worktree: detect from secondary worktree if main worktree is bare

Setup:
1. Have a bare repo with core.bare = true in config.worktree
2. Create a new worktree

Behavior:
From the secondary worktree the main worktree appears as non-bare.

Expected:
From the secondary worktree the main worktree should appear as bare.

Why current behavior is not good?
If the main worktree is detected as not bare it doesn't allow checking out the branch of the main worktree. There are possibly other problems associated with that behavior.

Why is it happening?
While we're inside the secondary worktree we don't initialize the main worktree's repository with its configuration.

How is it fixed?
Load actual configs of the main worktree. Also, skip the config loading step if we're already inside the current worktree because in that case we rely on is_bare_repository() to return the correct result.

Other solutions considered:
Alternatively, instead of incorrectly always using the_repository as the main worktree's repository, we can detect and load the actual repository of the main worktree and then use that repository's is_bare value extracted from correct configs. However, this approach is a bit riskier and could also affect performance. Since we had the assignment worktree->repo = the_repository for a long time already, I decided it's safe to keep it as it is for now; it can be still fixed separately from this change.

Real life use case:
1. Have a bare repo
2. Create a worktree from the bare repo
3. In the secondary worktree enable sparse-checkout - this enables extensions.worktreeConfig and keeps core.bare=true setting in config.worktree of the bare worktree
4. The secondary worktree or any other non-bare worktree created won't be able to use branch main (not even once), but it should be able to.

Signed-off-by Olga Pilipenco <[email protected]>

---
Changes since v1:

* no code changes
* rebased with maint
* CC added

Existing broken functionality forces our project to use hacks on bare repo that we'd like to avoid. I would really appreciate reviews of this patch to move closer towards fixing the issue. This is my first contribution to git/git, I apologize if I got lost in the instructions, but I tried my best to follow the rules.

Quoted 10 lines · Olga Pilipenco via GitGitGadget

Setup:
1. Have a bare repo with core.bare = true in config.worktree
2. Create a new worktree

Behavior:
From the secondary worktree the main worktree appears as non-bare.

Expected:
From the secondary worktree the main worktree should appear as bare.

Quoted 8 lines

Why current behavior is not good?
If the main worktree is detected as not bare it doesn't allow checking out the branch of the main worktree. There are possibly other problems associated with that behavior.

Why is it happening?
While we're inside the secondary worktree we don't initialize the main worktree's repository with its configuration.

Okay, this is clearly a very real problem and explains this comment added by f3534c98e4 (worktree: update is_bare heuristics, 2019-04-19):

NEEDSWORK: If this function is called from a secondary worktree and config.worktree is present, is_bare_repository_cfg will reflect the contents of config.worktree, not the contents of the main worktree. This means that worktree->is_bare may be set to 0 even if the main worktree is configured to be bare.

(Aside: I recall reading this comment when Jonathan added it but wasn't able to dig into it at the time to really understand it, and never got back around to it. Now, after studying your patch, I understand what it was about.

Quoted 4 lines

How is it fixed?
Load actual configs of the main worktree. Also, skip the config loading step if we're already inside the current worktree because in that case we rely on is_bare_repository() to return the correct result.

I found that I had to dig around a bit to fully understand the problem expressed by this commit message. Perhaps adding a bit more detail would help? Here's my attempt at rewriting the above (also in a way which is more idiomatic to this project):

When extensions.worktreeConfig is true and the main worktree is bare -- that is, its config.worktree file contains core.bare=true -- commands run from secondary worktrees incorrectly see the main worktree as not bare. As such, those commands incorrectly think that the repository's default branch (typically "main" or "master") is checked out in the bare repository even though it's not. This makes it impossible, for instance, to checkout or delete the default branch from a secondary worktree, among other shortcomings.

This problem occurs because, when extensions.worktreeConfig is true, commands run in secondary worktrees only consult $commondir/config and $commondir/worktrees/<id>/config.worktree, thus they never see the main worktree's core.bare=true setting in $commondir/config.worktree.

Fix this problem by consulting the main worktree's config.worktree file when checking whether it is bare. (This extra work is performed only when running from a secondary worktree.)

Quoted 11 lines

Other solutions considered:
Alternatively, instead of incorrectly always using the_repository as the main worktree's repository, we can detect and load the actual repository of the main worktree and then use that repository's is_bare value extracted from correct configs. However, this approach is a bit riskier and could also affect performance. Since we had the assignment worktree->repo = the_repository for a long time already, I decided it's safe to keep it as it is for now; it can be still fixed separately from this change.

