threads / patch / 63876

patchstatus: add --json output format to git status

Subject: [PATCH] status: add --json output format to git status

## tl;dr

6 messages between Jul 30, 2025 and Aug 4, 2025. Diffs are folded; open one to read it.

replies: 5people: 3as markdown or json

Tach via GitGitGadget· Jul 30, 2025, 06:27 UTC · lore
From: tacherasasi <tacherasasi@gmail.com>

Add a new --json flag to 'git status' that outputs repository state in a structured JSON format. This enables reliable machine parsing of status information for tools and automation.

The JSON output includes:
- Branch information (name, detached state, ahead/behind counts)
- Staged files array
- Unstaged files array
- Untracked files array
- Ignored files array (with --ignored flag)
Implementation details:
- Add STATUS_FORMAT_JSON to wt_status_format enum
- Add JSON output option to git status and git commit
- Implement JSON formatting helpers for arrays and branch info
- Structure output for easy parsing and future extensibility
Example:
  $ git status --json
  {
    "branch": {
      "current": "main",
      "detached": false
    },
    "staged": ["file1.txt"],
    "unstaged": ["file2.txt"],
    "untracked": ["file3.txt"]
  }

This provides a robust alternative to parsing traditional output formats, making it easier to build reliable tools and automation around Git status information.

Signed-off-by: Tachera Sasi <tachera@ekilie.com>
Signed-off-by: tacherasasi <tacherasasi@gmail.com>
---
    status: add --json output format to git status
Published-As: https://github.com/gitgitgadget/git/releases/tag/pr-1937%2FtacheraSasi%2Ffeature%2Fjson-status-output-v1
Fetch-It-Via: git fetch https://github.com/gitgitgadget/git pr-1937/tacheraSasi/feature/json-status-output-v1
Pull-Request: https://github.com/gitgitgadget/git/pull/1937
 builtin/commit.c |  9 ++++-
 wt-status.c      | 98 ++++++++++++++++++++++++++++++++++++++++++++++++
 wt-status.h      |  1 +
 3 files changed, 107 insertions(+), 1 deletion(-)
Show changes to 3 files +107 −1

builtin/commit.c, wt-status.c, wt-status.h

diff --git a/builtin/commit.c b/builtin/commit.c
index fba0dded64a..f1db4fdfd9a 100644
--- a/builtin/commit.c
+++ b/builtin/commit.c
@@ -1540,6 +1540,9 @@ struct repository *repo UNUSED)
 		OPT_SET_INT(0, "long", &status_format,
 			    N_("show status in long format (default)"),
 			    STATUS_FORMAT_LONG),
+		OPT_SET_INT(0, "json", &status_format,
+		    N_("show status in JSON format"),
+		    STATUS_FORMAT_JSON),
 		OPT_BOOL('z', "null", &s.null_termination,
 			 N_("terminate entries with NUL")),
 		{
@@ -1603,7 +1606,8 @@ struct repository *repo UNUSED)
 		       prefix, argv);
 
 	if (status_format != STATUS_FORMAT_PORCELAIN &&
-	    status_format != STATUS_FORMAT_PORCELAIN_V2)
+	    status_format != STATUS_FORMAT_PORCELAIN_V2 &&
+	    status_format != STATUS_FORMAT_JSON)
 		progress_flag = REFRESH_PROGRESS;
 	repo_read_index(the_repository);
 	refresh_index(the_repository->index,
@@ -1735,6 +1739,9 @@ int cmd_commit(int argc,
 		OPT_SET_INT(0, "long", &status_format,
 			    N_("show status in long format (default)"),
 			    STATUS_FORMAT_LONG),
+		OPT_SET_INT(0, "json", &status_format,
+		    N_("show status in JSON format"),
+		    STATUS_FORMAT_JSON),
 		OPT_BOOL('z', "null", &s.null_termination,
 			 N_("terminate entries with NUL")),
 		OPT_BOOL(0, "amend", &amend, N_("amend previous commit")),
diff --git a/wt-status.c b/wt-status.c
index 454601afa15..7192fb4d057 100644
--- a/wt-status.c
+++ b/wt-status.c
@@ -2564,6 +2564,101 @@ static void wt_porcelain_v2_print(struct wt_status *s)
 	}
 }
 
