Merge branch 'master' into sp/fast-import

I'm bringing master in early so that the OBJ_OFS_DELTA implementation
is available as part of the topic.  This way git-fast-import can
learn about this new slightly smaller and faster packfile format,
and can generate them directly rather than needing to have them be
repacked with git-pack-objects.

Due to the API changes in master during the period of development
of git-fast-import, a few minor tweaks to fast-import.c are needed
to produce a working merge.  I've done them here as part of the
merge to ensure bisection always works.

Signed-off-by: Shawn O. Pearce <spearce@spearce.org>

1 files changed, 270 insertions, 101 deletions

diff --git a/upload-pack.c b/upload-pack.c
index 51ce936b0..3a466c6a3 100644
--- a/upload-pack.c
+++ b/upload-pack.c
@@ -1,24 +1,38 @@
-#include <signal.h>
-#include <sys/wait.h>
-#include <sys/poll.h>
 #include "cache.h"
 #include "refs.h"
 #include "pkt-line.h"
+#include "sideband.h"
 #include "tag.h"
 #include "object.h"
 #include "commit.h"
 #include "exec_cmd.h"
+#include "diff.h"
+#include "revision.h"
+#include "list-objects.h"
 
 static const char upload_pack_usage[] = "git-upload-pack [--strict] [--timeout=nn] <dir>";
 
-#define THEY_HAVE (1U << 0)
-#define OUR_REF (1U << 1)
-#define WANTED (1U << 2)
+/* bits #0..7 in revision.h, #8..10 in commit.c */
+#define THEY_HAVE	(1u << 11)
+#define OUR_REF		(1u << 12)
+#define WANTED		(1u << 13)
+#define COMMON_KNOWN	(1u << 14)
+#define REACHABLE	(1u << 15)
+
+#define SHALLOW		(1u << 16)
+#define NOT_SHALLOW	(1u << 17)
+#define CLIENT_SHALLOW	(1u << 18)
+
+static unsigned long oldest_have;
+
 static int multi_ack, nr_our_refs;
-static int use_thin_pack;
+static int use_thin_pack, use_ofs_delta;
 static struct object_array have_obj;
 static struct object_array want_obj;
 static unsigned int timeout;
+/* 0 for no sideband,
+ * otherwise maximum packet size (up to 65520 bytes).
+ */
 static int use_sideband;
 
 static void reset_timeout(void)
@@ -33,45 +47,53 @@ static int strip(char *line, int len)
 	return len;
 }
 
-#define PACKET_MAX 1000
 static ssize_t send_client_data(int fd, const char *data, ssize_t sz)
 {
-	ssize_t ssz;
-	const char *p;
-
-	if (!data) {
-		if (!use_sideband)
-			return 0;
-		packet_flush(1);
+	if (use_sideband)
+		return send_sideband(1, fd, data, sz, use_sideband);
+	if (fd == 3)
+		/* emergency quit */
+		fd = 2;
+	if (fd == 2) {
+		/* XXX: are we happy to lose stuff here? */
+		xwrite(fd, data, sz);
+		return sz;
 	}
+	return safe_write(fd, data, sz);
+}
 
-	if (!use_sideband) {
-		if (fd == 3)
-			/* emergency quit */
-			fd = 2;
-		if (fd == 2) {
-			xwrite(fd, data, sz);
-			return sz;
-		}
-		return safe_write(fd, data, sz);
-	}
-	p = data;
-	ssz = sz;
-	while (sz) {
-		unsigned n;
-		char hdr[5];
-
-		n = sz;
-		if (PACKET_MAX - 5 < n)
-			n = PACKET_MAX - 5;
-		sprintf(hdr, "%04x", n + 5);
-		hdr[4] = fd;
-		safe_write(1, hdr, 5);
-		safe_write(1, p, n);
-		p += n;
-		sz -= n;
+FILE *pack_pipe = NULL;
+static void show_commit(struct commit *commit)
+{
+	if (commit->object.flags & BOUNDARY)
+		fputc('-', pack_pipe);
+	if (fputs(sha1_to_hex(commit->object.sha1), pack_pipe) < 0)
+		die("broken output pipe");
+	fputc('\n', pack_pipe);
+	fflush(pack_pipe);
+	free(commit->buffer);
+	commit->buffer = NULL;
+}
+
+static void show_object(struct object_array_entry *p)
+{
+	/* An object with name "foo\n0000000..." can be used to
+	 * confuse downstream git-pack-objects very badly.
+	 */
+	const char *ep = strchr(p->name, '\n');
+	if (ep) {
+		fprintf(pack_pipe, "%s %.*s\n", sha1_to_hex(p->item->sha1),
+		       (int) (ep - p->name),
+		       p->name);
 	}
-	return ssz;
+	else
+		fprintf(pack_pipe, "%s %s\n",
+				sha1_to_hex(p->item->sha1), p->name);
+}
+
+static void show_edge(struct commit *commit)
+{
+	fprintf(pack_pipe, "-%s\n", sha1_to_hex(commit->object.sha1));
 }
 
 static void create_pack_file(void)
@@ -95,48 +117,40 @@ static void create_pack_file(void)
 
 	if (!pid_rev_list) {
 		int i;
-		int args;
-		const char **argv;
-		const char **p;
-		char *buf;
+		struct rev_info revs;
 
