diff --git a/dulwich/object_store.py b/dulwich/object_store.py index 0d0b92fc..18751bde 100644 --- a/dulwich/object_store.py +++ b/dulwich/object_store.py @@ -1,1316 +1,1318 @@ # object_store.py -- Object store for git objects # Copyright (C) 2008-2013 Jelmer Vernooij # and others # # Dulwich is dual-licensed under the Apache License, Version 2.0 and the GNU # General Public License as public by the Free Software Foundation; version 2.0 # or (at your option) any later version. You can redistribute it and/or # modify it under the terms of either of these two licenses. # # Unless required by applicable law or agreed to in writing, software # distributed under the License is distributed on an "AS IS" BASIS, # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. # See the License for the specific language governing permissions and # limitations under the License. # # You should have received a copy of the licenses; if not, see # for a copy of the GNU General Public License # and for a copy of the Apache # License, Version 2.0. # """Git object store interfaces and implementation.""" from io import BytesIO import errno from itertools import chain import os import stat import sys import tempfile import time from dulwich.diff_tree import ( tree_changes, walk_trees, ) from dulwich.errors import ( NotTreeError, ) from dulwich.file import GitFile from dulwich.objects import ( Commit, ShaFile, Tag, Tree, ZERO_SHA, hex_to_sha, sha_to_hex, hex_to_filename, S_ISGITLINK, object_class, ) from dulwich.pack import ( Pack, PackData, PackInflater, iter_sha1, pack_objects_to_data, write_pack_header, write_pack_index_v2, write_pack_data, write_pack_object, compute_file_sha, PackIndexer, PackStreamCopier, ) INFODIR = 'info' PACKDIR = 'pack' class BaseObjectStore(object): """Object store interface.""" def determine_wants_all(self, refs): return [sha for (ref, sha) in refs.items() if sha not in self and not ref.endswith(b"^{}") and not sha == ZERO_SHA] def iter_shas(self, shas): """Iterate over the objects for the specified shas. :param shas: Iterable object with SHAs :return: Object iterator """ return ObjectStoreIterator(self, shas) def contains_loose(self, sha): """Check if a particular object is present by SHA1 and is loose.""" raise NotImplementedError(self.contains_loose) def contains_packed(self, sha): """Check if a particular object is present by SHA1 and is packed.""" raise NotImplementedError(self.contains_packed) def __contains__(self, sha): """Check if a particular object is present by SHA1. This method makes no distinction between loose and packed objects. """ return self.contains_packed(sha) or self.contains_loose(sha) @property def packs(self): """Iterable of pack objects.""" raise NotImplementedError def get_raw(self, name): """Obtain the raw text for an object. :param name: sha for the object. :return: tuple with numeric type and object contents. """ raise NotImplementedError(self.get_raw) 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 __iter__(self): """Iterate over the SHAs that are present in this store.""" raise NotImplementedError(self.__iter__) def add_object(self, obj): """Add a single object to this object store. """ raise NotImplementedError(self.add_object) - def add_objects(self, objects): + def add_objects(self, objects, progress=None): """Add a set of objects to this object store. :param objects: Iterable over a list of (object, path) tuples """ raise NotImplementedError(self.add_objects) - def add_pack_data(self, count, pack_data): + def add_pack_data(self, count, pack_data, progress=None): """Add pack data to this object store. :param num_items: Number of items to add :param pack_data: Iterator over pack data tuples """ if count == 0: # Don't bother writing an empty pack file return f, commit, abort = self.add_pack() try: - write_pack_data(f, count, pack_data) + write_pack_data(f, count, pack_data, progress) except BaseException: abort() raise else: return commit() def tree_changes(self, source, target, want_unchanged=False, include_trees=False, change_type_same=False): """Find the differences between the contents of two trees :param source: SHA1 of the source tree :param target: SHA1 of the target tree :param want_unchanged: Whether unchanged files should be reported :param include_trees: Whether to include trees :param change_type_same: Whether to report files changing type in the same entry. :return: Iterator over tuples with (oldpath, newpath), (oldmode, newmode), (oldsha, newsha) """ for change in tree_changes(self, source, target, want_unchanged=want_unchanged, include_trees=include_trees, change_type_same=change_type_same): yield ((change.old.path, change.new.path), (change.old.mode, change.new.mode), (change.old.sha, change.new.sha)) def iter_tree_contents(self, tree_id, include_trees=False): """Iterate the contents of a tree and all subtrees. Iteration is depth-first pre-order, as in e.g. os.walk. :param tree_id: SHA1 of the tree. :param include_trees: If True, include tree objects in the iteration. :return: Iterator over TreeEntry namedtuples for all the objects in a tree. """ for entry, _ in walk_trees(self, tree_id, None): if ((entry.mode is not None and not stat.S_ISDIR(entry.mode)) or include_trees): yield entry def find_missing_objects(self, haves, wants, progress=None, get_tagged=None, get_parents=lambda commit: commit.parents): """Find the missing objects required for a set of revisions. :param haves: Iterable over SHAs already in common. :param wants: Iterable over SHAs of objects to fetch. :param progress: Simple progress function that will be called with updated progress strings. :param get_tagged: Function that returns a dict of pointed-to sha -> tag sha for including tags. :param get_parents: Optional function for getting the parents of a commit. :return: Iterator over (sha, path) pairs. """ finder = MissingObjectFinder(self, haves, wants, progress, get_tagged, get_parents=get_parents) return iter(finder.next, None) def find_common_revisions(self, graphwalker): """Find which revisions this store has in common using graphwalker. :param graphwalker: A graphwalker object. :return: List of SHAs that are in common """ haves = [] sha = next(graphwalker) while sha: if sha in self: haves.append(sha) graphwalker.ack(sha) sha = next(graphwalker) return haves def generate_pack_contents(self, have, want, progress=None): """Iterate over the contents of a pack file. :param have: List of SHA1s of objects that should not be sent :param want: List of SHA1s of objects that should be sent :param progress: Optional progress reporting method """ return self.iter_shas(self.find_missing_objects(have, want, progress)) def generate_pack_data(self, have, want, progress=None, ofs_delta=True): """Generate pack data objects for a set of wants/haves. :param have: List of SHA1s of objects that should not be sent :param want: List of SHA1s of objects that should be sent :param ofs_delta: Whether OFS deltas can be included :param progress: Optional progress reporting method """ # TODO(jelmer): More efficient implementation return pack_objects_to_data( self.generate_pack_contents(have, want, progress)) def peel_sha(self, sha): """Peel all tags from a SHA. :param sha: The object SHA to peel. :return: The fully-peeled SHA1 of a tag object, after peeling all intermediate tags; if the original ref does not point to a tag, this will equal the original SHA1. """ obj = self[sha] obj_class = object_class(obj.type_name) while obj_class is Tag: obj_class, sha = obj.object obj = self[sha] return obj def _collect_ancestors(self, heads, common=set(), get_parents=lambda commit: commit.parents): """Collect all ancestors of heads up to (excluding) those in common. :param heads: commits to start from :param common: commits to end at, or empty set to walk repository completely :param get_parents: Optional function for getting the parents of a commit. :return: a tuple (A, B) where A - all commits reachable from heads but not present in common, B - common (shared) elements that are directly reachable from heads """ bases = set() commits = set() queue = [] queue.extend(heads) while queue: e = queue.pop(0) if e in common: bases.add(e) elif e not in commits: commits.add(e) cmt = self[e] queue.extend(get_parents(cmt)) return (commits, bases) def close(self): """Close any files opened by this object store.""" # Default implementation is a NO-OP class PackBasedObjectStore(BaseObjectStore): def __init__(self): self._pack_cache = {} @property def alternates(self): return [] def contains_packed(self, sha): """Check if a particular object is present by SHA1 and is packed. This does not check alternates. """ for pack in self.packs: if sha in pack: return True return False def __contains__(self, sha): """Check if a particular object is present by SHA1. This method makes no distinction between loose and packed objects. """ if self.contains_packed(sha) or self.contains_loose(sha): return True for alternate in self.alternates: if sha in alternate: return True return False def _pack_cache_stale(self): """Check whether the pack cache is stale.""" raise NotImplementedError(self._pack_cache_stale) def _add_known_pack(self, base_name, pack): """Add a newly appeared pack to the cache by path. """ prev_pack = self._pack_cache.get(base_name) if prev_pack is not pack: self._pack_cache[base_name] = pack if prev_pack: prev_pack.close() def _flush_pack_cache(self): pack_cache = self._pack_cache self._pack_cache = {} while pack_cache: (name, pack) = pack_cache.popitem() pack.close() def close(self): self._flush_pack_cache() @property def packs(self): """List with pack objects.""" if self._pack_cache is None or self._pack_cache_stale(): self._update_pack_cache() return self._pack_cache.values() def _iter_alternate_objects(self): """Iterate over the SHAs of all the objects in alternate stores.""" for alternate in self.alternates: for alternate_object in alternate: yield alternate_object def _iter_loose_objects(self): """Iterate over the SHAs of all loose objects.""" raise NotImplementedError(self._iter_loose_objects) def _get_loose_object(self, sha): raise NotImplementedError(self._get_loose_object) def _remove_loose_object(self, sha): raise NotImplementedError(self._remove_loose_object) def _remove_pack(self, name): raise NotImplementedError(self._remove_pack) def pack_loose_objects(self): """Pack loose objects. :return: Number of objects packed """ objects = set() for sha in self._iter_loose_objects(): objects.add((self._get_loose_object(sha), None)) self.add_objects(list(objects)) for obj, path in objects: self._remove_loose_object(obj.id) return len(objects) def repack(self): """Repack the packs in this repository. Note that this implementation is fairly naive and currently keeps all objects in memory while it repacks. """ loose_objects = set() for sha in self._iter_loose_objects(): loose_objects.add(self._get_loose_object(sha)) objects = {(obj, None) for obj in loose_objects} old_packs = {p.name(): p for p in self.packs} for name, pack in old_packs.items(): objects.update((obj, None) for obj in pack.iterobjects()) self._flush_pack_cache() # The name of the consolidated pack might match the name of a # pre-existing pack. Take care not to remove the newly created # consolidated pack. consolidated = self.add_objects(objects) old_packs.pop(consolidated.name(), None) for obj in loose_objects: self._remove_loose_object(obj.id) for name, pack in old_packs.items(): self._remove_pack(pack) self._update_pack_cache() return len(objects) def __iter__(self): """Iterate over the SHAs that are present in this store.""" iterables = (list(self.packs) + [self._iter_loose_objects()] + [self._iter_alternate_objects()]) return chain(*iterables) def contains_loose(self, sha): """Check if a particular object is present by SHA1 and is loose. This does not check alternates. """ return self._get_loose_object(sha) is not None def get_raw(self, name): """Obtain the raw fulltext for an object. :param name: sha for the object. :return: tuple with numeric type and object contents. """ if len(name) == 40: sha = hex_to_sha(name) hexsha = name elif len(name) == 20: sha = name hexsha = None else: raise AssertionError("Invalid object name %r" % name) for pack in self.packs: try: return pack.get_raw(sha) except KeyError: pass if hexsha is None: hexsha = sha_to_hex(name) ret = self._get_loose_object(hexsha) if ret is not None: return ret.type_num, ret.as_raw_string() for alternate in self.alternates: try: return alternate.get_raw(hexsha) except KeyError: pass raise KeyError(hexsha) - def add_objects(self, objects): + def add_objects(self, objects, progress=None): """Add a set of objects to this object store. :param objects: Iterable over (object, path) tuples, should support __len__. :return: Pack object of the objects written. """ - return self.add_pack_data(*pack_objects_to_data(objects)) + return self.add_pack_data( + *pack_objects_to_data(objects), + progress=progress) class DiskObjectStore(PackBasedObjectStore): """Git-style object store that exists on disk.""" def __init__(self, path): """Open an object store. :param path: Path of the object store. """ super(DiskObjectStore, self).__init__() self.path = path self.pack_dir = os.path.join(self.path, PACKDIR) self._pack_cache_time = 0 self._pack_cache = {} self._alternates = None def __repr__(self): return "<%s(%r)>" % (self.__class__.