+
+static void wt_json_print_string_array(struct wt_status *s, const char *name, struct string_list *list)
+{
+	int i;
+	fprintf(s->fp, "  \"%s\": [", name);
+	for (i = 0; i < list->nr; i++) {
+		if (i > 0)
+			fprintf(s->fp, ", ");
+		fprintf(s->fp, "\"%s\"", list->items[i].string);
+	}
+	fprintf(s->fp, "]");
+}
+
+static void wt_json_print_change_array(struct wt_status *s, const char *name, int change_type)
+{
+	int i;
+	struct string_list files = STRING_LIST_INIT_DUP;
+
+	for (i = 0; i < s->change.nr; i++) {
+		struct wt_status_change_data *d;
+		struct string_list_item *it;
+		it = &(s->change.items[i]);
+		d = it->util;
+
+		if ((change_type == WT_STATUS_UPDATED && d->index_status &&
+		     d->index_status != DIFF_STATUS_UNMERGED) ||
+		    (change_type == WT_STATUS_CHANGED && d->worktree_status &&
+		     d->worktree_status != DIFF_STATUS_UNMERGED)) {
+			string_list_append(&files, it->string);
+		}
+	}
+
+	wt_json_print_string_array(s, name, &files);
+	string_list_clear(&files, 0);
+}
+
+static void wt_json_print_branch_info(struct wt_status *s)
+{
+	struct branch *branch;
+	const char *branch_name;
+	int ahead = 0, behind = 0;
+
+	fprintf(s->fp, "  \"branch\": {\n");
+
+	if (s->branch && !s->is_initial) {
+		if (!strcmp(s->branch, "HEAD")) {
+			fprintf(s->fp, "    \"current\": \"HEAD\",\n");
+			fprintf(s->fp, "    \"detached\": true");
+		} else {
+			if (skip_prefix(s->branch, "refs/heads/", &branch_name)) {
+				fprintf(s->fp, "    \"current\": \"%s\",\n", branch_name);
+				fprintf(s->fp, "    \"detached\": false");
+
+				branch = branch_get(branch_name);
+				if (branch && branch->merge && branch->merge[0] && branch->merge[0]->dst) {
+					if (!stat_tracking_info(branch, &ahead, &behind, NULL, 0, 0)) {
+						fprintf(s->fp, ",\n    \"ahead\": %d,\n    \"behind\": %d", ahead, behind);
+					}
+				}
+			} else {
+				fprintf(s->fp, "    \"current\": \"%s\",\n", s->branch);
+				fprintf(s->fp, "    \"detached\": false");
+			}
+		}
+	} else {
+		fprintf(s->fp, "    \"current\": null,\n");
+		fprintf(s->fp, "    \"detached\": false");
+	}
+
+	fprintf(s->fp, "\n  }");
+}
+
+static void wt_json_status_print(struct wt_status *s)
+{
+	fprintf(s->fp, "{\n");
+
+	wt_json_print_branch_info(s);
+	fprintf(s->fp, ",\n");
+
+	wt_json_print_change_array(s, "staged", WT_STATUS_UPDATED);
+	fprintf(s->fp, ",\n");
+
+	wt_json_print_change_array(s, "unstaged", WT_STATUS_CHANGED);
+	fprintf(s->fp, ",\n");
+
+	wt_json_print_string_array(s, "untracked", &s->untracked);
+
+	if (s->ignored.nr > 0) {
+		fprintf(s->fp, ",\n");
+		wt_json_print_string_array(s, "ignored", &s->ignored);
+	}
+
+	fprintf(s->fp, "\n}\n");
+}
+
 void wt_status_print(struct wt_status *s)
 {
 	trace2_data_intmax("status", s->repo, "count/changed", s->change.nr);
@@ -2583,6 +2678,9 @@ void wt_status_print(struct wt_status *s)
 	case STATUS_FORMAT_PORCELAIN_V2:
 		wt_porcelain_v2_print(s);
 		break;
+	case STATUS_FORMAT_JSON:
+		wt_json_status_print(s);
+		break;
 	case STATUS_FORMAT_UNSPECIFIED:
 		BUG("finalize_deferred_config() should have been called");
 		break;
diff --git a/wt-status.h b/wt-status.h
index 4e377ce62b8..e929af832b2 100644
--- a/wt-status.h
+++ b/wt-status.h
@@ -74,6 +74,7 @@ enum wt_status_format {
 	STATUS_FORMAT_SHORT,
 	STATUS_FORMAT_PORCELAIN,
 	STATUS_FORMAT_PORCELAIN_V2,
+	STATUS_FORMAT_JSON,
 
 	STATUS_FORMAT_UNSPECIFIED
 };

base-commit: cb3b40381e1d5ee32dde96521ad7cfd68eb308a6
-- 
gitgitgadget
Phillip Wood· Jul 30, 2025, 09:01 UTC · re: Tach via GitGitGadget · lore