I found this paragraph somewhat confusing because it seems to conflate a repository (i.e. the shared object database) with the struct repository type, and the configuration which happens to get loaded and stored (as one of *many* members) of the repository structure. I had to read it several times to understand that this was talking about instantiating a separate struct repository initialized from the main worktree configuration. I agree that doing so would likely be overkill and could impact performance negatively. I understand that you added this paragraph because SubmittingPatches suggests to do so, but I think it can probably be omitted in this case unless it can be rewritten to be more clear (but even then I doubt it is necessary to keep it).

Quoted 10 lines

Real life use case:
1. Have a bare repo
2. Create a worktree from the bare repo
3. In the secondary worktree enable sparse-checkout - this enables extensions.worktreeConfig and keeps core.bare=true setting in config.worktree of the bare worktree
4. The secondary worktree or any other non-bare worktree created won't be able to use branch main (not even once), but it should be able to.

This is mostly repeating what was said earlier, thus probably isn't adding any value to the commit message. I'd probably drop it.

Quoted 5 lines

Signed-off-by Olga Pilipenco <[email protected]>

---
Changes since v1:

Quoted 3 lines

* no code changes
* rebased with maint
* CC added

Sorry. I've had your v1 sitting in my ever-increasingly-large backlog of patches to look at, but have been extra busy the last many months and never managed to get to it.

Quoted 6 lines

Existing broken functionality forces our project to use hacks on bare repo that we'd like to avoid. I would really appreciate reviews of this patch to move closer towards fixing the issue. This is my first contribution to git/git, I apologize if I got lost in the instructions, but I tried my best to follow the rules.

Your submission is fine. Unfortunately, the project has a lack of reviewers but no lack of submitters, so sometimes patches get overlooked or simply buried.

t/t3200-branch.sh test_expect_success 'bare main worktree has HEAD at branch deleted by secondary · line 410

Quoted 3 lines

t/t3200-branch.sh +1 -0

@@ -410,6 +410,20 @@ test_expect_success 'bare main worktree has HEAD at branch deleted by secondary
+test_expect_success 'secondary worktree can switch to main if common dir is bare worktree' '

The use of "common dir" is a bit confusing. Also, this patch is fixing the more general problem that secondary worktrees think that the bare main worktree has a branch checked out. So, perhaps a better title would be:

secondary worktrees recognize core.bare=true in main config.worktree

or something?

t/t3200-branch.sh test_expect_success 'bare main worktree has HEAD at branch deleted by secondary · line 425

Quoted 13 lines

t/t3200-branch.sh +12 -0

+       test_when_finished "rm -rf bare_repo non_bare_repo secondary_worktree" &&
+       git init -b main non_bare_repo &&
+       test_commit -C non_bare_repo x &&
+
+       git clone --bare non_bare_repo bare_repo &&
+       git -C bare_repo config extensions.worktreeConfig true &&
+       git -C bare_repo config unset core.bare &&
+       git -C bare_repo config --worktree core.bare true &&
+
+       git -C bare_repo worktree add ../secondary_worktree &&
+       git -C secondary_worktree checkout main
+'

Very straightforward and exactly what I expected to see once I understood the problem.

worktree.c static int is_current_worktree(struct worktree *wt) · line 65

Quoted 3 lines

worktree.c +1 -0

@@ -65,6 +65,28 @@ static int is_current_worktree(struct worktree *wt)
+static int is_bare_git_dir(const char *git_dir)

Nit: I wonder if a name such as is_main_worktree_bare() would clue readers in a bit more?

worktree.c static int is_current_worktree(struct worktree *wt) · line 71

Quoted 4 lines

worktree.c +3 -0

+{
+       int bare = 0;
+       struct config_set cs = { { 0 } };

This is not your fault since this construct is used elsewhere in this file (from which I presume you copied it), but project consensus is that using the notation {{0}} to work around a complaint from the Apple compiler (and only the Apple compiler) should be avoided, and that {0} is preferred. So, if you reroll, changing this to {0} may make other reviewers happy (or you can leave it as is to be consistent with existing precedence in this file; I don't feel strongly about it).

