Properly split out PackIndex1 and PackIndex2.

diff --git a/dulwich/object_store.py b/dulwich/object_store.py
index ee92aad550489a5b6da31f82289fee6b8f32d10c..5ef52861ea8eb4a4bfd216b4c761cf9da52a65a6 100644 (file)
--- a/dulwich/object_store.py
+++ b/dulwich/object_store.py
@@ -30,6 +30,7 @@ from dulwich.pack import (
     PackData, 
     iter_sha1, 
     load_packs, 
+    load_pack_index,
     write_pack,
     write_pack_data,
     write_pack_index_v2,
@@ -145,7 +146,7 @@ class ObjectStore(object):
         temppath = os.path.join(self.pack_dir, 
             sha_to_hex(urllib2.randombytes(20))+".temppack")
         write_pack(temppath, p.iterobjects(self.get_raw), len(p))
-        pack_sha = PackIndex(temppath+".idx").objects_sha1()
+        pack_sha = load_pack_index(temppath+".idx").objects_sha1()
         newbasename = os.path.join(self.pack_dir, "pack-%s" % pack_sha)
         os.rename(temppath+".pack", newbasename+".pack")
         os.rename(temppath+".idx", newbasename+".idx")

diff --git a/dulwich/pack.py b/dulwich/pack.py
index e754274b261549b89ca81f8f7082899b8373eef9..2006dae615257d1251528ef7ff1a47af00fb5827 100644 (file)
--- a/dulwich/pack.py
+++ b/dulwich/pack.py
@@ -90,6 +90,7 @@ def read_zlib(data, offset, dec_size):
 
 
 def iter_sha1(iter):
+    """Return the hexdigest of the SHA1 over a set of names."""
     sha1 = make_sha()
     for name in iter:
         sha1.update(name)
@@ -117,6 +118,20 @@ def simple_mmap(f, offset, size, access=mmap.ACCESS_READ):
         return mem, offset
 
 
+def load_pack_index(filename):
+    f = open(filename, 'r')
+    if f.read(4) == '\377tOc':
+        version = struct.unpack(">L", f.read(4))[0]
+        if version == 2:
+            f.seek(0)
+            return PackIndex2(filename, file=f)
+        else:
+            raise KeyError("Unknown pack index format %d" % version)
+    else:
+        f.seek(0)
+        return PackIndex1(filename, file=f)
+
+
 class PackIndex(object):
     """An index in to a packfile.
   
@@ -131,7 +146,7 @@ class PackIndex(object):
     the start and end offset and then bisect in to find if the value is present.
     """
   
-    def __init__(self, filename):
+    def __init__(self, filename, file=None):
         """Create a pack index object.
     
         Provide it with the name of the index file to consider, and it will map
@@ -141,22 +156,15 @@ class PackIndex(object):
         # Take the size now, so it can be checked each time we map the file to
         # ensure that it hasn't changed.
         self._size = os.path.getsize(filename)
-        self._file = open(filename, 'r')
+        if file is None:
+            self._file = open(filename, 'r')
+        else:
+            self._file = file
         self._contents, map_offset = simple_mmap(self._file, 0, self._size)
         assert map_offset == 0
-        if self._contents[:4] != '\377tOc':
-            self.version = 1
-            self._fan_out_table = self._read_fan_out_table(0)
-        else:
-            (self.version, ) = unpack_from(">L", self._contents, 4)
-            assert self.version in (2,), "Version was %d" % self.version
-            self._fan_out_table = self._read_fan_out_table(8)
-            self._name_table_offset = 8 + 0x100 * 4
-            self._crc32_table_offset = self._name_table_offset + 20 * len(self)
-            self._pack_offset_table_offset = self._crc32_table_offset + 4 * len(self)
   
     def __eq__(self, other):
-        if type(self) != type(other):
+        if not isinstance(other, PackIndex):
             return False
     
         if self._fan_out_table != other._fan_out_table:
@@ -179,34 +187,19 @@ class PackIndex(object):
     