Re: [PATCH] status: add --json output format to git status

On 30/07/2025 07:27, Tach via GitGitGadget wrote:
Show 34 quoted lines
> From: tacherasasi <tacherasasi@gmail.com>
> 
> Add a new --json flag to 'git status' that outputs repository state
> in a structured JSON format. This enables reliable machine parsing
> of status information for tools and automation.
> 
> The JSON output includes:
> - Branch information (name, detached state, ahead/behind counts)
> - Staged files array
> - Unstaged files array
> - Untracked files array
> - Ignored files array (with --ignored flag)
> 
> Implementation details:
> - Add STATUS_FORMAT_JSON to wt_status_format enum
> - Add JSON output option to git status and git commit
> - Implement JSON formatting helpers for arrays and branch info
> - Structure output for easy parsing and future extensibility
> 
> Example:
>    $ git status --json
>    {
>      "branch": {
>        "current": "main",
>        "detached": false
>      },
>      "staged": ["file1.txt"],
>      "unstaged": ["file2.txt"],
>      "untracked": ["file3.txt"]
>    }
> 
> This provides a robust alternative to parsing traditional output
> formats, making it easier to build reliable tools and automation
> around Git status information.

What problem are you having paring the current output? You say this patch provides a robust alternative but I'm afraid it does not seem to handle filenames containing newlines or double quotes and it assumes they are utf-8 encoded. It also seems to print a trailing comma if an array has more than one item which is not allowed by the json spec. We already have support for generating json output as documented json-writer.h which will help with most of that but not the filename encoding problem. New features should also be accompanied by tests to prevent future regressions. I'm not opposed to adding json output if it is robust and handles non utf-8 filenames but it would be helpful to understand what problems you are having parsing the output of "git status --porcelain[=v2]".