__name__, self.path) @property def alternates(self): if self._alternates is not None: return self._alternates self._alternates = [] for path in self._read_alternate_paths(): self._alternates.append(DiskObjectStore(path)) return self._alternates def _read_alternate_paths(self): try: f = GitFile(os.path.join(self.path, INFODIR, "alternates"), 'rb') except (OSError, IOError) as e: if e.errno == errno.ENOENT: return raise with f: for line in f.readlines(): line = line.rstrip(b"\n") if line[0] == b"#": continue if os.path.isabs(line): yield line.decode(sys.getfilesystemencoding()) else: yield os.path.join(self.path, line).decode( sys.getfilesystemencoding()) def add_alternate_path(self, path): """Add an alternate path to this object store. """ try: os.mkdir(os.path.join(self.path, INFODIR)) except OSError as e: if e.errno != errno.EEXIST: raise alternates_path = os.path.join(self.path, INFODIR, "alternates") with GitFile(alternates_path, 'wb') as f: try: orig_f = open(alternates_path, 'rb') except (OSError, IOError) as e: if e.errno != errno.ENOENT: raise else: with orig_f: f.write(orig_f.read()) f.write(path.encode(sys.getfilesystemencoding()) + b"\n") if not os.path.isabs(path): path = os.path.join(self.path, path) self.alternates.append(DiskObjectStore(path)) def _update_pack_cache(self): try: pack_dir_contents = os.listdir(self.pack_dir) except OSError as e: if e.errno == errno.ENOENT: self._pack_cache_time = 0 self.close() return raise self._pack_cache_time = max( os.stat(self.pack_dir).st_mtime, time.time()) pack_files = set() for name in pack_dir_contents: if name.startswith("pack-") and name.endswith(".pack"): # verify that idx exists first (otherwise the pack was not yet # fully written) idx_name = os.path.splitext(name)[0] + ".idx" if idx_name in pack_dir_contents: pack_name = name[:-len(".pack")] pack_files.add(pack_name) # Open newly appeared pack files for f in pack_files: if f not in self._pack_cache: self._pack_cache[f] = Pack(os.path.join(self.pack_dir, f)) # Remove disappeared pack files for f in set(self._pack_cache) - pack_files: self._pack_cache.pop(f).close() def _pack_cache_stale(self): try: return os.stat(self.pack_dir).st_mtime >= self._pack_cache_time except OSError as e: if e.errno == errno.ENOENT: return True raise def _get_shafile_path(self, sha): # Check from object dir return hex_to_filename(self.path, sha) def _iter_loose_objects(self): for base in os.listdir(self.path): if len(base) != 2: continue for rest in os.listdir(os.path.join(self.path, base)): yield (base+rest).encode(sys.getfilesystemencoding()) 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 _remove_loose_object(self, sha): os.remove(self._get_shafile_path(sha)) def _remove_pack(self, pack): os.remove(pack.data.path) os.remove(pack.index.path) def _get_pack_basepath(self, entries): suffix = iter_sha1(entry[0] for entry in entries) # TODO: Handle self.pack_dir being bytes suffix = suffix.decode('ascii') return os.path.join(self.pack_dir, "pack-" + suffix) def _complete_thin_pack(self, f, path, copier, indexer): """Move a specific file containing a pack into the pack directory. :note: The file should be on the same file system as the packs directory. :param f: Open file object for the pack. :param path: Path to the pack file. :param copier: A PackStreamCopier to use for writing pack data. :param indexer: A PackIndexer for indexing the pack. """ entries = list(indexer) # Update the header with the new number of objects. f.seek(0) write_pack_header(f, len(entries) + len(indexer.ext_refs())) # Must flush before reading (http://bugs.python.org/issue3207) f.flush() # Rescan the rest of the pack, computing the SHA with the new header. new_sha = compute_file_sha(f, end_ofs=-20) # Must reposition before writing (http://bugs.python.org/issue3207) f.seek(0, os.SEEK_CUR) # Complete the pack. for ext_sha in indexer.ext_refs(): assert len(ext_sha) == 20 type_num, data = self.get_raw(ext_sha) offset = f.tell() crc32 = write_pack_object(f, type_num, data, sha=new_sha) entries.append((ext_sha, offset, crc32)) pack_sha = new_sha.digest() f.write(pack_sha) f.close() # Move the pack in. entries.sort() pack_base_name = self._get_pack_basepath(entries) if sys.platform == 'win32': try: os.rename(path, pack_base_name + '.pack') except WindowsError: os.remove(pack_base_name + '.pack') os.rename(path, pack_base_name + '.pack') else: os.rename(path, pack_base_name + '.pack') # Write the index. index_file = GitFile(pack_base_name + '.idx', 'wb') try: write_pack_index_v2(index_file, entries, pack_sha) index_file.close() finally: index_file.abort() # Add the pack to the store and return it. final_pack = Pack(pack_base_name) final_pack.check_length_and_checksum() self._add_known_pack(pack_base_name, final_pack) return final_pack def add_thin_pack(self, read_all, read_some): """Add a new thin pack to this object store. Thin packs are packs that contain deltas with parents that exist outside the pack. They should never be placed in the object store directly, and always indexed and completed as they are copied. :param read_all: Read function that blocks until the number of requested bytes are read. :param read_some: Read function that returns at least one byte, but may not return the number of bytes requested. :return: A Pack object pointing at the now-completed thin pack in the objects/pack directory. """ fd, path = tempfile.mkstemp(dir=self.path, prefix='tmp_pack_') with os.fdopen(fd, 'w+b') as f: indexer = PackIndexer(f, resolve_ext_ref=self.get_raw) copier = PackStreamCopier(read_all, read_some, f, delta_iter=indexer) copier.verify() return self._complete_thin_pack(f, path, copier, indexer) def move_in_pack(self, path): """Move a specific file containing a pack into the pack directory. :note: The file should be on the same file system as the packs directory. :param path: Path to the pack file. """ with PackData(path) as p: entries = p.sorted_entries() basename = self._get_pack_basepath(entries) with GitFile(basename+".idx", "wb") as f: write_pack_index_v2(f, entries, p.get_stored_checksum()) if self._pack_cache is None or self._pack_cache_stale(): self._update_pack_cache() try: return self._pack_cache[basename] except KeyError: pass else: os.unlink(path) try: os.rename(path, basename + ".pack") except OSError as e: if e.errno == errno.EEXIST: # This can happen on Windows.. # It's safe to ignore this, since if the file already exists, # it should have the same contents as the one we just # generated. os.unlink(path) else: raise final_pack = Pack(basename) self._add_known_pack(basename, final_pack) return final_pack def add_pack(self): """Add a new pack to this object store. :return: Fileobject to write to, a commit function to call when the pack is finished and an abort function. """ fd, path = tempfile.mkstemp(dir=self.pack_dir, suffix=".pack") f = os.fdopen(fd, 'wb') def commit(): f.flush() os.fsync(fd) f.close() if os.path.getsize(path) > 0: return self.move_in_pack(path) else: os.remove(path) return None def abort(): f.close() os.remove(path) return f, commit, abort def add_object(self, obj): """Add a single object to this object store. :param obj: Object to add """ path = self._get_shafile_path(obj.id) dir = os.path.dirname(path) try: os.mkdir(dir) except OSError as e: if e.errno != errno.EEXIST: raise if os.path.exists(path): return # Already there, no need to write again with GitFile(path, 'wb') as f: f.write(obj.as_legacy_object()) @classmethod def init(cls, path): try: os.mkdir(path) except OSError as e: if e.errno != errno.EEXIST: raise os.mkdir(os.path.join(path, "info")) os.mkdir(os.path.join(path, PACKDIR)) return cls(path) class MemoryObjectStore(BaseObjectStore): """Object store that keeps all objects in memory.""" def __init__(self): super(MemoryObjectStore, self).__init__() self._data = {} def _to_hexsha(self, sha): if len(sha) == 40: return sha elif len(sha) == 20: return sha_to_hex(sha) else: raise ValueError("Invalid sha %r" % (sha,)) def contains_loose(self, sha): """Check if a particular object is present by SHA1 and is loose.""" return self._to_hexsha(sha) in self._data def contains_packed(self, sha): """Check if a particular object is present by SHA1 and is packed.""" return False def __iter__(self): """Iterate over the SHAs that are present in this store.""" return iter(self._data.keys()) @property def packs(self): """List with pack objects.""" return [] def get_raw(self, name): """Obtain the raw text for an object. :param name: sha for the object. :return: tuple with numeric type and object contents. """ obj = self[self._to_hexsha(name)] return obj.type_num, obj.as_raw_string() def __getitem__(self, name): return self._data[self._to_hexsha(name)].copy() def __delitem__(self, name): """Delete an object from this store, for testing only.""" del self._data[self._to_hexsha(name)] def add_object(self, obj): """Add a single object to this object store. """ self._data[obj.id] = obj.copy() - def add_objects(self, objects): + def add_objects(self, objects, progress=None): """Add a set of objects to this object store. :param objects: Iterable over a list of (object, path) tuples """ for obj, path in objects: self.add_object(obj) def add_pack(self): """Add a new pack to this object store. Because this object store doesn't support packs, we extract and add the individual objects. :return: Fileobject to write to and a commit function to call when the pack is finished. """ f = BytesIO() def commit(): p = PackData.from_file(BytesIO(f.getvalue()), f.tell()) f.close() for obj in PackInflater.for_pack_data(p, self.get_raw): self.add_object(obj) def abort(): pass return f, commit, abort def _complete_thin_pack(self, f, indexer): """Complete a thin pack by adding external references. :param f: Open file object for the pack. :param indexer: A PackIndexer for indexing the pack. """ entries = list(indexer) # Update the header with the new number of objects. f.seek(0) write_pack_header(f, len(entries) + len(indexer.ext_refs())) # Rescan the rest of the pack, computing the SHA with the new header. new_sha = compute_file_sha(f, end_ofs=-20) # Complete the pack. for ext_sha in indexer.ext_refs(): assert len(ext_sha) == 20 type_num, data = self.get_raw(ext_sha) write_pack_object(f, type_num, data, sha=new_sha) pack_sha = new_sha.digest() f.write(pack_sha) def add_thin_pack(self, read_all, read_some): """Add a new thin pack to this object store. Thin packs are packs that contain deltas with parents that exist outside the pack. Because this object store doesn't support packs, we extract and add the individual objects. :param read_all: Read function that blocks until the number of requested bytes are read. :param read_some: Read function that returns at least one byte, but may not return the number of bytes requested. """ f, commit, abort = self.add_pack() try: indexer = PackIndexer(f, resolve_ext_ref=self.get_raw) copier = PackStreamCopier(read_all, read_some, f, delta_iter=indexer) copier.verify() self._complete_thin_pack(f, indexer) except BaseException: abort() raise else: commit() class ObjectIterator(object): """Interface for iterating over objects.""" def iterobjects(self): raise NotImplementedError(self.iterobjects) class ObjectStoreIterator(ObjectIterator): """ObjectIterator that works on top of an ObjectStore.""" def __init__(self, store, sha_iter): """Create a new ObjectIterator. :param store: Object store to retrieve from :param sha_iter: Iterator over (sha, path) tuples """ self.store = store self.sha_iter = sha_iter self._shas = [] def __iter__(self): """Yield tuple with next object and path.""" for sha, path in self.itershas(): yield self.store[sha], path def iterobjects(self): """Iterate over just the objects.""" for o, path in self: yield o def itershas(self): """Iterate over the SHAs.""" for sha in self._shas: yield sha for sha in self.sha_iter: self._shas.append(sha) yield sha def __contains__(self, needle): """Check if an object is present. :note: This checks if the object is present in the underlying object store, not if it would be yielded by the iterator. :param needle: SHA1 of the object to check for """ return needle in self.store def __getitem__(self, key): """Find an object by SHA1. :note: This retrieves the object from the underlying object store. It will also succeed if the object would not be returned by the iterator. """ return self.store[key] def __len__(self): """Return the number of objects.""" return len(list(self.itershas())) def empty(self): iter = self.itershas() try: iter() except StopIteration: return True else: return False def __bool__(self): """Indicate whether this object has contents.""" return not self.empty() def tree_lookup_path(lookup_obj, root_sha, path): """Look up an object in a Git tree. :param lookup_obj: Callback for retrieving object by SHA1 :param root_sha: SHA1 of the root tree :param path: Path to lookup :return: A tuple of (mode, SHA) of the resulting path. """ tree = lookup_obj(root_sha) if not isinstance(tree, Tree): raise NotTreeError(root_sha) return tree.lookup_path(lookup_obj, path) def _collect_filetree_revs(obj_store, tree_sha, kset): """Collect SHA1s of files and directories for specified tree. :param obj_store: Object store to get objects by SHA from :param tree_sha: tree reference to walk :param kset: set to fill with references to files and directories """ filetree = obj_store[tree_sha] for name, mode, sha in filetree.iteritems(): if not S_ISGITLINK(mode) and sha not in kset: kset.add(sha) if stat.S_ISDIR(mode): _collect_filetree_revs(obj_store, sha, kset) def _split_commits_and_tags(obj_store, lst, ignore_unknown=False): """Split object id list into three lists with commit, tag, and other SHAs. Commits referenced by tags are included into commits list as well. Only SHA1s known in this repository will get through, and unless ignore_unknown argument is True, KeyError is thrown for SHA1 missing in the repository :param obj_store: Object store to get objects by SHA1 from :param lst: Collection of commit and tag SHAs :param ignore_unknown: True to skip SHA1 missing in the repository silently. :return: A tuple of (commits, tags, others) SHA1s """ commits = set() tags = set() others = set() for e in lst: try: o = obj_store[e] except KeyError: if not ignore_unknown: raise else: if isinstance(o, Commit): commits.add(e) elif isinstance(o, Tag): tags.add(e) tagged = o.object[1] c, t, o = _split_commits_and_tags( obj_store, [tagged], ignore_unknown=ignore_unknown) commits |= c tags |= t others |= o else: others.add(e) return (commits, tags, others) class MissingObjectFinder(object): """Find the objects missing from another object store. :param object_store: Object store containing at least all objects to be sent :param haves: SHA1s of commits not to send (already present in target) :param wants: SHA1s of commits to send :param progress: Optional function to report progress to. :param get_tagged: Function that returns a dict of pointed-to sha -> tag sha for including tags. :param get_parents: Optional function for getting the parents of a commit. :param tagged: dict of pointed-to sha -> tag sha for including tags """ def __init__(self, object_store, haves, wants, progress=None, get_tagged=None, get_parents=lambda commit: commit.parents): self.object_store = object_store self._get_parents = get_parents # process Commits and Tags differently # Note, while haves may list commits/tags not available locally, # and such SHAs would get filtered out by _split_commits_and_tags, # wants shall list only known SHAs, and otherwise # _split_commits_and_tags fails with KeyError have_commits, have_tags, have_others = ( _split_commits_and_tags(object_store, haves, True)) want_commits, want_tags, want_others = ( _split_commits_and_tags(object_store, wants, False)) # all_ancestors is a set of commits that shall not be sent # (complete repository up to 'haves') all_ancestors = object_store._collect_ancestors( have_commits, get_parents=self._get_parents)[0] # all_missing - complete set of commits between haves and wants # common - commits from all_ancestors we hit into while # traversing parent hierarchy of wants missing_commits, common_commits = object_store._collect_ancestors( want_commits, all_ancestors, get_parents=self._get_parents) self.sha_done = set() # Now, fill sha_done with commits and revisions of # files and directories known to be both locally # and on target. Thus these commits and files # won't get selected for fetch for h in common_commits: self.sha_done.add(h) cmt = object_store[h] _collect_filetree_revs(object_store, cmt.tree, self.sha_done) # record tags we have as visited, too for t in have_tags: self.sha_done.add(t) missing_tags = want_tags.difference(have_tags) missing_others = want_others.difference(have_others) # in fact, what we 'want' is commits, tags, and others # we've found missing wants = missing_commits.union(missing_tags) wants = wants.union(missing_others) self.objects_to_send = set([(w, None, False) for w in wants]) if progress is None: self.progress = lambda x: None else: self.progress = progress self._tagged = get_tagged and get_tagged() or {} def add_todo(self, entries): self.objects_to_send.update([e for e in entries if not e[0] in self.sha_done]) def next(self): while True: if not self.objects_to_send: return None (sha, name, leaf) = self.objects_to_send.pop() if sha not in self.sha_done: break if not leaf: o = self.object_store[sha] if isinstance(o, Commit): self.add_todo([(o.tree, "", False)]) elif