diff mbox series

[v2] ls-files: introduce "--format" option

Message ID pull.1262.v2.git.1655629990185.gitgitgadget@gmail.com (mailing list archive)
State Superseded
Headers show
Series [v2] ls-files: introduce "--format" option | expand

Commit Message

ZheNing Hu June 19, 2022, 9:13 a.m. UTC
From: ZheNing Hu <adlternative@gmail.com>

Add a new option --format that output index enties
informations with custom format, taking inspiration
from the option with the same name in the `git ls-tree`
command.

--format cannot used with -s, -o, -k, --resolve-undo,
--deduplicate and --debug.

Signed-off-by: ZheNing Hu <adlternative@gmail.com>
---
    ls-files: introduce "--format" options
    
    v1->v2:
    
     1. do some code style fix suggected by Ævar.
     2. remove --object-only option (I have tried to use fast path for it,
        but cannot see any performance promote compare with
        --format=%(objectname))

Published-As: https://github.com/gitgitgadget/git/releases/tag/pr-1262%2Fadlternative%2Fzh%2Fls-file-format-v2
Fetch-It-Via: git fetch https://github.com/gitgitgadget/git pr-1262/adlternative/zh/ls-file-format-v2
Pull-Request: https://github.com/gitgitgadget/git/pull/1262

Range-diff vs v1:

 1:  432d80b8c78 ! 1:  67f2c3b8ebe ls-files: introduce "--format" option
     @@ Commit message
          command.
      
          --format cannot used with -s, -o, -k, --resolve-undo,
     -    --deduplicate, --debug.
     +    --deduplicate and --debug.
      
          Signed-off-by: ZheNing Hu <adlternative@gmail.com>
      
     @@ Documentation/git-ls-files.txt: followed by the  ("attr/<eolattr>").
       	trailing slash, such as "x/" for a sparse directory "x".
       
      +--format=<format>::
     -+	A string that interpolates %(fieldname) from the result being shown.
     ++	A string that interpolates `%(fieldname)` from the result being shown.
      +	It also interpolates `%%` to `%`, and `%xx` where `xx` are hex digits
      +	interpolates to character with hex code `xx`; for example `%00`
      +	interpolates to `\0` (NUL), `%09` to `\t` (TAB) and %0a to `\n` (LF).
     -+	--format cannot be combined with `-s`, `-o`, `-k`, `--resolve-undo`,
     ++	--format cannot be combined with `-s`, `-o`, `-k`, `--resolve-undo` and
      +	`--debug`.
       \--::
       	Do not interpret any more arguments as options.
     @@ builtin/ls-files.c: static void write_name(const char *name)
       
