Move pack functions to their own file

1 files changed, 490 insertions, 9 deletions

diff --git a/src/pack.c b/src/pack.c
index 71319a794..dca1903bd 100644
--- a/src/pack.c
+++ b/src/pack.c
@@ -27,12 +27,185 @@
 #include "odb.h"
 #include "pack.h"
 #include "delta-apply.h"
+#include "sha1_lookup.h"
 
 #include "git2/oid.h"
 #include "git2/zlib.h"
 
-unsigned char *pack_window_open(
-		struct pack_file *p,
+static int packfile_open(struct git_pack_file *p);
+static off_t nth_packed_object_offset(const struct git_pack_file *p, uint32_t n);
+int packfile_unpack_compressed(
+		git_rawobj *obj,
+		struct git_pack_file *p,
+		git_mwindow **w_curs,
+		off_t curpos,
+		size_t size,
+		git_otype type);
+
+/* Can find the offset of an object given
+ * a prefix of an identifier.
+ * Throws GIT_EAMBIGUOUSOIDPREFIX if short oid
+ * is ambiguous within the pack.
+ * This method assumes that len is between
+ * GIT_OID_MINPREFIXLEN and GIT_OID_HEXSZ.
+ */
+static int pack_entry_find_offset(
+		off_t *offset_out,
+		git_oid *found_oid,
+		struct git_pack_file *p,
+		const git_oid *short_oid,
+		unsigned int len);
+
+/***********************************************************
+ *
+ * PACK INDEX METHODS
+ *
+ ***********************************************************/
+
+static void pack_index_free(struct git_pack_file *p)
+{
+	if (p->index_map.data) {
+		git_futils_mmap_free(&p->index_map);
+		p->index_map.data = NULL;
+	}
+}
+
+static int pack_index_check(const char *path,  struct git_pack_file *p)
+{
+	struct git_pack_idx_header *hdr;
+	uint32_t version, nr, i, *index;
+
+	void *idx_map;
+	size_t idx_size;
+
+	struct stat st;
+
+	/* TODO: properly open the file without access time */
+	git_file fd = p_open(path, O_RDONLY /*| O_NOATIME */);
+
+	int error;
+
+	if (fd < 0)
+		return git__throw(GIT_EOSERR, "Failed to check index. File missing or corrupted");
+
+	if (p_fstat(fd, &st) < GIT_SUCCESS) {
+		p_close(fd);
+		return git__throw(GIT_EOSERR, "Failed to check index. File appears to be corrupted");
+	}
+
+	if (!git__is_sizet(st.st_size))
+		return GIT_ENOMEM;
+
+	idx_size = (size_t)st.st_size;
+
+	if (idx_size < 4 * 256 + 20 + 20) {
+		p_close(fd);
+		return git__throw(GIT_EOBJCORRUPTED, "Failed to check index. Object is corrupted");
+	}
+
+	error = git_futils_mmap_ro(&p->index_map, fd, 0, idx_size);
+	p_close(fd);
+
+	if (error < GIT_SUCCESS)
+		return git__rethrow(error, "Failed to check index");
+
+	hdr = idx_map = p->index_map.data;
+
+	if (hdr->idx_signature == htonl(PACK_IDX_SIGNATURE)) {
+		version = ntohl(hdr->idx_version);
+
+		if (version < 2 || version > 2) {
+			git_futils_mmap_free(&p->index_map);
+			return git__throw(GIT_EOBJCORRUPTED, "Failed to check index. Unsupported index version");
+		}
+
+	} else
+		version = 1;
+
+	nr = 0;
+	index = idx_map;
+
+	if (version > 1)
+		index += 2;  /* skip index header */
+
+	for (i = 0; i < 256; i++) {
+		uint32_t n = ntohl(index[i]);
+		if (n < nr) {
+			git_futils_mmap_free(&p->index_map);
+			return git__throw(GIT_EOBJCORRUPTED, "Failed to check index. Index is non-monotonic");
+		}
+		nr = n;
+	}
+
+	if (version == 1) {
+		/*
+		 * Total size:
+		 *  - 256 index entries 4 bytes each
+		 *  - 24-byte entries * nr (20-byte sha1 + 4-byte offset)
+		 *  - 20-byte SHA1 of the packfile
+		 *  - 20-byte SHA1 file checksum
+		 */
+		if (idx_size != 4*256 + nr * 24 + 20 + 20) {
+			git_futils_mmap_free(&p->index_map);
+			return git__throw(GIT_EOBJCORRUPTED, "Failed to check index. Object is corrupted");
+		}
+	} else if (version == 2) {
+		/*
+		 * Minimum size:
+		 *  - 8 bytes of header
+		 *  - 256 index entries 4 bytes each
+		 *  - 20-byte sha1 entry * nr
+		 *  - 4-byte crc entry * nr
+		 *  - 4-byte offset entry * nr
+		 *  - 20-byte SHA1 of the packfile
+		 *  - 20-byte SHA1 file checksum
+		 * And after the 4-byte offset table might be a
+		 * variable sized table containing 8-byte entries
+		 * for offsets larger than 2^31.
+		 */
+		unsigned long min_size = 8 + 4*256 + nr*(20 + 4 + 4) + 20 + 20;
+		unsigned long max_size = min_size;
+
+		if (nr)
+			max_size += (nr - 1)*8;
+
+		if (idx_size < min_size || idx_size > max_size) {
+			git_futils_mmap_free(&p->index_map);
+			return git__throw(GIT_EOBJCORRUPTED, "Failed to check index. Wrong index size");
+		}
+
+		/* Make sure that off_t is big enough to access the whole pack...
+		 * Is this an issue in libgit2? It shouldn't. */
+		if (idx_size != min_size && (sizeof(off_t) <= 4)) {
+			git_futils_mmap_free(&p->index_map);
+			return git__throw(GIT_EOSERR, "Failed to check index. off_t not big enough to access the whole pack");
+		}
+	}
+
+	p->index_version = version;
+	p->num_objects = nr;
+	return GIT_SUCCESS;
+}
+
+static int pack_index_open(struct git_pack_file *p)
+{
+	char *idx_name;
+	int error;
+
+	if (p->index_map.data)
+		return GIT_SUCCESS;
+
+	idx_name = git__strdup(p->pack_name);
+	strcpy(idx_name + strlen(idx_name) - STRLEN(".pack"), ".idx");
+
+	error = pack_index_check(idx_name, p);
+	free(idx_name);
+
+	return error == GIT_SUCCESS ? GIT_SUCCESS : git__rethrow(error, "Failed to open index");
+}
+
+static unsigned char *pack_window_open(
+		struct git_pack_file *p,
 		git_mwindow **w_cursor,
 		off_t offset,
 		unsigned int *left)
@@ -109,9 +282,9 @@ int git_packfile_unpack_header(
 	return GIT_SUCCESS;
 }
 
