Merge branch 'jt/rev-list-z' into seen

"git rev-list" learns machine-parsable output format that delimits
each field with NUL.

* jt/rev-list-z:
  rev-list: support NUL-delimited --missing option
  rev-list: support NUL-delimited --boundary option
  rev-list: support delimiting objects with NUL bytes
  revision: support NUL-delimited --stdin mode
  rev-list: refactor early option parsing
  rev-list: inline `show_object_with_name()` in `show_object()`

Author: gitster

Documentation/rev-list-options.adoc

@@ -361,6 +361,32 @@ ifdef::git-rev-list[]
 --progress=<header>::
 	Show progress reports on stderr as objects are considered. The
 	`<header>` text will be printed with each progress update.
+
+-z::
+	Instead of being newline-delimited, each outputted object and its
+	accompanying metadata is delimited using NUL bytes in the following
+	form:
++
+-----------------------------------------------------------------------
+<OID> NUL [<token>=<value> NUL]...
+-----------------------------------------------------------------------
++
+Additional object metadata, such as object paths or boundary/missing objects,
+is printed using the `<token>=<value>` form. Token values are printed as-is
+without any encoding/truncation. An OID entry never contains a '=' character
+and thus is used to signal the start of a new object record. Examples:
++
+-----------------------------------------------------------------------
+<OID> NUL
+<OID> NUL path=<path> NUL
+<OID> NUL boundary=yes NUL
+<OID> NUL missing=yes NUL [<token>=<value> NUL]...
+-----------------------------------------------------------------------
++
+This mode is only compatible with the `--objects`, `--boundary`, and
+`--missing` output options. Also, revision and pathspec argument parsing on
+stdin with the `--stdin` option is NUL byte delimited instead of using newlines
+while in this mode.
 endif::git-rev-list[]
 
 History Simplification

builtin/rev-list.c

@@ -16,6 +16,7 @@
 #include "object-file.h"
 #include "object-store-ll.h"
 #include "pack-bitmap.h"
+#include "parse-options.h"
 #include "log-tree.h"
 #include "graph.h"
 #include "bisect.h"
@@ -64,6 +65,7 @@ static const char rev_list_usage[] =
 "    --abbrev-commit\n"
 "    --left-right\n"
 "    --count\n"
+"    -z\n"
 "  special purpose:\n"
 "    --bisect\n"
 "    --bisect-vars\n"
@@ -96,6 +98,9 @@ static int arg_show_object_names = 1;
 
 #define DEFAULT_OIDSET_SIZE     (16*1024)
 
+static char line_term = '\n';
+static char info_term = ' ';
+
 static int show_disk_usage;
 static off_t total_disk_usage;
 static int human_readable;
@@ -131,24 +136,39 @@ static void print_missing_object(struct missing_objects_map_entry *entry,
 {
 	struct strbuf sb = STRBUF_INIT;
 
+	if (line_term)
+		putchar('?');
+
+	printf("%s", oid_to_hex(&entry->entry.oid));
+
+	if (!line_term)
+		printf("%cmissing=yes", info_term);
+
 	if (!print_missing_info) {
-		printf("?%s\n", oid_to_hex(&entry->entry.oid));
+		putchar(line_term);
 		return;
 	}
 
 	if (entry->path && *entry->path) {
 		struct strbuf path = STRBUF_INIT;
 
-		strbuf_addstr(&sb, " path=");
-		quote_path(entry->path, NULL, &path, QUOTE_PATH_QUOTE_SP);
-		strbuf_addbuf(&sb, &path);
+		strbuf_addf(&sb, "%cpath=", info_term);
+
+		if (line_term) {
+			quote_path(entry->path, NULL, &path, QUOTE_PATH_QUOTE_SP);
+			strbuf_addbuf(&sb, &path);
+		} else {
+			strbuf_addstr(&sb, entry->path);
+		}
 
 		strbuf_release(&path);
 	}
 	if (entry->type)
-		strbuf_addf(&sb, " type=%s", type_name(entry->type));
+		strbuf_addf(&sb, "%ctype=%s", info_term, type_name(entry->type));
+
+	fwrite(sb.buf, sizeof(char), sb.len, stdout);
+	putchar(line_term);
 
-	printf("?%s%s\n", oid_to_hex(&entry->entry.oid), sb.buf);
 	strbuf_release(&sb);
 }
 
