Merge branch 'bc/rev-parse-path-format'

"git rev-parse" can be explicitly told to give output as absolute
or relative path with the `--path-format=(absolute|relative)` option.

* bc/rev-parse-path-format:
  rev-parse: add option for absolute or relative path formatting
  abspath: add a function to resolve paths with missing components

Documentation/git-rev-parse.txt

@@ -212,6 +212,18 @@ Options for Files
 	Only the names of the variables are listed, not their value,
 	even if they are set.
 
+--path-format=(absolute|relative)::
+	Controls the behavior of certain other options. If specified as absolute, the
+	paths printed by those options will be absolute and canonical. If specified as
+	relative, the paths will be relative to the current working directory if that
+	is possible.  The default is option specific.
++
+This option may be specified multiple times and affects only the arguments that
+follow it on the command line, either to the end of the command line or the next
+instance of this option.
+
+The following options are modified by `--path-format`:
+
 --git-dir::
 	Show `$GIT_DIR` if defined. Otherwise show the path to
 	the .git directory. The path shown, when relative, is
@@ -221,13 +233,42 @@ If `$GIT_DIR` is not defined and the current directory
 is not detected to lie in a Git repository or work tree
 print a message to stderr and exit with nonzero status.
 
+--git-common-dir::
+	Show `$GIT_COMMON_DIR` if defined, else `$GIT_DIR`.
+
+--resolve-git-dir <path>::
+	Check if <path> is a valid repository or a gitfile that
+	points at a valid repository, and print the location of the
+	repository.  If <path> is a gitfile then the resolved path
+	to the real repository is printed.
+
+--git-path <path>::
+	Resolve "$GIT_DIR/<path>" and takes other path relocation
+	variables such as $GIT_OBJECT_DIRECTORY,
+	$GIT_INDEX_FILE... into account. For example, if
+	$GIT_OBJECT_DIRECTORY is set to /foo/bar then "git rev-parse
+	--git-path objects/abc" returns /foo/bar/abc.
+
+--show-toplevel::
+	Show the (by default, absolute) path of the top-level directory
+	of the working tree. If there is no working tree, report an error.
+
+--show-superproject-working-tree::
+	Show the absolute path of the root of the superproject's
+	working tree (if exists) that uses the current repository as
+	its submodule.  Outputs nothing if the current repository is
+	not used as a submodule by any project.
+
+--shared-index-path::
+	Show the path to the shared index file in split index mode, or
+	empty if not in split-index mode.
+
+The following options are unaffected by `--path-format`:
+
 --absolute-git-dir::
 	Like `--git-dir`, but its output is always the canonicalized
 	absolute path.
 
---git-common-dir::
-	Show `$GIT_COMMON_DIR` if defined, else `$GIT_DIR`.
-
 --is-inside-git-dir::
 	When the current working directory is below the repository
 	directory print "true", otherwise "false".
@@ -242,19 +283,6 @@ print a message to stderr and exit with nonzero status.
 --is-shallow-repository::
 	When the repository is shallow print "true", otherwise "false".
 
---resolve-git-dir <path>::
-	Check if <path> is a valid repository or a gitfile that
-	points at a valid repository, and print the location of the
-	repository.  If <path> is a gitfile then the resolved path
-	to the real repository is printed.
-
---git-path <path>::
-	Resolve "$GIT_DIR/<path>" and takes other path relocation
-	variables such as $GIT_OBJECT_DIRECTORY,
-	$GIT_INDEX_FILE... into account. For example, if
-	$GIT_OBJECT_DIRECTORY is set to /foo/bar then "git rev-parse
-	--git-path objects/abc" returns /foo/bar/abc.
-
 --show-cdup::
 	When the command is invoked from a subdirectory, show the
 	path of the top-level directory relative to the current
@@ -265,20 +293,6 @@ print a message to stderr and exit with nonzero status.
 	path of the current directory relative to the top-level
 	directory.
 