Thanks
Phillip
Show 178 quoted lines
> Signed-off-by: Tachera Sasi <tachera@ekilie.com>
> Signed-off-by: tacherasasi <tacherasasi@gmail.com>
> ---
>      status: add --json output format to git status
> 
> Published-As: https://github.com/gitgitgadget/git/releases/tag/pr-1937%2FtacheraSasi%2Ffeature%2Fjson-status-output-v1
> Fetch-It-Via: git fetch https://github.com/gitgitgadget/git pr-1937/tacheraSasi/feature/json-status-output-v1
> Pull-Request: https://github.com/gitgitgadget/git/pull/1937
> 
>   builtin/commit.c |  9 ++++-
>   wt-status.c      | 98 ++++++++++++++++++++++++++++++++++++++++++++++++
>   wt-status.h      |  1 +
>   3 files changed, 107 insertions(+), 1 deletion(-)
> 
> diff --git a/builtin/commit.c b/builtin/commit.c
> index fba0dded64a..f1db4fdfd9a 100644
> --- a/builtin/commit.c
> +++ b/builtin/commit.c
> @@ -1540,6 +1540,9 @@ struct repository *repo UNUSED)
>   		OPT_SET_INT(0, "long", &status_format,
>   			    N_("show status in long format (default)"),
>   			    STATUS_FORMAT_LONG),
> +		OPT_SET_INT(0, "json", &status_format,
> +		    N_("show status in JSON format"),
> +		    STATUS_FORMAT_JSON),
>   		OPT_BOOL('z', "null", &s.null_termination,
>   			 N_("terminate entries with NUL")),
>   		{
> @@ -1603,7 +1606,8 @@ struct repository *repo UNUSED)
>   		       prefix, argv);
>   
>   	if (status_format != STATUS_FORMAT_PORCELAIN &&
> -	    status_format != STATUS_FORMAT_PORCELAIN_V2)
> +	    status_format != STATUS_FORMAT_PORCELAIN_V2 &&
> +	    status_format != STATUS_FORMAT_JSON)
>   		progress_flag = REFRESH_PROGRESS;
>   	repo_read_index(the_repository);
>   	refresh_index(the_repository->index,
> @@ -1735,6 +1739,9 @@ int cmd_commit(int argc,
>   		OPT_SET_INT(0, "long", &status_format,
>   			    N_("show status in long format (default)"),
>   			    STATUS_FORMAT_LONG),
> +		OPT_SET_INT(0, "json", &status_format,
> +		    N_("show status in JSON format"),
> +		    STATUS_FORMAT_JSON),
>   		OPT_BOOL('z', "null", &s.null_termination,
>   			 N_("terminate entries with NUL")),
>   		OPT_BOOL(0, "amend", &amend, N_("amend previous commit")),
> diff --git a/wt-status.c b/wt-status.c
> index 454601afa15..7192fb4d057 100644
> --- a/wt-status.c
> +++ b/wt-status.c
> @@ -2564,6 +2564,101 @@ static void wt_porcelain_v2_print(struct wt_status *s)
>   	}
>   }
>   
> +
> +static void wt_json_print_string_array(struct wt_status *s, const char *name, struct string_list *list)
> +{
> +	int i;
> +	fprintf(s->fp, "  \"%s\": [", name);
> +	for (i = 0; i < list->nr; i++) {
> +		if (i > 0)
> +			fprintf(s->fp, ", ");
> +		fprintf(s->fp, "\"%s\"", list->items[i].string);
> +	}
> +	fprintf(s->fp, "]");
> +}
> +
> +static void wt_json_print_change_array(struct wt_status *s, const char *name, int change_type)
> +{
> +	int i;
> +	struct string_list files = STRING_LIST_INIT_DUP;
> +
> +	for (i = 0; i < s->change.nr; i++) {
> +		struct wt_status_change_data *d;
> +		struct string_list_item *it;
> +		it = &(s->change.items[i]);
> +		d = it->util;
> +
> +		if ((change_type == WT_STATUS_UPDATED && d->index_status &&
> +		     d->index_status != DIFF_STATUS_UNMERGED) ||
> +		    (change_type == WT_STATUS_CHANGED && d->worktree_status &&
> +		     d->worktree_status != DIFF_STATUS_UNMERGED)) {
> +			string_list_append(&files, it->string);
> +		}
> +	}
> +
> +	wt_json_print_string_array(s, name, &files);
> +	string_list_clear(&files, 0);
> +}
> +
> +static void wt_json_print_branch_info(struct wt_status *s)
> +{
> +	struct branch *branch;
> +	const char *branch_name;
> +	int ahead = 0, behind = 0;
> +
> +	fprintf(s->fp, "  \"branch\": {\n");
> +
> +	if (s->branch && !s->is_initial) {
> +		if (!strcmp(s->branch, "HEAD")) {
> +			fprintf(s->fp, "    \"current\": \"HEAD\",\n");
> +			fprintf(s->fp, "    \"detached\": true");
> +		} else {
> +			if (skip_prefix(s->branch, "refs/heads/", &branch_name)) {
> +				fprintf(s->fp, "    \"current\": \"%s\",\n", branch_name);
> +				fprintf(s->fp, "    \"detached\": false");
> +
> +				branch = branch_get(branch_name);
> +				if (branch && branch->merge && branch->merge[0] && branch->merge[0]->dst) {
> +					if (!stat_tracking_info(branch, &ahead, &behind, NULL, 0, 0)) {
> +						fprintf(s->fp, ",\n    \"ahead\": %d,\n    \"behind\": %d", ahead, behind);
> +					}
> +				}
> +			} else {
> +				fprintf(s->fp, "    \"current\": \"%s\",\n", s->branch);
> +				fprintf(s->fp, "    \"detached\": false");
> +			}
> +		}
> +	} else {
> +		fprintf(s->fp, "    \"current\": null,\n");
> +		fprintf(s->fp, "    \"detached\": false");
> +	}
> +
> +	fprintf(s->fp, "\n  }");
> +}
> +
> +static void wt_json_status_print(struct wt_status *s)
> +{
> +	fprintf(s->fp, "{\n");
> +
> +	wt_json_print_branch_info(s);
> +	fprintf(s->fp, ",\n");
> +
> +	wt_json_print_change_array(s, "staged", WT_STATUS_UPDATED);
> +	fprintf(s->fp, ",\n");
> +
> +	wt_json_print_change_array(s, "unstaged", WT_STATUS_CHANGED);
> +	fprintf(s->fp, ",\n");
> +
> +	wt_json_print_string_array(s, "untracked", &s->untracked);
> +
> +	if (s->ignored.nr > 0) {
> +		fprintf(s->fp, ",\n");
> +		wt_json_print_string_array(s, "ignored", &s->ignored);
> +	}
> +
> +	fprintf(s->fp, "\n}\n");
> +}
> +
>   void wt_status_print(struct wt_status *s)
>   {
>   	trace2_data_intmax("status", s->repo, "count/changed", s->change.nr);
> @@ -2583,6 +2678,9 @@ void wt_status_print(struct wt_status *s)
>   	case STATUS_FORMAT_PORCELAIN_V2:
>   		wt_porcelain_v2_print(s);
>   		break;
> +	case STATUS_FORMAT_JSON:
> +		wt_json_status_print(s);
> +		break;
>   	case STATUS_FORMAT_UNSPECIFIED:
>   		BUG("finalize_deferred_config() should have been called");
>   		break;
> diff --git a/wt-status.h b/wt-status.h
> index 4e377ce62b8..e929af832b2 100644
> --- a/wt-status.h
> +++ b/wt-status.h
> @@ -74,6 +74,7 @@ enum wt_status_format {
>   	STATUS_FORMAT_SHORT,
>   	STATUS_FORMAT_PORCELAIN,
>   	STATUS_FORMAT_PORCELAIN_V2,
> +	STATUS_FORMAT_JSON,
>   
>   	STATUS_FORMAT_UNSPECIFIED
>   };
> 
> base-commit: cb3b40381e1d5ee32dde96521ad7cfd68eb308a6
Junio C Hamano· Jul 30, 2025, 14:58 UTC · re: Tach via GitGitGadget · lore

Re: [PATCH] status: add --json output format to git status

"Tach via GitGitGadget" <gitgitgadget@gmail.com> writes:
> From: tacherasasi <tacherasasi@gmail.com>
> Add a new --json flag to 'git status' that outputs repository state
> in a structured JSON format. This enables reliable machine parsing
> of status information for tools and automation.