      +static void write_name_to_buf(struct strbuf *sb, const char *name)
      +{
     -+	name = relative_path(name, prefix_len ? prefix : NULL, sb);
     -+	if (line_terminator) {
     -+		quote_c_style(name, sb, NULL, 0);
     -+	} else {
     -+		strbuf_add(sb, name, strlen(name));
     -+	}
     ++	const char *rel = relative_path(name, prefix_len ? prefix : NULL, sb);
     ++	if (line_terminator)
     ++		quote_c_style(rel, sb, NULL, 0);
     ++	else
     ++		strbuf_add(sb, rel, strlen(rel));
      +}
      +
       static const char *get_tag(const struct cache_entry *ce, const char *tag)
     @@ builtin/ls-files.c: static void show_submodule(struct repository *superproject,
      +	if (len)
      +		return len;
      +	if (*start != '(')
     -+		die(_("bad ls-files format: element '%s' does not start with '('"), start);
     ++		die(_("bad ls-files format: element '%s' "
     ++		      "does not start with '('"), start);
      +
      +	end = strchr(start + 1, ')');
      +	if (!end)
     -+		die(_("bad ls-files format: element '%s' does not end in ')'"), start);
     ++		die(_("bad ls-files format: element '%s'"
     ++		      "does not end in ')'"), start);
      +
      +	len = end - start + 1;
     -+	if (skip_prefix(start, "(tag)", &p)) {
     ++	if (skip_prefix(start, "(tag)", &p))
      +		strbuf_addstr(sb, get_tag(data->ce, data->tag));
     -+	} else if (skip_prefix(start, "(objectmode)", &p)) {
     ++	else if (skip_prefix(start, "(objectmode)", &p))
      +		strbuf_addf(sb, "%06o", data->ce->ce_mode);
     -+	} else if (skip_prefix(start, "(objectname)", &p)) {
     ++	else if (skip_prefix(start, "(objectname)", &p))
      +		strbuf_add_unique_abbrev(sb, &data->ce->oid, abbrev);
     -+	} else if (skip_prefix(start, "(stage)", &p)) {
     ++	else if (skip_prefix(start, "(stage)", &p))
      +		strbuf_addf(sb, "%d", ce_stage(data->ce));
     -+	} else if (skip_prefix(start, "(eol)", &p)) {
     -+		write_eolinfo_to_buf(sb, data->istate, data->ce, data->pathname);
     -+	} else if (skip_prefix(start, "(path)", &p)) {
     ++	else if (skip_prefix(start, "(eol)", &p))
     ++		write_eolinfo_to_buf(sb, data->istate,
     ++				     data->ce, data->pathname);
     ++	else if (skip_prefix(start, "(path)", &p))
      +		write_name_to_buf(sb, data->pathname);
     -+	} else if (skip_prefix(start, "(ctime)", &p)) {
     -+		strbuf_addf(sb, "ctime: %u:%u", sd->sd_ctime.sec, sd->sd_ctime.nsec);
     -+	} else if (skip_prefix(start, "(mtime)", &p)) {
     -+		strbuf_addf(sb, "mtime: %u:%u", sd->sd_mtime.sec, sd->sd_mtime.nsec);
     -+	} else if (skip_prefix(start, "(dev)", &p)) {
     ++	else if (skip_prefix(start, "(ctime)", &p))
     ++		strbuf_addf(sb, "ctime: %u:%u",
     ++			    sd->sd_ctime.sec, sd->sd_ctime.nsec);
     ++	else if (skip_prefix(start, "(mtime)", &p))
     ++		strbuf_addf(sb, "mtime: %u:%u",
     ++			    sd->sd_mtime.sec, sd->sd_mtime.nsec);
     ++	else if (skip_prefix(start, "(dev)", &p))
      +		strbuf_addf(sb, "dev: %u", sd->sd_dev);
     -+	} else if (skip_prefix(start, "(ino)", &p)) {
     ++	else if (skip_prefix(start, "(ino)", &p))
      +		strbuf_addf(sb, "ino: %u", sd->sd_ino);
     -+	} else if (skip_prefix(start, "(uid)", &p)) {
     ++	else if (skip_prefix(start, "(uid)", &p))
      +		strbuf_addf(sb, "uid: %u", sd->sd_uid);
     -+	} else if (skip_prefix(start, "(gid)", &p)) {
     ++	else if (skip_prefix(start, "(gid)", &p))
      +		strbuf_addf(sb, "gid: %u", sd->sd_gid);
     -+	} else if (skip_prefix(start, "(size)", &p)) {
     ++	else if (skip_prefix(start, "(size)", &p))
      +		strbuf_addf(sb, "size: %u", sd->sd_size);
     -+	} else if (skip_prefix(start, "(flags)", &p)) {
     ++	else if (skip_prefix(start, "(flags)", &p))
      +		strbuf_addf(sb, "flags: %x", data->ce->ce_flags);
     -+	} else {
     ++	else {
      +		errlen = (unsigned long)len;
      +		die(_("bad ls-files format: %%%.*s"), errlen, start);
      +	}
     @@ builtin/ls-files.c: int cmd_ls_files(int argc, const char **argv, const char *cm
       		OPT_BOOL(0, "sparse", &show_sparse_dirs,
       			 N_("show sparse directories in the presence of a sparse index")),
      +		OPT_STRING_F(0, "format", &format, N_("format"),
     -+					 N_("format to use for the output"),
     -+					 PARSE_OPT_NONEG),
     ++			     N_("format to use for the output"),
     ++			     PARSE_OPT_NONEG),
       		OPT_END()
       	};
       	int ret = 0;
     @@ t/t3013-ls-files-format.sh (new)
      +'
      +
      +test_expect_success 'git ls-files --format objectmode' '
     -+	cat >expect <<-EOF &&
     ++	cat >expect <<-\EOF &&
      +	100755
      +	100644
      +	EOF
     @@ t/t3013-ls-files-format.sh (new)
      +'
      +
      +test_expect_success 'git ls-files --format path' '
     -+	cat >expect <<-EOF &&
     ++	cat >expect <<-\EOF &&
      +	o1
      +	o2
      +	EOF
     @@ t/t3013-ls-files-format.sh (new)
      +'
      +
      +test_expect_success 'git ls-files --format ctime' '
     -+	git ls-files --debug | grep ctime >expect &&
     ++	git ls-files --debug >out &&
     ++	grep ctime out >expect &&
      +	git ls-files --format="  %(ctime)" >actual &&
      +	test_cmp expect actual
      +'
      +
      +test_expect_success 'git ls-files --format mtime' '
     -+	git ls-files --debug | grep mtime >expect &&
     ++	git ls-files --debug >out &&
     ++	grep mtime out >expect &&
      +	git ls-files --format="  %(mtime)" >actual &&
      +	test_cmp expect actual
      +'
      +
      +test_expect_success 'git ls-files --format dev and ino' '
     -+	git ls-files --debug | grep dev >expect &&
     ++	git ls-files --debug >out &&
     ++	grep dev out >expect &&
      +	git ls-files --format="  %(dev)%x09%(ino)" >actual &&
      +	test_cmp expect actual
      +'
      +
      +test_expect_success 'git ls-files --format uid and gid' '
     -+	git ls-files --debug | grep uid >expect &&
     ++	git ls-files --debug >out &&
     ++	grep uid out >expect &&
      +	git ls-files --format="  %(uid)%x09%(gid)" >actual &&
      +	test_cmp expect actual
      +'
      +
      +test_expect_success 'git ls-files --format with -m' '
      +	echo change >o1 &&
     -+	cat >expect <<-EOF &&
     ++	cat >expect <<-\EOF &&
      +	o1
      +	EOF
      +	git ls-files --format="%(path)" -m >actual &&
     @@ t/t3013-ls-files-format.sh (new)
      +'
      +
      +test_expect_success 'git ls-files --format with -d' '
     -+	rm o1 &&
     -+	test_when_finished "git restore o1" &&
     -+	cat >expect <<-EOF &&
     -+	o1
     ++	echo o3 >o3 &&
     ++	git add o3 &&
     ++	rm o3 &&
     ++	cat >expect <<-\EOF &&
     ++	o3
      +	EOF
      +	git ls-files --format="%(path)" -d >actual &&
      +	test_cmp expect actual
      +'
      +
      +test_expect_success 'git ls-files --format size and flags' '
     -+	git ls-files --debug | grep size >expect &&
     ++	git ls-files --debug >out &&
     ++	grep size out >expect &&
      +	git ls-files --format="  %(size)%x09%(flags)" >actual &&
      +	test_cmp expect actual
      +'
     @@ t/t3013-ls-files-format.sh (new)
      +	test_cmp expect actual
      +'
      +
     -+test_expect_success 'git ls-files --format with -s must fail' '
     -+	test_must_fail git ls-files --format="%(objectname)" -s
     -+'
     -+
     -+test_expect_success 'git ls-files --format with -o must fail' '
     -+	test_must_fail git ls-files --format="%(objectname)" -o
     -+'
     -+
     -+test_expect_success 'git ls-files --format with -k must fail' '
     -+	test_must_fail git ls-files --format="%(objectname)" -k
     -+'
     -+
     -+test_expect_success 'git ls-files --format with --resolve-undo must fail' '
     -+	test_must_fail git ls-files --format="%(objectname)" --resolve-undo
     -+'
     -+
     -+test_expect_success 'git ls-files --format with --deduplicate must fail' '
     -+	test_must_fail git ls-files --format="%(objectname)" --deduplicate
     -+'
     -+
     -+test_expect_success 'git ls-files --format with --debug must fail' '
     -+	test_must_fail git ls-files --format="%(objectname)" --debug
     -+'
     -+
     ++for flag in -s -o -k --resolve-undo --deduplicate --debug
     ++do
     ++	test_expect_success "git ls-files --format is incompatible with $flag" '
     ++		test_must_fail git ls-files --format="%(objectname)" $flag
     ++	'
     ++done
      +test_done
 2:  81ae1280e8e < -:  ----------- ls-files: introduce "--object-only" option