---show-toplevel::
-	Show the absolute path of the top-level directory of the working
-	tree. If there is no working tree, report an error.
-
---show-superproject-working-tree::
-	Show the absolute path of the root of the superproject's
-	working tree (if exists) that uses the current repository as
-	its submodule.  Outputs nothing if the current repository is
-	not used as a submodule by any project.
-
---shared-index-path::
-	Show the path to the shared index file in split index mode, or
-	empty if not in split-index mode.
-
 --show-object-format[=(storage|input|output)]::
 	Show the object format (hash algorithm) used for the repository
 	for storage inside the `.git` directory, input, or output. For

abspath.c

@@ -67,19 +67,15 @@ static void get_root_part(struct strbuf *resolved, struct strbuf *remaining)
 #endif
 
 /*
- * Return the real path (i.e., absolute path, with symlinks resolved
+ * If set, any number of trailing components may be missing; otherwise, only one
- * and extra slashes removed) equivalent to the specified path.  (If
+ * may be.
- * you want an absolute path but don't mind links, use
- * absolute_path().)  Places the resolved realpath in the provided strbuf.
- *
- * The directory part of path (i.e., everything up to the last
- * dir_sep) must denote a valid, existing directory, but the last
- * component need not exist.  If die_on_error is set, then die with an
- * informative error message if there is a problem.  Otherwise, return
- * NULL on errors (without generating any output).
  */
-char *strbuf_realpath(struct strbuf *resolved, const char *path,
+#define REALPATH_MANY_MISSING (1 << 0)
-		      int die_on_error)
+/* Should we die if there's an error? */
+#define REALPATH_DIE_ON_ERROR (1 << 1)
+
+static char *strbuf_realpath_1(struct strbuf *resolved, const char *path,
+			       int flags)
 {
 	struct strbuf remaining = STRBUF_INIT;
 	struct strbuf next = STRBUF_INIT;
@@ -89,7 +85,7 @@ char *strbuf_realpath(struct strbuf *resolved, const char *path,
 	struct stat st;
 
 	if (!*path) {
-		if (die_on_error)
+		if (flags & REALPATH_DIE_ON_ERROR)
 			die("The empty string is not a valid path");
 		else
 			goto error_out;
@@ -101,7 +97,7 @@ char *strbuf_realpath(struct strbuf *resolved, const char *path,
 	if (!resolved->len) {
 		/* relative path; can use CWD as the initial resolved path */
 		if (strbuf_getcwd(resolved)) {
-			if (die_on_error)
+			if (flags & REALPATH_DIE_ON_ERROR)
 				die_errno("unable to get current working directory");
 			else
 				goto error_out;
@@ -129,8 +125,9 @@ char *strbuf_realpath(struct strbuf *resolved, const char *path,
 
 		if (lstat(resolved->buf, &st)) {
 			/* error out unless this was the last component */
-			if (errno != ENOENT || remaining.len) {
+			if (errno != ENOENT ||
-				if (die_on_error)
+			   (!(flags & REALPATH_MANY_MISSING) && remaining.len)) {
+				if (flags & REALPATH_DIE_ON_ERROR)
 					die_errno("Invalid path '%s'",
 						  resolved->buf);
 				else
@@ -143,7 +140,7 @@ char *strbuf_realpath(struct strbuf *resolved, const char *path,
 			if (num_symlinks++ > MAXSYMLINKS) {
 				errno = ELOOP;
 
-				if (die_on_error)
+				if (flags & REALPATH_DIE_ON_ERROR)
 					die("More than %d nested symlinks "
 					    "on path '%s'", MAXSYMLINKS, path);
 				else
@@ -153,7 +150,7 @@ char *strbuf_realpath(struct strbuf *resolved, const char *path,
 			len = strbuf_readlink(&symlink, resolved->buf,
 					      st.st_size);
 			if (len < 0) {
-				if (die_on_error)
+				if (flags & REALPATH_DIE_ON_ERROR)
 					die_errno("Invalid symlink '%s'",
 						  resolved->buf);
 				else
@@ -202,6 +199,37 @@ error_out:
 	return retval;
 }
 
+/*
+ * Return the real path (i.e., absolute path, with symlinks resolved
+ * and extra slashes removed) equivalent to the specified path.  (If
+ * you want an absolute path but don't mind links, use
+ * absolute_path().)  Places the resolved realpath in the provided strbuf.
+ *
+ * The directory part of path (i.e., everything up to the last
+ * dir_sep) must denote a valid, existing directory, but the last
+ * component need not exist.  If die_on_error is set, then die with an
+ * informative error message if there is a problem.  Otherwise, return
+ * NULL on errors (without generating any output).
+ */
+char *strbuf_realpath(struct strbuf *resolved, const char *path,
+		      int die_on_error)
+{
+	return strbuf_realpath_1(resolved, path,
+				 die_on_error ? REALPATH_DIE_ON_ERROR : 0);
+}
+
+/*
+ * Just like strbuf_realpath, but allows an arbitrary number of path
+ * components to be missing.
+ */
+char *strbuf_realpath_forgiving(struct strbuf *resolved, const char *path,
+				int die_on_error)
+{
+	return strbuf_realpath_1(resolved, path,
+				 ((die_on_error ? REALPATH_DIE_ON_ERROR : 0) |
+				  REALPATH_MANY_MISSING));
+}
+
 char *real_pathdup(const char *path, int die_on_error)
 {
 	struct strbuf realpath = STRBUF_INIT;

