diff --git a/swh/vault/cookers/git_bare.py b/swh/vault/cookers/git_bare.py index 9de6a75..6f057ef 100644 --- a/swh/vault/cookers/git_bare.py +++ b/swh/vault/cookers/git_bare.py @@ -1,509 +1,517 @@ # Copyright (C) 2021 The Software Heritage developers # See the AUTHORS file at the top-level directory of this distribution # License: GNU General Public License version 3, or any later version # See top-level LICENSE file for more information """ This cooker creates tarballs containing a bare .git directory, that can be unpacked and cloned like any git repository. It works in three steps: 1. Write objects one by one in :file:`.git/objects/` 2. Calls ``git repack`` to pack all these objects into git packfiles. 3. Creates a tarball of the resulting repository It keeps a set of all written (or about-to-be-written) object hashes in memory to avoid downloading and writing the same objects twice. """ import datetime import logging import os.path import re import subprocess import tarfile import tempfile from typing import Any, Dict, Iterable, Iterator, List, Optional, Set, Tuple import zlib from swh.core.api.classes import stream_results_optional from swh.model import identifiers from swh.model.hashutil import hash_to_bytehex, hash_to_hex from swh.model.model import ( Content, DirectoryEntry, ObjectType, Person, Revision, RevisionType, Sha1Git, TargetType, TimestampWithTimezone, ) from swh.storage.algos.revisions_walker import DFSRevisionsWalker from swh.storage.algos.snapshot import snapshot_get_all_branches from swh.vault.cookers.base import BaseVaultCooker from swh.vault.to_disk import HIDDEN_MESSAGE, SKIPPED_MESSAGE RELEASE_BATCH_SIZE = 10000 REVISION_BATCH_SIZE = 10000 DIRECTORY_BATCH_SIZE = 10000 CONTENT_BATCH_SIZE = 100 logger = logging.getLogger(__name__) class GitBareCooker(BaseVaultCooker): use_fsck = True def cache_type_key(self) -> str: return self.obj_type def check_exists(self): obj_type = self.obj_type.split("_")[0] if obj_type == "revision": return not list(self.storage.revision_missing([self.obj_id])) elif obj_type == "directory": return not list(self.storage.directory_missing([self.obj_id])) if obj_type == "snapshot": return not list(self.storage.snapshot_missing([self.obj_id])) else: raise NotImplementedError(f"GitBareCooker for {obj_type}") def obj_swhid(self) -> identifiers.CoreSWHID: obj_type = self.obj_type.split("_")[0] return identifiers.CoreSWHID( object_type=identifiers.ObjectType[obj_type.upper()], object_id=self.obj_id, ) def _push(self, stack: List[Sha1Git], obj_ids: Iterable[Sha1Git]) -> None: assert not isinstance(obj_ids, bytes) revision_ids = [id_ for id_ in obj_ids if id_ not in self._seen] self._seen.update(revision_ids) stack.extend(revision_ids) def _pop(self, stack: List[Sha1Git], n: int) -> List[Sha1Git]: obj_ids = stack[-n:] stack[-n:] = [] return obj_ids def prepare_bundle(self): # Objects we will visit soon: self._rel_stack: List[Sha1Git] = [] self._rev_stack: List[Sha1Git] = [] self._dir_stack: List[Sha1Git] = [] self._cnt_stack: List[Sha1Git] = [] # Set of objects already in any of the stacks: self._seen: Set[Sha1Git] = set() self._walker_state: Optional[Any] = None # Set of errors we expect git-fsck to raise at the end: self._expected_fsck_errors = set() with tempfile.TemporaryDirectory(prefix="swh-vault-gitbare-") as workdir: # Initialize a Git directory self.workdir = workdir self.gitdir = os.path.join(workdir, "clone.git") os.mkdir(self.gitdir) self.init_git() # Add the root object to the stack of objects to visit self.push_subgraph(self.obj_type.split("_")[0], self.obj_id) # Load and write all the objects to disk self.load_objects() # Write the root object as a ref (this step is skipped if it's a snapshot) # This must be done before repacking; git-repack ignores orphan objects. self.write_refs() self.repack() self.write_archive() def init_git(self) -> None: subprocess.run(["git", "-C", self.gitdir, "init", "--bare"], check=True) # Create all possible dirs ahead of time, so we don't have to check for # existence every time. for byte in