 Documentation/git-ls-files.txt |  51 ++++++++++++-
 builtin/ls-files.c             | 130 ++++++++++++++++++++++++++++++++-
 t/t3013-ls-files-format.sh     | 130 +++++++++++++++++++++++++++++++++
 3 files changed, 307 insertions(+), 4 deletions(-)
 create mode 100755 t/t3013-ls-files-format.sh


base-commit: ab336e8f1c8009c8b1aab8deb592148e69217085

Comments

Phillip Wood June 19, 2022, 1:50 p.m. UTC | #1
Hi ZheNing

On 19/06/2022 10:13, ZheNing Hu via GitGitGadget wrote:
> From: ZheNing Hu <adlternative@gmail.com>
> 
> Add a new option --format that output index enties
> informations with custom format, taking inspiration
> from the option with the same name in the `git ls-tree`
> command.
> 
> --format cannot used with -s, -o, -k, --resolve-undo,
> --deduplicate and --debug.

I think this is an interesting feature that provides functionality that 
is not available by feeding index entries into cat-file.

> Signed-off-by: ZheNing Hu <adlternative@gmail.com>
>   Documentation/git-ls-files.txt |  51 ++++++++++++-
>   builtin/ls-files.c             | 130 ++++++++++++++++++++++++++++++++-
>   t/t3013-ls-files-format.sh     | 130 +++++++++++++++++++++++++++++++++
>   3 files changed, 307 insertions(+), 4 deletions(-)
>   create mode 100755 t/t3013-ls-files-format.sh
> 
> diff --git a/Documentation/git-ls-files.txt b/Documentation/git-ls-files.txt
> index 0dabf3f0ddc..9a88c92f1ad 100644
> --- a/Documentation/git-ls-files.txt
> +++ b/Documentation/git-ls-files.txt
> @@ -20,7 +20,7 @@ SYNOPSIS
>   		[--exclude-standard]
>   		[--error-unmatch] [--with-tree=<tree-ish>]
>   		[--full-name] [--recurse-submodules]
> -		[--abbrev[=<n>]] [--] [<file>...]
> +		[--abbrev[=<n>]] [--format=<format>] [--] [<file>...]
>   
>   DESCRIPTION
>   -----------
> @@ -192,6 +192,13 @@ followed by the  ("attr/<eolattr>").
>   	to the contained files. Sparse directories will be shown with a
>   	trailing slash, such as "x/" for a sparse directory "x".
>   
> +--format=<format>::
> +	A string that interpolates `%(fieldname)` from the result being shown.
> +	It also interpolates `%%` to `%`, and `%xx` where `xx` are hex digits
> +	interpolates to character with hex code `xx`; for example `%00`
> +	interpolates to `\0` (NUL), `%09` to `\t` (TAB) and %0a to `\n` (LF).
> +	--format cannot be combined with `-s`, `-o`, `-k`, `--resolve-undo` and
> +	`--debug`.
>   \--::
>   	Do not interpret any more arguments as options.
>   
> @@ -223,6 +230,48 @@ quoted as explained for the configuration variable `core.quotePath`
>   (see linkgit:git-config[1]).  Using `-z` the filename is output
>   verbatim and the line is terminated by a NUL byte.
>   
> +It is possible to print in a custom format by using the `--format`
> +option, which is able to interpolate different fields using
> +a `%(fieldname)` notation. For example, if you only care about the
> +"objectname" and "path" fields, you can execute with a specific
> +"--format" like
> +
> +	git ls-files --format='%(objectname) %(path)'
> +
> +FIELD NAMES
> +-----------
> +Various values from structured fields can be used to interpolate
> +into the resulting output. For each outputting line, the following
> +names can be used:
> +
> +tag::
> +	The tag of file status.

