [1] https://gitgitgadget.github.io/ [2] https://github.com/gitgitgadget/gitgitgadget/issues/173 [3] https://github.com/gitgitgadget/gitgitgadget/issues ^ permalink raw reply [flat|nested] 20+ messages in thread
* Re: [PATCH 0/1] Improve automatic setup of tracking for new branches 2021-07-30 16:01 ` Philippe Blain @ 2021-07-30 17:45 ` Ben Boeckel 2021-08-02 21:17 ` Johannes Schindelin 1 sibling, 0 replies; 20+ messages in thread From: Ben Boeckel @ 2021-07-30 17:45 UTC (permalink / raw) To: Philippe Blain Cc: git, Martin Ågren, Junio C Hamano, Jeff King, Jeff Hostetler, Nguyễn Thái Ngọc Duy, Taylor Blau On Fri, Jul 30, 2021 at 12:01:36 -0400, Philippe Blain wrote: > Hi again, > > Le 2021-07-30 à 09:57, Ben Boeckel a écrit : > > On Fri, Jul 30, 2021 at 09:35:39 -0400, Philippe Blain wrote: > >> Le 2021-07-28 à 22:01, Ben Boeckel a écrit : > >> > >> Small nit: usually when sending a second version of a patch, you would use > >> the '-v2' argument to 'git format-patch' so that the patch and cover letter > >> is prefixed [PATCH v2]. > > > > Yes, I realized that I had forgotten the `--reroll-count=` argument when > > making this patch (I suppose a way to store the Cc list for a topic > > somewhere would be nice so I didn't lean so heavily on shell history > > would help this). > > There is 'format.cc', but it's not branch-specific, so you would have to > use one worktree per branch with extension.worktreeConfig... I suppose a `branch.<name>.cc` configuration might be in order :) . Though at that point, `branch.<name>.rerollCount` and `branch.<name>.sendEmailTo` also make sense… More brainstorming below too. > > FWIW, my main gripe with the email-based workflow is the lack of > > coordinated metadata (LWN has numerous comments by me about my views if > > you're curious, but I should really formalize them into blog posts). But > > when in Rome :) . > > I agree. I almost always use Gitgitgadget [1], which keeps track of the CC list for > me, of updating the re-roll count, of adding the in-reply-to header such that subsequent > versions of the series are sent as a response to the cover letter of the previous version, > generating a range-diff against the previous version, > commenting on the PR when the series is mentioned in "What's cooking", etc. That's nifty. I guess since I started here, things are a bit messy for this patch though? > Some things it does not support are: sending a patch as a response to some random > mail on the list, which is sometimes useful, reading the commit notes to generate > in-patch commentaries [2], customizing the diff generated by 'format-patch'. > Other things are listed at [3]. > > Recently I've even been using only the terminal with Gitgitgadget: > I use 'git branch --edit-description' > to write my cover letter, and then use the 'gh' GitHub CLI to open my PR: I think having `branch.<name>.coverLetter` and/or `branch.coverLetterDirectory` (defaulting to reading `<name>` or `<name>.txt`) to store the cover letter contents would be nice. I know I've blown away enough `0000-*.patch` edits with an ill-timed `git format-patch` before… Forges could use this to hook in with their push options[1] through their wrappers or other tools. --Ben [1]https://docs.gitlab.com/ee/user/project/push_options.html#push-options-for-merge-requests ^ permalink raw reply [flat|nested] 20+ messages in thread
* Re: [PATCH 0/1] Improve automatic setup of tracking for new branches 2021-07-30 16:01 ` Philippe Blain 2021-07-30 17:45 ` Ben Boeckel @ 2021-08-02 21:17 ` Johannes Schindelin 1 sibling, 0 replies; 20+ messages in thread From: Johannes Schindelin @ 2021-08-02 21:17 UTC (permalink / raw) To: Philippe Blain Cc: Ben Boeckel, git, Martin Ågren, Junio C Hamano, Jeff King, Jeff Hostetler, Nguyễn Thái Ngọc Duy, Taylor Blau Hi Philippe, On Fri, 30 Jul 2021, Philippe Blain wrote: > Recently I've even been using only the terminal with Gitgitgadget: > I use 'git branch --edit-description' > to write my cover letter, and then use the 'gh' GitHub CLI to open my PR: > > $ git config --get-regexp alias.desc* > alias.desc-title !git config branch.$(git branch --show-current).description | > head -1 > alias.desc-body !git config branch.$(git branch --show-current).description | > tail -n+3 > $ gh pr create --title "$(git desc-title)" --body "$(git desc-body)" --head > phil-blain:$(git branch --show-current) I bet you could streamline this even further by defining a `gh` alias: https://cli.github.com/manual/gh_alias_set Ciao, Dscho ^ permalink raw reply [flat|nested] 20+ messages in thread
end of thread, other threads:[~2021-08-02 21:17 UTC | newest] Thread overview: 20+ messages (download: mbox.gz / follow: Atom feed) -- links below jump to the message on this page -- 2021-07-28 13:50 [PATCH 0/1] Improve automatic setup of tracking for new branches Ben Boeckel 2021-07-28 13:50 ` [PATCH 1/1] config: support setting up a remote tracking branch upon creation Ben Boeckel 2021-07-28 17:48 ` Junio C Hamano 2021-07-28 18:26 ` Ben Boeckel 2021-07-28 18:39 ` Junio C Hamano 2021-07-29 2:01 ` [PATCH 0/1] Improve automatic setup of tracking for new branches Ben Boeckel 2021-07-29 2:01 ` [PATCH 1/1] config: support a default remote tracking setup upon branch creation Ben Boeckel 2021-07-30 13:35 ` Philippe Blain 2021-07-30 14:07 ` Ben Boeckel 2021-07-30 17:32 ` Junio C Hamano 2021-08-02 13:02 ` Ævar Arnfjörð Bjarmason 2021-08-02 13:16 ` Ben Boeckel 2021-08-02 15:20 ` Ævar Arnfjörð Bjarmason 2021-07-30 1:04 ` [PATCH 0/1] Improve automatic setup of tracking for new branches Junio C Hamano 2021-07-30 1:33 ` Ben Boeckel 2021-07-30 13:35 ` Philippe Blain 2021-07-30 13:57 ` Ben Boeckel 2021-07-30 16:01 ` Philippe Blain 2021-07-30 17:45 ` Ben Boeckel 2021-08-02 21:17 ` Johannes Schindelin
This is an external index of several public inboxes, see mirroring instructions on how to clone and mirror all data and code used by this external index.