@@ -235,13 +255,18 @@ static void show_commit(struct commit *commit, void *data)
 		fputs(info->header_prefix, stdout);
 
 	if (revs->include_header) {
-		if (!revs->graph)
+		if (!revs->graph && line_term)
 			fputs(get_revision_mark(revs, commit), stdout);
 		if (revs->abbrev_commit && revs->abbrev)
 			fputs(repo_find_unique_abbrev(the_repository, &commit->object.oid, revs->abbrev),
 			      stdout);
 		else
 			fputs(oid_to_hex(&commit->object.oid), stdout);
+
+		if (!line_term) {
+			if (commit->object.flags & BOUNDARY)
+				printf("%cboundary=yes", info_term);
+		}
 	}
 	if (revs->print_parents) {
 		struct commit_list *parents = commit->parents;
@@ -263,7 +288,7 @@ static void show_commit(struct commit *commit, void *data)
 	if (revs->commit_format == CMIT_FMT_ONELINE)
 		putchar(' ');
 	else if (revs->include_header)
-		putchar('\n');
+		putchar(line_term);
 
 	if (revs->verbose_header) {
 		struct strbuf buf = STRBUF_INIT;
@@ -357,10 +382,19 @@ static void show_object(struct object *obj, const char *name, void *cb_data)
 		return;
 	}
 
-	if (arg_show_object_names)
-		show_object_with_name(stdout, obj, name);
-	else
-		printf("%s\n", oid_to_hex(&obj->oid));
+	printf("%s", oid_to_hex(&obj->oid));
+
+	if (arg_show_object_names) {
+		if (line_term) {
+			putchar(info_term);
+			for (const char *p = name; *p && *p != '\n'; p++)
+				putchar(*p);
+		} else if (*name) {
+			printf("%cpath=%s", info_term, name);
+		}
+	}
+
+	putchar(line_term);
 }
 
 static void show_edge(struct commit *commit)
@@ -634,19 +668,19 @@ int cmd_rev_list(int argc,
 		if (!strcmp(arg, "--exclude-promisor-objects")) {
 			fetch_if_missing = 0;
 			revs.exclude_promisor_objects = 1;
-			break;
-		}
-	}
-	for (i = 1; i < argc; i++) {
-		const char *arg = argv[i];
-		if (skip_prefix(arg, "--missing=", &arg)) {
-			if (revs.exclude_promisor_objects)
-				die(_("options '%s' and '%s' cannot be used together"), "--exclude-promisor-objects", "--missing");
-			if (parse_missing_action_value(arg))
-				break;
+		} else if (skip_prefix(arg, "--missing=", &arg)) {
+			parse_missing_action_value(arg);
+		} else if (!strcmp(arg, "-z")) {
+			s_r_opt.nul_delim_stdin = 1;
+			line_term = '\0';
+			info_term = '\0';
 		}
 	}
 
+	die_for_incompatible_opt2(revs.exclude_promisor_objects,
+				  "--exclude_promisor_objects",
+				  arg_missing_action, "--missing");
+
 	if (arg_missing_action)
 		revs.do_not_die_on_missing_objects = 1;
 
@@ -755,6 +789,20 @@ int cmd_rev_list(int argc,
 		usage(rev_list_usage);
 
 	}
+
+	/*
+	 * Reject options currently incompatible with -z. For some options, this
+	 * is not an inherent limitation and support may be implemented in the
+	 * future.
+	 */
+	if (!line_term) {
+		if (revs.graph || revs.verbose_header || show_disk_usage ||
+		    info.show_timestamp || info.header_prefix || bisect_list ||
+		    use_bitmap_index || revs.edge_hint || revs.left_right ||
+		    revs.cherry_mark)
+			die(_("-z option used with unsupported option"));
+	}
+
 	if (revs.commit_format != CMIT_FMT_USERFORMAT)
 		revs.include_header = 1;
 	if (revs.commit_format != CMIT_FMT_UNSPECIFIED) {

revision.c

@@ -59,14 +59,6 @@ implement_shared_commit_slab(revision_sources, char *);
 
 static inline int want_ancestry(const struct rev_info *revs);
 
-void show_object_with_name(FILE *out, struct object *obj, const char *name)
-{
-	fprintf(out, "%s ", oid_to_hex(&obj->oid));
-	for (const char *p = name; *p && *p != '\n'; p++)
-		fputc(*p, out);
-	fputc('\n', out);
-}
-
 static void mark_blob_uninteresting(struct blob *blob)
 {
 	if (!blob)
@@ -2283,10 +2275,10 @@ int handle_revision_arg(const char *arg, struct rev_info *revs, int flags, unsig
 	return ret;
 }
 
-static void read_pathspec_from_stdin(struct strbuf *sb,
-				     struct strvec *prune)
+static void read_pathspec_from_stdin(struct strbuf *sb, struct strvec *prune,
+				     int line_term)
 {
-	while (strbuf_getline(sb, stdin) != EOF)
+	while (strbuf_getdelim_strip_crlf(sb, stdin, line_term) != EOF)
 		strvec_push(prune, sb->buf);
 }
 
@@ -2913,8 +2905,8 @@ static int handle_revision_pseudo_opt(struct rev_info *revs,
 	return 1;
 }
 