-		if (create_full_pack) {
-			args = 10;
-			use_thin_pack = 0; /* no point doing it */
-		}
-		else
-			args = have_obj.nr + want_obj.nr + 5;
-		p = xmalloc(args * sizeof(char *));
-		argv = (const char **) p;
-		buf = xmalloc(args * 45);
+		pack_pipe = fdopen(lp_pipe[1], "w");
 
-		dup2(lp_pipe[1], 1);
-		close(0);
-		close(lp_pipe[0]);
-		close(lp_pipe[1]);
-		*p++ = "rev-list";
-		*p++ = use_thin_pack ? "--objects-edge" : "--objects";
 		if (create_full_pack)
-			*p++ = "--all";
-		else {
+			use_thin_pack = 0; /* no point doing it */
+		init_revisions(&revs, NULL);
+		revs.tag_objects = 1;
+		revs.tree_objects = 1;
+		revs.blob_objects = 1;
+		if (use_thin_pack)
+			revs.edge_hint = 1;
+
+		if (create_full_pack) {
+			const char *args[] = {"rev-list", "--all", NULL};
+			setup_revisions(2, args, &revs, NULL);
+		} else {
 			for (i = 0; i < want_obj.nr; i++) {
 				struct object *o = want_obj.objects[i].item;
-				*p++ = buf;
-				memcpy(buf, sha1_to_hex(o->sha1), 41);
-				buf += 41;
+				/* why??? */
+				o->flags &= ~UNINTERESTING;
+				add_pending_object(&revs, o, NULL);
 			}
-		}
-		if (!create_full_pack)
 			for (i = 0; i < have_obj.nr; i++) {
 				struct object *o = have_obj.objects[i].item;
-				*p++ = buf;
-				*buf++ = '^';
-				memcpy(buf, sha1_to_hex(o->sha1), 41);
-				buf += 41;
+				o->flags |= UNINTERESTING;
+				add_pending_object(&revs, o, NULL);
 			}
-		*p++ = NULL;
-		execv_git_cmd(argv);
-		die("git-upload-pack: unable to exec git-rev-list");
+			setup_revisions(0, NULL, &revs, NULL);
+		}
+		prepare_revision_walk(&revs);
+		mark_edges_uninteresting(revs.commits, &revs, show_edge);
+		traverse_commit_list(&revs, show_commit, show_object);
+		exit(0);
 	}
 
 	if (pipe(pu_pipe) < 0)
@@ -160,7 +174,9 @@ static void create_pack_file(void)
 		close(pu_pipe[1]);
 		close(pe_pipe[0]);
 		close(pe_pipe[1]);
-		execl_git_cmd("pack-objects", "--stdout", "--progress", NULL);
+		execl_git_cmd("pack-objects", "--stdout", "--progress",
+			      use_ofs_delta ? "--delta-base-offset" : NULL,
+			      NULL);
 		kill(pid_rev_list, SIGKILL);
 		die("git-upload-pack: unable to exec git-pack-objects");
 	}
@@ -227,7 +243,7 @@ static void create_pack_file(void)
 					*cp++ = buffered;
 					outsz++;
 				}
-				sz = read(pu_pipe[0], cp,
+				sz = xread(pu_pipe[0], cp,
 					  sizeof(data) - outsz);
 				if (0 < sz)
 						;
@@ -252,7 +268,7 @@ static void create_pack_file(void)
 				/* Status ready; we ship that in the side-band
 				 * or dump to the standard error.
 				 */
-				sz = read(pe_pipe[0], progress,
+				sz = xread(pe_pipe[0], progress,
 					  sizeof(progress));
 				if (0 < sz)
 					send_client_data(2, progress, sz);
@@ -308,7 +324,8 @@ static void create_pack_file(void)
 				goto fail;
 			fprintf(stderr, "flushed.\n");
 		}
-		send_client_data(1, NULL, 0);
+		if (use_sideband)
+			packet_flush(1);
 		return;
 	}
  fail:
@@ -323,11 +340,12 @@ static void create_pack_file(void)
 static int got_sha1(char *hex, unsigned char *sha1)
 {
 	struct object *o;
+	int we_knew_they_have = 0;
 