worktree.c static int is_current_worktree(struct worktree *wt) · line 80

Quoted 10 lines

worktree.c +9 -0

+       char *config_file;
+       char *worktree_config_file;
+
+       config_file = xstrfmt("%s/config", git_dir);
+       worktree_config_file = xstrfmt("%s/config.worktree",  git_dir);
+
+       git_configset_init(&cs);
+       git_configset_add_file(&cs, config_file);
+       git_configset_add_file(&cs, worktree_config_file);

Genuine question: I haven't thought too deeply about it, but do we gain anything by loading $commondir/config here -- which is shared by the main worktree and all secondary worktrees -- considering that it was already loaded and consulted by the earlier is-bare check before this function was even called?

worktree.c static int is_current_worktree(struct worktree *wt) · line 87

Quoted 7 lines

worktree.c +6 -0

+       git_configset_get_bool(&cs, "core.bare", &bare);
+
+       git_configset_clear(&cs);
+       free(config_file);
+       free(worktree_config_file);
+       return bare;

Everything gets cleaned up correctly. Good.

worktree.c static struct worktree *get_main_worktree(int skip_reading_head) · line 103

Quoted 6 lines

worktree.c +3 -0

@@ -77,18 +99,16 @@ static struct worktree *get_main_worktree(int skip_reading_head)
+       /*
+        * NEEDSWORK: the_repository is not always main worktree's repository
+       */
       worktree->repo = the_repository;
       worktree->path = strbuf_detach(&worktree_path, NULL);

I found this new NEEDSWORK comment rather confusing the first several times I read the patch. It wasn't until I finally realized that the reference to the_repository here is the same reference to

the_repository in the commit message -- which confused me, as well -- that I understood what this was trying to say. The actual problem, of course, is that the _configuration_ stored in the_repository is the secondary worktree's configuration, not the main worktree's configuration. Considering that this patch addresses that problem, I'd probably just drop this new comment altogether (unless, perhaps, you rewrite it to talk about the _configuration_ stored in the_repository).

Quoted 4 lines

worktree.c +0 -3

-       /*
-        * NEEDSWORK: If this function is called from a secondary worktree and
-        * config.worktree is present, is_bare_repository_cfg will reflect the

worktree.c static struct worktree *get_main_worktree(int skip_reading_head) · line 110

Quoted 10 lines

worktree.c +3 -6

-        * contents of config.worktree, not the contents of the main worktree.
-        * This means that worktree->is_bare may be set to 0 even if the main
-        * worktree is configured to be bare.
-        */
-       worktree->is_bare = (is_bare_repository_cfg == 1) ||
-               is_bare_repository();
       worktree->is_current = is_current_worktree(worktree);
+       worktree->is_bare = (is_bare_repository_cfg == 1) ||
+               is_bare_repository() ||
+               (!worktree->is_current && is_bare_git_dir(repo_get_common_dir(the_repository)));

This is performing the expensive check only if the earlier checks left the question unanswered. Good.

Quoted 24 lines · Eric Sunshine

On Thu, Jan 16, 2025 at 4:35 PM Olga Pilipenco via GitGitGadget <[email protected]> wrote:
I found that I had to dig around a bit to fully understand the problem expressed by this commit message. Perhaps adding a bit more detail would help? Here's my attempt at rewriting the above (also in a way which is more idiomatic to this project):

When extensions.worktreeConfig is true and the main worktree is bare -- that is, its config.worktree file contains core.bare=true -- commands run from secondary worktrees incorrectly see the main worktree as not bare. As such, those commands incorrectly think that the repository's default branch (typically "main" or "master") is checked out in the bare repository even though it's not. This makes it impossible, for instance, to checkout or delete the default branch from a secondary worktree, among other shortcomings.

This problem occurs because, when extensions.worktreeConfig is true, commands run in secondary worktrees only consult $commondir/config and $commondir/worktrees/<id>/config.worktree, thus they never see the main worktree's core.bare=true setting in $commondir/config.worktree.

Quoted 3 lines

Fix this problem by consulting the main worktree's config.worktree file when checking whether it is bare. (This extra work is performed only when running from a secondary worktree.)

Wow, your explanation is so much better than mine.Thank you for “translating" it for the world :) I’m still trying to get used to the terminology used in this codebase. I’ll steal your description for sure (if you don’t mind).