isinstance(o, Tree): self.add_todo([(s, n, not stat.S_ISDIR(m)) for n, m, s in o.iteritems() if not S_ISGITLINK(m)]) elif isinstance(o, Tag): self.add_todo([(o.object[1], None, False)]) if sha in self._tagged: self.add_todo([(self._tagged[sha], None, True)]) self.sha_done.add(sha) self.progress(("counting objects: %d\r" % len(self.sha_done)).encode('ascii')) return (sha, name) __next__ = next class ObjectStoreGraphWalker(object): """Graph walker that finds what commits are missing from an object store. :ivar heads: Revisions without descendants in the local repo :ivar get_parents: Function to retrieve parents in the local repo """ def __init__(self, local_heads, get_parents): """Create a new instance. :param local_heads: Heads to start search with :param get_parents: Function for finding the parents of a SHA1. """ self.heads = set(local_heads) self.get_parents = get_parents self.parents = {} def ack(self, sha): """Ack that a revision and its ancestors are present in the source.""" if len(sha) != 40: raise ValueError("unexpected sha %r received" % sha) ancestors = set([sha]) # stop if we run out of heads to remove while self.heads: for a in ancestors: if a in self.heads: self.heads.remove(a) # collect all ancestors new_ancestors = set() for a in ancestors: ps = self.parents.get(a) if ps is not None: new_ancestors.update(ps) self.parents[a] = None # no more ancestors; stop if not new_ancestors: break ancestors = new_ancestors def next(self): """Iterate over ancestors of heads in the target.""" if self.heads: ret = self.heads.pop() ps = self.get_parents(ret) self.parents[ret] = ps self.heads.update( [p for p in ps if p not in self.parents]) return ret return None __next__ = next def commit_tree_changes(object_store, tree, changes): """Commit a specified set of changes to a tree structure. This will apply a set of changes on top of an existing tree, storing new objects in object_store. changes are a list of tuples with (path, mode, object_sha). Paths can be both blobs and trees. See the mode and object sha to None deletes the path. This method works especially well if there are only a small number of changes to a big tree. For a large number of changes to a large tree, use e.g. commit_tree. :param object_store: Object store to store new objects in and retrieve old ones from. :param tree: Original tree root :param changes: changes to apply :return: New tree root object """ # TODO(jelmer): Save up the objects and add them using .add_objects # rather than with individual calls to .add_object. nested_changes = {} for (path, new_mode, new_sha) in changes: try: (dirname, subpath) = path.split(b'/', 1) except ValueError: if new_sha is None: del tree[path] else: tree[path] = (new_mode, new_sha) else: nested_changes.setdefault(dirname, []).append( (subpath, new_mode, new_sha)) for name, subchanges in nested_changes.items(): try: orig_subtree = object_store[tree[name][1]] except KeyError: orig_subtree = Tree() subtree = commit_tree_changes(object_store, orig_subtree, subchanges) if len(subtree) == 0: del tree[name] else: tree[name] = (stat.S_IFDIR, subtree.id) object_store.add_object(tree) return tree class OverlayObjectStore(BaseObjectStore): """Object store that can overlay multiple object stores.""" def __init__(self, bases, add_store=None): self.bases = bases self.add_store = add_store def add_object(self, object): if self.add_store is None: raise NotImplementedError(self.add_object) return self.add_store.add_object(object) - def add_objects(self, objects): + def add_objects(self, objects, progress=None): if self.add_store is None: raise NotImplementedError(self.add_object) - return self.add_store.add_objects(objects) + return self.add_store.add_objects(objects, progress) @property def packs(self): ret = [] for b in self.bases: ret.extend(b.packs) return ret def __iter__(self): done = set() for b in self.bases: for o_id in b: if o_id not in done: yield o_id done.add(o_id) def get_raw(self, sha_id): for b in self.bases: try: return b.get_raw(sha_id) except KeyError: pass else: raise KeyError(sha_id) def contains_packed(self, sha): for b in self.bases: if b.contains_packed(sha): return True else: return False def contains_loose(self, sha): for b in self.bases: if b.contains_loose(sha): return True else: return False diff --git a/dulwich/pack.py b/dulwich/pack.py index 96f3b9bc..c2a5bdfa 100644 --- a/dulwich/pack.py +++ b/dulwich/pack.py @@ -1,2035 +1,2040 @@ # pack.py -- For dealing with packed git objects. # Copyright (C) 2007 James Westby # Copyright (C) 2008-2013 Jelmer Vernooij # # Dulwich is dual-licensed under the Apache License, Version 2.0 and the GNU # General Public License as public by the Free Software Foundation; version 2.0 # or (at your option) any later version. You can redistribute it and/or # modify it under the terms of either of these two licenses. # # Unless required by applicable law or agreed to in writing, software # distributed under the License is distributed on an "AS IS" BASIS, # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. # See the License for the specific language governing permissions and # limitations under the License. # # You should have received a copy of the licenses; if not, see # for a copy of the GNU General Public License # and for a copy of the Apache # License, Version 2.0. # """Classes for dealing with packed git objects. A pack is a compact representation of a bunch of objects, stored using deltas where possible. They have two parts, the pack file, which stores the data, and an index that tells you where the data is. To find an object you look in all of the index files 'til you find a match for the object name. You then use the pointer got from this as a pointer in to the corresponding packfile. """ from collections import defaultdict import binascii from io import BytesIO, UnsupportedOperation from collections import ( deque, ) import difflib import struct from itertools import chain try: from itertools import imap, izip except ImportError: # Python3 imap = map izip = zip import os import sys from hashlib import sha1 from os import ( SEEK_CUR, SEEK_END, ) from struct import unpack_from import zlib try: import mmap except ImportError: has_mmap = False else: has_mmap = True # For some reason the above try, except fails to set has_mmap = False for plan9 if sys.platform == 'Plan9': has_mmap = False from dulwich.errors import ( # noqa: E402 ApplyDeltaError, ChecksumMismatch, ) from dulwich.file import GitFile # noqa: E402 from dulwich.lru_cache import ( # noqa: E402 LRUSizeCache, ) from dulwich.objects import ( # noqa: E402 ShaFile, hex_to_sha, sha_to_hex, object_header, ) OFS_DELTA = 6 REF_DELTA = 7 DELTA_TYPES = (OFS_DELTA, REF_DELTA) DEFAULT_PACK_DELTA_WINDOW_SIZE = 10 def take_msb_bytes(read, crc32=None): """Read bytes marked with most significant bit. :param read: Read function """ ret = [] while len(ret) == 0 or ret[-1] & 0x80: b = read(1) if crc32 is not None: crc32 = binascii.crc32(b, crc32) ret.append(ord(b[:1])) return ret, crc32 class UnpackedObject(object): """Class encapsulating an object unpacked from a pack file. These objects should only be created from within unpack_object. Most members start out as empty and are filled in at various points by read_zlib_chunks, unpack_object, DeltaChainIterator, etc. End users of this object should take care that the function they're getting this object from is guaranteed to set the members they need. """ __slots__ = [ 'offset', # Offset in its pack. '_sha', # Cached binary SHA. 'obj_type_num', # Type of this object. 'obj_chunks', # Decompressed and delta-resolved chunks. 'pack_type_num', # Type of this object in the pack (may be a delta). 'delta_base', # Delta base offset or SHA. 'comp_chunks', # Compressed object chunks. 'decomp_chunks', # Decompressed object chunks. 'decomp_len', # Decompressed length of this object. 'crc32', # CRC32. ] # TODO(dborowitz): read_zlib_chunks and unpack_object could very well be # methods of this object. def __init__(self, pack_type_num, delta_base, decomp_len, crc32): self.offset = None self._sha = None self.pack_type_num = pack_type_num self.delta_base = delta_base self.comp_chunks = None self.decomp_chunks = [] self.decomp_len = decomp_len self.crc32 = crc32 if pack_type_num in DELTA_TYPES: self.obj_type_num = None self.obj_chunks = None else: self.obj_type_num = pack_type_num self.obj_chunks = self.decomp_chunks self.delta_base = delta_base def sha(self): """Return the binary SHA of this object.""" if self._sha is None: self._sha = obj_sha(self.obj_type_num, self.obj_chunks) return self._sha def sha_file(self): """Return a ShaFile from this object.""" return ShaFile.from_raw_chunks(self.obj_type_num, self.obj_chunks) # Only provided for backwards compatibility with code that expects either # chunks or a delta tuple. def _obj(self): """Return the decompressed chunks, or (delta base, delta chunks).""" if self.pack_type_num in DELTA_TYPES: return (self.delta_base, self.decomp_chunks) else: return self.decomp_chunks def __eq__(self, other): if not isinstance(other, UnpackedObject): return False for slot in self.__slots__: if getattr(self, slot) != getattr(other, slot): return False return True def __ne__(self, other): return not (self == other) def __repr__(self): data = ['%s=%r' % (s, getattr(self, s)) for s in self.__slots__] return '%s(%s)' % (self.__class__.__name__, ', '.join(data)) _ZLIB_BUFSIZE = 4096 def read_zlib_chunks(read_some, unpacked, include_comp=False, buffer_size=_ZLIB_BUFSIZE): """Read zlib data from a buffer. This function requires that the buffer have additional data following the compressed data, which is guaranteed to be the case for git pack files. :param read_some: Read function that returns at least one byte, but may return less than the requested size. :param unpacked: An UnpackedObject to write result data to. If its crc32 attr is not None, the CRC32 of the compressed bytes will be computed using this starting CRC32. After this function, will have the following attrs set: * comp_chunks (if include_comp is True) * decomp_chunks * decomp_len * crc32 :param include_comp: If True, include compressed data in the result. :param buffer_size: Size of the read buffer. :return: Leftover unused data from the decompression. :raise zlib.error: if a decompression error occurred. """ if unpacked.decomp_len <= -1: raise ValueError('non-negative zlib data stream size expected') decomp_obj = zlib.decompressobj() comp_chunks = [] decomp_chunks = unpacked.decomp_chunks decomp_len = 0 crc32 = unpacked.crc32 while True: add = read_some(buffer_size) if not add: raise zlib.error('EOF before end of zlib stream') comp_chunks.append(add) decomp = decomp_obj.decompress(add) decomp_len += len(decomp) decomp_chunks.append(decomp) unused = decomp_obj.unused_data if unused: left = len(unused) if crc32 is not None: crc32 = binascii.crc32(add[:-left], crc32) if include_comp: comp_chunks[-1] = add[:-left] break elif crc32 is not None: crc32 = binascii.crc32(add, crc32) if crc32 is not None: crc32 &= 0xffffffff if decomp_len != unpacked.decomp_len: raise zlib.error('decompressed data does not match expected size') unpacked.crc32 = crc32 if include_comp: unpacked.comp_chunks = comp_chunks return unused def iter_sha1(iter): """Return the hexdigest of the SHA1 over a set of names. :param iter: Iterator over string objects :return: 40-byte hex sha1 digest """ sha = sha1() for name in iter: sha.update(name) return sha.hexdigest().encode('ascii') def load_pack_index(path): """Load an index file by path. :param filename: Path to the index file :return: A PackIndex loaded from the given path """ with GitFile(path, 'rb') as f: return load_pack_index_file(path, f) def _load_file_contents(f, size=None): try: fd = f.fileno() except (UnsupportedOperation, AttributeError): fd = None # Attempt to use mmap if possible if fd is not None: if size is None: size = os.fstat(fd).st_size if has_mmap: try: contents = mmap.mmap(fd, size, access=mmap.ACCESS_READ) except mmap.error: # Perhaps a socket? pass else: return contents, size contents = f.read() size = len(contents) return contents, size def load_pack_index_file(path, f): """Load an index file from a file-like object. :param path: Path for the index file :param f: File-like object :return: A PackIndex loaded from the given file """ contents, size = _load_file_contents(f) if contents[:4] == b'\377tOc': version = struct.unpack(b'>L', contents[4:8])[0] if version == 2: return PackIndex2( path, file=f, contents=contents, size=size) else: raise KeyError('Unknown pack index format %d' % version) else: return PackIndex1(path, file=f, contents=contents, size=size) def bisect_find_sha(start, end, sha, unpack_name): """Find a SHA in a data blob with sorted SHAs. :param start: Start index of range to search :param end: End index of range to search :param sha: Sha to find :param unpack_name: Callback to retrieve SHA by index :return: Index of the SHA, or None if it wasn't found """ assert start <= end while start <= end: i = (start + end) // 2 file_sha = unpack_name(i) if file_sha < sha: start = i + 1 elif file_sha > sha: end = i - 1 else: return i return None class PackIndex(object): """An index in to a packfile. Given a sha id of an object a pack index can tell you the location in the packfile of that object if it has it. """ def __eq__(self, other): if not isinstance(other, PackIndex): return False for (name1, _, _), (name2, _, _) in izip(self.iterentries(), other.iterentries()): if name1 != name2: return False return True def __ne__(self, other): return not self.__eq__(other) def __len__(self): """Return the number of entries in this pack index.""" raise NotImplementedError(self.