The documentation for -t strong discourages its use, so I wonder if we 
really want to expose it here.

> +objectmode::
> +	The mode of the object.
> +objectname::
> +	The name of the object.
> +stage::
> +	The stage of the file.
> +eol::
> +	The line endings of files.

Every other option refers to either a "file" or "object" but here we 
have "files". Looking at the implementation below this will print the 
line ending from both the index and the worktree, it would be useful to 
clarify that here.

> +path::
> +	The pathname of the object.
> +ctime::
> +	The create time of file.

It is not clear from this whether this (and all the file attributes 
below) are coming from the worktree or the index or both like eol?

> +mtime::
> +	The modify time of file.
> +dev::
> +	The ID of device containing file.
> +ino::
> +	The inode number of file.
> +uid::
> +	The user id of file owner.
> +gid::
> +	The group id of file owner.
> +size::
> +	The size of the file.
> +flags::
> +	The flags of the file.

What are the flags?

> [...]  
> +static size_t expand_show_index(struct strbuf *sb, const char *start,
> +			       void *context)
> +{
> +	struct show_index_data *data = context;
> +	const char *end;
> +	const char *p;
> +	unsigned int errlen;
 > [...]
> +	else if (skip_prefix(start, "(flags)", &p))
> +		strbuf_addf(sb, "flags: %x", data->ce->ce_flags);
> +	else {
> +		errlen = (unsigned long)len;
> +		die(_("bad ls-files format: %%%.*s"), errlen, start);

errlen is declared as an unsigned int, but you cast len which is a 
size_t to unsigned long when assigning to errlen. Then errlen is used 
where a signed int is required by die. There is also a style violation 
as if any branch of an if needs braces then they should all be braced. I 
think that the best solution would be to drop errlen and just write

	else
		die(_("bad ls-files format: %%%.*s"), (int)len, start);

It would be interesting to check the performance of this implementation 
on a large repository as it is doing a lot of branching inside a loop. I 
don't think we should change it unless it turns out to be a problem. 
Then we could try switching on the first character of the format 
specifier or some other optimization.

Best Wishes

Phillip
ZheNing Hu June 20, 2022, 1:32 p.m. UTC | #2
Phillip Wood <phillip.wood123@gmail.com> 于2022年6月19日周日 21:50写道:
>
> Hi ZheNing
>
> On 19/06/2022 10:13, ZheNing Hu via GitGitGadget wrote:
> > From: ZheNing Hu <adlternative@gmail.com>
> >
> > Add a new option --format that output index enties
> > informations with custom format, taking inspiration
> > from the option with the same name in the `git ls-tree`
> > command.
> >
> > --format cannot used with -s, -o, -k, --resolve-undo,
> > --deduplicate and --debug.
>
> I think this is an interesting feature that provides functionality that
> is not available by feeding index entries into cat-file.
>

Yeah, it cares about index state. Having this feature, maybe we can
easier check index/work-tree state.