         :return: Tuple with object name (SHA), offset in pack file and 
               CRC32 checksum (if known)."""
-        if self.version == 1:
-            (offset, name) = unpack_from(">L20s", self._contents, 
-                (0x100 * 4) + (i * 24))
-            return (name, offset, None)
-        else:
-            return (self._unpack_name(i), self._unpack_offset(i), 
-                    self._unpack_crc32_checksum(i))
+        raise NotImplementedError(self._unpack_entry)
   
     def _unpack_name(self, i):
-        if self.version == 1:
-            offset = (0x100 * 4) + (i * 24) + 4
-        else:
-            offset = self._name_table_offset + i * 20
-        return self._contents[offset:offset+20]
+        """Unpack the i-th name from the index file."""
+        raise NotImplementedError(self._unpack_name)
   
     def _unpack_offset(self, i):
-        if self.version == 1:
-            offset = (0x100 * 4) + (i * 24)
-        else:
-            offset = self._pack_offset_table_offset + i * 4
-        return unpack_from(">L", self._contents, offset)[0]
-  
+        """Unpack the i-th object offset from the index file."""
+        raise NotImplementedError(self._unpack_offset)
+
     def _unpack_crc32_checksum(self, i):
-        if self.version == 1:
-            return None
-        else:
-            return unpack_from(">L", self._contents, 
-                                      self._crc32_table_offset + i * 4)[0]
+        """Unpack the crc32 checksum for the i-th object from the index file."""
+        raise NotImplementedError(self._unpack_crc32_checksum)
   
     def __iter__(self):
         return imap(sha_to_hex, self._itersha())
@@ -216,6 +209,10 @@ class PackIndex(object):
             yield self._unpack_name(i)
   
     def objects_sha1(self):
+        """Return the hex SHA1 over all the shas of all objects in this pack.
+        
+        :note: This is used for the filename of the pack.
+        """
         return iter_sha1(self._itersha())
   
     def iterentries(self):
@@ -287,6 +284,63 @@ class PackIndex(object):
         return None
 
 
+class PackIndex1(PackIndex):
+    """Version 1 Pack Index."""
+
+    def __init__(self, filename, file=None):
+        PackIndex.__init__(self, filename, file)
+        self.version = 1
+        self._fan_out_table = self._read_fan_out_table(0)
+
+    def _unpack_entry(self, i):
+        (offset, name) = unpack_from(">L20s", self._contents, 
+            (0x100 * 4) + (i * 24))
+        return (name, offset, None)
+ 
+    def _unpack_name(self, i):
+        offset = (0x100 * 4) + (i * 24) + 4
+        return self._contents[offset:offset+20]
+  
+    def _unpack_offset(self, i):
+        offset = (0x100 * 4) + (i * 24)
+        return unpack_from(">L", self._contents, offset)[0]
+  
+    def _unpack_crc32_checksum(self, i):
+        # Not stored in v1 index files
+        return None 
+  
+
+class PackIndex2(PackIndex):
+    """Version 2 Pack Index."""
+
+    def __init__(self, filename, file=None):
+        PackIndex.__init__(self, filename, file)
+        assert self._contents[:4] == '\377tOc', "Not a v2 pack index file"
+        (self.version, ) = unpack_from(">L", self._contents, 4)
+        assert self.version == 2, "Version was %d" % self.version
+        self._fan_out_table = self._read_fan_out_table(8)
+        self._name_table_offset = 8 + 0x100 * 4
+        self._crc32_table_offset = self._name_table_offset + 20 * len(self)
+        self._pack_offset_table_offset = self._crc32_table_offset + 4 * len(self)
+
+    def _unpack_entry(self, i):
+        return (self._unpack_name(i), self._unpack_offset(i), 
+                self._unpack_crc32_checksum(i))
+ 
+    def _unpack_name(self, i):
+        offset = self._name_table_offset + i * 20
+        return self._contents[offset:offset+20]
+  
+    def _unpack_offset(self, i):
+        offset = self._pack_offset_table_offset + i * 4
+        return unpack_from(">L", self._contents, offset)[0]
+  
+    def _unpack_crc32_checksum(self, i):
+        return unpack_from(">L", self._contents, 
+                          self._crc32_table_offset + i * 4)[0]
+  
+
+
 def read_pack_header(f):
     header = f.read(12)
     assert header[:4] == "PACK"
@@ -846,7 +900,7 @@ class Pack(object):
     @property
     def idx(self):
         if self._idx is None:
-            self._idx = PackIndex(self._idx_path)
+            self._idx = load_pack_index(self._idx_path)
         return self._idx
 