builtin/rev-parse.c

@@ -583,6 +583,75 @@ static void handle_ref_opt(const char *pattern, const char *prefix)
 	clear_ref_exclusion(&ref_excludes);
 }
 
+enum format_type {
+	/* We would like a relative path. */
+	FORMAT_RELATIVE,
+	/* We would like a canonical absolute path. */
+	FORMAT_CANONICAL,
+	/* We would like the default behavior. */
+	FORMAT_DEFAULT,
+};
+
+enum default_type {
+	/* Our default is a relative path. */
+	DEFAULT_RELATIVE,
+	/* Our default is a relative path if there's a shared root. */
+	DEFAULT_RELATIVE_IF_SHARED,
+	/* Our default is a canonical absolute path. */
+	DEFAULT_CANONICAL,
+	/* Our default is not to modify the item. */
+	DEFAULT_UNMODIFIED,
+};
+
+static void print_path(const char *path, const char *prefix, enum format_type format, enum default_type def)
+{
+	char *cwd = NULL;
+	/*
+	 * We don't ever produce a relative path if prefix is NULL, so set the
+	 * prefix to the current directory so that we can produce a relative
+	 * path whenever possible.  If we're using RELATIVE_IF_SHARED mode, then
+	 * we want an absolute path unless the two share a common prefix, so don't
+	 * set it in that case, since doing so causes a relative path to always
+	 * be produced if possible.
+	 */
+	if (!prefix && (format != FORMAT_DEFAULT || def != DEFAULT_RELATIVE_IF_SHARED))
+		prefix = cwd = xgetcwd();
+	if (format == FORMAT_DEFAULT && def == DEFAULT_UNMODIFIED) {
+		puts(path);
+	} else if (format == FORMAT_RELATIVE ||
+		  (format == FORMAT_DEFAULT && def == DEFAULT_RELATIVE)) {
+		/*
+		 * In order for relative_path to work as expected, we need to
+		 * make sure that both paths are absolute paths.  If we don't,
+		 * we can end up with an unexpected absolute path that the user
+		 * didn't want.
+		 */
+		struct strbuf buf = STRBUF_INIT, realbuf = STRBUF_INIT, prefixbuf = STRBUF_INIT;
+		if (!is_absolute_path(path)) {
+			strbuf_realpath_forgiving(&realbuf, path,  1);
+			path = realbuf.buf;
+		}
+		if (!is_absolute_path(prefix)) {
+			strbuf_realpath_forgiving(&prefixbuf, prefix, 1);
+			prefix = prefixbuf.buf;
+		}
+		puts(relative_path(path, prefix, &buf));
+		strbuf_release(&buf);
+		strbuf_release(&realbuf);
+		strbuf_release(&prefixbuf);
+	} else if (format == FORMAT_DEFAULT && def == DEFAULT_RELATIVE_IF_SHARED) {
+		struct strbuf buf = STRBUF_INIT;
+		puts(relative_path(path, prefix, &buf));
+		strbuf_release(&buf);
+	} else {
+		struct strbuf buf = STRBUF_INIT;
+		strbuf_realpath_forgiving(&buf, path, 1);
+		puts(buf.buf);
+		strbuf_release(&buf);
+	}
+	free(cwd);
+}
+
 int cmd_rev_parse(int argc, const char **argv, const char *prefix)
 {
 	int i, as_is = 0, verify = 0, quiet = 0, revs_count = 0, type = 0;
@@ -596,6 +665,7 @@ int cmd_rev_parse(int argc, const char **argv, const char *prefix)
 	struct strbuf buf = STRBUF_INIT;
 	const int hexsz = the_hash_algo->hexsz;
 	int seen_end_of_options = 0;
+	enum format_type format = FORMAT_DEFAULT;
 
