Add "git-patch-id" program to generate patch ID's.

A "patch ID" is nothing but a SHA1 of the diff associated with a patch,
with whitespace and line numbers ignored.  As such, it's "reasonably
stable", but at the same time also reasonably unique, ie two patches
that have the same "patch ID" are almost guaranteed to be the same
thing.

IOW, you can use this thing to look for likely duplicate commits.

2 files changed, 81 insertions, 1 deletions

diff --git a/Makefile b/Makefile
index e3405dc3a..fd5c80f75 100644
--- a/Makefile
+++ b/Makefile
@@ -35,7 +35,7 @@ PROG=   git-update-cache git-diff-files git-init-db git-write-tree \
 	git-http-pull git-ssh-push git-ssh-pull git-rev-list git-mktag \
 	git-diff-helper git-tar-tree git-local-pull git-write-blob \
 	git-get-tar-commit-id git-mkdelta git-apply git-stripspace \
-	git-cvs2git git-diff-stages git-rev-parse
+	git-cvs2git git-diff-stages git-rev-parse git-patch-id
 
 all: $(PROG)
 
@@ -121,6 +121,7 @@ git-stripspace: stripspace.c
 git-cvs2git: cvs2git.c
 git-diff-stages: diff-stages.c
 git-rev-parse: rev-parse.c
+git-patch-id: patch-id.c
 
 git-http-pull: LIBS += -lcurl
 git-rev-list: LIBS += -lssl
diff --git a/patch-id.c b/patch-id.c
new file mode 100644
index 000000000..0bd640cc4
--- /dev/null
+++ b/patch-id.c
@@ -0,0 +1,79 @@
+#include <ctype.h>
+#include "cache.h"
+
+static void flush_current_id(int patchlen, unsigned char *id, SHA_CTX *c)
+{
+	unsigned char result[20];
+	char name[50];
+
+	if (!patchlen)
+		return;
+
+	SHA1_Final(result, c);
+	memcpy(name, sha1_to_hex(id), 41);
+	printf("%s %s\n", sha1_to_hex(result), name);
+	SHA1_Init(c);
+}
+
+static int remove_space(char *line)
+{
+	char *src = line;
+	char *dst = line;
+	unsigned char c;
+
+	while ((c = *src++) != '\0') {
+		if (!isspace(c))
+			*dst++ = c;
+	}
+	return dst - line;
+}
+
+static void generate_id_list(void)
+{
+	static unsigned char sha1[20];
+	static char line[1000];
+	SHA_CTX ctx;
+	int patchlen = 0;
+
+	SHA1_Init(&ctx);
+	while (fgets(line, sizeof(line), stdin) != NULL) {
+		unsigned char n[20];
+		char *p = line;
+		int len;
+
+		if (!memcmp(line, "diff-tree ", 10))
+			p += 10;
+
+		if (!get_sha1_hex(p, n)) {
+			flush_current_id(patchlen, sha1, &ctx);
+			memcpy(sha1, n, 20);
+			patchlen = 0;
+			continue;
+		}
+
+		/* Ignore commit comments */
+		if (!patchlen && memcmp(line, "diff ", 5))
+			continue;
+
+		/* Ignore line numbers when computing the SHA1 of the patch */
+		if (!memcmp(line, "@@ -", 4))
+			continue;
+
+		/* Compute the sha without whitespace */
+		len = remove_space(line);
+		patchlen += len;
+		SHA1_Update(&ctx, line, len);
+	}
+	flush_current_id(patchlen, sha1, &ctx);
+}
+
+static const char patch_id_usage[] = "usage: git-patch-id < patch";
+
+int main(int argc, char **argv)
+{
+	if (argc != 1)
+		usage(patch_id_usage);
+
+	generate_id_list();
+	return 0;
+}