teach git-unpack-objects about deltas with offset to base

For delta resolution to be possible, a list of sha1/offset tuple must
be constructed in memory in order to load the appropriate base object.

Signed-off-by: Nicolas Pitre <nico@cam.org>
Signed-off-by: Junio C Hamano <junkio@cox.net>

diff --git a/builtin-unpack-objects.c b/builtin-unpack-objects.c
index c6c636895605815373b00b64223362c2308cfed0..e70a71163d18dff1dd182c0d0b02967044693884 100644 (file)
--- a/builtin-unpack-objects.c
+++ b/builtin-unpack-objects.c
@@ -15,7 +15,7 @@ static const char unpack_usage[] = "git-unpack-objects [-n] [-q] [-r] < pack-fil
 
 /* We always read in 4kB chunks. */
 static unsigned char buffer[4096];
 
 /* We always read in 4kB chunks. */
 static unsigned char buffer[4096];

-static unsigned long offset, len;
+static unsigned long offset, len, consumed_bytes;

 static SHA_CTX ctx;
 
 /*
 static SHA_CTX ctx;
 
 /*


@@ -51,6 +51,7 @@ static void use(int bytes)
                die("used more bytes than were available");
        len -= bytes;
        offset += bytes;
                die("used more bytes than were available");
        len -= bytes;
        offset += bytes;

+       consumed_bytes += bytes;

 }
 
 static void *get_data(unsigned long size)
 }
 
 static void *get_data(unsigned long size)


@@ -89,35 +90,49 @@ static void *get_data(unsigned long size)
 
 struct delta_info {
        unsigned char base_sha1[20];
 
 struct delta_info {
        unsigned char base_sha1[20];

+       unsigned long base_offset;

        unsigned long size;
        void *delta;
        unsigned long size;
        void *delta;

+       unsigned nr;

        struct delta_info *next;
 };
 
 static struct delta_info *delta_list;
 
        struct delta_info *next;
 };
 
 static struct delta_info *delta_list;
 

-static void add_delta_to_list(unsigned char *base_sha1, void *delta, unsigned long size)
+static void add_delta_to_list(unsigned nr, unsigned const char *base_sha1,
+                             unsigned long base_offset,
+                             void *delta, unsigned long size)

 {
        struct delta_info *info = xmalloc(sizeof(*info));
 
        hashcpy(info->base_sha1, base_sha1);
 {
        struct delta_info *info = xmalloc(sizeof(*info));
 
        hashcpy(info->base_sha1, base_sha1);

+       info->base_offset = base_offset;

        info->size = size;
        info->delta = delta;
        info->size = size;
        info->delta = delta;

+       info->nr = nr;

        info->next = delta_list;
        delta_list = info;
 }
 
        info->next = delta_list;
        delta_list = info;
 }
 

-static void added_object(unsigned char *sha1, const char *type, void *data, unsigned long size);
+struct obj_info {
+       unsigned long offset;
+       unsigned char sha1[20];
+};
+
+static struct obj_info *obj_list;

 
 

-static void write_object(void *buf, unsigned long size, const char *type)
+static void added_object(unsigned nr, const char *type, void *data,
+                        unsigned long size);
+
+static void write_object(unsigned nr, void *buf, unsigned long size,
+                        const char *type)

 {
 {

-       unsigned char sha1[20];
-       if (write_sha1_file(buf, size, type, sha1) < 0)
+       if (write_sha1_file(buf, size, type, obj_list[nr].sha1) < 0)

                die("failed to write object");
                die("failed to write object");

-       added_object(sha1, type, buf, size);
+       added_object(nr, type, buf, size);

 }
 
 }
 

-static void resolve_delta(const char *type,
+static void resolve_delta(unsigned nr, const char *type,

                          void *base, unsigned long base_size,
                          void *delta, unsigned long delta_size)
 {
                          void *base, unsigned long base_size,
                          void *delta, unsigned long delta_size)
 {


@@ -130,20 +145,23 @@ static void resolve_delta(const char *type,
        if (!result)
                die("failed to apply delta");
        free(delta);
        if (!result)
                die("failed to apply delta");
        free(delta);

-       write_object(result, result_size, type);
+       write_object(nr, result, result_size, type);

        free(result);
 }
 
        free(result);
 }
 

-static void added_object(unsigned char *sha1, const char *type, void *data, unsigned long size)
+static void added_object(unsigned nr, const char *type, void *data,
+                        unsigned long size)