 	if (argc > 1 && !strcmp("--parseopt", argv[1]))
 		return cmd_parseopt(argc - 1, argv + 1, prefix);
@@ -668,8 +738,9 @@ int cmd_rev_parse(int argc, const char **argv, const char *prefix)
 				if (!argv[i + 1])
 					die("--git-path requires an argument");
 				strbuf_reset(&buf);
-				puts(relative_path(git_path("%s", argv[i + 1]),
+				print_path(git_path("%s", argv[i + 1]), prefix,
-						   prefix, &buf));
+						format,
+						DEFAULT_RELATIVE_IF_SHARED);
 				i++;
 				continue;
 			}
@@ -687,6 +758,16 @@ int cmd_rev_parse(int argc, const char **argv, const char *prefix)
 					show(arg);
 				continue;
 			}
+			if (opt_with_value(arg, "--path-format", &arg)) {
+				if (!strcmp(arg, "absolute")) {
+					format = FORMAT_CANONICAL;
+				} else if (!strcmp(arg, "relative")) {
+					format = FORMAT_RELATIVE;
+				} else {
+					die("unknown argument to --path-format: %s", arg);
+				}
+				continue;
+			}
 			if (!strcmp(arg, "--default")) {
 				def = argv[++i];
 				if (!def)
@@ -807,7 +888,7 @@ int cmd_rev_parse(int argc, const char **argv, const char *prefix)
 			if (!strcmp(arg, "--show-toplevel")) {
 				const char *work_tree = get_git_work_tree();
 				if (work_tree)
-					puts(work_tree);
+					print_path(work_tree, prefix, format, DEFAULT_UNMODIFIED);
 				else
 					die("this operation must be run in a work tree");
 				continue;
@@ -815,7 +896,7 @@ int cmd_rev_parse(int argc, const char **argv, const char *prefix)
 			if (!strcmp(arg, "--show-superproject-working-tree")) {
 				struct strbuf superproject = STRBUF_INIT;
 				if (get_superproject_working_tree(&superproject))
-					puts(superproject.buf);
+					print_path(superproject.buf, prefix, format, DEFAULT_UNMODIFIED);
 				strbuf_release(&superproject);
 				continue;
 			}
@@ -850,16 +931,18 @@ int cmd_rev_parse(int argc, const char **argv, const char *prefix)
 				const char *gitdir = getenv(GIT_DIR_ENVIRONMENT);
 				char *cwd;
 				int len;
+				enum format_type wanted = format;
 				if (arg[2] == 'g') {	/* --git-dir */
 					if (gitdir) {
-						puts(gitdir);
+						print_path(gitdir, prefix, format, DEFAULT_UNMODIFIED);
 						continue;
 					}
 					if (!prefix) {
-						puts(".git");
+						print_path(".git", prefix, format, DEFAULT_UNMODIFIED);
 						continue;
 					}
 				} else {		/* --absolute-git-dir */
+					wanted = FORMAT_CANONICAL;
 					if (!gitdir && !prefix)
 						gitdir = ".git";
 					if (gitdir) {
@@ -872,14 +955,14 @@ int cmd_rev_parse(int argc, const char **argv, const char *prefix)
 				}
 				cwd = xgetcwd();
 				len = strlen(cwd);
-				printf("%s%s.git\n", cwd, len && cwd[len-1] != '/' ? "/" : "");
+				strbuf_reset(&buf);
+				strbuf_addf(&buf, "%s%s.git", cwd, len && cwd[len-1] != '/' ? "/" : "");
 				free(cwd);
+				print_path(buf.buf, prefix, wanted, DEFAULT_CANONICAL);
 				continue;
 			}
 			if (!strcmp(arg, "--git-common-dir")) {
-				strbuf_reset(&buf);
+				print_path(get_git_common_dir(), prefix, format, DEFAULT_RELATIVE_IF_SHARED);
-				puts(relative_path(get_git_common_dir(),
-						   prefix, &buf));
 				continue;
 			}
 			if (!strcmp(arg, "--is-inside-git-dir")) {
@@ -909,8 +992,7 @@ int cmd_rev_parse(int argc, const char **argv, const char *prefix)
 				if (the_index.split_index) {
 					const struct object_id *oid = &the_index.split_index->base_oid;
 					const char *path = git_path("sharedindex.%s", oid_to_hex(oid));
-					strbuf_reset(&buf);
+					print_path(path, prefix, format, DEFAULT_RELATIVE);
-					puts(relative_path(path, prefix, &buf));
 				}
 				continue;
 			}

