git/list[1] front-page[2] threads[3] people[4] search[5] about
 

Re: [PATCH] builtin/describe: introduce --broken flag

From
Junio C Hamano <gitster@pobox.com>
Date
Mar 21, 2017, 21:51 UTC
Message-ID
<xmqq7f3invr6.fsf@gitster.mtv.corp.google.com>
In-Reply-To
<20170321185139.8300-1-sbeller@google.com>
Stefan Beller <sbeller@google.com> writes:
> This patch helps to fix the root cause in [1], which tries to work around
> this situation.

I do not necessarily think it is reasonable to give $version-dirty and proceed when a repository corruption is detected; if there is a breakage in the repository, "git describe" is correct to die when a populated submodule is broken. IOW, I do not agree that [1] below is working with a sensible expectation.

This is a tangent, but how does the Gerrit repository get corrupted in the way described in [1] in the first place? That might be what needs to be corrected, perhaps?

Show 19 quoted lines
>  Documentation/git-describe.txt |  7 +++++
>  builtin/describe.c             | 59 ++++++++++++++++++++++++++++++------------
>  t/t6120-describe.sh            | 20 ++++++++++++++
>  3 files changed, 70 insertions(+), 16 deletions(-)
>
> diff --git a/Documentation/git-describe.txt b/Documentation/git-describe.txt
> index 8755f3af7b..b71fa7a4ad 100644
> --- a/Documentation/git-describe.txt
> +++ b/Documentation/git-describe.txt
> @@ -34,6 +34,13 @@ OPTIONS
>  	It means describe HEAD and appends <mark> (`-dirty` by
>  	default) if the working tree is dirty.
>  
> +--broken[=<mark>]::
> +	Describe the working tree.
> +	It means describe HEAD and appends <mark> (`-broken` by
> +	default) if the working tree cannot be examined for dirtiness.
> +	This implies `--dirty`, which is the fallback behavior when
> +	the working tree examination works.

The wording for the "--dirty" is already awkward, but this one is even more so ("Describe the working tree. It means" conveys no useful information). I however cannot come up with something much better. This is the best I could come up with:

	Describe the state of the working tree.  When the working
	tree matches HEAD, the output is the same as "git describe
	HEAD" and "-dirty" is appended to it if the working tree has
	local modification.  When a repository is corrupt and Git
	cannot determine if there is local modification, instead of
	dying, append "-broken" instead.

The last sentence can be tweaked to replace the description for "--dirty".

Show 23 quoted lines
> diff --git a/builtin/describe.c b/builtin/describe.c
> index 76c18059bf..37a83520c9 100644
> --- a/builtin/describe.c
> +++ b/builtin/describe.c
> @@ -9,6 +9,7 @@
>  #include "diff.h"
>  #include "hashmap.h"
>  #include "argv-array.h"
> +#include "run-command.h"
>  
>  #define SEEN		(1u << 0)
>  #define MAX_TAGS	(FLAG_BITS - 1)
> @@ -31,12 +32,7 @@ static int have_util;
>  static struct string_list patterns = STRING_LIST_INIT_NODUP;
>  static struct string_list exclude_patterns = STRING_LIST_INIT_NODUP;
>  static int always;
> -static const char *dirty;
> -
> -/* diff-index command arguments to check if working tree is dirty. */
> -static const char *diff_index_args[] = {
> -	"diff-index", "--quiet", "HEAD", "--", NULL
> -};
> +static const char *append, *dirty, *broken;
Perhaps call it "suffix" or something?
Show 11 quoted lines
> +		if (broken) {
> +			struct child_process cp = CHILD_PROCESS_INIT;
> +			argv_array_pushl(&cp.args, "diff-index", "--quiet", "HEAD", "--", NULL);
> ...
> +			}
> +		} else if (dirty) {
> +			struct argv_array args = ARGV_ARRAY_INIT;
>  			static struct lock_file index_lock;
>  			int fd;
>  
> +			argv_array_pushl(&args, "diff-index", "--quiet", "HEAD", "--", NULL);

Somehow it looks like the patch is making the code a lot worse by losing diff_index_args[] and duplicating the same command line twice in the code. Wouldn't argv_array_pushv() into these two different args array from the same template work better?

Previous: Stefan BellerNext: Stefan Beller
Message 8 of 15 in “git-describe deals gracefully with broken submodules”
  1. 0/3 git-describe deals gracefully with broken submodulesStefan Beller, Mar 21, 2017
  2. 1/3 submodule.c: port is_submodule_modified to use porcelain 2Stefan Beller, Mar 21, 2017
  3. 3/3 builtin/describe: introduce --submodule-error-as-dirty flagStefan Beller, Mar 21, 2017
  4. 2/3 revision machinery: gentle submodule errorsStefan Beller, Mar 21, 2017
  5. Junio C HamanoMar 21, 2017
  6. Junio C HamanoMar 21, 2017
  7. builtin/describe: introduce --broken flagStefan Beller, Mar 21, 2017
  8. Junio C HamanoMar 21, 2017
  9. Stefan BellerMar 21, 2017
  10. Junio C HamanoMar 21, 2017
  11. Stefan BellerMar 21, 2017
  12. builtin/describe: introduce --broken flagStefan Beller, Mar 21, 2017
  13. Junio C HamanoMar 22, 2017
  14. Jakub NarębskiMar 22, 2017
  15. Stefan BellerMar 21, 2017

Read the whole thread, see it on lore, or plain text.

$ cat FOOTERMessages come from the public archive at lore.kernel.org/git, fetched every hour. The front page is chosen and written each morning by an AI editor and can be wrong; the threads themselves are the record. About and API. For agents: an MCP server at https://gitlist.dev/mcp, and any thread, story or person page as Markdown by adding .md to its URL (or sending Accept: text/markdown). Details in /llms.txt.