 	if (get_sha1_hex(hex, sha1))
 		die("git-upload-pack: expected SHA1 object, got '%s'", hex);
 	if (!has_sha1_file(sha1))
-		return 0;
+		return -1;
 
 	o = lookup_object(sha1);
 	if (!(o && o->parsed))
@@ -336,22 +354,92 @@ static int got_sha1(char *hex, unsigned char *sha1)
 		die("oops (%s)", sha1_to_hex(sha1));
 	if (o->type == OBJ_COMMIT) {
 		struct commit_list *parents;
+		struct commit *commit = (struct commit *)o;
 		if (o->flags & THEY_HAVE)
-			return 0;
-		o->flags |= THEY_HAVE;
-		for (parents = ((struct commit*)o)->parents;
+			we_knew_they_have = 1;
+		else
+			o->flags |= THEY_HAVE;
+		if (!oldest_have || (commit->date < oldest_have))
+			oldest_have = commit->date;
+		for (parents = commit->parents;
 		     parents;
 		     parents = parents->next)
 			parents->item->object.flags |= THEY_HAVE;
 	}
-	add_object_array(o, NULL, &have_obj);
+	if (!we_knew_they_have) {
+		add_object_array(o, NULL, &have_obj);
+		return 1;
+	}
+	return 0;
+}
+
+static int reachable(struct commit *want)
+{
+	struct commit_list *work = NULL;
+
+	insert_by_date(want, &work);
+	while (work) {
+		struct commit_list *list = work->next;
+		struct commit *commit = work->item;
+		free(work);
+		work = list;
+
+		if (commit->object.flags & THEY_HAVE) {
+			want->object.flags |= COMMON_KNOWN;
+			break;
+		}
+		if (!commit->object.parsed)
+			parse_object(commit->object.sha1);
+		if (commit->object.flags & REACHABLE)
+			continue;
+		commit->object.flags |= REACHABLE;
+		if (commit->date < oldest_have)
+			continue;
+		for (list = commit->parents; list; list = list->next) {
+			struct commit *parent = list->item;
+			if (!(parent->object.flags & REACHABLE))
+				insert_by_date(parent, &work);
+		}
+	}
+	want->object.flags |= REACHABLE;
+	clear_commit_marks(want, REACHABLE);
+	free_commit_list(work);
+	return (want->object.flags & COMMON_KNOWN);
+}
+
+static int ok_to_give_up(void)
+{
+	int i;
+
+	if (!have_obj.nr)
+		return 0;
+
+	for (i = 0; i < want_obj.nr; i++) {
+		struct object *want = want_obj.objects[i].item;
+
+		if (want->flags & COMMON_KNOWN)
+			continue;
+		want = deref_tag(want, "a want line", 0);
+		if (!want || want->type != OBJ_COMMIT) {
+			/* no way to tell if this is reachable by
+			 * looking at the ancestry chain alone, so
+			 * leave a note to ourselves not to worry about
+			 * this object anymore.
+			 */
+			want_obj.objects[i].item->flags |= COMMON_KNOWN;
+			continue;
+		}
+		if (!reachable((struct commit *)want))
+			return 0;
+	}
 	return 1;
 }
 
 static int get_common_commits(void)
 {
 	static char line[1000];
-	unsigned char sha1[20], last_sha1[20];
+	unsigned char sha1[20];
+	char hex[41], last_hex[41];
 	int len;
 
 	track_object_refs = 0;
@@ -368,21 +456,29 @@ static int get_common_commits(void)
 		}
 		len = strip(line, len);
 		if (!strncmp(line, "have ", 5)) {
-			if (got_sha1(line+5, sha1) &&
-			    (multi_ack || have_obj.nr == 1)) {
-				packet_write(1, "ACK %s%s\n",
-					     sha1_to_hex(sha1),
-					     multi_ack ?  " continue" : "");
-				if (multi_ack)
-					hashcpy(last_sha1, sha1);
+			switch (got_sha1(line+5, sha1)) {
+			case -1: /* they have what we do not */
+				if (multi_ack && ok_to_give_up())
+					packet_write(1, "ACK %s continue\n",
+						     sha1_to_hex(sha1));
+				break;
+			default:
+				memcpy(hex, sha1_to_hex(sha1), 41);
+				if (multi_ack) {
+					const char *msg = "ACK %s continue\n";
+					packet_write(1, msg, hex);
+					memcpy(last_hex, hex, 41);
+				}
+				else if (have_obj.nr == 1)
+					packet_write(1, "ACK %s\n", hex);
+				break;
 			}
 			continue;
 		}
 		if (!strcmp(line, "done")) {
 			if (have_obj.nr > 0) {
 				if (multi_ack)
-					packet_write(1, "ACK %s\n",
-							sha1_to_hex(last_sha1));
+					packet_write(1, "ACK %s\n", last_hex);
 				return 0;
 			}
 			packet_write(1, "NAK\n");
@@ -394,8 +490,9 @@ static int get_common_commits(void)
 
 static void receive_needs(void)
 {
+	struct object_array shallows = {0, 0, NULL};
 	static char line[1000];
-	int len;
+	int len, depth = 0;
 
 	for (;;) {
 		struct object *o;
@@ -403,8 +500,29 @@ static void receive_needs(void)
 		len = packet_read_line(0, line, sizeof(line));
 		reset_timeout();
 		if (!len)
-			return;
+			break;
 