__len__) def __iter__(self): """Iterate over the SHAs in this pack.""" return imap(sha_to_hex, self._itersha()) def iterentries(self): """Iterate over the entries in this pack index. :return: iterator over tuples with object name, offset in packfile and crc32 checksum. """ raise NotImplementedError(self.iterentries) def get_pack_checksum(self): """Return the SHA1 checksum stored for the corresponding packfile. :return: 20-byte binary digest """ raise NotImplementedError(self.get_pack_checksum) def object_index(self, sha): """Return the index in to the corresponding packfile for the object. Given the name of an object it will return the offset that object lives at within the corresponding pack file. If the pack file doesn't have the object then None will be returned. """ if len(sha) == 40: sha = hex_to_sha(sha) return self._object_index(sha) def object_sha1(self, index): """Return the SHA1 corresponding to the index in the pack file. """ # PERFORMANCE/TODO(jelmer): Avoid scanning entire index for (name, offset, crc32) in self.iterentries(): if offset == index: return name else: raise KeyError(index) def _object_index(self, sha): """See object_index. :param sha: A *binary* SHA string. (20 characters long)_ """ raise NotImplementedError(self._object_index) 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 _itersha(self): """Yield all the SHA1's of the objects in the index, sorted.""" raise NotImplementedError(self._itersha) class MemoryPackIndex(PackIndex): """Pack index that is stored entirely in memory.""" def __init__(self, entries, pack_checksum=None): """Create a new MemoryPackIndex. :param entries: Sequence of name, idx, crc32 (sorted) :param pack_checksum: Optional pack checksum """ self._by_sha = {} self._by_index = {} for name, idx, crc32 in entries: self._by_sha[name] = idx self._by_index[idx] = name self._entries = entries self._pack_checksum = pack_checksum def get_pack_checksum(self): return self._pack_checksum def __len__(self): return len(self._entries) def _object_index(self, sha): return self._by_sha[sha][0] def object_sha1(self, index): return self._by_index[index] def _itersha(self): return iter(self._by_sha) def iterentries(self): return iter(self._entries) class FilePackIndex(PackIndex): """Pack index that is based on a file. To do the loop it opens the file, and indexes first 256 4 byte groups with the first byte of the sha id. The value in the four byte group indexed is the end of the group that shares the same starting byte. Subtract one from the starting byte and index again to find the start of the group. The values are sorted by sha id within the group, so do the math to find the start and end offset and then bisect in to find if the value is present. """ def __init__(self, filename, file=None, contents=None, size=None): """Create a pack index object. Provide it with the name of the index file to consider, and it will map it whenever required. """ self._filename = filename # Take the size now, so it can be checked each time we map the file to # ensure that it hasn't changed. if file is None: self._file = GitFile(filename, 'rb') else: self._file = file if contents is None: self._contents, self._size = _load_file_contents(self._file, size) else: self._contents, self._size = (contents, size) @property def path(self): return self._filename def __eq__(self, other): # Quick optimization: if (isinstance(other, FilePackIndex) and self._fan_out_table != other._fan_out_table): return False return super(FilePackIndex, self).__eq__(other) def close(self): self._file.close() if getattr(self._contents, "close", None) is not None: self._contents.close() def __len__(self): """Return the number of entries in this pack index.""" return self._fan_out_table[-1] def _unpack_entry(self, i): """Unpack the i-th entry in the index file. :return: Tuple with object name (SHA), offset in pack file and CRC32 checksum (if known). """ raise NotImplementedError(self._unpack_entry) def _unpack_name(self, i): """Unpack the i-th name from the index file.""" raise NotImplementedError(self._unpack_name) def _unpack_offset(self, i): """Unpack the i-th object offset from the index file.""" raise NotImplementedError(self._unpack_offset) def _unpack_crc32_checksum(self, i): """Unpack the crc32 checksum for the ith object from the index file. """ raise NotImplementedError(self._unpack_crc32_checksum) def _itersha(self): for i in range(len(self)): yield self._unpack_name(i) def iterentries(self): """Iterate over the entries in this pack index. :return: iterator over tuples with object name, offset in packfile and crc32 checksum. """ for i in range(len(self)): yield self._unpack_entry(i) def _read_fan_out_table(self, start_offset): ret = [] for i in range(0x100): fanout_entry = self._contents[ start_offset+i*4:start_offset+(i+1)*4] ret.append(struct.unpack('>L', fanout_entry)[0]) return ret def check(self): """Check that the stored checksum matches the actual checksum.""" actual = self.calculate_checksum() stored = self.get_stored_checksum() if actual != stored: raise ChecksumMismatch(stored, actual) def calculate_checksum(self): """Calculate the SHA1 checksum over this pack index. :return: This is a 20-byte binary digest """ return sha1(self._contents[:-20]).digest() def get_pack_checksum(self): """Return the SHA1 checksum stored for the corresponding packfile. :return: 20-byte binary digest """ return bytes(self._contents[-40:-20]) def get_stored_checksum(self): """Return the SHA1 checksum stored for this index. :return: 20-byte binary digest """ return bytes(self._contents[-20:]) def _object_index(self, sha): """See object_index. :param sha: A *binary* SHA string. (20 characters long)_ """ assert len(sha) == 20 idx = ord(sha[:1]) if idx == 0: start = 0 else: start = self._fan_out_table[idx-1] end = self._fan_out_table[idx] i = bisect_find_sha(start, end, sha, self._unpack_name) if i is None: raise KeyError(sha) return self._unpack_offset(i) class PackIndex1(FilePackIndex): """Version 1 Pack Index file.""" def __init__(self, filename, file=None, contents=None, size=None): super(PackIndex1, self).__init__(filename, file, contents, size) 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(FilePackIndex): """Version 2 Pack Index file.""" def __init__(self, filename, file=None, contents=None, size=None): super(PackIndex2, self).__init__(filename, file, contents, size) if self._contents[:4] != b'\377tOc': raise AssertionError('Not a v2 pack index file') (self.version, ) = unpack_from(b'>L', self._contents, 4) if self.version != 2: raise AssertionError('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)) self._pack_offset_largetable_offset = ( self._pack_offset_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 offset = unpack_from('>L', self._contents, offset)[0] if offset & (2**31): offset = ( self._pack_offset_largetable_offset + (offset & (2 ** 31 - 1)) * 8) offset = unpack_from('>Q', self._contents, offset)[0] return offset def _unpack_crc32_checksum(self, i): return unpack_from('>L', self._contents, self._crc32_table_offset + i * 4)[0] def read_pack_header(read): """Read the header of a pack file. :param read: Read function :return: Tuple of (pack version, number of objects). If no data is available to read, returns (None, None). """ header = read(12) if not header: return None, None if header[:4] != b'PACK': raise AssertionError('Invalid pack header %r' % header) (version,) = unpack_from(b'>L', header, 4) if version not in (2, 3): raise AssertionError('Version was %d' % version) (num_objects,) = unpack_from(b'>L', header, 8) return (version, num_objects) def chunks_length(chunks): if isinstance(chunks, bytes): return len(chunks) else: return sum(imap(len, chunks)) def unpack_object(read_all, read_some=None, compute_crc32=False, include_comp=False, zlib_bufsize=_ZLIB_BUFSIZE): """Unpack a Git object. :param read_all: Read function that blocks until the number of requested bytes are read. :param read_some: Read function that returns at least one byte, but may not return the number of bytes requested. :param compute_crc32: If True, compute the CRC32 of the compressed data. If False, the returned CRC32 will be None. :param include_comp: If True, include compressed data in the result. :param zlib_bufsize: An optional buffer size for zlib operations. :return: A tuple of (unpacked, unused), where unused is the unused data leftover from decompression, and unpacked in an UnpackedObject with the following attrs set: * obj_chunks (for non-delta types) * pack_type_num * delta_base (for delta types) * comp_chunks (if include_comp is True) * decomp_chunks * decomp_len * crc32 (if compute_crc32 is True) """ if read_some is None: read_some = read_all if compute_crc32: crc32 = 0 else: crc32 = None bytes, crc32 = take_msb_bytes(read_all, crc32=crc32) type_num = (bytes[0] >> 4) & 0x07 size = bytes[0] & 0x0f for i, byte in enumerate(bytes[1:]): size += (byte & 0x7f) << ((i * 7) + 4) raw_base = len(bytes) if type_num == OFS_DELTA: bytes, crc32 = take_msb_bytes(read_all, crc32=crc32) raw_base += len(bytes) if bytes[-1] & 0x80: raise AssertionError delta_base_offset = bytes[0] & 0x7f for byte in bytes[1:]: delta_base_offset += 1 delta_base_offset <<= 7 delta_base_offset += (byte & 0x7f) delta_base = delta_base_offset elif type_num == REF_DELTA: delta_base = read_all(20) if compute_crc32: crc32 = binascii.crc32(delta_base, crc32) raw_base += 20 else: delta_base = None unpacked = UnpackedObject(type_num, delta_base, size, crc32) unused = read_zlib_chunks(read_some, unpacked, buffer_size=zlib_bufsize, include_comp=include_comp) return unpacked, unused def _compute_object_size(value): """Compute the size of a unresolved object for use with LRUSizeCache.""" (num, obj) = value if num in DELTA_TYPES: return chunks_length(obj[1]) return chunks_length(obj) class PackStreamReader(object): """Class to read a pack stream. The pack is read from a ReceivableProtocol using read() or recv() as appropriate. """ def __init__(self, read_all, read_some=None, zlib_bufsize=_ZLIB_BUFSIZE): self.read_all = read_all if read_some is None: self.read_some = read_all else: self.read_some = read_some self.sha = sha1() self._offset = 0 self._rbuf = BytesIO() # trailer is a deque to avoid memory allocation on small reads self._trailer = deque() self._zlib_bufsize = zlib_bufsize def _read(self, read, size): """Read up to size bytes using the given callback. As a side effect, update the verifier's hash (excluding the last 20 bytes read). :param read: The read callback to read from. :param size: The maximum number of bytes to read; the particular behavior is callback-specific. """ data = read(size) # maintain a trailer of the last 20 bytes we've read n = len(data) self._offset += n tn = len(self._trailer) if n >= 20: to_pop = tn to_add = 20 else: to_pop = max(n + tn - 20, 0) to_add = n self.sha.update( bytes(bytearray([self._trailer.popleft() for _ in range(to_pop)]))) self._trailer.extend(data[-to_add:]) # hash everything but the trailer self.sha.update(data[:-to_add]) return data def _buf_len(self): buf = self._rbuf start = buf.tell() buf.seek(0, SEEK_END) end = buf.tell() buf.seek(start) return end - start @property def offset(self): return self._offset - self._buf_len() def read(self, size): """Read, blocking until size bytes are read.""" buf_len = self._buf_len() if buf_len >= size: return self._rbuf.read(size) buf_data = self._rbuf.read() self._rbuf = BytesIO() return buf_data + self._read(self.read_all, size - buf_len) def recv(self, size): """Read up to size bytes, blocking until one byte is read.""" buf_len = self._buf_len() if buf_len: data = self._rbuf.read(size) if size >= buf_len: self._rbuf = BytesIO() return data return self._read(self.read_some, size) def __len__(self): return self._num_objects def read_objects(self, compute_crc32=False): """Read the objects in this pack file. :param compute_crc32: If True, compute the CRC32 of the compressed data. If False, the returned CRC32 will be None. :return: Iterator over UnpackedObjects with the following members set: offset obj_type_num obj_chunks (for non-delta types) delta_base (for delta types) decomp_chunks decomp_len crc32 (if compute_crc32 is True) :raise ChecksumMismatch: if the checksum of the pack contents does not match the checksum in the pack trailer. :raise zlib.error: if an error occurred during zlib decompression. :raise IOError: if an error occurred writing to the output file. """ pack_version, self._num_objects = read_pack_header(self.read) if pack_version is None: return for i in range(self._num_objects): offset = self.offset unpacked, unused = unpack_object( self.read, read_some=self.recv, compute_crc32=compute_crc32, zlib_bufsize=self._zlib_bufsize) unpacked.offset = offset # prepend any unused data to current read buffer buf = BytesIO() buf.write(unused) buf.write(self._rbuf.read()) buf.seek(0) self._rbuf = buf yield unpacked if self._buf_len() < 20: # If the read buffer is full, then the last read() got the whole # trailer off the wire. If not, it means there is still some of the # trailer to read. We need to read() all 20 bytes; N come from the # read buffer and (20 - N) come from the wire. self.read(20) pack_sha = bytearray(self._trailer) if pack_sha != self.sha.digest(): raise ChecksumMismatch(sha_to_hex(pack_sha), self.sha.hexdigest()) class PackStreamCopier(PackStreamReader): """Class to verify a pack stream as it is being read. The pack is read from a ReceivableProtocol using read() or recv() as appropriate and written out to the given file-like object. """ def __init__(self, read_all, read_some, outfile, delta_iter=None): """Initialize the copier. :param read_all: Read function that blocks until the number of requested bytes are read. :param read_some: Read function that returns at least one byte, but may not return the number of bytes requested. :param outfile: File-like object to write output through. :param delta_iter: Optional DeltaChainIterator to record deltas as we read them. """ super(PackStreamCopier, self).__init__(read_all, read_some=read_some) self.outfile = outfile self._delta_iter = delta_iter def _read(self, read, size): """Read data from the read callback and write it to the file.""" data = super(PackStreamCopier, self)._read(read, size) self.outfile.write(data) return data def verify(self): """Verify a pack stream and write it to the output file. See PackStreamReader.iterobjects for a list of exceptions this may throw. """ if self._delta_iter: for unpacked in self.read_objects(): self._delta_iter.record(unpacked) else: for _ in self.read_objects(): pass def obj_sha(type, chunks): """Compute the SHA for a numeric type and object chunks.""" sha = sha1() sha.update(object_header(type, chunks_length(chunks))) if isinstance(chunks, bytes): sha.update(chunks) else: for chunk in chunks: sha.update(chunk) return sha.digest() def compute_file_sha(f, start_ofs=0, end_ofs=0, buffer_size=1 << 16): """Hash a portion of a file into a new SHA. :param f: A file-like object to read from that supports seek(). :param start_ofs: The offset in the file to start reading at. :param end_ofs: The offset in the file to end reading at, relative to the end of the file. :param buffer_size: A buffer size for reading. :return: A new SHA object updated with data read from the file. """ sha = sha1() f.seek(0, SEEK_END) length = f.tell() if (end_ofs < 0 and length + end_ofs < start_ofs) or end_ofs > length: raise AssertionError( "Attempt to read beyond file length. " "start_ofs: %d, end_ofs: %d, file length: %d" % ( start_ofs, end_ofs, length)) todo = length + end_ofs - start_ofs f.seek(start_ofs) while todo: data = f.read(min(todo, buffer_size)) sha.update(data) todo -= len(data) return sha class PackData(object): """The data contained in a packfile. Pack files can be accessed both sequentially for exploding a pack, and directly with the help of an index to retrieve a specific object. The objects within are either complete or a delta against another. The header is variable length. If the MSB of each byte is set then it indicates that the subsequent byte is still part of the header. For the first byte the next MS bits are the type, which tells you the type of object, and whether it is a delta. The LS byte is the lowest bits of the size. For each subsequent byte the LS 7 bits are the next MS bits of the size, i.e. the last byte of the header contains the MS bits of the size. For the complete objects the data is stored as zlib deflated data. The size in the header is the uncompressed object size, so to uncompress you need to just keep feeding data to zlib until you get an object back, or it errors on bad data. This is done here by just giving the complete buffer from the start of the deflated object on. This is bad, but until I get mmap sorted out it will have to do. Currently there are no integrity checks done. Also no attempt is made to try and detect the delta case, or a request for an object at the wrong position. It will all just throw a zlib or KeyError. """ def __init__(self, filename, file=None, size=None): """Create a PackData object representing