> > Signed-off-by: ZheNing Hu <adlternative@gmail.com>
> >   Documentation/git-ls-files.txt |  51 ++++++++++++-
> >   builtin/ls-files.c             | 130 ++++++++++++++++++++++++++++++++-
> >   t/t3013-ls-files-format.sh     | 130 +++++++++++++++++++++++++++++++++
> >   3 files changed, 307 insertions(+), 4 deletions(-)
> >   create mode 100755 t/t3013-ls-files-format.sh
> >
> > diff --git a/Documentation/git-ls-files.txt b/Documentation/git-ls-files.txt
> > index 0dabf3f0ddc..9a88c92f1ad 100644
> > --- a/Documentation/git-ls-files.txt
> > +++ b/Documentation/git-ls-files.txt
> > @@ -20,7 +20,7 @@ SYNOPSIS
> >               [--exclude-standard]
> >               [--error-unmatch] [--with-tree=<tree-ish>]
> >               [--full-name] [--recurse-submodules]
> > -             [--abbrev[=<n>]] [--] [<file>...]
> > +             [--abbrev[=<n>]] [--format=<format>] [--] [<file>...]
> >
> >   DESCRIPTION
> >   -----------
> > @@ -192,6 +192,13 @@ followed by the  ("attr/<eolattr>").
> >       to the contained files. Sparse directories will be shown with a
> >       trailing slash, such as "x/" for a sparse directory "x".
> >
> > +--format=<format>::
> > +     A string that interpolates `%(fieldname)` from the result being shown.
> > +     It also interpolates `%%` to `%`, and `%xx` where `xx` are hex digits
> > +     interpolates to character with hex code `xx`; for example `%00`
> > +     interpolates to `\0` (NUL), `%09` to `\t` (TAB) and %0a to `\n` (LF).
> > +     --format cannot be combined with `-s`, `-o`, `-k`, `--resolve-undo` and
> > +     `--debug`.
> >   \--::
> >       Do not interpret any more arguments as options.
> >
> > @@ -223,6 +230,48 @@ quoted as explained for the configuration variable `core.quotePath`
> >   (see linkgit:git-config[1]).  Using `-z` the filename is output
> >   verbatim and the line is terminated by a NUL byte.
> >
> > +It is possible to print in a custom format by using the `--format`
> > +option, which is able to interpolate different fields using
> > +a `%(fieldname)` notation. For example, if you only care about the
> > +"objectname" and "path" fields, you can execute with a specific
> > +"--format" like
> > +
> > +     git ls-files --format='%(objectname) %(path)'
> > +
> > +FIELD NAMES
> > +-----------
> > +Various values from structured fields can be used to interpolate
> > +into the resulting output. For each outputting line, the following
> > +names can be used:
> > +
> > +tag::
> > +     The tag of file status.
>
> The documentation for -t strong discourages its use, so I wonder if we
> really want to expose it here.
>

I think it's ok to remove it.

> > +objectmode::
> > +     The mode of the object.
> > +objectname::
> > +     The name of the object.
> > +stage::
> > +     The stage of the file.
> > +eol::
> > +     The line endings of files.
>
> Every other option refers to either a "file" or "object" but here we
> have "files". Looking at the implementation below this will print the
> line ending from both the index and the worktree, it would be useful to
> clarify that here.
>

Sure...

> > +path::
> > +     The pathname of the object.
> > +ctime::
> > +     The create time of file.
>
> It is not clear from this whether this (and all the file attributes
> below) are coming from the worktree or the index or both like eol?
>

...I think they are basically index cache_entry attributes, except eol
cares about both
worktree and index. I will fix them.

> > +mtime::
> > +     The modify time of file.
> > +dev::
> > +     The ID of device containing file.
> > +ino::
> > +     The inode number of file.
> > +uid::
> > +     The user id of file owner.
> > +gid::
> > +     The group id of file owner.
> > +size::
> > +     The size of the file.
> > +flags::
> > +     The flags of the file.
>
> What are the flags?
>

It is cache entry flags which include In-memory only flags and some
extended on-disk flags.

> > [...]
> > +static size_t expand_show_index(struct strbuf *sb, const char *start,
> > +                            void *context)
> > +{
> > +     struct show_index_data *data = context;
> > +     const char *end;
> > +     const char *p;
> > +     unsigned int errlen;
>  > [...]
> > +     else if (skip_prefix(start, "(flags)", &p))
> > +             strbuf_addf(sb, "flags: %x", data->ce->ce_flags);
> > +     else {
> > +             errlen = (unsigned long)len;
> > +             die(_("bad ls-files format: %%%.*s"), errlen, start);
>
> errlen is declared as an unsigned int, but you cast len which is a
> size_t to unsigned long when assigning to errlen. Then errlen is used
> where a signed int is required by die. There is also a style violation
> as if any branch of an if needs braces then they should all be braced. I
> think that the best solution would be to drop errlen and just write
>
>         else
>                 die(_("bad ls-files format: %%%.*s"), (int)len, start);
>

This piece of code is copying from ls-tree. Maybe we should fix it too.

> It would be interesting to check the performance of this implementation
> on a large repository as it is doing a lot of branching inside a loop. I
> don't think we should change it unless it turns out to be a problem.
> Then we could try switching on the first character of the format
> specifier or some other optimization.
>

Just like ref-filter or something else does, it parses atoms
and then fills buffers with information. Maybe we need such performance
optimization later, but for now, it's just easier to implement this patch :)

> Best Wishes
>
> Phillip

Thanks

ZheNing Hu
diff mbox series

Patch

diff --git a/Documentation/git-ls-files.txt b/Documentation/git-ls-files.txt
index 0dabf3f0ddc..9a88c92f1ad 100644
--- a/Documentation/git-ls-files.txt
+++ b/Documentation/git-ls-files.txt
@@ -20,7 +20,7 @@  SYNOPSIS
 		[--exclude-standard]
 		[--error-unmatch] [--with-tree=<tree-ish>]
 		[--full-name] [--recurse-submodules]
-		[--abbrev[=<n>]] [--] [<file>...]
+		[--abbrev[=<n>]] [--format=<format>] [--] [<file>...]
 