-int packfile_unpack_delta(
+static int packfile_unpack_delta(
 		git_rawobj *obj,
-		struct pack_file *p,
+		struct git_pack_file *p,
 		git_mwindow **w_curs,
 		off_t curpos,
 		size_t delta_size,
@@ -127,7 +300,7 @@ int packfile_unpack_delta(
 		return git__throw(GIT_EOBJCORRUPTED, "Delta offset is zero");
 
 	git_mwindow_close(w_curs);
-	error = packfile_unpack(&base, p, base_offset);
+	error = git_packfile_unpack(&base, p, base_offset);
 
 	/*
 	 * TODO: git.git tries to load the base from other packfiles
@@ -157,9 +330,9 @@ int packfile_unpack_delta(
 	return error; /* error set by git__delta_apply */
 }
 
-int packfile_unpack(
+int git_packfile_unpack(
 		git_rawobj *obj,
-		struct pack_file *p,
+		struct git_pack_file *p,
 		off_t obj_offset)
 {
 	git_mwindow *w_curs = NULL;
@@ -209,7 +382,7 @@ int packfile_unpack(
 
 int packfile_unpack_compressed(
 		git_rawobj *obj,
-		struct pack_file *p,
+		struct git_pack_file *p,
 		git_mwindow **w_curs,
 		off_t curpos,
 		size_t size,
@@ -257,7 +430,7 @@ int packfile_unpack_compressed(
 }
 
 off_t get_delta_base(
-		struct pack_file *p,
+		struct git_pack_file *p,
 		git_mwindow **w_curs,
 		off_t *curpos,
 		git_otype type,
@@ -298,3 +471,311 @@ off_t get_delta_base(
 