the pack in the given filename. The file must exist and stay readable until the object is disposed of. It must also stay the same size. It will be mapped whenever needed. Currently there is a restriction on the size of the pack as the python mmap implementation is flawed. """ self._filename = filename self._size = size self._header_size = 12 if file is None: self._file = GitFile(self._filename, 'rb') else: self._file = file (version, self._num_objects) = read_pack_header(self._file.read) self._offset_cache = LRUSizeCache( 1024*1024*20, compute_size=_compute_object_size) self.pack = None @property def filename(self): return os.path.basename(self._filename) @property def path(self): return self._filename @classmethod def from_file(cls, file, size): return cls(str(file), file=file, size=size) @classmethod def from_path(cls, path): return cls(filename=path) def close(self): self._file.close() def __enter__(self): return self def __exit__(self, exc_type, exc_val, exc_tb): self.close() def _get_size(self): if self._size is not None: return self._size self._size = os.path.getsize(self._filename) if self._size < self._header_size: errmsg = ('%s is too small for a packfile (%d < %d)' % (self._filename, self._size, self._header_size)) raise AssertionError(errmsg) return self._size def __len__(self): """Returns the number of objects in this pack.""" return self._num_objects def calculate_checksum(self): """Calculate the checksum for this pack. :return: 20-byte binary SHA1 digest """ return compute_file_sha(self._file, end_ofs=-20).digest() def get_ref(self, sha): """Get the object for a ref SHA, only looking in this pack.""" # TODO: cache these results if self.pack is None: raise KeyError(sha) try: offset = self.pack.index.object_index(sha) except KeyError: offset = None if offset: type, obj = self.get_object_at(offset) elif self.pack is not None and self.pack.resolve_ext_ref: type, obj = self.pack.resolve_ext_ref(sha) else: raise KeyError(sha) return offset, type, obj def resolve_object(self, offset, type, obj, get_ref=None): """Resolve an object, possibly resolving deltas when necessary. :return: Tuple with object type and contents. """ # Walk down the delta chain, building a stack of deltas to reach # the requested object. base_offset = offset base_type = type base_obj = obj delta_stack = [] while base_type in DELTA_TYPES: prev_offset = base_offset if get_ref is None: get_ref = self.get_ref if base_type == OFS_DELTA: (delta_offset, delta) = base_obj # TODO: clean up asserts and replace with nicer error messages base_offset = base_offset - delta_offset base_type, base_obj = self.get_object_at(base_offset) assert isinstance(base_type, int) elif base_type == REF_DELTA: (basename, delta) = base_obj assert isinstance(basename, bytes) and len(basename) == 20 base_offset, base_type, base_obj = get_ref(basename) assert isinstance(base_type, int) delta_stack.append((prev_offset, base_type, delta)) # Now grab the base object (mustn't be a delta) and apply the # deltas all the way up the stack. chunks = base_obj for prev_offset, delta_type, delta in reversed(delta_stack): chunks = apply_delta(chunks, delta) # TODO(dborowitz): This can result in poor performance if # large base objects are separated from deltas in the pack. # We should reorganize so that we apply deltas to all # objects in a chain one after the other to optimize cache # performance. if prev_offset is not None: self._offset_cache[prev_offset] = base_type, chunks return base_type, chunks def iterobjects(self, progress=None, compute_crc32=True): self._file.seek(self._header_size) for i in range(1, self._num_objects + 1): offset = self._file.tell() unpacked, unused = unpack_object( self._file.read, compute_crc32=compute_crc32) if progress is not None: progress(i, self._num_objects) yield (offset, unpacked.pack_type_num, unpacked._obj(), unpacked.crc32) # Back up over unused data. self._file.seek(-len(unused), SEEK_CUR) def _iter_unpacked(self): # TODO(dborowitz): Merge this with iterobjects, if we can change its # return type. self._file.seek(self._header_size) if self._num_objects is None: return for _ in range(self._num_objects): offset = self._file.tell() unpacked, unused = unpack_object( self._file.read, compute_crc32=False) unpacked.offset = offset yield unpacked # Back up over unused data. self._file.seek(-len(unused), SEEK_CUR) def iterentries(self, progress=None): """Yield entries summarizing the contents of this pack. :param progress: Progress function, called with current and total object count. :return: iterator of tuples with (sha, offset, crc32) """ num_objects = self._num_objects resolve_ext_ref = ( self.pack.resolve_ext_ref if self.pack is not None else None) indexer = PackIndexer.for_pack_data( self, resolve_ext_ref=resolve_ext_ref) for i, result in enumerate(indexer): if progress is not None: progress(i, num_objects) yield result def sorted_entries(self, progress=None): """Return entries in this pack, sorted by SHA. :param progress: Progress function, called with current and total object count :return: List of tuples with (sha, offset, crc32) """ ret = sorted(self.iterentries(progress=progress)) return ret def create_index_v1(self, filename, progress=None): """Create a version 1 file for this data file. :param filename: Index filename. :param progress: Progress report function :return: Checksum of index file """ entries = self.sorted_entries(progress=progress) with GitFile(filename, 'wb') as f: return write_pack_index_v1(f, entries, self.calculate_checksum()) def create_index_v2(self, filename, progress=None): """Create a version 2 index file for this data file. :param filename: Index filename. :param progress: Progress report function :return: Checksum of index file """ entries = self.sorted_entries(progress=progress) with GitFile(filename, 'wb') as f: return write_pack_index_v2(f, entries, self.calculate_checksum()) def create_index(self, filename, progress=None, version=2): """Create an index file for this data file. :param filename: Index filename. :param progress: Progress report function :return: Checksum of index file """ if version == 1: return self.create_index_v1(filename, progress) elif version == 2: return self.create_index_v2(filename, progress) else: raise ValueError('unknown index format %d' % version) def get_stored_checksum(self): """Return the expected checksum stored in this pack.""" self._file.seek(-20, SEEK_END) return self._file.read(20) def check(self): """Check the consistency of this pack.""" actual = self.calculate_checksum() stored = self.get_stored_checksum() if actual != stored: raise ChecksumMismatch(stored, actual) def get_compressed_data_at(self, offset): """Given offset in the packfile return compressed data that is there. Using the associated index the location of an object can be looked up, and then the packfile can be asked directly for that object using this function. """ assert offset >= self._header_size self._file.seek(offset) unpacked, _ = unpack_object(self._file.read, include_comp=True) return (unpacked.pack_type_num, unpacked.delta_base, unpacked.comp_chunks) def get_object_at(self, offset): """Given an offset in to the packfile return the object that is there. Using the associated index the location of an object can be looked up, and then the packfile can be asked directly for that object using this function. """ try: return self._offset_cache[offset] except KeyError: pass assert offset >= self._header_size self._file.seek(offset) unpacked, _ = unpack_object(self._file.read) return (unpacked.pack_type_num, unpacked._obj()) class DeltaChainIterator(object): """Abstract iterator over pack data based on delta chains. Each object in the pack is guaranteed to be inflated exactly once, regardless of how many objects reference it as a delta base. As a result, memory usage is proportional to the length of the longest delta chain. Subclasses can override _result to define the result type of the iterator. By default, results are UnpackedObjects with the following members set: * offset * obj_type_num * obj_chunks * pack_type_num * delta_base (for delta types) * comp_chunks (if _include_comp is True) * decomp_chunks * decomp_len * crc32 (if _compute_crc32 is True) """ _compute_crc32 = False _include_comp = False def __init__(self, file_obj, resolve_ext_ref=None): self._file = file_obj self._resolve_ext_ref = resolve_ext_ref self._pending_ofs = defaultdict(list) self._pending_ref = defaultdict(list) self._full_ofs = [] self._shas = {} self._ext_refs = [] @classmethod def for_pack_data(cls, pack_data, resolve_ext_ref=None): walker = cls(None, resolve_ext_ref=resolve_ext_ref) walker.set_pack_data(pack_data) for unpacked in pack_data._iter_unpacked(): walker.record(unpacked) return walker def record(self, unpacked): type_num = unpacked.pack_type_num offset = unpacked.offset if type_num == OFS_DELTA: base_offset = offset - unpacked.delta_base self._pending_ofs[base_offset].append(offset) elif type_num == REF_DELTA: self._pending_ref[unpacked.delta_base].append(offset) else: self._full_ofs.append((offset, type_num)) def set_pack_data(self, pack_data): self._file = pack_data._file def _walk_all_chains(self): for offset, type_num in self._full_ofs: for result in self._follow_chain(offset, type_num, None): yield result for result in self._walk_ref_chains(): yield result assert not self._pending_ofs def _ensure_no_pending(self): if self._pending_ref: raise KeyError([sha_to_hex(s) for s in self._pending_ref]) def _walk_ref_chains(self): if not self._resolve_ext_ref: self._ensure_no_pending() return for base_sha, pending in sorted(self._pending_ref.items()): if base_sha not in self._pending_ref: continue try: type_num, chunks = self._resolve_ext_ref(base_sha) except KeyError: # Not an external ref, but may depend on one. Either it will # get popped via a _follow_chain call, or we will raise an # error below. continue self._ext_refs.append(base_sha) self._pending_ref.pop(base_sha) for new_offset in pending: for result in self._follow_chain(new_offset, type_num, chunks): yield result self._ensure_no_pending() def _result(self, unpacked): return unpacked def _resolve_object(self, offset, obj_type_num, base_chunks): self._file.seek(offset) unpacked, _ = unpack_object( self._file.read, include_comp=self._include_comp, compute_crc32=self._compute_crc32) unpacked.offset = offset if base_chunks is None: assert unpacked.pack_type_num == obj_type_num else: assert unpacked.pack_type_num in DELTA_TYPES unpacked.obj_type_num = obj_type_num unpacked.obj_chunks = apply_delta(base_chunks, unpacked.decomp_chunks) return unpacked def _follow_chain(self, offset, obj_type_num, base_chunks): # Unlike PackData.get_object_at, there is no need to cache offsets as # this approach by design inflates each object exactly once. todo = [(offset, obj_type_num, base_chunks)] for offset, obj_type_num, base_chunks in todo: unpacked = self._resolve_object(offset, obj_type_num, base_chunks) yield self._result(unpacked) unblocked = chain(self._pending_ofs.pop(unpacked.offset, []), self._pending_ref.pop(unpacked.sha(), [])) todo.extend( (new_offset, unpacked.obj_type_num, unpacked.obj_chunks) for new_offset in unblocked) def __iter__(self): return self._walk_all_chains() def ext_refs(self): return self._ext_refs class PackIndexer(DeltaChainIterator): """Delta chain iterator that yields index entries.""" _compute_crc32 = True def _result(self, unpacked): return unpacked.sha(), unpacked.offset, unpacked.crc32 class PackInflater(DeltaChainIterator): """Delta chain iterator that yields ShaFile objects.""" def _result(self, unpacked): return unpacked.sha_file() class SHA1Reader(object): """Wrapper for file-like object that remembers the SHA1 of its data.""" def __init__(self, f): self.f = f self.sha1 = sha1(b'') def read(self, num=None): data = self.f.read(num) self.sha1.update(data) return data def check_sha(self): stored = self.f.read(20) if stored != self.sha1.digest(): raise ChecksumMismatch(self.sha1.hexdigest(), sha_to_hex(stored)) def close(self): return self.f.close() def tell(self): return self.f.tell() class SHA1Writer(object): """Wrapper for file-like object that remembers the SHA1 of its data.""" def __init__(self, f): self.f = f self.length = 0 self.sha1 = sha1(b'') def write(self, data): self.sha1.update(data) self.f.write(data) self.length += len(data) def write_sha(self): sha = self.sha1.digest() assert len(sha) == 20 self.f.write(sha) self.length += len(sha) return sha def close(self): sha = self.write_sha() self.f.close() return sha def offset(self): return self.length def tell(self): return self.f.tell() def pack_object_header(type_num, delta_base, size): """Create a pack object header for the given object info. :param type_num: Numeric type of the object. :param delta_base: Delta base offset or ref, or None for whole objects. :param size: Uncompressed object size. :return: A header for a packed object. """ header = [] c = (type_num << 4) | (size & 15) size >>= 4 while size: header.append(c | 0x80) c = size & 0x7f size >>= 7 header.append(c) if type_num == OFS_DELTA: ret = [delta_base & 0x7f] delta_base >>= 7 while delta_base: delta_base -= 1 ret.insert(0, 0x80 | (delta_base & 0x7f)) delta_base >>= 7 header.extend(ret) elif type_num == REF_DELTA: assert len(delta_base) == 20 header += delta_base return bytearray(header) def write_pack_object(f, type, object, sha=None): """Write pack object to a file. :param f: File to write to :param type: Numeric type of the object :param object: Object to write :return: Tuple with offset at which the object was written, and crc32 """ if type in DELTA_TYPES: delta_base, object = object else: delta_base = None header = bytes(pack_object_header(type, delta_base, len(object))) comp_data = zlib.compress(object) crc32 = 0 for data in (header, comp_data): f.write(data) if sha is not None: sha.update(data) crc32 = binascii.crc32(data, crc32) return crc32 & 0xffffffff def write_pack(filename, objects, deltify=None, delta_window_size=None): """Write a new pack data file. :param filename: Path to the new pack file (without .pack extension) :param objects: Iterable of (object, path) tuples to write. Should provide __len__ :param window_size: Delta window size :param deltify: Whether to deltify pack objects :return: Tuple with checksum of pack file and index file """ with GitFile(filename + '.pack', 'wb') as f: entries, data_sum = write_pack_objects( f, objects, delta_window_size=delta_window_size, deltify=deltify) entries = sorted([(k, v[0], v[1]) for (k, v) in entries.items()]) with GitFile(filename + '.idx', 'wb') as f: return data_sum, write_pack_index_v2(f, entries, data_sum) def write_pack_header(f, num_objects): """Write a pack header for the given number of objects.""" f.write(b'PACK') # Pack header f.write(struct.pack(b'>L', 2)) # Pack version f.write(struct.pack(b'>L', num_objects)) # Number of objects in pack def deltify_pack_objects(objects, window_size=None): """Generate deltas for pack objects. :param objects: An iterable of (object, path) tuples to deltify. :param window_size: Window size; None for default :return: Iterator over type_num, object id, delta_base, content delta_base is None for full text entries """ # TODO(jelmer): Use threads if window_size is None: window_size = DEFAULT_PACK_DELTA_WINDOW_SIZE # Build a list of objects ordered by the magic Linus heuristic # This helps us find good objects to diff