cache.h

@@ -1231,6 +1231,8 @@ static inline int is_absolute_path(const char *path)
 int is_directory(const char *);
 char *strbuf_realpath(struct strbuf *resolved, const char *path,
 		      int die_on_error);
+char *strbuf_realpath_forgiving(struct strbuf *resolved, const char *path,
+				int die_on_error);
 char *real_pathdup(const char *path, int die_on_error);
 const char *absolute_path(const char *path);
 char *absolute_pathdup(const char *path);

t/t1500-rev-parse.sh

@@ -3,6 +3,16 @@
 test_description='test git rev-parse'
 . ./test-lib.sh
 
+test_one () {
+	dir="$1" &&
+	expect="$2" &&
+	shift &&
+	shift &&
+	echo "$expect" >expect &&
+	git -C "$dir" rev-parse "$@" >actual &&
+	test_cmp expect actual
+}
+
 # usage: [options] label is-bare is-inside-git is-inside-work prefix git-dir absolute-git-dir
 test_rev_parse () {
 	d=
@@ -60,7 +70,13 @@ ROOT=$(pwd)
 
 test_expect_success 'setup' '
 	mkdir -p sub/dir work &&
-	cp -R .git repo.git
+	cp -R .git repo.git &&
+	git checkout -B main &&
+	test_commit abc &&
+	git checkout -b side &&
+	test_commit def &&
+	git checkout main &&
+	git worktree add worktree side
 '
 
 test_rev_parse toplevel false false true '' .git "$ROOT/.git"
@@ -88,6 +104,45 @@ test_rev_parse -C work -g ../repo.git -b t 'GIT_DIR=../repo.git, core.bare = tru
 
 test_rev_parse -C work -g ../repo.git -b u 'GIT_DIR=../repo.git, core.bare undefined' false false true ''
 
+test_expect_success 'rev-parse --path-format=absolute' '
+	test_one "." "$ROOT/.git" --path-format=absolute --git-dir &&
+	test_one "." "$ROOT/.git" --path-format=absolute --git-common-dir &&
+	test_one "sub/dir" "$ROOT/.git" --path-format=absolute --git-dir &&
+	test_one "sub/dir" "$ROOT/.git" --path-format=absolute --git-common-dir &&
+	test_one "worktree" "$ROOT/.git/worktrees/worktree" --path-format=absolute --git-dir &&
+	test_one "worktree" "$ROOT/.git" --path-format=absolute --git-common-dir &&
+	test_one "." "$ROOT" --path-format=absolute --show-toplevel &&
+	test_one "." "$ROOT/.git/objects" --path-format=absolute --git-path objects &&
+	test_one "." "$ROOT/.git/objects/foo/bar/baz" --path-format=absolute --git-path objects/foo/bar/baz
+'
+
+test_expect_success 'rev-parse --path-format=relative' '
+	test_one "." ".git" --path-format=relative --git-dir &&
+	test_one "." ".git" --path-format=relative --git-common-dir &&
+	test_one "sub/dir" "../../.git" --path-format=relative --git-dir &&
+	test_one "sub/dir" "../../.git" --path-format=relative --git-common-dir &&
+	test_one "worktree" "../.git/worktrees/worktree" --path-format=relative --git-dir &&
+	test_one "worktree" "../.git" --path-format=relative --git-common-dir &&
+	test_one "." "./" --path-format=relative --show-toplevel &&
+	test_one "." ".git/objects" --path-format=relative --git-path objects &&
+	test_one "." ".git/objects/foo/bar/baz" --path-format=relative --git-path objects/foo/bar/baz
+'
+
+test_expect_success '--path-format=relative does not affect --absolute-git-dir' '
+	git rev-parse --path-format=relative --absolute-git-dir >actual &&
+	echo "$ROOT/.git" >expect &&
+	test_cmp expect actual
+'
+
+test_expect_success '--path-format can change in the middle of the command line' '
+	git rev-parse --path-format=absolute --git-dir --path-format=relative --git-path objects/foo/bar >actual &&
+	cat >expect <<-EOF &&
+	$ROOT/.git
+	.git/objects/foo/bar
+	EOF
+	test_cmp expect actual
+'
+
 test_expect_success 'git-common-dir from worktree root' '
 	echo .git >expect &&
 	git rev-parse --git-common-dir >actual &&