Quoted 6 lines

worktree.c static int is_current_worktree(struct worktree *wt) · line 65

Quoted 3 lines

worktree.c +1 -0

@@ -65,6 +65,28 @@ static int is_current_worktree(struct worktree *wt)
+static int is_bare_git_dir(const char *git_dir)

Nit: I wonder if a name such as is_main_worktree_bare() would clue readers in a bit more?

I was about to explain how I wanted this function to be more generic and handle all sorts of bare and non-bare cases - whether it’s the main worktree or not. However, after seeing your comments and after revisiting the code, I realized that generalization doesn’t really provide much benefit here. It is much clearer if we're explicit that the bare check in this case is only performed on the main worktree. I’ll update it in the next version.

Quoted 7 lines

worktree.c static int is_current_worktree(struct worktree *wt) · line 80

Quoted 6 lines

worktree.c +6 -0

+    config_file = xstrfmt("%s/config", git_dir);
+    worktree_config_file = xstrfmt("%s/config.worktree", git_dir);
+
+    git_configset_init(&cs);
+    git_configset_add_file(&cs, config_file);
+    git_configset_add_file(&cs, worktree_config_file);

Quoted 5 lines

Genuine question: I haven't thought too deeply about it, but do we gain anything by loading $commondir/config here -- which is shared by the main worktree and all secondary worktrees -- considering that it was already loaded and consulted by the earlier is-bare check before this function was even called?

This function determines if a worktree is bare or not. I want this logic to work even when it’s called from a different context and not rely on other is-bare checks (that are a bit confusing tbh).

Quoted 6 lines

worktree.c static struct worktree *get_main_worktree(int skip_reading_head) · line 106

Quoted 5 lines

worktree.c +3 -0

+    /*
+    * NEEDSWORK: the_repository is not always main worktree's repository
+    */
   worktree->repo = the_repository;
   worktree->path = strbuf_detach(&worktree_path, NULL);

Quoted 11 lines

I found this new NEEDSWORK comment rather confusing the first several times I read the patch. It wasn't until I finally realized that the reference to the_repository here is the same reference to the_repository in the commit message -- which confused me, as well -- that I understood what this was trying to say. The actual problem, of course, is that the _configuration_ stored in the_repository is the secondary worktree's configuration, not the main worktree's configuration. Considering that this patch addresses that problem, I'd probably just drop this new comment altogether (unless, perhaps, you rewrite it to talk about the _configuration_ stored in the_repository).

This the_repository structure is soooo confusing, took me a while to figure out what it is! I would feel guilty not mentioning that under some circumstances the_repository assigned here could be not actual configuration of the worktree object. I don’t know if that will ever matter or not, but I find this assignment kinda “stinky” and want everyone to know about it. I don’t want to change this assignment in this patch because it didn’t bring any harm so far. I’ll try again to rephrase this comment, just to give a heads up in case someone experiences “weird” behaviour in this area (same way the previous NEEDSWORK comment gave me ideas why my workflow didn’t work and inspired me to try to fix it).

Thanks for the review. I’ll incorporate the changes in my next version and hopefully it will be good to go :tada: I hope I responded to all the comments, it’s a bit nerve-wrecking to contribute for the first time (so many rules and instructions!) :)

Quoted 4 lines

Wow, your explanation is so much better than mine.Thank you for “translating" it for the world :) I’m still trying to get used to the terminology used in this codebase. I’ll steal your description for sure (if you don’t mind).

You are more than welcome to use the proposed commit message rewrite.

(If you want to acknowledge assistance rendered, a Helped-by: trailer, preceding your Signed-off-by:, is the way to do so. Or not. It's up to you.)
...

Agreed about the is-bare checks -- and indeed the entire Git startup sequence -- being difficult to digest, however...
...

Quoted 9 lines

I see. When reviewing, I was wondering why the git-dir was being passed into the function. Your explanation above answers that question. On that note, in addition to renaming the function as suggested, for clarity, I would probably go a bit further and pass in a struct repository * rather than passing in the git-dir itself, just to make it clear that the function is checking main-worktree bareness of the repository in question, as opposed to merely checking bareness of any arbitrary directory. (At least, I would find the intention more clear at-a-glance with that additional change applied.)

Indeed, no need to pass git-dir anymore. There is actually no need to pass the_repository because it’s global. I like how this simplified things and made code clearer.