The --porcelain format is meant to serve this "parseable for tools and automation to read" use case. Was there anything your parser couldn't fathom out of the format?

> This provides a robust alternative to parsing traditional output
> formats, making it easier to build reliable tools and automation
> around Git status information.

The writer is obviously biased ;-) but I find this a bit hand-wavy and unconvincing.

> Signed-off-by: Tachera Sasi <tachera@ekilie.com>
> Signed-off-by: tacherasasi <tacherasasi@gmail.com>

It is not like two separate people worked on this patch, or somebody at ekilie worked on the patch but asked a different person with a very similar name to send it on their behalf, right?

Perhaps update the user.{name,email} on the in-body From: to match the one used on the first Sign-off, and drop the second Sign-off?

Show 11 quoted lines
> diff --git a/builtin/commit.c b/builtin/commit.c
> index fba0dded64a..f1db4fdfd9a 100644
> --- a/builtin/commit.c
> +++ b/builtin/commit.c
> @@ -1540,6 +1540,9 @@ struct repository *repo UNUSED)
>  		OPT_SET_INT(0, "long", &status_format,
>  			    N_("show status in long format (default)"),
>  			    STATUS_FORMAT_LONG),
> +		OPT_SET_INT(0, "json", &status_format,
> +		    N_("show status in JSON format"),
> +		    STATUS_FORMAT_JSON),

There are now several ways to set to status_format now. It could be a good idea to make a preliminary UI clean-up to introduce a new option "--format={short,long,porcelain,porcelain-v1,porcelain-v2}" as a separate step, declare that any new format will use this new option without adding a separate option on its own, before adding any new features. And then on top that, a new format can be introduced without cluttering the set of options.

Not that I am enthused to see json here, given that we have introduced porcelain format for the same purpose eons ago.

Show 10 quoted lines
> @@ -1603,7 +1606,8 @@ struct repository *repo UNUSED)
>  		       prefix, argv);
>  
>  	if (status_format != STATUS_FORMAT_PORCELAIN &&
> -	    status_format != STATUS_FORMAT_PORCELAIN_V2)
> +	    status_format != STATUS_FORMAT_PORCELAIN_V2 &&
> +	    status_format != STATUS_FORMAT_JSON)
>  		progress_flag = REFRESH_PROGRESS;
>  	repo_read_index(the_repository);
>  	refresh_index(the_repository->index,

Also, we might want to tweak the bit-assignment for the status_format constants to make this part easier to maintain (e.g., ones below 100 are for human consumption and gives progress bar and ones above 100 are for machine consumption and suppresses progress, or something like that), again as a separate preliminary clean-up step before adding any new features.

Show 7 quoted lines
> @@ -1735,6 +1739,9 @@ int cmd_commit(int argc,
>  		OPT_SET_INT(0, "long", &status_format,
>  			    N_("show status in long format (default)"),
>  			    STATUS_FORMAT_LONG),
> +		OPT_SET_INT(0, "json", &status_format,
> +		    N_("show status in JSON format"),
> +		    STATUS_FORMAT_JSON),

And if we are to clean-up with --format=<name> and build on top, this part would become somewhat different. The clean-up would probably be somewhat like what opt_parse_porcelain() is doing and "json" or "yaml" or any new format name would join if/else if/ cascade there.

Show 11 quoted lines
> diff --git a/wt-status.c b/wt-status.c
> index 454601afa15..7192fb4d057 100644
> --- a/wt-status.c
> +++ b/wt-status.c
> @@ -2564,6 +2564,101 @@ static void wt_porcelain_v2_print(struct wt_status *s)
>  	}
>  }
>  
> +
> +static void wt_json_print_string_array(struct wt_status *s, const char *name, struct string_list *list)
> +{

We seem to have a json write-out library-ish helpers already; is it too tied to its current use (i.e. trace output) and too inflexible to be used here, or does the status output have so specialized and unusual need as a normal JSON producing application that we need a special purpose code here with bunch of fprintf() and manual formatting?

Thanks.
Phillip Wood· Aug 4, 2025, 13:49 UTC · re: Junio C Hamano · lore

Re: [PATCH] status: add --json output format to git status

On 30/07/2025 15:58, Junio C Hamano wrote:
Show 8 quoted lines
> "Tach via GitGitGadget" <gitgitgadget@gmail.com> writes:
> 
>> This provides a robust alternative to parsing traditional output
>> formats, making it easier to build reliable tools and automation
>> around Git status information.
> 
> The writer is obviously biased ;-) but I find this a bit hand-wavy
> and unconvincing.