     def close(self):

diff --git a/dulwich/tests/test_pack.py b/dulwich/tests/test_pack.py
index e74fa7da576c0516daa7ee9babee2220336bea46..bbaa8dc12c63d635795f2d12b9002dcb34086171 100644 (file)
--- a/dulwich/tests/test_pack.py
+++ b/dulwich/tests/test_pack.py
@@ -25,10 +25,10 @@ from dulwich.objects import (
     )
 from dulwich.pack import (
     Pack,
-    PackIndex,
     PackData,
     apply_delta,
     create_delta,
+    load_pack_index,
     hex_to_sha,
     read_zlib,
     sha_to_hex,
@@ -50,7 +50,7 @@ class PackTests(unittest.TestCase):
   
     def get_pack_index(self, sha):
         """Returns a PackIndex from the datadir with the given sha"""
-        return PackIndex(os.path.join(self.datadir, 'pack-%s.idx' % sha))
+        return load_pack_index(os.path.join(self.datadir, 'pack-%s.idx' % sha))
   
     def get_pack_data(self, sha):
         """Returns a PackData object from the datadir with the given sha"""
@@ -144,14 +144,14 @@ class TestPackData(PackTests):
     def test_create_index_v1(self):
         p = self.get_pack_data(pack1_sha)
         p.create_index_v1("v1test.idx")
-        idx1 = PackIndex("v1test.idx")
+        idx1 = load_pack_index("v1test.idx")
         idx2 = self.get_pack_index(pack1_sha)
         self.assertEquals(idx1, idx2)
   
     def test_create_index_v2(self):
         p = self.get_pack_data(pack1_sha)
         p.create_index_v2("v2test.idx")
-        idx1 = PackIndex("v2test.idx")
+        idx1 = load_pack_index("v2test.idx")
         idx2 = self.get_pack_index(pack1_sha)
         self.assertEquals(idx1, idx2)
 
@@ -229,7 +229,7 @@ class BaseTestPackIndexWriting(object):
     def test_empty(self):
         pack_checksum = 'r\x19\x80\xe8f\xaf\x9a_\x93\xadgAD\xe1E\x9b\x8b\xa3\xe7\xb7'
         self._write_fn("empty.idx", [], pack_checksum)
-        idx = PackIndex("empty.idx")
+        idx = load_pack_index("empty.idx")
         self.assertTrue(idx.check())
         self.assertEquals(idx.get_pack_checksum(), pack_checksum)
         self.assertEquals(0, len(idx))
@@ -239,7 +239,7 @@ class BaseTestPackIndexWriting(object):
         my_entries = [('og\x0c\x0f\xb5?\x94cv\x0br\x95\xfb\xb8\x14\xe9e\xfb \xc8', 178, 42)]
         my_entries.sort()
         self._write_fn("single.idx", my_entries, pack_checksum)
-        idx = PackIndex("single.idx")
+        idx = load_pack_index("single.idx")
         self.assertEquals(idx.version, self._expected_version)
         self.assertTrue(idx.check())
         self.assertEquals(idx.get_pack_checksum(), pack_checksum)