builtin: patch-id: add --verbatim as a command mode

There are situations where the user might not want the default
setting where patch-id strips all whitespace. They might be working
in a language where white space is syntactically important, or they
might have CI testing that enforces strict whitespace linting. In
these cases, a whitespace change would result in the patch
fundamentally changing, and thus deserving of a different id.

Add a new mode that is exclusive of --stable and --unstable called
--verbatim. It also corresponds to the config
patchid.verbatim = true. In this mode, the stable algorithm is
used and whitespace is not stripped from the patch text.

Users of --unstable mainly care about compatibility with old git
versions, which unstripping the whitespace would break. Thus there
isn't a usecase for the combination of --verbatim and --unstable,
and we don't expose this so as to not add maintainence burden.

Signed-off-by: Jerry Zhang <jerry@skydio.com>
fixes https://github.com/Skydio/revup/issues/2
Signed-off-by: Junio C Hamano <gitster@pobox.com>

1 files changed, 48 insertions, 25 deletions

diff --git a/builtin/patch-id.c b/builtin/patch-id.c
index e7a3112314..afdd472369 100644
--- a/builtin/patch-id.c
+++ b/builtin/patch-id.c
@@ -2,6 +2,7 @@
 #include "builtin.h"
 #include "config.h"
 #include "diff.h"
+#include "parse-options.h"
 
 static void flush_current_id(int patchlen, struct object_id *id, struct object_id *result)
 {
@@ -57,7 +58,7 @@ static int scan_hunk_header(const char *p, int *p_before, int *p_after)
 }
 
 static int get_one_patchid(struct object_id *next_oid, struct object_id *result,
-			   struct strbuf *line_buf, int stable)
+			   struct strbuf *line_buf, int stable, int verbatim)
 {
 	int patchlen = 0, found_next = 0;
 	int before = -1, after = -1;
@@ -76,8 +77,11 @@ static int get_one_patchid(struct object_id *next_oid, struct object_id *result,
 		if (!skip_prefix(line, "diff-tree ", &p) &&
 		    !skip_prefix(line, "commit ", &p) &&
 		    !skip_prefix(line, "From ", &p) &&
-		    starts_with(line, "\\ ") && 12 < strlen(line))
+		    starts_with(line, "\\ ") && 12 < strlen(line)) {
+			if (verbatim)
+				the_hash_algo->update_fn(&ctx, line, strlen(line));
 			continue;
+		}
 
 		if (!get_oid_hex(p, next_oid)) {
 			found_next = 1;
@@ -152,8 +156,8 @@ static int get_one_patchid(struct object_id *next_oid, struct object_id *result,
 		if (line[0] == '+' || line[0] == ' ')
 			after--;
 
-		/* Compute the sha without whitespace */
-		len = remove_space(line);
+		/* Add line to hash algo (possibly removing whitespace) */
+		len = verbatim ? strlen(line) : remove_space(line);
 		patchlen += len;
 		the_hash_algo->update_fn(&ctx, line, len);
 	}
@@ -166,7 +170,7 @@ static int get_one_patchid(struct object_id *next_oid, struct object_id *result,
 	return patchlen;
 }
 
-static void generate_id_list(int stable)
+static void generate_id_list(int stable, int verbatim)
 {
 	struct object_id oid, n, result;
 	int patchlen;
@@ -174,21 +178,32 @@ static void generate_id_list(int stable)
 
 	oidclr(&oid);
 	while (!feof(stdin)) {
-		patchlen = get_one_patchid(&n, &result, &line_buf, stable);
+		patchlen = get_one_patchid(&n, &result, &line_buf, stable, verbatim);
 		flush_current_id(patchlen, &oid, &result);
 		oidcpy(&oid, &n);
 	}
 	strbuf_release(&line_buf);
 }
 
-static const char patch_id_usage[] = "git patch-id [--stable | --unstable]";
+static const char *const patch_id_usage[] = {
+	N_("git patch-id [--stable | --unstable | --verbatim]"), NULL
+};
+
+struct patch_id_opts {
+	int stable;
+	int verbatim;
+};
 
 static int git_patch_id_config(const char *var, const char *value, void *cb)
 {
-	int *stable = cb;
+	struct patch_id_opts *opts = cb;
 
 	if (!strcmp(var, "patchid.stable")) {
-		*stable = git_config_bool(var, value);
+		opts->stable = git_config_bool(var, value);
+		return 0;
+	}
+	if (!strcmp(var, "patchid.verbatim")) {
+		opts->verbatim = git_config_bool(var, value);
 		return 0;
 	}
 
@@ -197,21 +212,29 @@ static int git_patch_id_config(const char *var, const char *value, void *cb)
 
 int cmd_patch_id(int argc, const char **argv, const char *prefix)
 {
-	int stable = -1;
-
-	git_config(git_patch_id_config, &stable);
-
-	/* If nothing is set, default to unstable. */
-	if (stable < 0)
-		stable = 0;
-
-	if (argc == 2 && !strcmp(argv[1], "--stable"))
-		stable = 1;
-	else if (argc == 2 && !strcmp(argv[1], "--unstable"))
-		stable = 0;
-	else if (argc != 1)
-		usage(patch_id_usage);
-
-	generate_id_list(stable);
+	/* if nothing is set, default to unstable */
+	struct patch_id_opts config = {0, 0};
+	int opts = 0;
+	struct option builtin_patch_id_options[] = {
+		OPT_CMDMODE(0, "unstable", &opts,
+		    N_("use the unstable patch-id algorithm"), 1),
+		OPT_CMDMODE(0, "stable", &opts,
+		    N_("use the stable patch-id algorithm"), 2),
+		OPT_CMDMODE(0, "verbatim", &opts,
+			N_("don't strip whitespace from the patch"), 3),
+		OPT_END()
+	};
+
+	git_config(git_patch_id_config, &config);
+
+	/* verbatim implies stable */
+	if (config.verbatim)
+		config.stable = 1;
+
+	argc = parse_options(argc, argv, prefix, builtin_patch_id_options,
+			     patch_id_usage, 0);
+
+	generate_id_list(opts ? opts > 1 : config.stable,
+			 opts ? opts == 3 : config.verbatim);
 	return 0;
 }