-static void read_revisions_from_stdin(struct rev_info *revs,
-				      struct strvec *prune)
+static void read_revisions_from_stdin(struct rev_info *revs, struct strvec *prune,
+				      int line_term)
 {
 	struct strbuf sb;
 	int seen_dashdash = 0;
@@ -2926,7 +2918,7 @@ static void read_revisions_from_stdin(struct rev_info *revs,
 	warn_on_object_refname_ambiguity = 0;
 
 	strbuf_init(&sb, 1000);
-	while (strbuf_getline(&sb, stdin) != EOF) {
+	while (strbuf_getdelim_strip_crlf(&sb, stdin, line_term) != EOF) {
 		if (!sb.len)
 			break;
 
@@ -2954,7 +2946,7 @@ static void read_revisions_from_stdin(struct rev_info *revs,
 			die("bad revision '%s'", sb.buf);
 	}
 	if (seen_dashdash)
-		read_pathspec_from_stdin(&sb, prune);
+		read_pathspec_from_stdin(&sb, prune, line_term);
 
 	strbuf_release(&sb);
 	warn_on_object_refname_ambiguity = save_warning;
@@ -3027,13 +3019,16 @@ int setup_revisions(int argc, const char **argv, struct rev_info *revs, struct s
 			}
 
 			if (!strcmp(arg, "--stdin")) {
+				int term = opt && opt->nul_delim_stdin ? '\0' : '\n';
+
 				if (revs->disable_stdin) {
 					argv[left++] = arg;
 					continue;
 				}
 				if (revs->read_from_stdin++)
 					die("--stdin given twice?");
-				read_revisions_from_stdin(revs, &prune_data);
+				read_revisions_from_stdin(revs, &prune_data,
+							  term);
 				continue;
 			}
 

revision.h

@@ -439,7 +439,8 @@ struct setup_revision_opt {
 	void (*tweak)(struct rev_info *);
 	unsigned int	assume_dashdash:1,
 			allow_exclude_promisor_objects:1,
-			free_removed_argv_elements:1;
+			free_removed_argv_elements:1,
+			nul_delim_stdin:1;
 	unsigned revarg_opt;
 };
 int setup_revisions(int argc, const char **argv, struct rev_info *revs,
@@ -489,8 +490,6 @@ void mark_parents_uninteresting(struct rev_info *revs, struct commit *commit);
 void mark_tree_uninteresting(struct repository *r, struct tree *tree);
 void mark_trees_uninteresting_sparse(struct repository *r, struct oidset *trees);
 
-void show_object_with_name(FILE *, struct object *, const char *);
-
 /**
  * Helpers to check if a reference should be excluded.
  */

t/t6000-rev-list-misc.sh

@@ -182,4 +182,55 @@ test_expect_success 'rev-list --unpacked' '
 	test_cmp expect actual
 '
 
+test_expect_success 'rev-list -z' '
+	test_when_finished rm -rf repo &&
+
+	git init repo &&
+	test_commit -C repo 1 &&
+	test_commit -C repo 2 &&
+
+	oid1=$(git -C repo rev-parse HEAD) &&
+	oid2=$(git -C repo rev-parse HEAD~) &&
+
+	printf "%s\0%s\0" "$oid1" "$oid2" >expect &&
+	git -C repo rev-list -z HEAD >actual &&
+
+	test_cmp expect actual
+'
+
+test_expect_success 'rev-list -z --objects' '
+	test_when_finished rm -rf repo &&
+
+	git init repo &&
+	test_commit -C repo 1 &&
+	test_commit -C repo 2 &&
+
+	oid1=$(git -C repo rev-parse HEAD:1.t) &&
+	oid2=$(git -C repo rev-parse HEAD:2.t) &&
+	path1=1.t &&
+	path2=2.t &&
+
+	printf "%s\0path=%s\0%s\0path=%s\0" "$oid1" "$path1" "$oid2" "$path2" \
+		>expect &&
+	git -C repo rev-list -z --objects HEAD:1.t HEAD:2.t >actual &&
+
+	test_cmp expect actual
+'
+
+test_expect_success 'rev-list -z --boundary' '
+	test_when_finished rm -rf repo &&
+
+	git init repo &&
+	test_commit -C repo 1 &&
+	test_commit -C repo 2 &&
+
+	oid1=$(git -C repo rev-parse HEAD) &&
+	oid2=$(git -C repo rev-parse HEAD~) &&
+
+	printf "%s\0%s\0boundary=yes\0" "$oid1" "$oid2" >expect &&
+	git -C repo rev-list -z --boundary HEAD~.. >actual &&
+
+	test_cmp expect actual
+'
+
 test_done

t/t6017-rev-list-stdin.sh

@@ -148,4 +148,13 @@ test_expect_success '--not via stdin does not influence revisions from command l
 	test_cmp expect actual
 '
 
+test_expect_success 'NUL-delimited stdin' '
+	printf "%s\0%s\0%s\0" "HEAD" "--" "file-1" > input &&
+
+	git rev-list -z --objects HEAD -- file-1 >expect &&
+	git rev-list -z --objects --stdin <input >actual &&
+
+	test_cmp expect actual
+'
+
 test_done