 {
        struct delta_info **p = &delta_list;
        struct delta_info *info;
 
        while ((info = *p) != NULL) {
 {
        struct delta_info **p = &delta_list;
        struct delta_info *info;
 
        while ((info = *p) != NULL) {

-               if (!hashcmp(info->base_sha1, sha1)) {
+               if (!hashcmp(info->base_sha1, obj_list[nr].sha1) ||
+                   info->base_offset == obj_list[nr].offset) {

                        *p = info->next;
                        p = &delta_list;
                        *p = info->next;
                        p = &delta_list;

-                       resolve_delta(type, data, size, info->delta, info->size);
+                       resolve_delta(info->nr, type, data, size,
+                                     info->delta, info->size);

                        free(info);
                        continue;
                }
                        free(info);
                        continue;
                }


@@ -151,7 +169,8 @@ static void added_object(unsigned char *sha1, const char *type, void *data, unsi
        }
 }
 
        }
 }
 

-static void unpack_non_delta_entry(enum object_type kind, unsigned long size)
+static void unpack_non_delta_entry(enum object_type kind, unsigned long size,
+                                  unsigned nr)

 {
        void *buf = get_data(size);
        const char *type;
 {
        void *buf = get_data(size);
        const char *type;


@@ -164,30 +183,80 @@ static void unpack_non_delta_entry(enum object_type kind, unsigned long size)
        default: die("bad type %d", kind);
        }
        if (!dry_run && buf)
        default: die("bad type %d", kind);
        }
        if (!dry_run && buf)

-               write_object(buf, size, type);
+               write_object(nr, buf, size, type);

        free(buf);
 }
 
        free(buf);
 }
 

-static void unpack_delta_entry(unsigned long delta_size)
+static void unpack_delta_entry(enum object_type kind, unsigned long delta_size,
+                              unsigned nr)

 {
        void *delta_data, *base;
        unsigned long base_size;
        char type[20];
        unsigned char base_sha1[20];
 
 {
        void *delta_data, *base;
        unsigned long base_size;
        char type[20];
        unsigned char base_sha1[20];
 

-       hashcpy(base_sha1, fill(20));
-       use(20);
+       if (kind == OBJ_REF_DELTA) {
+               hashcpy(base_sha1, fill(20));
+               use(20);
+               delta_data = get_data(delta_size);
+               if (dry_run || !delta_data) {
+                       free(delta_data);
+                       return;
+               }
+               if (!has_sha1_file(base_sha1)) {
+                       hashcpy(obj_list[nr].sha1, null_sha1);
+                       add_delta_to_list(nr, base_sha1, 0, delta_data, delta_size);
+                       return;
+               }
+       } else {
+               unsigned base_found = 0;
+               unsigned char *pack, c;
+               unsigned long base_offset;
+               unsigned lo, mid, hi;

 
 

-       delta_data = get_data(delta_size);
-       if (dry_run || !delta_data) {
-               free(delta_data);
-               return;
-       }
+               pack = fill(1);
+               c = *pack;
+               use(1);
+               base_offset = c & 127;
+               while (c & 128) {
+                       base_offset += 1;
+                       if (!base_offset || base_offset & ~(~0UL >> 7))
+                               die("offset value overflow for delta base object");
+                       pack = fill(1);
+                       c = *pack;
+                       use(1);
+                       base_offset = (base_offset << 7) + (c & 127);
+               }
+               base_offset = obj_list[nr].offset - base_offset;

 
 

-       if (!has_sha1_file(base_sha1)) {
-               add_delta_to_list(base_sha1, delta_data, delta_size);
-               return;
+               delta_data = get_data(delta_size);
+               if (dry_run || !delta_data) {
+                       free(delta_data);
+                       return;
+               }
+               lo = 0;
+               hi = nr;
+               while (lo < hi) {
+                       mid = (lo + hi)/2;
+                       if (base_offset < obj_list[mid].offset) {
+                               hi = mid;
+                       } else if (base_offset > obj_list[mid].offset) {
+                               lo = mid + 1;
+                       } else {
+                               hashcpy(base_sha1, obj_list[mid].sha1);
+                               base_found = !is_null_sha1(base_sha1);
+                               break;
+                       }
+               }
+               if (!base_found) {
+                       /* The delta base object is itself a delta that
+                          has not been resolved yet. */
+                       hashcpy(obj_list[nr].sha1, null_sha1);
+                       add_delta_to_list(nr, null_sha1, base_offset, delta_data, delta_size);
+                       return;
+               }