I agree this isn't a very convincing reason. To me a more convincing reason is that virtually every scripting language comes with a way to parse json and base64 so if we emit json with path and ref names encoded as base64 there is no need for script authors to write a parser to parse our porcelain output. To me the fact that we have to encode names using base64 is a sign that json is not particularly well suited to the task but it is widely supported format.

Thanks
Phillip
Junio C Hamano· Aug 4, 2025, 13:54 UTC · re: Phillip Wood · lore

Re: [PATCH] status: add --json output format to git status

Phillip Wood <phillip.wood123@gmail.com> writes:
Show 16 quoted lines
> On 30/07/2025 15:58, Junio C Hamano wrote:
>> "Tach via GitGitGadget" <gitgitgadget@gmail.com> writes:
>> 
>>> This provides a robust alternative to parsing traditional output
>>> formats, making it easier to build reliable tools and automation
>>> around Git status information.
>> The writer is obviously biased ;-) but I find this a bit hand-wavy
>> and unconvincing.
>
> I agree this isn't a very convincing reason. To me a more convincing
> reason is that virtually every scripting language comes with a way to
> parse json and base64 so if we emit json with path and ref names
> encoded as base64 there is no need for script authors to write a
> parser to parse our porcelain output. To me the fact that we have to
> encode names using base64 is a sign that json is not particularly well
> suited to the task but it is widely supported format.
Very well put.  Thanks.
phillip.wood123@gmail.com· Jul 30, 2025, 10:08 UTC · lore

Re: [PATCH] status: add --json output format to git status

Hi Tachera
On 30/07/2025 10:15, Tachera W sasi wrote:
Show 5 quoted lines
> 
> For v2, I will:
>    * Rework the patch to use `json-writer.h` so escaping and JSON 
> compliance are handled
>      properly.

That will handle the escaping problems but we still have the problem that filenames and refnames are not guaranteed to be utf-8 encoded. I think the only way to handle that is to base64 encode them so that the caller can retrieve the raw bytes. See https://lore.kernel.org/git/ZvXMSKaUWWA-MG9J@tapette.crustytoothpaste.net/ for a previous discussion about this.

>    * Add tests covering filenames with quotes, newlines, and non-UTF-8 
> cases.
>    * Include tests under `t/` to prevent future regressions.

That's great, you should update Documentation/git-status.adoc as well to document the new option and output format