+		if (!strncmp("shallow ", line, 8)) {
+			unsigned char sha1[20];
+			struct object *object;
+			use_thin_pack = 0;
+			if (get_sha1(line + 8, sha1))
+				die("invalid shallow line: %s", line);
+			object = parse_object(sha1);
+			if (!object)
+				die("did not find object for %s", line);
+			object->flags |= CLIENT_SHALLOW;
+			add_object_array(object, NULL, &shallows);
+			continue;
+		}
+		if (!strncmp("deepen ", line, 7)) {
+			char *end;
+			use_thin_pack = 0;
+			depth = strtol(line + 7, &end, 0);
+			if (end == line + 7 || depth <= 0)
+				die("Invalid deepen: %s", line);
+			continue;
+		}
 		if (strncmp("want ", line, 5) ||
 		    get_sha1_hex(line+5, sha1_buf))
 			die("git-upload-pack: protocol error, "
@@ -413,8 +531,12 @@ static void receive_needs(void)
 			multi_ack = 1;
 		if (strstr(line+45, "thin-pack"))
 			use_thin_pack = 1;
-		if (strstr(line+45, "side-band"))
-			use_sideband = 1;
+		if (strstr(line+45, "ofs-delta"))
+			use_ofs_delta = 1;
+		if (strstr(line+45, "side-band-64k"))
+			use_sideband = LARGE_PACKET_MAX;
+		else if (strstr(line+45, "side-band"))
+			use_sideband = DEFAULT_PACKET_MAX;
 
 		/* We have sent all our refs already, and the other end
 		 * should have chosen out of them; otherwise they are
@@ -432,11 +554,58 @@ static void receive_needs(void)
 			add_object_array(o, NULL, &want_obj);
 		}
 	}
+	if (depth == 0 && shallows.nr == 0)
+		return;
+	if (depth > 0) {
+		struct commit_list *result, *backup;
+		int i;
+		backup = result = get_shallow_commits(&want_obj, depth,
+			SHALLOW, NOT_SHALLOW);
+		while (result) {
+			struct object *object = &result->item->object;
+			if (!(object->flags & (CLIENT_SHALLOW|NOT_SHALLOW))) {
+				packet_write(1, "shallow %s",
+						sha1_to_hex(object->sha1));
+				register_shallow(object->sha1);
+			}
+			result = result->next;
+		}
+		free_commit_list(backup);
+		for (i = 0; i < shallows.nr; i++) {
+			struct object *object = shallows.objects[i].item;
+			if (object->flags & NOT_SHALLOW) {
+				struct commit_list *parents;
+				packet_write(1, "unshallow %s",
+					sha1_to_hex(object->sha1));
+				object->flags &= ~CLIENT_SHALLOW;
+				/* make sure the real parents are parsed */
+				unregister_shallow(object->sha1);
+				object->parsed = 0;
+				parse_commit((struct commit *)object);
+				parents = ((struct commit *)object)->parents;
+				while (parents) {
+					add_object_array(&parents->item->object,
+							NULL, &want_obj);
+					parents = parents->next;
+				}
+			}
+			/* make sure commit traversal conforms to client */
+			register_shallow(object->sha1);
+		}
+		packet_flush(1);
+	} else
+		if (shallows.nr > 0) {
+			int i;
+			for (i = 0; i < shallows.nr; i++)
+				register_shallow(shallows.objects[i].item->sha1);
+		}
+	free(shallows.objects);
 }
 
-static int send_ref(const char *refname, const unsigned char *sha1)
+static int send_ref(const char *refname, const unsigned char *sha1, int flag, void *cb_data)
 {
-	static const char *capabilities = "multi_ack thin-pack side-band";
+	static const char *capabilities = "multi_ack thin-pack side-band"
+		" side-band-64k ofs-delta shallow";
 	struct object *o = parse_object(sha1);
 
 	if (!o)
@@ -462,8 +631,8 @@ static int send_ref(const char *refname, const unsigned char *sha1)
 static void upload_pack(void)
 {
 	reset_timeout();
-	head_ref(send_ref);
-	for_each_ref(send_ref);
+	head_ref(send_ref, NULL);
+	for_each_ref(send_ref, NULL);
 	packet_flush(1);
 	receive_needs();
 	if (want_obj.nr) {