        }
        }

+

        base = read_sha1_file(base_sha1, type, &base_size);
        if (!base) {
                error("failed to read delta-pack base object %s",
        base = read_sha1_file(base_sha1, type, &base_size);
        if (!base) {
                error("failed to read delta-pack base object %s",


@@ -197,7 +266,7 @@ static void unpack_delta_entry(unsigned long delta_size)
                has_errors = 1;
                return;
        }
                has_errors = 1;
                return;
        }

-       resolve_delta(type, base, base_size, delta_data, delta_size);
+       resolve_delta(nr, type, base, base_size, delta_data, delta_size);

        free(base);
 }
 
        free(base);
 }
 


@@ -208,6 +277,8 @@ static void unpack_one(unsigned nr, unsigned total)
        unsigned long size;
        enum object_type type;
 
        unsigned long size;
        enum object_type type;
 

+       obj_list[nr].offset = consumed_bytes;
+

        pack = fill(1);
        c = *pack;
        use(1);
        pack = fill(1);
        c = *pack;
        use(1);


@@ -216,7 +287,7 @@ static void unpack_one(unsigned nr, unsigned total)
        shift = 4;
        while (c & 0x80) {
                pack = fill(1);
        shift = 4;
        while (c & 0x80) {
                pack = fill(1);

-               c = *pack++;
+               c = *pack;

                use(1);
                size += (c & 0x7f) << shift;
                shift += 7;
                use(1);
                size += (c & 0x7f) << shift;
                shift += 7;


@@ -225,13 +296,14 @@ static void unpack_one(unsigned nr, unsigned total)
                static unsigned long last_sec;
                static unsigned last_percent;
                struct timeval now;
                static unsigned long last_sec;
                static unsigned last_percent;
                struct timeval now;

-               unsigned percentage = (nr * 100) / total;
+               unsigned percentage = ((nr+1) * 100) / total;

 
                gettimeofday(&now, NULL);
                if (percentage != last_percent || now.tv_sec != last_sec) {
                        last_sec = now.tv_sec;
                        last_percent = percentage;
 
                gettimeofday(&now, NULL);
                if (percentage != last_percent || now.tv_sec != last_sec) {
                        last_sec = now.tv_sec;
                        last_percent = percentage;

-                       fprintf(stderr, "%4u%% (%u/%u) done\r", percentage, nr, total);
+                       fprintf(stderr, "%4u%% (%u/%u) done\r",
+                                       percentage, (nr+1), total);

                }
        }
        switch (type) {
                }
        }
        switch (type) {


@@ -239,10 +311,11 @@ static void unpack_one(unsigned nr, unsigned total)
        case OBJ_TREE:
        case OBJ_BLOB:
        case OBJ_TAG:
        case OBJ_TREE:
        case OBJ_BLOB:
        case OBJ_TAG:

-               unpack_non_delta_entry(type, size);
+               unpack_non_delta_entry(type, size, nr);

                return;
        case OBJ_REF_DELTA:
                return;
        case OBJ_REF_DELTA:

-               unpack_delta_entry(size);
+       case OBJ_OFS_DELTA:
+               unpack_delta_entry(type, size, nr);

                return;
        default:
                error("bad object type %d", type);
                return;
        default:
                error("bad object type %d", type);


@@ -265,9 +338,10 @@ static void unpack_all(void)
                die("unknown pack file version %d", ntohl(hdr->hdr_version));
        fprintf(stderr, "Unpacking %d objects\n", nr_objects);
 
                die("unknown pack file version %d", ntohl(hdr->hdr_version));
        fprintf(stderr, "Unpacking %d objects\n", nr_objects);
 

+       obj_list = xmalloc(nr_objects * sizeof(*obj_list));

        use(sizeof(struct pack_header));
        for (i = 0; i < nr_objects; i++)
        use(sizeof(struct pack_header));
        for (i = 0; i < nr_objects; i++)

-               unpack_one(i+1, nr_objects);
+               unpack_one(i, nr_objects);

        if (delta_list)
                die("unresolved deltas left after unpacking");
 }
        if (delta_list)
                die("unresolved deltas left after unpacking");
 }