against us magic = [] for obj, path in objects: magic.append((obj.type_num, path, -obj.raw_length(), obj)) magic.sort() possible_bases = deque() for type_num, path, neg_length, o in magic: raw = o.as_raw_string() winner = raw winner_base = None for base in possible_bases: if base.type_num != type_num: continue delta = create_delta(base.as_raw_string(), raw) if len(delta) < len(winner): winner_base = base.sha().digest() winner = delta yield type_num, o.sha().digest(), winner_base, winner possible_bases.appendleft(o) while len(possible_bases) > window_size: possible_bases.pop() def pack_objects_to_data(objects): """Create pack data from objects :param objects: Pack objects :return: Tuples with (type_num, hexdigest, delta base, object chunks) """ count = len(objects) return (count, ((o.type_num, o.sha().digest(), None, o.as_raw_string()) for (o, path) in objects)) def write_pack_objects(f, objects, delta_window_size=None, deltify=None): """Write a new pack data file. :param f: File to write to :param objects: Iterable of (object, path) tuples to write. Should provide __len__ :param window_size: Sliding window size for searching for deltas; Set to None for default window size. :param deltify: Whether to deltify objects :return: Dict mapping id -> (offset, crc32 checksum), pack checksum """ if deltify is None: # PERFORMANCE/TODO(jelmer): This should be enabled but is *much* too # slow at the moment. deltify = False if deltify: pack_contents = deltify_pack_objects(objects, delta_window_size) pack_contents_count = len(objects) else: pack_contents_count, pack_contents = pack_objects_to_data(objects) return write_pack_data(f, pack_contents_count, pack_contents) -def write_pack_data(f, num_records, records): +def write_pack_data(f, num_records, records, progress=None): """Write a new pack data file. :param f: File to write to :param num_records: Number of records :param records: Iterator over type_num, object_id, delta_base, raw + :param progress: Function to report progress to :return: Dict mapping id -> (offset, crc32 checksum), pack checksum """ # Write the pack entries = {} f = SHA1Writer(f) write_pack_header(f, num_records) - for type_num, object_id, delta_base, raw in records: + for i, (type_num, object_id, delta_base, raw) in enumerate(records): + if progress is not None: + progress(( + 'writing pack data: %d/%d\r' % + (i, num_records)).encode('ascii')) offset = f.offset() if delta_base is not None: try: base_offset, base_crc32 = entries[delta_base] except KeyError: type_num = REF_DELTA raw = (delta_base, raw) else: type_num = OFS_DELTA raw = (offset - base_offset, raw) crc32 = write_pack_object(f, type_num, raw) entries[object_id] = (offset, crc32) return entries, f.write_sha() def write_pack_index_v1(f, entries, pack_checksum): """Write a new pack index file. :param f: A file-like object to write to :param entries: List of tuples with object name (sha), offset_in_pack, and crc32_checksum. :param pack_checksum: Checksum of the pack file. :return: The SHA of the written index file """ f = SHA1Writer(f) fan_out_table = defaultdict(lambda: 0) for (name, offset, entry_checksum) in entries: fan_out_table[ord(name[:1])] += 1 # Fan-out table for i in range(0x100): f.write(struct.pack('>L', fan_out_table[i])) fan_out_table[i+1] += fan_out_table[i] for (name, offset, entry_checksum) in entries: if not (offset <= 0xffffffff): raise TypeError("pack format 1 only supports offsets < 2Gb") f.write(struct.pack('>L20s', offset, name)) assert len(pack_checksum) == 20 f.write(pack_checksum) return f.write_sha() def _delta_encode_size(size): ret = bytearray() c = size & 0x7f size >>= 7 while size: ret.append(c | 0x80) c = size & 0x7f size >>= 7 ret.append(c) return ret # The length of delta compression copy operations in version 2 packs is limited # to 64K. To copy more, we use several copy operations. Version 3 packs allow # 24-bit lengths in copy operations, but we always make version 2 packs. _MAX_COPY_LEN = 0xffff def _encode_copy_operation(start, length): scratch = [] op = 0x80 for i in range(4): if start & 0xff << i*8: scratch.append((start >> i*8) & 0xff) op |= 1 << i for i in range(2): if length & 0xff << i*8: scratch.append((length >> i*8) & 0xff) op |= 1 << (4+i) return bytearray([op] + scratch) def create_delta(base_buf, target_buf): """Use python difflib to work out how to transform base_buf to target_buf. :param base_buf: Base buffer :param target_buf: Target buffer """ assert isinstance(base_buf, bytes) assert isinstance(target_buf, bytes) out_buf = bytearray() # write delta header out_buf += _delta_encode_size(len(base_buf)) out_buf += _delta_encode_size(len(target_buf)) # write out delta opcodes seq = difflib.SequenceMatcher(a=base_buf, b=target_buf) for opcode, i1, i2, j1, j2 in seq.get_opcodes(): # Git patch opcodes don't care about deletes! # if opcode == 'replace' or opcode == 'delete': # pass if opcode == 'equal': # If they are equal, unpacker will use data from base_buf # Write out an opcode that says what range to use copy_start = i1 copy_len = i2 - i1 while copy_len > 0: to_copy = min(copy_len, _MAX_COPY_LEN) out_buf += _encode_copy_operation(copy_start, to_copy) copy_start += to_copy copy_len -= to_copy if opcode == 'replace' or opcode == 'insert': # If we are replacing a range or adding one, then we just # output it to the stream (prefixed by its size) s = j2 - j1 o = j1 while s > 127: out_buf.append(127) out_buf += bytearray(target_buf[o:o+127]) s -= 127 o += 127 out_buf.append(s) out_buf += bytearray(target_buf[o:o+s]) return bytes(out_buf) def apply_delta(src_buf, delta): """Based on the similar function in git's patch-delta.c. :param src_buf: Source buffer :param delta: Delta instructions """ if not isinstance(src_buf, bytes): src_buf = b''.join(src_buf) if not isinstance(delta, bytes): delta = b''.join(delta) out = [] index = 0 delta_length = len(delta) def get_delta_header_size(delta, index): size = 0 i = 0 while delta: cmd = ord(delta[index:index+1]) index += 1 size |= (cmd & ~0x80) << i i += 7 if not cmd & 0x80: break return size, index src_size, index = get_delta_header_size(delta, index) dest_size, index = get_delta_header_size(delta, index) assert src_size == len(src_buf), '%d vs %d' % (src_size, len(src_buf)) while index < delta_length: cmd = ord(delta[index:index+1]) index += 1 if cmd & 0x80: cp_off = 0 for i in range(4): if cmd & (1 << i): x = ord(delta[index:index+1]) index += 1 cp_off |= x << (i * 8) cp_size = 0 # Version 3 packs can contain copy sizes larger than 64K. for i in range(3): if cmd & (1 << (4+i)): x = ord(delta[index:index+1]) index += 1 cp_size |= x << (i * 8) if cp_size == 0: cp_size = 0x10000 if (cp_off + cp_size < cp_size or cp_off + cp_size > src_size or cp_size > dest_size): break out.append(src_buf[cp_off:cp_off+cp_size]) elif cmd != 0: out.append(delta[index:index+cmd]) index += cmd else: raise ApplyDeltaError('Invalid opcode 0') if index != delta_length: raise ApplyDeltaError('delta not empty: %r' % delta[index:]) if dest_size != chunks_length(out): raise ApplyDeltaError('dest size incorrect') return out def write_pack_index_v2(f, entries, pack_checksum): """Write a new pack index file. :param f: File-like object to write to :param entries: List of tuples with object name (sha), offset_in_pack, and crc32_checksum. :param pack_checksum: Checksum of the pack file. :return: The SHA of the index file written """ f = SHA1Writer(f) f.write(b'\377tOc') # Magic! f.write(struct.pack('>L', 2)) fan_out_table = defaultdict(lambda: 0) for (name, offset, entry_checksum) in entries: fan_out_table[ord(name[:1])] += 1 # Fan-out table largetable = [] for i in range(0x100): f.write(struct.pack(b'>L', fan_out_table[i])) fan_out_table[i+1] += fan_out_table[i] for (name, offset, entry_checksum) in entries: f.write(name) for (name, offset, entry_checksum) in entries: f.write(struct.pack(b'>L', entry_checksum)) for (name, offset, entry_checksum) in entries: if offset < 2**31: f.write(struct.pack(b'>L', offset)) else: f.write(struct.pack(b'>L', 2**31 + len(largetable))) largetable.append(offset) for offset in largetable: f.write(struct.pack(b'>Q', offset)) assert len(pack_checksum) == 20 f.write(pack_checksum) return f.write_sha() write_pack_index = write_pack_index_v2 class Pack(object): """A Git pack object.""" def __init__(self, basename, resolve_ext_ref=None): self._basename = basename self._data = None self._idx = None self._idx_path = self._basename + '.idx' self._data_path = self._basename + '.pack' self._data_load = lambda: PackData(self._data_path) self._idx_load = lambda: load_pack_index(self._idx_path) self.resolve_ext_ref = resolve_ext_ref @classmethod def from_lazy_objects(self, data_fn, idx_fn): """Create a new pack object from callables to load pack data and index objects.""" ret = Pack('') ret._data_load = data_fn ret._idx_load = idx_fn return ret @classmethod def from_objects(self, data, idx): """Create a new pack object from pack data and index objects.""" ret = Pack('') ret._data_load = lambda: data ret._idx_load = lambda: idx return ret def name(self): """The SHA over the SHAs of the objects in this pack.""" return self.index.objects_sha1() @property def data(self): """The pack data object being used.""" if self._data is None: self._data = self._data_load() self._data.pack = self self.check_length_and_checksum() return self._data @property def index(self): """The index being used. :note: This may be an in-memory index """ if self._idx is None: self._idx = self._idx_load() return self._idx def close(self): if self._data is not None: self._data.close() if self._idx is not None: self._idx.close() def __enter__(self): return self def __exit__(self, exc_type, exc_val, exc_tb): self.close() def __eq__(self, other): return isinstance(self, type(other)) and self.index == other.index def __len__(self): """Number of entries in this pack.""" return len(self.index) def __repr__(self): return '%s(%r)' % (self.__class__.__name__, self._basename) def __iter__(self): """Iterate over all the sha1s of the objects in this pack.""" return iter(self.index) def check_length_and_checksum(self): """Sanity check the length and checksum of the pack index and data.""" assert len(self.index) == len(self.data) idx_stored_checksum = self.index.get_pack_checksum() data_stored_checksum = self.data.get_stored_checksum() if idx_stored_checksum != data_stored_checksum: raise ChecksumMismatch(sha_to_hex(idx_stored_checksum), sha_to_hex(data_stored_checksum)) def check(self): """Check the integrity of this pack. :raise ChecksumMismatch: if a checksum for the index or data is wrong """ self.index.check() self.data.check() for obj in self.iterobjects(): obj.check() # TODO: object connectivity checks def get_stored_checksum(self): return self.data.get_stored_checksum() def __contains__(self, sha1): """Check whether this pack contains a particular SHA1.""" try: self.index.object_index(sha1) return True except KeyError: return False def get_raw_unresolved(self, sha1): """Get raw unresolved data for a SHA. :param sha1: SHA to return data for :return: Tuple with pack object type, delta base (if applicable), list of data chunks """ offset = self.index.object_index(sha1) (obj_type, delta_base, chunks) = self.data.get_compressed_data_at( offset) if obj_type == OFS_DELTA: delta_base = sha_to_hex( self.index.object_sha1(offset - delta_base)) obj_type = REF_DELTA return (obj_type, delta_base, chunks) def get_raw(self, sha1): offset = self.index.object_index(sha1) obj_type, obj = self.data.get_object_at(offset) type_num, chunks = self.data.resolve_object(offset, obj_type, obj) return type_num, b''.join(chunks) def __getitem__(self, sha1): """Retrieve the specified SHA1.""" type, uncomp = self.get_raw(sha1) return ShaFile.from_raw_string(type, uncomp, sha=sha1) def iterobjects(self): """Iterate over the objects in this pack.""" return iter(PackInflater.for_pack_data( self.data, resolve_ext_ref=self.resolve_ext_ref)) def pack_tuples(self): """Provide an iterable for use with write_pack_objects. :return: Object that can iterate over (object, path) tuples and provides __len__ """ class PackTupleIterable(object): def __init__(self, pack): self.pack = pack def __len__(self): return len(self.pack) def __iter__(self): return ((o, None) for o in self.pack.iterobjects()) return PackTupleIterable(self) def keep(self, msg=None): """Add a .keep file for the pack, preventing git from garbage collecting it. :param msg: A message written inside the .keep file; can be used later to determine whether or not a .keep file is obsolete. :return: The path of the .keep file, as a string. """ keepfile_name = '%s.keep' % self._basename with GitFile(keepfile_name, 'wb') as keepfile: if msg: keepfile.write(msg) keepfile.write(b'\n') return keepfile_name try: from dulwich._pack import apply_delta, bisect_find_sha # noqa: F811 except ImportError: pass diff --git a/dulwich/repo.py b/dulwich/repo.py index b3cc2cad..e9e20f0d 100644 --- a/dulwich/repo.py +++ b/dulwich/repo.py @@ -1,1247 +1,1247 @@ # repo.py -- For dealing with git repositories. # Copyright (C) 2007 James Westby # Copyright (C) 2008-2013 Jelmer Vernooij # # Dulwich is dual-licensed under the Apache License, Version 2.0 and the GNU # General Public License as public by the Free Software Foundation; version 2.0 # or (at your option) any later version. You can redistribute it and/or # modify it under the terms of either of these two licenses. # # Unless required by applicable law or agreed to in writing, software # distributed under the License is distributed on an "AS IS" BASIS, # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. # See the License for the specific language governing permissions and # limitations under the License. # # You should have received a copy of the licenses; if not, see # for a copy of the GNU General Public License # and for a copy of the Apache # License, Version 2.0. # """Repository access. This module contains the base class for git repositories (BaseRepo) and an implementation which uses a repository on local disk (Repo). """ from io import BytesIO import errno import os import sys import stat import time from dulwich.errors import ( NoIndexPresent, NotBlobError, NotCommitError, NotGitRepository, NotTreeError, NotTagError, CommitError, RefFormatError, HookError, ) from dulwich.file import ( GitFile, ) from dulwich.object_store import ( DiskObjectStore, MemoryObjectStore, ObjectStoreGraphWalker, ) from dulwich.objects import ( check_hexsha, Blob, Commit, ShaFile, Tag, Tree, ) from dulwich.hooks import ( PreCommitShellHook, PostCommitShellHook, CommitMsgShellHook, ) from dulwich.refs import ( # noqa: F401 check_ref_format, RefsContainer, DictRefsContainer, InfoRefsContainer, DiskRefsContainer, read_packed_refs, read_packed_refs_with_peeled, write_packed_refs, SYMREF, ) import warnings CONTROLDIR = '.git' OBJECTDIR = 'objects' REFSDIR = 'refs' REFSDIR_TAGS = 'tags' REFSDIR_HEADS = 'heads' INDEX_FILENAME = "index" COMMONDIR = 'commondir' GITDIR = 'gitdir' WORKTREES = 'worktrees' BASE_DIRECTORIES = [ ["branches"], [REFSDIR], [REFSDIR, REFSDIR_TAGS], [REFSDIR, REFSDIR_HEADS], ["hooks"], ["info"] ] DEFAULT_REF = b'refs/heads/master' class InvalidUserIdentity(Exception): """User identity is not of the format 'user '""" def __init__(self, identity): self.identity = identity def check_user_identity(identity): """Verify that a user identity is formatted correctly. :param