 	return base_offset;
 }
+
+/***********************************************************
+ *
+ * PACKFILE METHODS
+ *
+ ***********************************************************/
+
+static struct git_pack_file *packfile_alloc(int extra)
+{
+	struct git_pack_file *p = git__malloc(sizeof(*p) + extra);
+	memset(p, 0, sizeof(*p));
+	p->mwf.fd = -1;
+	return p;
+}
+
+
+void packfile_free(struct git_pack_file *p)
+{
+	assert(p);
+
+	/* clear_delta_base_cache(); */
+	git_mwindow_free_all(&p->mwf);
+
+	if (p->mwf.fd != -1)
+		p_close(p->mwf.fd);
+
+	pack_index_free(p);
+
+	free(p->bad_object_sha1);
+	free(p);
+}
+
+static int packfile_open(struct git_pack_file *p)
+{
+	struct stat st;
+	struct git_pack_header hdr;
+	git_oid sha1;
+	unsigned char *idx_sha1;
+
+	if (!p->index_map.data && pack_index_open(p) < GIT_SUCCESS)
+		return git__throw(GIT_ENOTFOUND, "Failed to open packfile. File not found");
+
+	/* TODO: open with noatime */
+	p->mwf.fd = p_open(p->pack_name, O_RDONLY);
+	if (p->mwf.fd < 0 || p_fstat(p->mwf.fd, &st) < GIT_SUCCESS)
+		return git__throw(GIT_EOSERR, "Failed to open packfile. File appears to be corrupted");
+
+	if (git_mwindow_file_register(&p->mwf) < GIT_SUCCESS) {
+		p_close(p->mwf.fd);
+		return git__throw(GIT_ERROR, "Failed to register packfile windows");
+	}
+
+	/* If we created the struct before we had the pack we lack size. */
+	if (!p->mwf.size) {
+		if (!S_ISREG(st.st_mode))
+			goto cleanup;
+		p->mwf.size = (off_t)st.st_size;
+	} else if (p->mwf.size != st.st_size)
+		goto cleanup;
+
+#if 0
+	/* We leave these file descriptors open with sliding mmap;
+	 * there is no point keeping them open across exec(), though.
+	 */
+	fd_flag = fcntl(p->mwf.fd, F_GETFD, 0);
+	if (fd_flag < 0)
+		return error("cannot determine file descriptor flags");
+
+	fd_flag |= FD_CLOEXEC;
+	if (fcntl(p->pack_fd, F_SETFD, fd_flag) == -1)
+		return GIT_EOSERR;
+#endif
+
+	/* Verify we recognize this pack file format. */
+	if (p_read(p->mwf.fd, &hdr, sizeof(hdr)) < GIT_SUCCESS)
+		goto cleanup;
+
+	if (hdr.hdr_signature != htonl(PACK_SIGNATURE))
+		goto cleanup;
+
+	if (!pack_version_ok(hdr.hdr_version))
+		goto cleanup;
+
+	/* Verify the pack matches its index. */
+	if (p->num_objects != ntohl(hdr.hdr_entries))
+		goto cleanup;
+
+	if (p_lseek(p->mwf.fd, p->mwf.size - GIT_OID_RAWSZ, SEEK_SET) == -1)
+		goto cleanup;
+
+	if (p_read(p->mwf.fd, sha1.id, GIT_OID_RAWSZ) < GIT_SUCCESS)
+		goto cleanup;
+
+	idx_sha1 = ((unsigned char *)p->index_map.data) + p->index_map.len - 40;
+
+	if (git_oid_cmp(&sha1, (git_oid *)idx_sha1) != 0)
+		goto cleanup;
+
+	return GIT_SUCCESS;
+
+cleanup:
+	p_close(p->mwf.fd);
+	p->mwf.fd = -1;
+	return git__throw(GIT_EPACKCORRUPTED, "Failed to open packfile. Pack is corrupted");
+}
+