range(256): os.mkdir(os.path.join(self.gitdir, "objects", f"{byte:02x}")) def repack(self) -> None: if self.use_fsck: self.git_fsck() # Add objects we wrote in a pack subprocess.run(["git", "-C", self.gitdir, "repack"], check=True) # Remove their non-packed originals subprocess.run(["git", "-C", self.gitdir, "prune-packed"], check=True) def git_fsck(self) -> None: proc = subprocess.run( ["git", "-C", self.gitdir, "fsck"], stdout=subprocess.PIPE, stderr=subprocess.STDOUT, env={"LANG": "C.utf8"}, ) # Split on newlines not followed by a space errors = re.split("\n(?! )", proc.stdout.decode()) errors = [ error for error in errors if error and not error.startswith("warning ") ] unexpected_errors = set(errors) - self._expected_fsck_errors if unexpected_errors: raise Exception( "\n".join( ["Unexpected errors from git-fsck:"] + sorted(unexpected_errors) ) ) def write_refs(self, snapshot=None): refs: Dict[bytes, bytes] # ref name -> target obj_type = self.obj_type.split("_")[0] if obj_type == "directory": # We need a synthetic revision pointing to the directory author = Person.from_fullname( b"swh-vault, git-bare cooker " ) dt = datetime.datetime.now(tz=datetime.timezone.utc) dt = dt.replace(microsecond=0) # not supported by git date = TimestampWithTimezone.from_datetime(dt) revision = Revision( author=author, committer=author, date=date, committer_date=date, message=b"Initial commit", type=RevisionType.GIT, directory=self.obj_id, synthetic=True, ) self.write_revision_node(revision.to_dict()) refs = {b"refs/heads/master": hash_to_bytehex(revision.id)} elif obj_type == "revision": refs = {b"refs/heads/master": hash_to_bytehex(self.obj_id)} elif obj_type == "snapshot": if snapshot is None: # refs were already written in a previous step return + branches = [] + for (branch_name, branch) in snapshot.branches.items(): + if branch is None: + logging.error( + "%s has dangling branch: %r", snapshot.swhid(), branch_name + ) + else: + branches.append((branch_name, branch)) refs = { branch_name: ( b"ref: " + branch.target if branch.target_type == TargetType.ALIAS else hash_to_bytehex(branch.target) ) - for (branch_name, branch) in snapshot.branches.items() + for (branch_name, branch) in branches } else: assert False, obj_type for (ref_name, ref_target) in refs.items(): path = os.path.join(self.gitdir.encode(), ref_name) os.makedirs(os.path.dirname(path), exist_ok=True) with open(path, "wb") as fd: fd.write(ref_target) def write_archive(self): with tarfile.TarFile(mode="w", fileobj=self.fileobj) as tf: tf.add(self.gitdir, arcname=f"{self.obj_swhid()}.git", recursive=True) def _obj_path(self, obj_id: Sha1Git): return os.path.join(self.gitdir, self._obj_relative_path(obj_id)) def _obj_relative_path(self, obj_id: Sha1Git): obj_id_hex = hash_to_hex(obj_id) directory = obj_id_hex[0:2] filename = obj_id_hex[2:] return os.path.join("objects", directory, filename) def object_exists(self, obj_id: Sha1Git) -> bool: return os.path.exists(self._obj_path(obj_id)) def write_object(self, obj_id: Sha1Git, obj: bytes) -> bool: """Writes a git object on disk. Returns whether it was already written.""" # Git requires objects to be zlib-compressed; but repacking decompresses and # removes them, so we don't need to compress them too much. data = zlib.compress(obj, level=1) with open(self._obj_path(obj_id), "wb") as fd: fd.write(data) return True def push_subgraph(self, obj_type, obj_id) -> None: if obj_type == "revision": self.push_revision_subgraph(obj_id) elif obj_type == "directory": self._push(self._dir_stack, [obj_id]) elif obj_type == "snapshot": self.push_snapshot_subgraph(obj_id) else: raise NotImplementedError( f"GitBareCooker.queue_subgraph({obj_type!r}, ...)" ) def load_objects(self) -> None: while self._rel_stack or self._rev_stack or self._dir_stack or self._cnt_stack: release_ids = self._pop(self._rel_stack, RELEASE_BATCH_SIZE) self.push_releases_subgraphs(release_ids) revision_ids = self._pop(self._rev_stack, REVISION_BATCH_SIZE) self.load_revisions(revision_ids) directory_ids = self._pop(self._dir_stack, DIRECTORY_BATCH_SIZE) self.load_directories(directory_ids) content_ids = self._pop(self._cnt_stack, CONTENT_BATCH_SIZE) self.load_contents(content_ids) def push_revision_subgraph(self, obj_id: Sha1Git) -> None: """Fetches a revision and all its children, and writes them to disk""" loaded_from_graph = False if self.graph: from swh.graph.client import GraphArgumentException # First, try to cook using swh-graph, as it is more efficient than # swh-storage for querying the history obj_swhid = identifiers.CoreSWHID( object_type=identifiers.ObjectType.REVISION, object_id=obj_id, ) try: revision_ids = ( swhid.object_id for swhid in map( identifiers.CoreSWHID.from_string, self.graph.visit_nodes(str(obj_swhid), edges="rev:rev"), ) ) self._push(self._rev_stack, revision_ids) except GraphArgumentException as e: logger.info( "Revision %s not found in swh-graph, falling back to fetching " "history using swh-storage. %s", hash_to_hex(obj_id), e.args[0], ) else: loaded_from_graph = True if not loaded_from_graph: # If swh-graph is not available, or the revision is not yet in # swh-graph, fall back to self.storage.revision_log. # self.storage.revision_log also gives us the full revisions, # so we load them right now instead of just pushing them on the stack. walker = DFSRevisionsWalker(self.storage, obj_id, state=self._walker_state) for revision in walker: self.write_revision_node(revision) self._push(self._dir_stack, [revision["directory"]]) # Save the state, so the next call to the walker won't return the same # revisions self._walker_state = walker.export_state() def push_snapshot_subgraph(self, obj_id: Sha1Git) -> None: """Fetches a snapshot and all its children, and writes them to disk""" loaded_from_graph = False if self.graph: revision_ids = [] release_ids = [] from swh.graph.client import GraphArgumentException # First, try to cook using swh-graph, as it is more efficient than # swh-storage for querying the history obj_swhid = identifiers.CoreSWHID( object_type=identifiers.ObjectType.SNAPSHOT, object_id=obj_id, ) try: swhids = map( identifiers.CoreSWHID.from_string, self.graph.visit_nodes(str(obj_swhid), edges="snp:*,rel:*,rev:rev"), ) for swhid in swhids: if swhid.object_type == identifiers.ObjectType.REVISION: revision_ids.append(swhid.object_id) elif swhid.object_type == identifiers.ObjectType.RELEASE: release_ids.append(swhid.object_id) elif swhid.object_type == identifiers.ObjectType.SNAPSHOT: assert ( swhid.object_id == obj_id ), f"Snapshot {obj_id.hex()} references a different snapshot" else: raise NotImplementedError( f"{swhid.object_type} objects in snapshot subgraphs." ) except GraphArgumentException as e: logger.info( "Snapshot %s not found in swh-graph, falling back to fetching " "history for each branch. %s", hash_to_hex(obj_id), e.args[0], ) else: self._push(self._rev_stack, revision_ids) self._push(self._rel_stack, release_ids) loaded_from_graph = True # TODO: when self.graph is available and supports edge labels, use it # directly to get branch names. snapshot = snapshot_get_all_branches(self.storage, obj_id) assert snapshot, "Unknown snapshot" # should have been caught by check_exists() for branch in snapshot.branches.values(): if not loaded_from_graph: if branch.target_type == TargetType.REVISION: self.push_revision_subgraph(branch.target) elif branch.target_type == TargetType.RELEASE: self.push_releases_subgraphs([branch.target]) elif branch.target_type == TargetType.ALIAS: # Nothing to do, this for loop also iterates on the target branch # (if it exists) pass else: raise NotImplementedError(f"{branch.target_type} branches") self.write_refs(snapshot=snapshot) def load_revisions(self, obj_ids: List[Sha1Git]) -> None: """Given a list of revision ids, loads these revisions and their directories; but not their parent revisions.""" ret: List[Optional[Revision]] = self.storage.revision_get(obj_ids) revisions: List[Revision] = list(filter(None, ret)) if len(ret) != len(revisions): logger.error("Missing revision(s), ignoring them.") for revision in revisions: self.write_revision_node(revision.to_dict()) self._push(self._dir_stack, (rev.directory for rev in revisions)) def write_revision_node(self, revision: Dict[str, Any]) -> bool: """Writes a revision object to disk""" git_object = identifiers.revision_git_object(revision) return self.write_object(revision["id"], git_object) def push_releases_subgraphs(self, obj_ids: List[Sha1Git]) -> None: """Given a list of release ids, loads these releases and adds their target to the list of objects to visit""" ret = self.storage.release_get(obj_ids) releases = list(filter(None, ret)) if len(ret) != len(releases): logger.error("Missing release(s), ignoring them.") revision_ids: List[Sha1Git] = [] for release in releases: self.write_release_node(release.to_dict()) if release.target_type == ObjectType.REVISION: assert release.target, "{release.swhid(}) has no target" self.push_revision_subgraph(release.target) else: raise NotImplementedError(f"{release.target_type} release targets") self._push(self._rev_stack, revision_ids) def write_release_node(self, release: Dict[str, Any]) -> bool: """Writes a release object to disk""" git_object = identifiers.release_git_object(release) return self.write_object(release["id"], git_object) def load_directories(self, obj_ids: List[Sha1Git]) -> None: for obj_id in obj_ids: self.load_directory(obj_id) def load_directory(self, obj_id: Sha1Git) -> None: # Load the directory entries_it: Optional[Iterable[DirectoryEntry]] = stream_results_optional( self.storage.directory_get_entries, obj_id ) if entries_it is None: logger.error("Missing swh:1:dir:%s, ignoring.", hash_to_hex(obj_id)) return entries = [entry.to_dict() for entry in entries_it] directory = {"id": obj_id, "entries": entries} git_object = identifiers.directory_git_object(directory) self.write_object(obj_id, git_object) # Add children to the stack entry_loaders: Dict[str, List[Sha1Git]] = { "file": self._cnt_stack, "dir": self._dir_stack, "rev": self._rev_stack, } for entry in directory["entries"]: stack = entry_loaders[entry["type"]] self._push(stack, [entry["target"]]) def load_contents(self, obj_ids: List[Sha1Git]) -> None: # TODO: add support of filtered objects, somehow? # It's tricky, because, by definition, we can't write a git object with # the expected hash, so git-fsck *will* choke on it. contents = self.storage.content_get(obj_ids, "sha1_git") visible_contents = [] for (obj_id, content) in zip(obj_ids, contents): if content is None: # FIXME: this may also happen for missing content self.write_content(obj_id, SKIPPED_MESSAGE) self._expect_mismatched_object_error(obj_id) elif content.status == "visible": visible_contents.append(content) elif content.status == "hidden": self.write_content(obj_id, HIDDEN_MESSAGE) self._expect_mismatched_object_error(obj_id) else: assert False, ( f"unexpected status {content.status!r} " f"for content {hash_to_hex(content.sha1_git)}" ) contents_and_data: Iterator[Tuple[Content, Optional[bytes]]] if self.objstorage is None: contents_and_data = ( (content, self.storage.content_get_data(content.sha1)) for content in visible_contents ) else: contents_and_data = zip( visible_contents, self.objstorage.get_batch(c.sha1 for c in visible_contents), ) for (content, datum) in contents_and_data: if datum is None: logger.error( "{content.swhid()} is visible, but is missing data. Skipping." ) continue self.write_content(content.sha1_git, datum) def write_content(self, obj_id: Sha1Git, content: bytes) -> None: header = identifiers.git_object_header("blob", len(content)) self.write_object(obj_id, header + content) def _expect_mismatched_object_error(self, obj_id): obj_id_hex = hash_to_hex(obj_id) obj_path = self._obj_relative_path(obj_id) # For Git < 2.21: self._expected_fsck_errors.add( f"error: sha1 mismatch for ./{obj_path} (expected {obj_id_hex})" ) # For Git >= 2.21: self._expected_fsck_errors.add( f"error: hash mismatch for ./{obj_path} (expected {obj_id_hex})" ) self._expected_fsck_errors.add( f"error: {obj_id_hex}: object corrupt or missing: ./{obj_path}" ) self._expected_fsck_errors.add(f"missing blob {obj_id_hex}")