Quoted 10 lines

One reason I asked the question was due to concern that future readers of this code may very well wonder (as I did) why $commondir/config is being loaded when doing so is (apparently) unnecessary in this particular context. The question is especially pertinent given that this is a private helper function with a single caller. A second reason was that, over the years, a good deal of effort has been put into optimizing Git's startup to avoid doing unnecessary work, and this appears to be unnecessary since $commondir/config would already have been consulted by earlier checks before this function gets called (assuming I'm correctly understanding the code-flow).

I trust your judgement and knowledge of the code and really like the reasons presented. I’ll change this function to only check for worktree config. However I’d like to give it a good name where it’s clear we only check worktree config. It’s a bit challenging to make it short-ish and not to include multiple “worktree” words in the name. Before I submit a new release, maybe I have time to quickly align on the name. What do you think about this one:

is_main_worktree_bare_in_worktree_config

(It will check if bare=true in the main worktree’s worktree.config)

Naming is harder than the code itself :)

Thank you for this thorough explanation. I’ll drop the comment completely. Less code to read. (To be honest I’m not a big fan of comments and definitely don’t want to introduce confusing comments :)

Fun fact: my email app kept changing “worktree” to “corktree”. Maybe another git feature?

(By the way, when replying, please use the normal ">", "> >", "> > >" markers to signify quoted portions of earlier messages in the thread rather than using only indentation. The reason I make this request is that when I replied to your message, my mailer stripped away all indentation from your message, leaving all earlier quoted portions flush with the left margin, which made it very difficult to figure out which quotes came from which authors from which earlier messages, and I ended up having to reinsert the "> >" markers manually to restore structure to my reply.)

Quoted 9 lines · Eric Sunshine

(By the way, when replying, please use the normal ">", "> >", "> > >" markers to signify quoted portions of earlier messages in the thread rather than using only indentation. The reason I make this request is that when I replied to your message, my mailer stripped away all indentation from your message, leaving all earlier quoted portions flush with the left margin, which made it very difficult to figure out which quotes came from which authors from which earlier messages, and I ended up having to reinsert the "> >" markers manually to restore structure to my reply.)

Examining more closely, I see that your message was in fact multipart MIME with an HTML portion which was using <blockquote> and whatnot, which my (plain text) mailer stripped out, hence lost all formatting. If you can configure your mailer to send plain text and use the normal ">" markers, that would be generally helpful on this list.

Quoted 5 lines · Eric Sunshine

Examining more closely, I see that your message was in fact multipart MIME with an HTML portion which was using <blockquote> and whatnot, which my (plain text) mailer stripped out, hence lost all formatting. If you can configure your mailer to send plain text and use the normal ">" markers, that would be generally helpful on this list.

Ufff, my bad, sorry about that (should be plain text now).

I'm having trouble responding to my previous email and keeping that quoting format,
so I'll not waste time on trying to fix that (please forgive me, everyone). Just want to let you know that I agree with your response, I'll adjust my patch. Enough talking, let's get the code out!

worktree.c +11 -0

+/*
+* When in a secondary worktree, and when extensions.worktreeConfig
+* is true, only $commondir/config and $commondir/worktrees/<id>/
+* config.worktree are consulted, hence any core.bare=true setting in
+* $commondir/config.worktree gets overlooked. Thus, check it manually
+* to determine if the repository is bare.
+*/
+static int is_main_worktree_bare(struct repository *repo)
+{
+	int bare = 0;
+	struct config_set cs = {0};

/**
* get the main worktree
*/

worktree.c +0 -3

@@ -79,16 +101,11 @@ static struct worktree *get_main_worktree(int skip_reading_head)
 	CALLOC_ARRAY(worktree, 1);
 	worktree->repo = the_repository;
 	worktree->path = strbuf_detach(&worktree_path, NULL);
-	/*
-	 * NEEDSWORK: If this function is called from a secondary worktree and
-	 * config.worktree is present, is_bare_repository_cfg will reflect the

Quoted 11 lines · Olga Pilipenco via GitGitGadget

worktree.c +11 -0

+/*
+* When in a secondary worktree, and when extensions.worktreeConfig
+* is true, only $commondir/config and $commondir/worktrees/<id>/
+* config.worktree are consulted, hence any core.bare=true setting in
+* $commondir/config.worktree gets overlooked. Thus, check it manually
+* to determine if the repository is bare.
+*/
+static int is_main_worktree_bare(struct repository *repo)
+{
+	int bare = 0;
+	struct config_set cs = {0};

worktree.c static int is_current_worktree(struct worktree *wt) · line 88

Quoted 10 lines

worktree.c +10 -0

+	char *worktree_config = xstrfmt("%s/config.worktree", repo_get_common_dir(repo));
+
+	git_configset_init(&cs);
+	git_configset_add_file(&cs, worktree_config);
+	git_configset_get_bool(&cs, "core.bare", &bare);
+
+	git_configset_clear(&cs);
+	free(worktree_config);
+	return bare;
+}

That is nicely described.

