1 files changed, 86 insertions, 29 deletions

diff --git a/builtin/patch-id.c b/builtin/patch-id.c
index 881fcf3273..9d5585d3a7 100644
--- a/builtin/patch-id.c
+++ b/builtin/patch-id.c
@@ -2,6 +2,9 @@
 #include "builtin.h"
 #include "config.h"
 #include "diff.h"
+#include "gettext.h"
+#include "hex.h"
+#include "parse-options.h"
 
 static void flush_current_id(int patchlen, struct object_id *id, struct object_id *result)
 {
@@ -57,10 +60,12 @@ 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;
+	int diff_is_binary = 0;
+	char pre_oid_str[GIT_MAX_HEXSZ + 1], post_oid_str[GIT_MAX_HEXSZ + 1];
 	git_hash_ctx ctx;
 
 	the_hash_algo->init_fn(&ctx);
@@ -71,11 +76,14 @@ static int get_one_patchid(struct object_id *next_oid, struct object_id *result,
 		const char *p = line;
 		int len;
 
-		if (!skip_prefix(line, "diff-tree ", &p) &&
-		    !skip_prefix(line, "commit ", &p) &&
+		/* Possibly skip over the prefix added by "log" or "format-patch" */
+		if (!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;
@@ -88,14 +96,44 @@ static int get_one_patchid(struct object_id *next_oid, struct object_id *result,
 
 		/* Parsing diff header?  */
 		if (before == -1) {
-			if (starts_with(line, "index "))
+			if (starts_with(line, "GIT binary patch") ||
+			    starts_with(line, "Binary files")) {
+				diff_is_binary = 1;
+				before = 0;
+				the_hash_algo->update_fn(&ctx, pre_oid_str,
+							 strlen(pre_oid_str));
+				the_hash_algo->update_fn(&ctx, post_oid_str,
+							 strlen(post_oid_str));
+				if (stable)
+					flush_one_hunk(result, &ctx);
+				continue;
+			} else if (skip_prefix(line, "index ", &p)) {
+				char *oid1_end = strstr(line, "..");
+				char *oid2_end = NULL;
+				if (oid1_end)
+					oid2_end = strstr(oid1_end, " ");
+				if (!oid2_end)
+					oid2_end = line + strlen(line) - 1;
+				if (oid1_end != NULL && oid2_end != NULL) {
+					*oid1_end = *oid2_end = '\0';
+					strlcpy(pre_oid_str, p, GIT_MAX_HEXSZ + 1);
+					strlcpy(post_oid_str, oid1_end + 2, GIT_MAX_HEXSZ + 1);
+				}
 				continue;
-			else if (starts_with(line, "--- "))
+			} else if (starts_with(line, "--- "))
 				before = after = 1;
 			else if (!isalpha(line[0]))
 				break;
 		}
 
+		if (diff_is_binary) {
+			if (starts_with(line, "diff ")) {
+				diff_is_binary = 0;
+				before = -1;
+			}
+			continue;
+		}
+
 		/* Looking for a valid hunk header?  */
 		if (before == 0 && after == 0) {
 			if (starts_with(line, "@@ -")) {
@@ -120,8 +158,8 @@ 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);
 	}
@@ -134,7 +172,7 @@ 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;
@@ -142,21 +180,32 @@ 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;
 	}
 
@@ -165,21 +214,29 @@ 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;
 }