def test_deflated_smaller_window_buffer(self):
# zlib on some systems uses smaller buffers,
# resulting in a different header.
# See https://github.com/libgit2/libgit2/pull/464
sf = ShaFile.from_file(BytesIO(small_buffer_zlib_object))
self.assertEqual(sf.type_name, "tag")
self.assertEqual(sf.tagger, " <@localhost>")
def test_deflated_smaller_window_buffer(self):
# zlib on some systems uses smaller buffers,
# resulting in a different header.
# See https://github.com/libgit2/libgit2/pull/464
sf = ShaFile.from_file(BytesIO(small_buffer_zlib_object))
self.assertEqual(sf.type_name, b'tag')
self.assertEqual(sf.tagger, b' <@localhost>')
def _get_loose_object(self, sha):
path = osutils.joinpath(self._split_loose_object(sha))
try:
with self.transport.get(urlutils.quote_from_bytes(path)) as f:
return ShaFile.from_file(f)
except NoSuchFile:
return None
def _get_loose_object(self, sha):
path = self._get_shafile_path(sha)
try:
return ShaFile.from_path(path)
except (OSError, IOError) as e:
if e.errno == errno.ENOENT:
return None
raise
def _get_loose_object(self, sha):
path = self._get_shafile_path(sha)
try:
return ShaFile.from_path(path)
except (OSError, IOError), e:
if e.errno == errno.ENOENT:
return None
raise
def iterobjects(self, get_raw=None):
"""Iterate over the objects in this pack."""
if get_raw is None:
get_raw = self.get_raw
for offset, type, obj, crc32 in self.data.iterobjects():
assert isinstance(offset, int)
yield ShaFile.from_raw_string(
*self.data.resolve_object(offset, type, obj, get_raw))
def iterobjects(self, get_raw=None):
"""Iterate over the objects in this pack."""
if get_raw is None:
get_raw = self.get_raw
for offset, type, obj, crc32 in self.data.iterobjects():
assert isinstance(offset, int)
yield ShaFile.from_raw_string(
*self.data.resolve_object(offset, type, obj, get_raw))
def _get_git_object(self, sha: bytes) -> ShaFile:
# try to get the object from a pack file first to avoid flooding
# git server with numerous HTTP requests
for pack in list(self.packs):
try:
if sha in pack:
return pack[sha]
except (NotGitRepository, struct.error):
# missing (dulwich http client raises NotGitRepository on 404)
# or invalid pack index/content, remove it from global packs list
logger.debug(
"A pack file is missing or its content is invalid")
self.packs.remove(pack)
# fetch it from objects/ directory otherwise
sha_hex = sha.decode()
object_path = f"objects/{sha_hex[:2]}/{sha_hex[2:]}"
return ShaFile.from_file(self._http_get(object_path))
def iterentries(self, ext_resolve_ref=None, progress=None):
"""Yield entries summarizing the contents of this pack.
:param ext_resolve_ref: Optional function to resolve base
objects (in case this is a thin pack)
:param progress: Progress function, called with current and
total object count.
This will yield tuples with (sha, offset, crc32)
"""
found = {}
postponed = defaultdict(list)
class Postpone(Exception):
"""Raised to postpone delta resolving."""
def get_ref_text(sha):
assert len(sha) == 20
if sha in found:
return self.get_object_at(found[sha])
if ext_resolve_ref:
try:
return ext_resolve_ref(sha)
except KeyError:
pass
raise Postpone, (sha, )
extra = []
todo = chain(self.iterobjects(progress=progress), extra)
for (offset, type, obj, crc32) in todo:
assert isinstance(offset, int)
assert isinstance(type, int)
assert isinstance(obj, tuple) or isinstance(obj, str)
try:
type, obj = self.resolve_object(offset, type, obj,
get_ref_text)
except Postpone, (sha, ):
postponed[sha].append((offset, type, obj))
else:
shafile = ShaFile.from_raw_string(type, obj)
sha = shafile.sha().digest()
found[sha] = offset
yield sha, offset, crc32
extra.extend(postponed.get(sha, []))
def iterentries(self, ext_resolve_ref=None, progress=None):
"""Yield entries summarizing the contents of this pack.
:param ext_resolve_ref: Optional function to resolve base
objects (in case this is a thin pack)
:param progress: Progress function, called with current and
total object count.