Quoted 17 lines

/**
* get the main worktree
*/

worktree.c +0 -9

@@ -79,16 +101,11 @@ static struct worktree *get_main_worktree(int skip_reading_head)
 	CALLOC_ARRAY(worktree, 1);
 	worktree->repo = the_repository;
 	worktree->path = strbuf_detach(&worktree_path, NULL);
-	/*
-	 * NEEDSWORK: If this function is called from a secondary worktree and
-	 * config.worktree is present, is_bare_repository_cfg will reflect the
-	 * contents of config.worktree, not the contents of the main worktree.
-	 * This means that worktree->is_bare may be set to 0 even if the main
-	 * worktree is configured to be bare.
-	 */
-	worktree->is_bare = (is_bare_repository_cfg == 1) ||
-		is_bare_repository();
 	worktree->is_current = is_current_worktree(worktree);

worktree.c static struct worktree *get_main_worktree(int skip_reading_head) · line 107

Quoted 3 lines

worktree.c +3 -0

+	worktree->is_bare = (is_bare_repository_cfg == 1) ||
+		is_bare_repository() ||
+		(!worktree->is_current && is_main_worktree_bare(the_repository));

Is "this worktree does not have is_current bit set" equivalent to "this worktree is the main one, so is_main_worktree_bare() needs to be consulted"? That linkage between "the is_current bit unset" and "is the main worktree" is not obvious to me.

Quoted 43 lines · Junio C Hamano

Quoted 11 lines · Olga Pilipenco via GitGitGadget

worktree.c +11 -0

+/*
+* When in a secondary worktree, and when extensions.worktreeConfig
+* is true, only $commondir/config and $commondir/worktrees/<id>/
+* config.worktree are consulted, hence any core.bare=true setting in
+* $commondir/config.worktree gets overlooked. Thus, check it manually
+* to determine if the repository is bare.
+*/
+static int is_main_worktree_bare(struct repository *repo)
+{
+    int bare = 0;
+    struct config_set cs = {0};

worktree.c static int is_current_worktree(struct worktree *wt) · line 88

Quoted 10 lines

worktree.c +10 -0

+    char *worktree_config = xstrfmt("%s/config.worktree", repo_get_common_dir(repo));
+
+    git_configset_init(&cs);
+    git_configset_add_file(&cs, worktree_config);
+    git_configset_get_bool(&cs, "core.bare", &bare);
+
+    git_configset_clear(&cs);
+    free(worktree_config);
+    return bare;
+}

That is nicely described.

Quoted 17 lines

/**
* get the main worktree
*/

worktree.c +0 -9

@@ -79,16 +101,11 @@ static struct worktree *get_main_worktree(int skip_reading_head)
     CALLOC_ARRAY(worktree, 1);
     worktree->repo = the_repository;
     worktree->path = strbuf_detach(&worktree_path, NULL);
-    /*
-     * NEEDSWORK: If this function is called from a secondary worktree and
-     * config.worktree is present, is_bare_repository_cfg will reflect the
-     * contents of config.worktree, not the contents of the main worktree.
-     * This means that worktree->is_bare may be set to 0 even if the main
-     * worktree is configured to be bare.
-     */
-    worktree->is_bare = (is_bare_repository_cfg == 1) ||
-            is_bare_repository();
     worktree->is_current = is_current_worktree(worktree);

Quoted 8 lines

worktree.c static struct worktree *get_main_worktree(int skip_reading_head) · line 107

Quoted 3 lines

worktree.c +3 -0

+    worktree->is_bare = (is_bare_repository_cfg == 1) ||
+            is_bare_repository() ||
+            (!worktree->is_current && is_main_worktree_bare(the_repository));

Is "this worktree does not have is_current bit set" equivalent to "this worktree is the main one, so is_main_worktree_bare() needs to be consulted"? That linkage between "the is_current bit unset" and "is the main worktree" is not obvious to me.