 DESCRIPTION
 -----------
@@ -192,6 +192,13 @@  followed by the  ("attr/<eolattr>").
 	to the contained files. Sparse directories will be shown with a
 	trailing slash, such as "x/" for a sparse directory "x".
 
+--format=<format>::
+	A string that interpolates `%(fieldname)` from the result being shown.
+	It also interpolates `%%` to `%`, and `%xx` where `xx` are hex digits
+	interpolates to character with hex code `xx`; for example `%00`
+	interpolates to `\0` (NUL), `%09` to `\t` (TAB) and %0a to `\n` (LF).
+	--format cannot be combined with `-s`, `-o`, `-k`, `--resolve-undo` and
+	`--debug`.
 \--::
 	Do not interpret any more arguments as options.
 
@@ -223,6 +230,48 @@  quoted as explained for the configuration variable `core.quotePath`
 (see linkgit:git-config[1]).  Using `-z` the filename is output
 verbatim and the line is terminated by a NUL byte.
 
+It is possible to print in a custom format by using the `--format`
+option, which is able to interpolate different fields using
+a `%(fieldname)` notation. For example, if you only care about the
+"objectname" and "path" fields, you can execute with a specific
+"--format" like
+
+	git ls-files --format='%(objectname) %(path)'
+
+FIELD NAMES
+-----------
+Various values from structured fields can be used to interpolate
+into the resulting output. For each outputting line, the following
+names can be used:
+
+tag::
+	The tag of file status.
+objectmode::
+	The mode of the object.
+objectname::
+	The name of the object.
+stage::
+	The stage of the file.
+eol::
+	The line endings of files.
+path::
+	The pathname of the object.
+ctime::
+	The create time of file.
+mtime::
+	The modify time of file.
+dev::
+	The ID of device containing file.
+ino::
+	The inode number of file.
+uid::
+	The user id of file owner.
+gid::
+	The group id of file owner.
+size::
+	The size of the file.
+flags::
+	The flags of the file.
 
 EXCLUDE PATTERNS
 ----------------
diff --git a/builtin/ls-files.c b/builtin/ls-files.c
index e791b65e7e9..f037ccb58b4 100644
--- a/builtin/ls-files.c
+++ b/builtin/ls-files.c
@@ -11,6 +11,7 @@ 
 #include "quote.h"
 #include "dir.h"
 #include "builtin.h"
+#include "strbuf.h"
 #include "tree.h"
 #include "cache-tree.h"
 #include "parse-options.h"
@@ -48,6 +49,7 @@  static char *ps_matched;
 static const char *with_tree;
 static int exc_given;
 static int exclude_args;
+static const char *format;
 
 static const char *tag_cached = "";
 static const char *tag_unmerged = "";
@@ -58,8 +60,8 @@  static const char *tag_modified = "";
 static const char *tag_skip_worktree = "";
 static const char *tag_resolve_undo = "";
 
-static void write_eolinfo(struct index_state *istate,
-			  const struct cache_entry *ce, const char *path)
+static void write_eolinfo_internal(struct strbuf *sb, struct index_state *istate,
+				   const struct cache_entry *ce, const char *path)
 {
 	if (show_eol) {
 		struct stat st;
@@ -71,10 +73,25 @@  static void write_eolinfo(struct index_state *istate,
 							       ce->name);
 		if (!lstat(path, &st) && S_ISREG(st.st_mode))
 			w_txt = get_wt_convert_stats_ascii(path);
-		printf("i/%-5s w/%-5s attr/%-17s\t", i_txt, w_txt, a_txt);
+		if (sb)
+			strbuf_addf(sb, "i/%-5s w/%-5s attr/%-17s\t", i_txt, w_txt, a_txt);
+		else
+			printf("i/%-5s w/%-5s attr/%-17s\t", i_txt, w_txt, a_txt);
 	}
 }
 
+static void write_eolinfo(struct index_state *istate,
+			  const struct cache_entry *ce, const char *path)
+{
+	write_eolinfo_internal(NULL, istate, ce, path);
+}
+
+static void write_eolinfo_to_buf(struct strbuf *sb, struct index_state *istate,
+				 const struct cache_entry *ce, const char *path)
+{
+	write_eolinfo_internal(sb, istate, ce, path);
+}
+
 static void write_name(const char *name)
 {
 	/*
@@ -85,6 +102,15 @@  static void write_name(const char *name)
 				   stdout, line_terminator);
 }
 
+static void write_name_to_buf(struct strbuf *sb, const char *name)
+{
+	const char *rel = relative_path(name, prefix_len ? prefix : NULL, sb);
+	if (line_terminator)
+		quote_c_style(rel, sb, NULL, 0);
+	else
+		strbuf_add(sb, rel, strlen(rel));
+}
+
 static const char *get_tag(const struct cache_entry *ce, const char *tag)
 {
 	static char alttag[4];
@@ -222,6 +248,91 @@  static void show_submodule(struct repository *superproject,
 	repo_clear(&subrepo);
 }
 