+int git_packfile_check(struct git_pack_file **pack_out, const char *path)
+{
+	struct stat st;
+	struct git_pack_file *p;
+	size_t path_len;
+
+	*pack_out = NULL;
+	path_len = strlen(path);
+	p = packfile_alloc(path_len + 2);
+
+	/*
+	 * Make sure a corresponding .pack file exists and that
+	 * the index looks sane.
+	 */
+	path_len -= STRLEN(".idx");
+	if (path_len < 1) {
+		free(p);
+		return git__throw(GIT_ENOTFOUND, "Failed to check packfile. Wrong path name");
+	}
+
+	memcpy(p->pack_name, path, path_len);
+
+	strcpy(p->pack_name + path_len, ".keep");
+	if (git_futils_exists(p->pack_name) == GIT_SUCCESS)
+		p->pack_keep = 1;
+
+	strcpy(p->pack_name + path_len, ".pack");
+	if (p_stat(p->pack_name, &st) < GIT_SUCCESS || !S_ISREG(st.st_mode)) {
+		free(p);
+		return git__throw(GIT_ENOTFOUND, "Failed to check packfile. File not found");
+	}
+
+	/* ok, it looks sane as far as we can check without
+	 * actually mapping the pack file.
+	 */
+	p->mwf.size = (off_t)st.st_size;
+	p->pack_local = 1;
+	p->mtime = (git_time_t)st.st_mtime;
+
+	/* see if we can parse the sha1 oid in the packfile name */
+	if (path_len < 40 ||
+		git_oid_fromstr(&p->sha1, path + path_len - GIT_OID_HEXSZ) < GIT_SUCCESS)
+		memset(&p->sha1, 0x0, GIT_OID_RAWSZ);
+
+	*pack_out = p;
+	return GIT_SUCCESS;
+}
+
+/***********************************************************
+ *
+ * PACKFILE ENTRY SEARCH INTERNALS
+ *
+ ***********************************************************/
+
+static off_t nth_packed_object_offset(const struct git_pack_file *p, uint32_t n)
+{
+	const unsigned char *index = p->index_map.data;
+	index += 4 * 256;
+	if (p->index_version == 1) {
+		return ntohl(*((uint32_t *)(index + 24 * n)));
+	} else {
+		uint32_t off;
+		index += 8 + p->num_objects * (20 + 4);
+		off = ntohl(*((uint32_t *)(index + 4 * n)));
+		if (!(off & 0x80000000))
+			return off;
+		index += p->num_objects * 4 + (off & 0x7fffffff) * 8;
+		return (((uint64_t)ntohl(*((uint32_t *)(index + 0)))) << 32) |
+				   ntohl(*((uint32_t *)(index + 4)));
+	}
+}
+
+static int pack_entry_find_offset(
+		off_t *offset_out,
+		git_oid *found_oid,
+		struct git_pack_file *p,
+		const git_oid *short_oid,
+		unsigned int len)
+{
+	const uint32_t *level1_ofs = p->index_map.data;
+	const unsigned char *index = p->index_map.data;
+	unsigned hi, lo, stride;
+	int pos, found = 0;
+	const unsigned char *current = 0;
+
+	*offset_out = 0;
+
+	if (index == NULL) {
+		int error;
+
+		if ((error = pack_index_open(p)) < GIT_SUCCESS)
+			return git__rethrow(error, "Failed to find offset for pack entry");
+
+		assert(p->index_map.data);
+
+		index = p->index_map.data;
+		level1_ofs = p->index_map.data;
+	}
+
+	if (p->index_version > 1) {
+		level1_ofs += 2;
+		index += 8;
+	}
+
+	index += 4 * 256;