Does the thinking behind it go like this?

We grabbed the "main" worktree object and stored it in worktree; it is either our current worktree (in which case is_current is true), or it is not (in which case, is_current is false). We know that the old logic failed when asking the "is it bare" question from a secondary worktree. !worktree->is_current tells us that we _are_ asking the question from a secondary worktree, so we need to make the extra call to check config.worktree file as well in that case.

Perhaps the logic is clear to those who diagnosed the problem, wrote the patch, and reviewed it, in which case there is no reason to reroll. Perhaps it was just me to whom it was not obvious that the purpose of "is_current" check was not about "are we looking at the main worktree" but was about "if we are not in the main worktree, we need this extra check".

Thanks.

Quoted 7 lines

Perhaps the logic is clear to those who diagnosed the problem, wrote the patch, and reviewed it, in which case there is no reason to reroll. Perhaps it was just me to whom it was not obvious that the purpose of "is_current" check was not about "are we looking at the main worktree" but was about "if we are not in the main worktree, we need this extra check".

Quoted 1 line

Thanks.

You did a great job figuring it out and I agree it's confusing at first, but we tried our best to make it less confusing. is_current check is actually not necessary there, but having it there saves extra unnecessary calculations, also describes & fixes the exact scenario that didn't work (not being able to see main worktree as bare from a secondary worktree).

Quoted 7 lines · Olga Pilipenco

Quoted 7 lines

Perhaps the logic is clear to those who diagnosed the problem, wrote the patch, and reviewed it, in which case there is no reason to reroll. Perhaps it was just me to whom it was not obvious that the purpose of "is_current" check was not about "are we looking at the main worktree" but was about "if we are not in the main worktree, we need this extra check".

Quoted 8 lines

Quoted 1 line

Thanks.

You did a great job figuring it out and I agree it's confusing at first, but we tried our best to make it less confusing. is_current check is actually not necessary there, but having it there saves extra unnecessary calculations, also describes & fixes the exact scenario that didn't work (not being able to see main worktree as bare from a secondary worktree).

If I had to do a great job there, then the code does deserve to be explained a bit better for later developers who wonder why it is written in the way it is, perhaps we a single-liner comment?

Thanks.

I have 2 versions for comment:

1. Since is_main_worktree_bare explains quite well what it does we can have a shorter explanation of !worktree->is_current part, something like:

/* Additional checks are needed if main worktree is not current (checking from secondary worktree) */ (!worktree->is_current && is_main_worktree_bare(the_repository));

2. Or a bit longer inline explanation that partially repeats the explanation of is_main_worktree_bare
+ adds explanation about efficiency:
/*
* When in a secondary worktree we have to also verify if the main worktree
* is bare in $commondir/config.worktree.
* This check is unnecessary if we're currently in the main worktree,
* as prior checks already consulted all configs of the current worktree.
*/
(!worktree->is_current && is_main_worktree_bare(the_repository));

Quoted 8 lines · Olga Pilipenco

I have 2 versions for comment:

1. Since is_main_worktree_bare explains quite well what it does we can have a shorter explanation of !worktree->is_current part, something like:

/* Additional checks are needed if main worktree is not current (checking from secondary worktree) */ (!worktree->is_current && is_main_worktree_bare(the_repository));

For somebody who thought about the issue themselves (like me, before writing the message you are responding to), this shorter form would suffice. I'd rephrase it more like so

    /* When a secondary worktree, an extra check is needed */

for brevity, though.

Quoted 10 lines

2. Or a bit longer inline explanation that partially repeats the explanation of is_main_worktree_bare
+ adds explanation about efficiency:
/*
* When in a secondary worktree we have to also verify if the main worktree
* is bare in $commondir/config.worktree.
* This check is unnecessary if we're currently in the main worktree,
* as prior checks already consulted all configs of the current worktree.
*/
(!worktree->is_current && is_main_worktree_bare(the_repository));

And this more extended version would have helped me by not having to ask

Is "this worktree does not have is_current bit set" equivalent to "this worktree is the main one, so is_main_worktree_bare() needs to be consulted"? That linkage between "the is_current bit unset" and "is the main worktree" is not obvious to me.

in the first place.

In short, both should work, and I personally find that the latter may be a bit more helpful to readers.

THanks.