This will yield tuples with (sha, offset, crc32)
"""
found = {}
postponed = defaultdict(list)
class Postpone(Exception):
"""Raised to postpone delta resolving."""
def get_ref_text(sha):
assert len(sha) == 20
if sha in found:
return self.get_object_at(found[sha])
if ext_resolve_ref:
try:
return ext_resolve_ref(sha)
except KeyError:
pass
raise Postpone, (sha, )
extra = []
todo = chain(self.iterobjects(progress=progress), extra)
for (offset, type, obj, crc32) in todo:
assert isinstance(offset, int)
assert isinstance(type, int)
assert isinstance(obj, tuple) or isinstance(obj, str)
try:
type, obj = self.resolve_object(offset, type, obj, get_ref_text)
except Postpone, (sha, ):
postponed[sha].append((offset, type, obj))
else:
shafile = ShaFile.from_raw_string(type, obj)
sha = shafile.sha().digest()
found[sha] = offset
yield sha, offset, crc32
extra.extend(postponed.get(sha, []))
def store(self, istream):
obj = ShaFile.from_raw_string(type_to_type_id_map[istream.type], istream.read())
self._dw_repo.object_store.add_object(obj)
istream.binsha = obj.sha().digest()
return istream
def _do_test_count_blocks_chunks(self, count_blocks):
blob = ShaFile.from_raw_chunks(Blob.type_num, [b"a\nb", b"\na\n"])
self.assertEqual({hash("a\n"): 4, hash("b\n"): 2}, count_blocks(blob))
def _do_test_count_blocks_chunks(self, count_blocks):
blob = ShaFile.from_raw_chunks(Blob.type_num, ['a\nb', '\na\n'])
self.assertEqual({hash('a\n'): 4, hash('b\n'): 2}, _count_blocks(blob))
def _get_shafile(self, sha):
path = self._get_shafile_path(sha)
if os.path.exists(path):
return ShaFile.from_file(path)
return None
def _do_test_count_blocks_chunks(self, count_blocks):
blob = ShaFile.from_raw_chunks(Blob.type_num, ['a\nb', '\na\n'])
self.assertEqual({hash('a\n'): 4, hash('b\n'): 2}, _count_blocks(blob))
def store(self, istream):
obj = ShaFile.from_raw_string(type_to_type_id_map[istream.type],
istream.read())
self._dw_repo.object_store.add_object(obj)
istream.binsha = obj.sha().digest()
return istream
def __getitem__(self, sha):
stream = self._vf.get_record_stream([(sha,)], 'unordered', True)
entry = next(stream)
if entry.storage_kind == 'absent':
raise KeyError(sha)
return ShaFile._parse_legacy_object(entry.get_bytes_as('fulltext'))
def _do_test_count_blocks_chunks(self, count_blocks):
blob = ShaFile.from_raw_chunks(Blob.type_num, [b'a\nb', b'\na\n'])
self.assertBlockCountEqual({b'a\n': 4, b'b\n': 2}, _count_blocks(blob))
def iterobjects(self):
"""Iterate over the objects in this pack."""
for offset, type, obj, crc32 in self.data.iterobjects():
assert isinstance(offset, int)
yield ShaFile.from_raw_chunks(
*self.data.resolve_object(offset, type, obj))
def __getitem__(self, sha):
"""Obtain an object by SHA1."""
type_num, uncomp = self.get_raw(sha)
return ShaFile.from_raw_string(type_num, uncomp)
def __getitem__(self, sha):
"""Obtain an object by SHA1."""
type_num, uncomp = self.get_raw(sha)
return ShaFile.from_raw_string(type_num, uncomp, sha=sha)
def __getitem__(self, sha1):
"""Retrieve the specified SHA1."""
type, uncomp = self.get_raw(sha1)
return ShaFile.from_raw_string(type, uncomp)
def _get_loose_object(self, sha):
path = self._get_shafile_path(sha)
try:
return ShaFile.from_path(path)
except FileNotFoundError:
return None
def _get_shafile(self, sha):
path = self._get_shafile_path(sha)
if os.path.exists(path):
return ShaFile.from_file(path)
return None
def _do_test_count_blocks_chunks(self, count_blocks):
blob = ShaFile.from_raw_chunks(Blob.type_num, [b"a\nb", b"\na\n"])
self.assertBlockCountEqual({b"a\n": 4, b"b\n": 2}, _count_blocks(blob))
def __getitem__(self, sha1):
"""Retrieve the specified SHA1."""
type, uncomp = self.get_raw(sha1)
return ShaFile.from_raw_string(type, uncomp)