+struct show_index_data {
+	const char *tag;
+	const char *pathname;
+	struct index_state *istate;
+	const struct cache_entry *ce;
+};
+
+static size_t expand_show_index(struct strbuf *sb, const char *start,
+			       void *context)
+{
+	struct show_index_data *data = context;
+	const char *end;
+	const char *p;
+	unsigned int errlen;
+	const struct stat_data *sd = &data->ce->ce_stat_data;
+	size_t len = strbuf_expand_literal_cb(sb, start, NULL);
+	if (len)
+		return len;
+	if (*start != '(')
+		die(_("bad ls-files format: element '%s' "
+		      "does not start with '('"), start);
+
+	end = strchr(start + 1, ')');
+	if (!end)
+		die(_("bad ls-files format: element '%s'"
+		      "does not end in ')'"), start);
+
+	len = end - start + 1;
+	if (skip_prefix(start, "(tag)", &p))
+		strbuf_addstr(sb, get_tag(data->ce, data->tag));
+	else if (skip_prefix(start, "(objectmode)", &p))
+		strbuf_addf(sb, "%06o", data->ce->ce_mode);
+	else if (skip_prefix(start, "(objectname)", &p))
+		strbuf_add_unique_abbrev(sb, &data->ce->oid, abbrev);
+	else if (skip_prefix(start, "(stage)", &p))
+		strbuf_addf(sb, "%d", ce_stage(data->ce));
+	else if (skip_prefix(start, "(eol)", &p))
+		write_eolinfo_to_buf(sb, data->istate,
+				     data->ce, data->pathname);
+	else if (skip_prefix(start, "(path)", &p))
+		write_name_to_buf(sb, data->pathname);
+	else if (skip_prefix(start, "(ctime)", &p))
+		strbuf_addf(sb, "ctime: %u:%u",
+			    sd->sd_ctime.sec, sd->sd_ctime.nsec);
+	else if (skip_prefix(start, "(mtime)", &p))
+		strbuf_addf(sb, "mtime: %u:%u",
+			    sd->sd_mtime.sec, sd->sd_mtime.nsec);
+	else if (skip_prefix(start, "(dev)", &p))
+		strbuf_addf(sb, "dev: %u", sd->sd_dev);
+	else if (skip_prefix(start, "(ino)", &p))
+		strbuf_addf(sb, "ino: %u", sd->sd_ino);
+	else if (skip_prefix(start, "(uid)", &p))
+		strbuf_addf(sb, "uid: %u", sd->sd_uid);
+	else if (skip_prefix(start, "(gid)", &p))
+		strbuf_addf(sb, "gid: %u", sd->sd_gid);
+	else if (skip_prefix(start, "(size)", &p))
+		strbuf_addf(sb, "size: %u", sd->sd_size);
+	else if (skip_prefix(start, "(flags)", &p))
+		strbuf_addf(sb, "flags: %x", data->ce->ce_flags);
+	else {
+		errlen = (unsigned long)len;
+		die(_("bad ls-files format: %%%.*s"), errlen, start);
+	}
+
+	return len;
+}
+
+static void show_ce_fmt(struct repository *repo, const struct cache_entry *ce,
+			const char *format, const char *fullname, const char *tag) {
+
+	struct show_index_data data = {
+		.tag = tag,
+		.pathname = fullname,
+		.istate = repo->index,
+		.ce = ce,
+	};
+
+	struct strbuf sb = STRBUF_INIT;
+	strbuf_expand(&sb, format, expand_show_index, &data);
+	strbuf_addch(&sb, line_terminator);
+	fwrite(sb.buf, sb.len, 1, stdout);
+	strbuf_release(&sb);
+	return;
+}
+
 static void show_ce(struct repository *repo, struct dir_struct *dir,
 		    const struct cache_entry *ce, const char *fullname,
 		    const char *tag)
@@ -236,6 +347,11 @@  static void show_ce(struct repository *repo, struct dir_struct *dir,
 				  max_prefix_len, ps_matched,
 				  S_ISDIR(ce->ce_mode) ||
 				  S_ISGITLINK(ce->ce_mode))) {
+		if (format) {
+			show_ce_fmt(repo, ce, format, fullname, tag);
+			return;
+		}
+
 		tag = get_tag(ce, tag);
 