Thanks
Phillip
Show 266 quoted lines
> Thanks again for your review and for pointing me towards the proper helpers.
> 
> Best regards,
> Tachera Sasi
> 
> On Wed, Jul 30, 2025 at 12:01 PM Phillip Wood <phillip.wood123@gmail.com 
> <mailto:phillip.wood123@gmail.com>> wrote:
> 
>     On 30/07/2025 07:27, Tach via GitGitGadget wrote:
>      > From: tacherasasi <tacherasasi@gmail.com
>     <mailto:tacherasasi@gmail.com>>
>      >
>      > Add a new --json flag to 'git status' that outputs repository state
>      > in a structured JSON format. This enables reliable machine parsing
>      > of status information for tools and automation.
>      >
>      > The JSON output includes:
>      > - Branch information (name, detached state, ahead/behind counts)
>      > - Staged files array
>      > - Unstaged files array
>      > - Untracked files array
>      > - Ignored files array (with --ignored flag)
>      >
>      > Implementation details:
>      > - Add STATUS_FORMAT_JSON to wt_status_format enum
>      > - Add JSON output option to git status and git commit
>      > - Implement JSON formatting helpers for arrays and branch info
>      > - Structure output for easy parsing and future extensibility
>      >
>      > Example:
>      >    $ git status --json
>      >    {
>      >      "branch": {
>      >        "current": "main",
>      >        "detached": false
>      >      },
>      >      "staged": ["file1.txt"],
>      >      "unstaged": ["file2.txt"],
>      >      "untracked": ["file3.txt"]
>      >    }
>      >
>      > This provides a robust alternative to parsing traditional output
>      > formats, making it easier to build reliable tools and automation
>      > around Git status information.
> 
>     What problem are you having paring the current output? You say this
>     patch provides a robust alternative but I'm afraid it does not seem to
>     handle filenames containing newlines or double quotes and it assumes
>     they are utf-8 encoded. It also seems to print a trailing comma if an
>     array has more than one item which is not allowed by the json spec. We
>     already have support for generating json output as documented
>     json-writer.h which will help with most of that but not the filename
>     encoding problem. New features should also be accompanied by tests to
>     prevent future regressions. I'm not opposed to adding json output if it
>     is robust and handles non utf-8 filenames but it would be helpful to
>     understand what problems you are having parsing the output of "git
>     status --porcelain[=v2]".
> 
>     Thanks
> 
>     Phillip
> 
>      > Signed-off-by: Tachera Sasi <tachera@ekilie.com
>     <mailto:tachera@ekilie.com>>
>      > Signed-off-by: tacherasasi <tacherasasi@gmail.com
>     <mailto:tacherasasi@gmail.com>>
>      > ---
>      >      status: add --json output format to git status
>      >
>      > Published-As: https://github.com/gitgitgadget/git/releases/tag/
>     pr-1937%2FtacheraSasi%2Ffeature%2Fjson-status-output-v1 <https://
>     github.com/gitgitgadget/git/releases/tag/
>     pr-1937%2FtacheraSasi%2Ffeature%2Fjson-status-output-v1>
>      > Fetch-It-Via: git fetch https://github.com/gitgitgadget/git
>     <https://github.com/gitgitgadget/git> pr-1937/tacheraSasi/feature/
>     json-status-output-v1
>      > Pull-Request: https://github.com/gitgitgadget/git/pull/1937
>     <https://github.com/gitgitgadget/git/pull/1937>
>      >
>      >   builtin/commit.c |  9 ++++-
>      >   wt-status.c      | 98 +++++++++++++++++++++++++++++++++++++++++
>     +++++++
>      >   wt-status.h      |  1 +
>      >   3 files changed, 107 insertions(+), 1 deletion(-)
>      >
>      > diff --git a/builtin/commit.c b/builtin/commit.c
>      > index fba0dded64a..f1db4fdfd9a 100644
>      > --- a/builtin/commit.c
>      > +++ b/builtin/commit.c
>      > @@ -1540,6 +1540,9 @@ struct repository *repo UNUSED)
>      >               OPT_SET_INT(0, "long", &status_format,
>      >                           N_("show status in long format (default)"),
>      >                           STATUS_FORMAT_LONG),
>      > +             OPT_SET_INT(0, "json", &status_format,
>      > +                 N_("show status in JSON format"),
>      > +                 STATUS_FORMAT_JSON),
>      >               OPT_BOOL('z', "null", &s.null_termination,
>      >                        N_("terminate entries with NUL")),
>      >               {
>      > @@ -1603,7 +1606,8 @@ struct repository *repo UNUSED)
>      >                      prefix, argv);
>      >
>      >       if (status_format != STATUS_FORMAT_PORCELAIN &&
>      > -         status_format != STATUS_FORMAT_PORCELAIN_V2)
>      > +         status_format != STATUS_FORMAT_PORCELAIN_V2 &&
>      > +         status_format != STATUS_FORMAT_JSON)
>      >               progress_flag = REFRESH_PROGRESS;
>      >       repo_read_index(the_repository);
>      >       refresh_index(the_repository->index,
>      > @@ -1735,6 +1739,9 @@ int cmd_commit(int argc,
>      >               OPT_SET_INT(0, "long", &status_format,
>      >                           N_("show status in long format (default)"),
>      >                           STATUS_FORMAT_LONG),
>      > +             OPT_SET_INT(0, "json", &status_format,
>      > +                 N_("show status in JSON format"),
>      > +                 STATUS_FORMAT_JSON),
>      >               OPT_BOOL('z', "null", &s.null_termination,
>      >                        N_("terminate entries with NUL")),
>      >               OPT_BOOL(0, "amend", &amend, N_("amend previous
>     commit")),
>      > diff --git a/wt-status.c b/wt-status.c
>      > index 454601afa15..7192fb4d057 100644