identity: User identity bytestring :raise InvalidUserIdentity: Raised when identity is invalid """ try: fst, snd = identity.split(b' <', 1) except ValueError: raise InvalidUserIdentity(identity) if b'>' not in snd: raise InvalidUserIdentity(identity) def parse_graftpoints(graftpoints): """Convert a list of graftpoints into a dict :param graftpoints: Iterator of graftpoint lines Each line is formatted as: []* Resulting dictionary is: : [*] https://git.wiki.kernel.org/index.php/GraftPoint """ grafts = {} for l in graftpoints: raw_graft = l.split(None, 1) commit = raw_graft[0] if len(raw_graft) == 2: parents = raw_graft[1].split() else: parents = [] for sha in [commit] + parents: check_hexsha(sha, 'Invalid graftpoint') grafts[commit] = parents return grafts def serialize_graftpoints(graftpoints): """Convert a dictionary of grafts into string The graft dictionary is: : [*] Each line is formatted as: []* https://git.wiki.kernel.org/index.php/GraftPoint """ graft_lines = [] for commit, parents in graftpoints.items(): if parents: graft_lines.append(commit + b' ' + b' '.join(parents)) else: graft_lines.append(commit) return b'\n'.join(graft_lines) class BaseRepo(object): """Base class for a git repository. :ivar object_store: Dictionary-like object for accessing the objects :ivar refs: Dictionary-like object with the refs in this repository """ def __init__(self, object_store, refs): """Open a repository. This shouldn't be called directly, but rather through one of the base classes, such as MemoryRepo or Repo. :param object_store: Object store to use :param refs: Refs container to use """ self.object_store = object_store self.refs = refs self._graftpoints = {} self.hooks = {} def _determine_file_mode(self): """Probe the file-system to determine whether permissions can be trusted. :return: True if permissions can be trusted, False otherwise. """ raise NotImplementedError(self._determine_file_mode) def _init_files(self, bare): """Initialize a default set of named files.""" from dulwich.config import ConfigFile self._put_named_file('description', b"Unnamed repository") f = BytesIO() cf = ConfigFile() cf.set("core", "repositoryformatversion", "0") if self._determine_file_mode(): cf.set("core", "filemode", True) else: cf.set("core", "filemode", False) cf.set("core", "bare", bare) cf.set("core", "logallrefupdates", True) cf.write_to_file(f) self._put_named_file('config', f.getvalue()) self._put_named_file(os.path.join('info', 'exclude'), b'') def get_named_file(self, path): """Get a file from the control dir with a specific name. Although the filename should be interpreted as a filename relative to the control dir in a disk-based Repo, the object returned need not be pointing to a file in that location. :param path: The path to the file, relative to the control dir. :return: An open file object, or None if the file does not exist. """ raise NotImplementedError(self.get_named_file) def _put_named_file(self, path, contents): """Write a file to the control dir with the given name and contents. :param path: The path to the file, relative to the control dir. :param contents: A string to write to the file. """ raise NotImplementedError(self._put_named_file) def open_index(self): """Open the index for this repository. :raise NoIndexPresent: If no index is present :return: The matching `Index` """ raise NotImplementedError(self.open_index) def fetch(self, target, determine_wants=None, progress=None): """Fetch objects into another repository. :param target: The target repository :param determine_wants: Optional function to determine what refs to fetch. :param progress: Optional progress function :return: The local refs """ if determine_wants is None: determine_wants = target.object_store.determine_wants_all target.object_store.add_objects( self.fetch_objects(determine_wants, target.get_graph_walker(), - progress)) + progress), progress=progress) return self.get_refs() def fetch_objects(self, determine_wants, graph_walker, progress, get_tagged=None): """Fetch the missing objects required for a set of revisions. :param determine_wants: Function that takes a dictionary with heads and returns the list of heads to fetch. :param graph_walker: Object that can iterate over the list of revisions to fetch and has an "ack" method that will be called to acknowledge that a revision is present. :param progress: Simple progress function that will be called with updated progress strings. :param get_tagged: Function that returns a dict of pointed-to sha -> tag sha for including tags. :return: iterator over objects, with __len__ implemented """ wants = determine_wants(self.get_refs()) if not isinstance(wants, list): raise TypeError("determine_wants() did not return a list") shallows = getattr(graph_walker, 'shallow', frozenset()) unshallows = getattr(graph_walker, 'unshallow', frozenset()) if wants == []: # TODO(dborowitz): find a way to short-circuit that doesn't change # this interface. if shallows or unshallows: # Do not send a pack in shallow short-circuit path return None return [] # If the graph walker is set up with an implementation that can # ACK/NAK to the wire, it will write data to the client through # this call as a side-effect. haves = self.object_store.find_common_revisions(graph_walker) # Deal with shallow requests separately because the haves do # not reflect what objects are missing if shallows or unshallows: # TODO: filter the haves commits from iter_shas. the specific # commits aren't missing. haves = [] def get_parents(commit): if commit.id in shallows: return [] return self.get_parents(commit.id, commit) return self.object_store.iter_shas( self.object_store.find_missing_objects( haves, wants, progress, get_tagged, get_parents=get_parents)) def get_graph_walker(self, heads=None): """Retrieve a graph walker. A graph walker is used by a remote repository (or proxy) to find out which objects are present in this repository. :param heads: Repository heads to use (optional) :return: A graph walker object """ if heads is None: heads = self.refs.as_dict(b'refs/heads').values() return ObjectStoreGraphWalker(heads, self.get_parents) def get_refs(self): """Get dictionary with all refs. :return: A ``dict`` mapping ref names to SHA1s """ return self.refs.as_dict() def head(self): """Return the SHA1 pointed at by HEAD.""" return self.refs[b'HEAD'] def _get_object(self, sha, cls): assert len(sha) in (20, 40) ret = self.get_object(sha) if not isinstance(ret, cls): if cls is Commit: raise NotCommitError(ret) elif cls is Blob: raise NotBlobError(ret) elif cls is Tree: raise NotTreeError(ret) elif cls is Tag: raise NotTagError(ret) else: raise Exception("Type invalid: %r != %r" % ( ret.type_name, cls.type_name)) return ret def get_object(self, sha): """Retrieve the object with the specified SHA. :param sha: SHA to retrieve :return: A ShaFile object :raise KeyError: when the object can not be found """ return self.object_store[sha] def get_parents(self, sha, commit=None): """Retrieve the parents of a specific commit. If the specific commit is a graftpoint, the graft parents will be returned instead. :param sha: SHA of the commit for which to retrieve the parents :param commit: Optional commit matching the sha :return: List of parents """ try: return self._graftpoints[sha] except KeyError: if commit is None: commit = self[sha] return commit.parents def get_config(self): """Retrieve the config object. :return: `ConfigFile` object for the ``.git/config`` file. """ raise NotImplementedError(self.get_config) def get_description(self): """Retrieve the description for this repository. :return: String with the description of the repository as set by the user. """ raise NotImplementedError(self.get_description) def set_description(self, description): """Set the description for this repository. :param description: Text to set as description for this repository. """ raise NotImplementedError(self.set_description) def get_config_stack(self): """Return a config stack for this repository. This stack accesses the configuration for both this repository itself (.git/config) and the global configuration, which usually lives in ~/.gitconfig. :return: `Config` instance for this repository """ from dulwich.config import StackedConfig backends = [self.get_config()] + StackedConfig.default_backends() return StackedConfig(backends, writable=backends[0]) def get_peeled(self, ref): """Get the peeled value of a ref. :param ref: The refname to peel. :return: The fully-peeled SHA1 of a tag object, after peeling all intermediate tags; if the original ref does not point to a tag, this will equal the original SHA1. """ cached = self.refs.get_peeled(ref) if cached is not None: return cached return self.object_store.peel_sha(self.refs[ref]).id def get_walker(self, include=None, *args, **kwargs): """Obtain a walker for this repository. :param include: Iterable of SHAs of commits to include along with their ancestors. Defaults to [HEAD] :param exclude: Iterable of SHAs of commits to exclude along with their ancestors, overriding includes. :param order: ORDER_* constant specifying the order of results. Anything other than ORDER_DATE may result in O(n) memory usage. :param reverse: If True, reverse the order of output, requiring O(n) memory. :param max_entries: The maximum number of entries to yield, or None for no limit. :param paths: Iterable of file or subtree paths to show entries for. :param rename_detector: diff.RenameDetector object for detecting renames. :param follow: If True, follow path across renames/copies. Forces a default rename_detector. :param since: Timestamp to list commits after. :param until: Timestamp to list commits before. :param queue_cls: A class to use for a queue of commits, supporting the iterator protocol. The constructor takes a single argument, the Walker. :return: A `Walker` object """ from dulwich.walk import Walker if include is None: include = [self.head()] if isinstance(include, str): include = [include] kwargs['get_parents'] = lambda commit: self.get_parents( commit.id, commit) return Walker(self.object_store, include, *args, **kwargs) def __getitem__(self, name): """Retrieve a Git object by SHA1 or ref. :param name: A Git object SHA1 or a ref name :return: A `ShaFile` object, such as a Commit or Blob :raise KeyError: when the specified ref or object does not exist """ if not isinstance(name, bytes): raise TypeError("'name' must be bytestring, not %.80s" % type(name).__name__) if len(name) in (20, 40): try: return self.object_store[name] except (KeyError, ValueError): pass try: return self.object_store[self.refs[name]] except RefFormatError: raise KeyError(name) def __contains__(self, name): """Check if a specific Git object or ref is present. :param name: Git object SHA1 or ref name """ if len(name) in (20, 40): return name in self.object_store or name in self.refs else: return name in self.refs def __setitem__(self, name, value): """Set a ref. :param name: ref name :param value: Ref value - either a ShaFile object, or a hex sha """ if name.startswith(b"refs/") or name == b'HEAD': if isinstance(value, ShaFile): self.refs[name] = value.id elif isinstance(value, bytes): self.refs[name] = value else: raise TypeError(value) else: raise ValueError(name) def __delitem__(self, name): """Remove a ref. :param name: Name of the ref to remove """ if name.startswith(b"refs/") or name == b"HEAD": del self.refs[name] else: raise ValueError(name) def _get_user_identity(self): """Determine the identity to use for new commits. """ user = os.environ.get("GIT_COMMITTER_NAME") email = os.environ.get("GIT_COMMITTER_EMAIL") config = self.get_config_stack() if user is None: try: user = config.get(("user", ), "name") except KeyError: user = None if email is None: try: email = config.get(("user", ), "email") except KeyError: email = None if user is None: import getpass user = getpass.getuser().encode(sys.getdefaultencoding()) if email is None: import getpass import socket email = ("{}@{}".format(getpass.getuser(), socket.gethostname()) .encode(sys.getdefaultencoding())) return (user + b" <" + email + b">") def _add_graftpoints(self, updated_graftpoints): """Add or modify graftpoints :param updated_graftpoints: Dict of commit shas to list of parent shas """ # Simple validation for commit, parents in updated_graftpoints.items(): for sha in [commit] + parents: check_hexsha(sha, 'Invalid graftpoint') self._graftpoints.update(updated_graftpoints) def _remove_graftpoints(self, to_remove=[]): """Remove graftpoints :param to_remove: List of commit shas """ for sha in to_remove: del self._graftpoints[sha] def do_commit(self, message=None, committer=None, author=None, commit_timestamp=None, commit_timezone=None, author_timestamp=None, author_timezone=None, tree=None, encoding=None, ref=b'HEAD', merge_heads=None): """Create a new commit. :param message: Commit message :param committer: Committer fullname :param author: Author fullname (defaults to committer) :param commit_timestamp: Commit timestamp (defaults to now) :param commit_timezone: Commit timestamp timezone (defaults to GMT) :param author_timestamp: Author timestamp (defaults to commit timestamp) :param author_timezone: Author timestamp timezone (defaults to commit timestamp timezone) :param tree: SHA1 of the tree root to use (if not specified the current index will be committed). :param encoding: Encoding :param ref: Optional ref to commit to (defaults to current branch) :param merge_heads: Merge heads (defaults to .git/MERGE_HEADS) :return: New commit SHA1 """ import time c = Commit() if tree is None: index = self.open_index() c.tree = index.commit(self.object_store) else: if len(tree) != 40: raise ValueError("tree must be a 40-byte hex sha string") c.tree = tree try: self.hooks['pre-commit'].execute() except HookError as e: raise CommitError(e) except KeyError: # no hook defined, silent fallthrough pass if merge_heads is None: # FIXME: Read merge heads from .git/MERGE_HEADS merge_heads = [] if committer is None: committer = self._get_user_identity() check_user_identity(committer) c.committer = committer if commit_timestamp is None: # FIXME: Support GIT_COMMITTER_DATE environment variable commit_timestamp = time.time() c.commit_time = int(commit_timestamp) if commit_timezone is None: # FIXME: Use current user timezone rather than UTC commit_timezone = 0 c.commit_timezone = commit_timezone if author is None: # FIXME: Support GIT_AUTHOR_NAME/GIT_AUTHOR_EMAIL environment # variables author = committer c.author = author check_user_identity(author) if author_timestamp is None: # FIXME: Support GIT_AUTHOR_DATE environment variable author_timestamp = commit_timestamp c.author_time = int(author_timestamp) if author_timezone is None: author_timezone = commit_timezone c.author_timezone = author_timezone if encoding is not None: c.encoding = encoding if message is None: # FIXME: Try to read commit message from .git/MERGE_MSG raise ValueError("No commit message specified") try: c.message = self.hooks['commit-msg'].execute(message) if c.message is None: c.message = message except HookError as e: raise CommitError(e) except KeyError: # no hook defined, message not modified c.message = message if ref is None: # Create a dangling commit c.parents = merge_heads self.object_store.add_object(c) else: try: old_head = self.refs[ref] c.parents = [old_head] + merge_heads self.object_store.add_object(c) ok = self.refs.set_if_equals( ref, old_head, c.id, message=b"commit: " + message, committer=committer, timestamp=commit_timestamp, timezone=commit_timezone) except KeyError: c.parents = merge_heads self.object_store.add_object(c) ok = self.refs.add_if_new( ref, c.id, message=b"commit: " + message, committer=committer, timestamp=commit_timestamp, timezone=commit_timezone) if not ok: # Fail if the atomic compare-and-swap failed, leaving the # commit and all its objects as garbage. raise CommitError("%s changed during commit" % (ref,)) try: self.hooks['post-commit'].execute() except HookError as e: # silent failure warnings.warn("post-commit hook failed: %s" % e, UserWarning) except KeyError: # no hook defined, silent fallthrough pass return c.id def read_gitfile(f): """Read a ``.git`` file. The first line of the file should start with "gitdir: " :param f: File-like object to read from :return: A path """ cs = f.read() if not cs.startswith("gitdir: "): raise ValueError("Expected file to start with 'gitdir: '") return cs[len("gitdir: "):].rstrip("\n") class Repo(BaseRepo): """A git repository backed by local disk. To open an existing repository, call the contructor with the path of the repository. To create a new repository, use the Repo.init class method. """ def __init__(self, root): hidden_path = os.path.join(root, CONTROLDIR) if os.path.isdir(os.path.join(hidden_path, OBJECTDIR)): self.bare = False self._controldir = hidden_path elif (os.path.isdir(os.path.join(root, OBJECTDIR)) and os.path.isdir(os.path.join(root, REFSDIR))): self.bare = True self._controldir = root elif os.path.isfile(hidden_path): self.bare = False with open(hidden_path, 'r') as f: path = read_gitfile(f) self.bare = False self._controldir = os.path.join(root, path) else: raise NotGitRepository( "No git repository was found at %(path)s" % dict(path=root) ) commondir = self.get_named_file(COMMONDIR) if commondir is not None: with commondir: self._commondir = os.path.join( self.controldir(), commondir.read().rstrip(b"\r\n").decode( sys.getfilesystemencoding())) else: self._commondir = self._controldir self.path = root object_store = DiskObjectStore( os.path.join(self.commondir(), OBJECTDIR)) refs = DiskRefsContainer(self.commondir(), self._controldir, logger=self._write_reflog) BaseRepo.