def _assert_pack_file(self, pack, version, size): assert pack.version() == 2 assert pack.size() == size assert len(pack.checksum()) == 20 num_obj = 0 for obj in pack.stream_iter(): num_obj += 1 info = pack.info(obj.pack_offset) stream = pack.stream(obj.pack_offset) assert info.pack_offset == stream.pack_offset assert info.type_id == stream.type_id assert hasattr(stream, 'read') # it should be possible to read from both streams assert obj.read() == stream.read() streams = pack.collect_streams(obj.pack_offset) assert streams # read the stream try: dstream = DeltaApplyReader.new(streams) except ValueError: # ignore these, old git versions use only ref deltas, # which we havent resolved ( as we are without an index ) # Also ignore non-delta streams continue # END get deltastream # read all data = dstream.read() assert len(data) == dstream.size # test seek dstream.seek(0) assert dstream.read() == data # read chunks # NOTE: the current implementation is safe, it basically transfers # all calls to the underlying memory map # END for each object assert num_obj == size
def _assert_pack_file(self, pack, version, size): assert pack.version() == 2 assert pack.size() == size assert len(pack.checksum()) == 20 num_obj = 0 for obj in pack.stream_iter(): num_obj += 1 info = pack.info(obj.pack_offset) stream = pack.stream(obj.pack_offset) assert info.pack_offset == stream.pack_offset assert info.type_id == stream.type_id assert hasattr(stream, 'read') # it should be possible to read from both streams assert obj.read() == stream.read() streams = pack.collect_streams(obj.pack_offset) assert streams # read the stream try: dstream = DeltaApplyReader.new(streams) except ValueError: # ignore these, old git versions use only ref deltas, # which we havent resolved ( as we are without an index ) # Also ignore non-delta streams continue # END get deltastream # read all data = dstream.read() assert len(data) == dstream.size # test seek dstream.seek(0) assert dstream.read() == data # read chunks # NOTE: the current implementation is safe, it basically transfers # all calls to the underlying memory map # END for each object assert num_obj == size
def _object(self, sha, as_stream, index=-1):
""":return: OInfo or OStream object providing information about the given sha
:param index: if not -1, its assumed to be the sha's index in the IndexFile"""
# its a little bit redundant here, but it needs to be efficient
if index < 0:
index = self._sha_to_index(sha)
if sha is None:
sha = self._index.sha(index)
# END assure sha is present ( in output )
offset = self._index.offset(index)
type_id, uncomp_size, data_rela_offset = pack_object_header_info(
self._pack._cursor.use_region(offset).buffer())
if as_stream:
if type_id not in delta_types:
packstream = self._pack.stream(offset)
return OStream(sha, packstream.type, packstream.size,
packstream.stream)
# END handle non-deltas
# produce a delta stream containing all info
# To prevent it from applying the deltas when querying the size,
# we extract it from the delta stream ourselves
streams = self.collect_streams_at_offset(offset)
dstream = DeltaApplyReader.new(streams)
return ODeltaStream(sha, dstream.type, None, dstream)
else:
if type_id not in delta_types:
return OInfo(sha, type_id_to_type_map[type_id], uncomp_size)
# END handle non-deltas
# deltas are a little tougher - unpack the first bytes to obtain
# the actual target size, as opposed to the size of the delta data
streams = self.collect_streams_at_offset(offset)
buf = streams[0].read(512)
offset, src_size = msb_size(buf)
offset, target_size = msb_size(buf, offset)
# collect the streams to obtain the actual object type
if streams[-1].type_id in delta_types:
raise BadObject(sha, "Could not resolve delta object")
return OInfo(sha, streams[-1].type, target_size)
def _object(self, sha, as_stream, index=-1):
""":return: OInfo or OStream object providing information about the given sha
:param index: if not -1, its assumed to be the sha's index in the IndexFile"""
# its a little bit redundant here, but it needs to be efficient
if index < 0:
index = self._sha_to_index(sha)
if sha is None:
sha = self._index.sha(index)
# END assure sha is present ( in output )
offset = self._index.offset(index)
type_id, uncomp_size, data_rela_offset = pack_object_header_info(self._pack._cursor.use_region(offset).buffer())
if as_stream:
if type_id not in delta_types:
packstream = self._pack.stream(offset)
return OStream(sha, packstream.type, packstream.size, packstream.stream)
# END handle non-deltas
# produce a delta stream containing all info
# To prevent it from applying the deltas when querying the size,
# we extract it from the delta stream ourselves
streams = self.collect_streams_at_offset(offset)
dstream = DeltaApplyReader.new(streams)
return ODeltaStream(sha, dstream.type, None, dstream)
else:
if type_id not in delta_types:
return OInfo(sha, type_id_to_type_map[type_id], uncomp_size)
# END handle non-deltas
# deltas are a little tougher - unpack the first bytes to obtain
# the actual target size, as opposed to the size of the delta data
streams = self.collect_streams_at_offset(offset)
buf = streams[0].read(512)
offset, src_size = msb_size(buf)
offset, target_size = msb_size(buf, offset)
# collect the streams to obtain the actual object type
if streams[-1].type_id in delta_types:
raise BadObject(sha, "Could not resolve delta object")
return OInfo(sha, streams[-1].type, target_size)