 		if (!show_stage) {
@@ -675,6 +791,9 @@  int cmd_ls_files(int argc, const char **argv, const char *cmd_prefix)
 			 N_("suppress duplicate entries")),
 		OPT_BOOL(0, "sparse", &show_sparse_dirs,
 			 N_("show sparse directories in the presence of a sparse index")),
+		OPT_STRING_F(0, "format", &format, N_("format"),
+			     N_("format to use for the output"),
+			     PARSE_OPT_NONEG),
 		OPT_END()
 	};
 	int ret = 0;
@@ -699,6 +818,11 @@  int cmd_ls_files(int argc, const char **argv, const char *cmd_prefix)
 	for (i = 0; i < exclude_list.nr; i++) {
 		add_pattern(exclude_list.items[i].string, "", 0, pl, --exclude_args);
 	}
+
+	if (format && (show_stage || show_others || show_killed ||
+		show_resolve_undo || skipping_duplicates || debug_mode))
+			die(_("ls-files --format cannot used with -s, -o, -k, --resolve-undo, --deduplicate, --debug"));
+
 	if (show_tag || show_valid_bit || show_fsmonitor_bit) {
 		tag_cached = "H ";
 		tag_unmerged = "M ";
diff --git a/t/t3013-ls-files-format.sh b/t/t3013-ls-files-format.sh
new file mode 100755
index 00000000000..1a1b09e7b3c
--- /dev/null
+++ b/t/t3013-ls-files-format.sh
@@ -0,0 +1,130 @@ 
+#!/bin/sh
+
+test_description='git ls-files --format test'
+
+. ./test-lib.sh
+
+test_expect_success 'setup' '
+	echo o1 >o1 &&
+	echo o2 >o2 &&
+	git add o1 o2 &&
+	git add --chmod +x o1 &&
+	git commit -m base
+'
+
+test_expect_success 'git ls-files --format tag' '
+	printf "H \nH \n" >expect &&
+	git ls-files --format="%(tag)" -t >actual &&
+	test_cmp expect actual
+'
+
+test_expect_success 'git ls-files --format objectmode' '
+	cat >expect <<-\EOF &&
+	100755
+	100644
+	EOF
+	git ls-files --format="%(objectmode)" -t >actual &&
+	test_cmp expect actual
+'
+
+test_expect_success 'git ls-files --format objectname' '
+	oid1=$(git hash-object o1) &&
+	oid2=$(git hash-object o2) &&
+	cat >expect <<-EOF &&
+	$oid1
+	$oid2
+	EOF
+	git ls-files --format="%(objectname)" >actual &&
+	test_cmp expect actual
+'
+
+test_expect_success 'git ls-files --format eol' '
+	printf "i/lf    w/lf    attr/                 \t\n" >expect &&
+	printf "i/lf    w/lf    attr/                 \t\n" >>expect &&
+	git ls-files --format="%(eol)" --eol >actual &&
+	test_cmp expect actual
+'
+
+test_expect_success 'git ls-files --format path' '
+	cat >expect <<-\EOF &&
+	o1
+	o2
+	EOF
+	git ls-files --format="%(path)" >actual &&
+	test_cmp expect actual
+'
+
+test_expect_success 'git ls-files --format ctime' '
+	git ls-files --debug >out &&
+	grep ctime out >expect &&
+	git ls-files --format="  %(ctime)" >actual &&
+	test_cmp expect actual
+'
+
+test_expect_success 'git ls-files --format mtime' '
+	git ls-files --debug >out &&
+	grep mtime out >expect &&
+	git ls-files --format="  %(mtime)" >actual &&
+	test_cmp expect actual
+'
+
+test_expect_success 'git ls-files --format dev and ino' '
+	git ls-files --debug >out &&
+	grep dev out >expect &&
+	git ls-files --format="  %(dev)%x09%(ino)" >actual &&
+	test_cmp expect actual
+'
+
+test_expect_success 'git ls-files --format uid and gid' '
+	git ls-files --debug >out &&
+	grep uid out >expect &&
+	git ls-files --format="  %(uid)%x09%(gid)" >actual &&
+	test_cmp expect actual
+'
+
+test_expect_success 'git ls-files --format with -m' '
+	echo change >o1 &&
+	cat >expect <<-\EOF &&
+	o1
+	EOF
+	git ls-files --format="%(path)" -m >actual &&
+	test_cmp expect actual
+'
+
+test_expect_success 'git ls-files --format with -d' '
+	echo o3 >o3 &&
+	git add o3 &&
+	rm o3 &&
+	cat >expect <<-\EOF &&
+	o3
+	EOF
+	git ls-files --format="%(path)" -d >actual &&
+	test_cmp expect actual
+'
+
+test_expect_success 'git ls-files --format size and flags' '
+	git ls-files --debug >out &&
+	grep size out >expect &&
+	git ls-files --format="  %(size)%x09%(flags)" >actual &&
+	test_cmp expect actual
+'
+
+test_expect_success 'git ls-files --format imitate --stage' '
+	git ls-files --stage >expect &&
+	git ls-files --format="%(objectmode) %(objectname) %(stage)%x09%(path)" >actual &&
+	test_cmp expect actual
+'
+
+test_expect_success 'git ls-files --format imitate --debug' '
+	git ls-files --debug >expect &&
+	git ls-files --format="%(path)%x0a  %(ctime)%x0a  %(mtime)%x0a  %(dev)%x09%(ino)%x0a  %(uid)%x09%(gid)%x0a  %(size)%x09%(flags)" >actual &&
+	test_cmp expect actual
+'
+
+for flag in -s -o -k --resolve-undo --deduplicate --debug
+do
+	test_expect_success "git ls-files --format is incompatible with $flag" '
+		test_must_fail git ls-files --format="%(objectname)" $flag
+	'
+done
+test_done