__init__(self, object_store, refs) self._graftpoints = {} graft_file = self.get_named_file(os.path.join("info", "grafts"), basedir=self.commondir()) if graft_file: with graft_file: self._graftpoints.update(parse_graftpoints(graft_file)) graft_file = self.get_named_file("shallow", basedir=self.commondir()) if graft_file: with graft_file: self._graftpoints.update(parse_graftpoints(graft_file)) self.hooks['pre-commit'] = PreCommitShellHook(self.controldir()) self.hooks['commit-msg'] = CommitMsgShellHook(self.controldir()) self.hooks['post-commit'] = PostCommitShellHook(self.controldir()) def _write_reflog(self, ref, old_sha, new_sha, committer, timestamp, timezone, message): from .reflog import format_reflog_line path = os.path.join( self.controldir(), 'logs', ref.decode(sys.getfilesystemencoding())) try: os.makedirs(os.path.dirname(path)) except OSError as e: if e.errno != errno.EEXIST: raise if committer is None: committer = self._get_user_identity() check_user_identity(committer) if timestamp is None: timestamp = int(time.time()) if timezone is None: timezone = 0 # FIXME with open(path, 'ab') as f: f.write(format_reflog_line(old_sha, new_sha, committer, timestamp, timezone, message) + b'\n') @classmethod def discover(cls, start='.'): """Iterate parent directories to discover a repository Return a Repo object for the first parent directory that looks like a Git repository. :param start: The directory to start discovery from (defaults to '.') """ remaining = True path = os.path.abspath(start) while remaining: try: return cls(path) except NotGitRepository: path, remaining = os.path.split(path) raise NotGitRepository( "No git repository was found at %(path)s" % dict(path=start) ) def controldir(self): """Return the path of the control directory.""" return self._controldir def commondir(self): """Return the path of the common directory. For a main working tree, it is identical to controldir(). For a linked working tree, it is the control directory of the main working tree.""" return self._commondir def _determine_file_mode(self): """Probe the file-system to determine whether permissions can be trusted. :return: True if permissions can be trusted, False otherwise. """ fname = os.path.join(self.path, '.probe-permissions') with open(fname, 'w') as f: f.write('') st1 = os.lstat(fname) os.chmod(fname, st1.st_mode ^ stat.S_IXUSR) st2 = os.lstat(fname) os.unlink(fname) mode_differs = st1.st_mode != st2.st_mode st2_has_exec = (st2.st_mode & stat.S_IXUSR) != 0 return mode_differs and st2_has_exec def _put_named_file(self, path, contents): """Write a file to the control dir with the given name and contents. :param path: The path to the file, relative to the control dir. :param contents: A string to write to the file. """ path = path.lstrip(os.path.sep) with GitFile(os.path.join(self.controldir(), path), 'wb') as f: f.write(contents) def get_named_file(self, path, basedir=None): """Get a file from the control dir with a specific name. Although the filename should be interpreted as a filename relative to the control dir in a disk-based Repo, the object returned need not be pointing to a file in that location. :param path: The path to the file, relative to the control dir. :param basedir: Optional argument that specifies an alternative to the control dir. :return: An open file object, or None if the file does not exist. """ # TODO(dborowitz): sanitize filenames, since this is used directly by # the dumb web serving code. if basedir is None: basedir = self.controldir() path = path.lstrip(os.path.sep) try: return open(os.path.join(basedir, path), 'rb') except (IOError, OSError) as e: if e.errno == errno.ENOENT: return None raise def index_path(self): """Return path to the index file.""" return os.path.join(self.controldir(), INDEX_FILENAME) def open_index(self): """Open the index for this repository. :raise NoIndexPresent: If no index is present :return: The matching `Index` """ from dulwich.index import Index if not self.has_index(): raise NoIndexPresent() return Index(self.index_path()) def has_index(self): """Check if an index is present.""" # Bare repos must never have index files; non-bare repos may have a # missing index file, which is treated as empty. return not self.bare def stage(self, fs_paths): """Stage a set of paths. :param fs_paths: List of paths, relative to the repository path """ root_path_bytes = self.path.encode(sys.getfilesystemencoding()) if not isinstance(fs_paths, list): fs_paths = [fs_paths] from dulwich.index import ( blob_from_path_and_stat, index_entry_from_stat, _fs_to_tree_path, ) index = self.open_index() for fs_path in fs_paths: if not isinstance(fs_path, bytes): fs_path = fs_path.encode(sys.getfilesystemencoding()) if os.path.isabs(fs_path): raise ValueError( "path %r should be relative to " "repository root, not absolute" % fs_path) tree_path = _fs_to_tree_path(fs_path) full_path = os.path.join(root_path_bytes, fs_path) try: st = os.lstat(full_path) except OSError: # File no longer exists try: del index[tree_path] except KeyError: pass # already removed else: if not stat.S_ISDIR(st.st_mode): blob = blob_from_path_and_stat(full_path, st) self.object_store.add_object(blob) index[tree_path] = index_entry_from_stat(st, blob.id, 0) else: try: del index[tree_path] except KeyError: pass index.write() def clone(self, target_path, mkdir=True, bare=False, origin=b"origin"): """Clone this repository. :param target_path: Target path :param mkdir: Create the target directory :param bare: Whether to create a bare repository :param origin: Base name for refs in target repository cloned from this repository :return: Created repository as `Repo` """ if not bare: target = self.init(target_path, mkdir=mkdir) else: target = self.init_bare(target_path, mkdir=mkdir) self.fetch(target) encoded_path = self.path if not isinstance(encoded_path, bytes): encoded_path = encoded_path.encode(sys.getfilesystemencoding()) ref_message = b"clone: from " + encoded_path target.refs.import_refs( b'refs/remotes/' + origin, self.refs.as_dict(b'refs/heads'), message=ref_message) target.refs.import_refs( b'refs/tags', self.refs.as_dict(b'refs/tags'), message=ref_message) try: target.refs.add_if_new( DEFAULT_REF, self.refs[DEFAULT_REF], message=ref_message) except KeyError: pass target_config = target.get_config() target_config.set(('remote', 'origin'), 'url', encoded_path) target_config.set(('remote', 'origin'), 'fetch', '+refs/heads/*:refs/remotes/origin/*') target_config.write_to_path() # Update target head head_chain, head_sha = self.refs.follow(b'HEAD') if head_chain and head_sha is not None: target.refs.set_symbolic_ref(b'HEAD', head_chain[-1], message=ref_message) target[b'HEAD'] = head_sha if not bare: # Checkout HEAD to target dir target.reset_index() return target def reset_index(self, tree=None): """Reset the index back to a specific tree. :param tree: Tree SHA to reset to, None for current HEAD tree. """ from dulwich.index import ( build_index_from_tree, validate_path_element_default, validate_path_element_ntfs, ) if tree is None: tree = self[b'HEAD'].tree config = self.get_config() honor_filemode = config.get_boolean( b'core', b'filemode', os.name != "nt") if config.get_boolean(b'core', b'core.protectNTFS', os.name == "nt"): validate_path_element = validate_path_element_ntfs else: validate_path_element = validate_path_element_default return build_index_from_tree( self.path, self.index_path(), self.object_store, tree, honor_filemode=honor_filemode, validate_path_element=validate_path_element) def get_config(self): """Retrieve the config object. :return: `ConfigFile` object for the ``.git/config`` file. """ from dulwich.config import ConfigFile path = os.path.join(self._controldir, 'config') try: return ConfigFile.from_path(path) except (IOError, OSError) as e: if e.errno != errno.ENOENT: raise ret = ConfigFile() ret.path = path return ret def get_description(self): """Retrieve the description of this repository. :return: A string describing the repository or None. """ path = os.path.join(self._controldir, 'description') try: with GitFile(path, 'rb') as f: return f.read() except (IOError, OSError) as e: if e.errno != errno.ENOENT: raise return None def __repr__(self): return "" % self.path def set_description(self, description): """Set the description for this repository. :param description: Text to set as description for this repository. """ self._put_named_file('description', description) @classmethod def _init_maybe_bare(cls, path, bare): for d in BASE_DIRECTORIES: os.mkdir(os.path.join(path, *d)) DiskObjectStore.init(os.path.join(path, OBJECTDIR)) ret = cls(path) ret.refs.set_symbolic_ref(b'HEAD', DEFAULT_REF) ret._init_files(bare) return ret @classmethod def init(cls, path, mkdir=False): """Create a new repository. :param path: Path in which to create the repository :param mkdir: Whether to create the directory :return: `Repo` instance """ if mkdir: os.mkdir(path) controldir = os.path.join(path, CONTROLDIR) os.mkdir(controldir) cls._init_maybe_bare(controldir, False) return cls(path) @classmethod def _init_new_working_directory(cls, path, main_repo, identifier=None, mkdir=False): """Create a new working directory linked to a repository. :param path: Path in which to create the working tree. :param main_repo: Main repository to reference :param identifier: Worktree identifier :param mkdir: Whether to create the directory :return: `Repo` instance """ if mkdir: os.mkdir(path) if identifier is None: identifier = os.path.basename(path) main_worktreesdir = os.path.join(main_repo.controldir(), WORKTREES) worktree_controldir = os.path.join(main_worktreesdir, identifier) gitdirfile = os.path.join(path, CONTROLDIR) with open(gitdirfile, 'wb') as f: f.write(b'gitdir: ' + worktree_controldir.encode(sys.getfilesystemencoding()) + b'\n') try: os.mkdir(main_worktreesdir) except OSError as e: if e.errno != errno.EEXIST: raise try: os.mkdir(worktree_controldir) except OSError as e: if e.errno != errno.EEXIST: raise with open(os.path.join(worktree_controldir, GITDIR), 'wb') as f: f.write(gitdirfile.encode(sys.getfilesystemencoding()) + b'\n') with open(os.path.join(worktree_controldir, COMMONDIR), 'wb') as f: f.write(b'../..\n') with open(os.path.join(worktree_controldir, 'HEAD'), 'wb') as f: f.write(main_repo.head() + b'\n') r = cls(path) r.reset_index() return r @classmethod def init_bare(cls, path, mkdir=False): """Create a new bare repository. ``path`` should already exist and be an empty directory. :param path: Path to create bare repository in :return: a `Repo` instance """ if mkdir: os.mkdir(path) return cls._init_maybe_bare(path, True) create = init_bare def close(self): """Close any files opened by this repository.""" self.object_store.close() def __enter__(self): return self def __exit__(self, exc_type, exc_val, exc_tb): self.close() class MemoryRepo(BaseRepo): """Repo that stores refs, objects, and named files in memory. MemoryRepos are always bare: they have no working tree and no index, since those have a stronger dependency on the filesystem. """ def __init__(self): from dulwich.config import ConfigFile self._reflog = [] refs_container = DictRefsContainer({}, logger=self._append_reflog) BaseRepo.__init__(self, MemoryObjectStore(), refs_container) self._named_files = {} self.bare = True self._config = ConfigFile() self._description = None def _append_reflog(self, *args): self._reflog.append(args) def set_description(self, description): self._description = description def get_description(self): return self._description def _determine_file_mode(self): """Probe the file-system to determine whether permissions can be trusted. :return: True if permissions can be trusted, False otherwise. """ return sys.platform != 'win32' def _put_named_file(self, path, contents): """Write a file to the control dir with the given name and contents. :param path: The path to the file, relative to the control dir. :param contents: A string to write to the file. """ self._named_files[path] = contents def get_named_file(self, path): """Get a file from the control dir with a specific name. Although the filename should be interpreted as a filename relative to the control dir in a disk-baked Repo, the object returned need not be pointing to a file in that location. :param path: The path to the file, relative to the control dir. :return: An open file object, or None if the file does not exist. """ contents = self._named_files.get(path, None) if contents is None: return None return BytesIO(contents) def open_index(self): """Fail to open index for this repo, since it is bare. :raise NoIndexPresent: Raised when no index is present """ raise NoIndexPresent() def get_config(self): """Retrieve the config object. :return: `ConfigFile` object. """ return self._config @classmethod def init_bare(cls, objects, refs): """Create a new bare repository in memory. :param objects: Objects for the new repository, as iterable :param refs: Refs as dictionary, mapping names to object SHA1s """ ret = cls() for obj in objects: ret.object_store.add_object(obj) for refname, sha in refs.items(): ret.refs.add_if_new(refname, sha) ret._init_files(bare=True) return ret