+	hi = ntohl(level1_ofs[(int)short_oid->id[0]]);
+	lo = ((short_oid->id[0] == 0x0) ? 0 : ntohl(level1_ofs[(int)short_oid->id[0] - 1]));
+
+	if (p->index_version > 1) {
+		stride = 20;
+	} else {
+		stride = 24;
+		index += 4;
+	}
+
+#ifdef INDEX_DEBUG_LOOKUP
+	printf("%02x%02x%02x... lo %u hi %u nr %d\n",
+		short_oid->id[0], short_oid->id[1], short_oid->id[2], lo, hi, p->num_objects);
+#endif
+
+	/* Use git.git lookup code */
+	pos =  sha1_entry_pos(index, stride, 0, lo, hi, p->num_objects, short_oid->id);
+
+	if (pos >= 0) {
+		/* An object matching exactly the oid was found */
+		found = 1;
+		current = index + pos * stride;
+	} else {
+		/* No object was found */
+		/* pos refers to the object with the "closest" oid to short_oid */
+		pos = - 1 - pos;
+		if (pos < (int)p->num_objects) {
+			current = index + pos * stride;
+
+			if (!git_oid_ncmp(short_oid, (const git_oid *)current, len)) {
+				found = 1;
+			}
+		}
+	}
+
+	if (found && pos + 1 < (int)p->num_objects) {
+		/* Check for ambiguousity */
+		const unsigned char *next = current + stride;
+
+		if (!git_oid_ncmp(short_oid, (const git_oid *)next, len)) {
+			found = 2;
+		}
+	}
+
+	if (!found) {
+		return git__throw(GIT_ENOTFOUND, "Failed to find offset for pack entry. Entry not found");
+	} else if (found > 1) {
+		return git__throw(GIT_EAMBIGUOUSOIDPREFIX, "Failed to find offset for pack entry. Ambiguous sha1 prefix within pack");
+	} else {
+		*offset_out = nth_packed_object_offset(p, pos);
+		git_oid_fromraw(found_oid, current);
+
+#ifdef INDEX_DEBUG_LOOKUP
+		unsigned char hex_sha1[GIT_OID_HEXSZ + 1];
+		git_oid_fmt(hex_sha1, found_oid);
+		hex_sha1[GIT_OID_HEXSZ] = '\0';
+		printf("found lo=%d %s\n", lo, hex_sha1);
+#endif
+		return GIT_SUCCESS;
+	}
+}
+
+int git_pack_entry_find(
+		struct git_pack_entry *e,
+		struct git_pack_file *p,
+		const git_oid *short_oid,
+		unsigned int len)
+{
+	off_t offset;
+	git_oid found_oid;
+	int error;
+
+	assert(p);
+
+	if (len == GIT_OID_HEXSZ && p->num_bad_objects) {
+		unsigned i;
+		for (i = 0; i < p->num_bad_objects; i++)
+			if (git_oid_cmp(short_oid, &p->bad_object_sha1[i]) == 0)
+				return git__throw(GIT_ERROR, "Failed to find pack entry. Bad object found");
+	}
+
+	error = pack_entry_find_offset(&offset, &found_oid, p, short_oid, len);
+	if (error < GIT_SUCCESS)
+		return git__rethrow(error, "Failed to find pack entry. Couldn't find offset");
+
+	/* we found a unique entry in the index;
+	 * make sure the packfile backing the index
+	 * still exists on disk */
+	if (p->mwf.fd == -1 && packfile_open(p) < GIT_SUCCESS)
+		return git__throw(GIT_EOSERR, "Failed to find pack entry. Packfile doesn't exist on disk");
+
+	e->offset = offset;
+	e->p = p;
+
+	git_oid_cpy(&e->sha1, &found_oid);
+	return GIT_SUCCESS;
+}