>      > --- a/wt-status.c
>      > +++ b/wt-status.c
>      > @@ -2564,6 +2564,101 @@ static void wt_porcelain_v2_print(struct
>     wt_status *s)
>      >       }
>      >   }
>      >
>      > +
>      > +static void wt_json_print_string_array(struct wt_status *s,
>     const char *name, struct string_list *list)
>      > +{
>      > +     int i;
>      > +     fprintf(s->fp, "  \"%s\": [", name);
>      > +     for (i = 0; i < list->nr; i++) {
>      > +             if (i > 0)
>      > +                     fprintf(s->fp, ", ");
>      > +             fprintf(s->fp, "\"%s\"", list->items[i].string);
>      > +     }
>      > +     fprintf(s->fp, "]");
>      > +}
>      > +
>      > +static void wt_json_print_change_array(struct wt_status *s,
>     const char *name, int change_type)
>      > +{
>      > +     int i;
>      > +     struct string_list files = STRING_LIST_INIT_DUP;
>      > +
>      > +     for (i = 0; i < s->change.nr <http://change.nr>; i++) {
>      > +             struct wt_status_change_data *d;
>      > +             struct string_list_item *it;
>      > +             it = &(s->change.items[i]);
>      > +             d = it->util;
>      > +
>      > +             if ((change_type == WT_STATUS_UPDATED && d-
>      >index_status &&
>      > +                  d->index_status != DIFF_STATUS_UNMERGED) ||
>      > +                 (change_type == WT_STATUS_CHANGED && d-
>      >worktree_status &&
>      > +                  d->worktree_status != DIFF_STATUS_UNMERGED)) {
>      > +                     string_list_append(&files, it->string);
>      > +             }
>      > +     }
>      > +
>      > +     wt_json_print_string_array(s, name, &files);
>      > +     string_list_clear(&files, 0);
>      > +}
>      > +
>      > +static void wt_json_print_branch_info(struct wt_status *s)
>      > +{
>      > +     struct branch *branch;
>      > +     const char *branch_name;
>      > +     int ahead = 0, behind = 0;
>      > +
>      > +     fprintf(s->fp, "  \"branch\": {\n");
>      > +
>      > +     if (s->branch && !s->is_initial) {
>      > +             if (!strcmp(s->branch, "HEAD")) {
>      > +                     fprintf(s->fp, "    \"current\": \"HEAD\",\n");
>      > +                     fprintf(s->fp, "    \"detached\": true");
>      > +             } else {
>      > +                     if (skip_prefix(s->branch, "refs/heads/",
>     &branch_name)) {
>      > +                             fprintf(s->fp, "    \"current\":
>     \"%s\",\n", branch_name);
>      > +                             fprintf(s->fp, "    \"detached\":
>     false");
>      > +
>      > +                             branch = branch_get(branch_name);
>      > +                             if (branch && branch->merge &&
>     branch->merge[0] && branch->merge[0]->dst) {
>      > +                                     if (!
>     stat_tracking_info(branch, &ahead, &behind, NULL, 0, 0)) {
>      > +                                             fprintf(s->fp, ",
>     \n    \"ahead\": %d,\n    \"behind\": %d", ahead, behind);
>      > +                                     }
>      > +                             }
>      > +                     } else {
>      > +                             fprintf(s->fp, "    \"current\":
>     \"%s\",\n", s->branch);
>      > +                             fprintf(s->fp, "    \"detached\":
>     false");
>      > +                     }
>      > +             }
>      > +     } else {
>      > +             fprintf(s->fp, "    \"current\": null,\n");
>      > +             fprintf(s->fp, "    \"detached\": false");
>      > +     }
>      > +
>      > +     fprintf(s->fp, "\n  }");
>      > +}
>      > +
>      > +static void wt_json_status_print(struct wt_status *s)
>      > +{
>      > +     fprintf(s->fp, "{\n");
>      > +
>      > +     wt_json_print_branch_info(s);
>      > +     fprintf(s->fp, ",\n");
>      > +
>      > +     wt_json_print_change_array(s, "staged", WT_STATUS_UPDATED);
>      > +     fprintf(s->fp, ",\n");
>      > +
>      > +     wt_json_print_change_array(s, "unstaged", WT_STATUS_CHANGED);
>      > +     fprintf(s->fp, ",\n");
>      > +
>      > +     wt_json_print_string_array(s, "untracked", &s->untracked);
>      > +
>      > +     if (s->ignored.nr <http://ignored.nr> > 0) {
>      > +             fprintf(s->fp, ",\n");
>      > +             wt_json_print_string_array(s, "ignored", &s->ignored);
>      > +     }
>      > +
>      > +     fprintf(s->fp, "\n}\n");
>      > +}
>      > +
>      >   void wt_status_print(struct wt_status *s)
>      >   {
>      >       trace2_data_intmax("status", s->repo, "count/changed", s-
>      >change.nr <http://change.nr>);
>      > @@ -2583,6 +2678,9 @@ void wt_status_print(struct wt_status *s)
>      >       case STATUS_FORMAT_PORCELAIN_V2:
>      >               wt_porcelain_v2_print(s);
>      >               break;
>      > +     case STATUS_FORMAT_JSON:
>      > +             wt_json_status_print(s);
>      > +             break;
>      >       case STATUS_FORMAT_UNSPECIFIED:
>      >               BUG("finalize_deferred_config() should have been
>     called");
>      >               break;
>      > diff --git a/wt-status.h b/wt-status.h
>      > index 4e377ce62b8..e929af832b2 100644
>      > --- a/wt-status.h
>      > +++ b/wt-status.h
>      > @@ -74,6 +74,7 @@ enum wt_status_format {
>      >       STATUS_FORMAT_SHORT,
>      >       STATUS_FORMAT_PORCELAIN,
>      >       STATUS_FORMAT_PORCELAIN_V2,
>      > +     STATUS_FORMAT_JSON,
>      >
>      >       STATUS_FORMAT_UNSPECIFIED
>      >   };
>      >
>      > base-commit: cb3b40381e1d5ee32dde96521ad7cfd68eb308a6
> 

← back to recent threads