diff --git a/swh/loader/git/loader.py b/swh/loader/git/loader.py index 67fd267..51dd375 100644 --- a/swh/loader/git/loader.py +++ b/swh/loader/git/loader.py @@ -1,308 +1,307 @@ # Copyright (C) 2015-2018 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 import datetime import dulwich.repo import os import shutil from dulwich.errors import ObjectFormatException, EmptyFileException from collections import defaultdict from swh.model import hashutil from swh.loader.core.loader import SWHStatelessLoader from . import converters, utils class GitLoader(SWHStatelessLoader): """Load a git repository from a directory. """ CONFIG_BASE_FILENAME = 'loader/git-loader' def __init__(self, config=None): super().__init__(logging_class='swh.loader.git.Loader', config=config) - def prepare_origin(self, *args, **kwargs): - self.origin_url, *_ = args + def prepare_origin_visit(self, origin_url, directory, visit_date): + self.origin_url = origin_url self.origin = converters.origin_url_to_origin(self.origin_url) - return super().prepare_origin(*args, **kwargs) + self.visit_date = visit_date def prepare(self, origin_url, directory, visit_date): self.repo = dulwich.repo.Repo(directory) - self.visit_date = visit_date def iter_objects(self): object_store = self.repo.object_store for pack in object_store.packs: objs = list(pack.index.iterentries()) objs.sort(key=lambda x: x[1]) for sha, offset, crc32 in objs: yield hashutil.hash_to_bytehex(sha) yield from object_store._iter_loose_objects() yield from object_store._iter_alternate_objects() def _check(self, obj): """Check the object's repository representation. If any errors in check exists, an ObjectFormatException is raised. Args: obj (object): Dulwich object read from the repository. """ obj.check() from dulwich.objects import Commit, Tag try: # For additional checks on dulwich objects with date # for now, only checks on *time if isinstance(obj, Commit): commit_time = obj._commit_time utils.check_date_time(commit_time) author_time = obj._author_time utils.check_date_time(author_time) elif isinstance(obj, Tag): tag_time = obj._tag_time utils.check_date_time(tag_time) except Exception as e: raise ObjectFormatException(e) def get_object(self, oid): """Given an object id, return the object if it is found and not malformed in some way. Args: oid (bytes): the object's identifier Returns: The object if found without malformation """ try: # some errors are raised when reading the object obj = self.repo[oid] # some we need to check ourselves self._check(obj) except KeyError: _id = oid.decode('utf-8') self.log.warn('object %s not found, skipping' % _id, extra={ 'swh_type': 'swh_loader_git_missing_object', 'swh_object_id': _id, 'origin_id': self.origin_id, }) return None except ObjectFormatException: _id = oid.decode('utf-8') self.log.warn('object %s malformed, skipping' % _id, extra={ 'swh_type': 'swh_loader_git_missing_object', 'swh_object_id': _id, 'origin_id': self.origin_id, }) return None except EmptyFileException: _id = oid.decode('utf-8') self.log.warn('object %s corrupted (empty file), skipping' % _id, extra={ 'swh_type': 'swh_loader_git_missing_object', 'swh_object_id': _id, 'origin_id': self.origin_id, }) else: return obj def fetch_data(self): """Fetch the data from the data source""" self.previous_snapshot = self.storage.snapshot_get_latest( self.origin_id ) type_to_ids = defaultdict(list) for oid in self.iter_objects(): obj = self.get_object(oid) if not obj: continue type_name = obj.type_name type_to_ids[type_name].append(oid) self.type_to_ids = type_to_ids def has_contents(self): """Checks whether we need to load contents""" return bool(self.type_to_ids[b'blob']) def get_content_ids(self): """Get the content identifiers from the git repository""" for oid in self.type_to_ids[b'blob']: yield converters.dulwich_blob_to_content_id(self.repo[oid]) def get_contents(self): """Get the contents that need to be loaded""" max_content_size = self.config['content_size_limit'] missing_contents = set(self.storage.content_missing( self.get_content_ids(), 'sha1_git')) for oid in missing_contents: yield converters.dulwich_blob_to_content( self.repo[hashutil.hash_to_bytehex(oid)], log=self.log, max_content_size=max_content_size, origin_id=self.origin_id) def has_directories(self): """Checks whether we need to load directories""" return bool(self.type_to_ids[b'tree']) def get_directory_ids(self): """Get the directory identifiers from the git repository""" return (hashutil.hash_to_bytes(id.decode()) for id in self.type_to_ids[b'tree']) def get_directories(self): """Get the directories that need to be loaded""" missing_dirs = set(self.storage.directory_missing( sorted(self.get_directory_ids()))) for oid in missing_dirs: yield converters.dulwich_tree_to_directory( self.repo[hashutil.hash_to_bytehex(oid)], log=self.log) def has_revisions(self): """Checks whether we need to load revisions""" return bool(self.type_to_ids[b'commit']) def get_revision_ids(self): """Get the revision identifiers from the git repository""" return (hashutil.hash_to_bytes(id.decode()) for id in self.type_to_ids[b'commit']) def get_revisions(self): """Get the revisions that need to be loaded""" missing_revs = set(self.storage.revision_missing( sorted(self.get_revision_ids()))) for oid in missing_revs: yield converters.dulwich_commit_to_revision( self.repo[hashutil.hash_to_bytehex(oid)], log=self.log) def has_releases(self): """Checks whether we need to load releases""" return bool(self.type_to_ids[b'tag']) def get_release_ids(self): """Get the release identifiers from the git repository""" return (hashutil.hash_to_bytes(id.decode()) for id in self.type_to_ids[b'tag']) def get_releases(self): """Get the releases that need to be loaded""" missing_rels = set(self.storage.release_missing( sorted(self.get_release_ids()))) for oid in missing_rels: yield converters.dulwich_tag_to_release( self.repo[hashutil.hash_to_bytehex(oid)], log=self.log) def get_snapshot(self): """Turn the list of branches into a snapshot to load""" branches = {} for ref, target in self.repo.refs.as_dict().items(): obj = self.get_object(target) if obj: branches[ref] = { 'target': hashutil.bytehex_to_hash(target), 'target_type': converters.DULWICH_TYPES[obj.type_name], } else: branches[ref] = None self.snapshot = converters.branches_to_snapshot(branches) return self.snapshot def get_fetch_history_result(self): """Return the data to store in fetch_history for the current loader""" return { 'contents': len(self.type_to_ids[b'blob']), 'directories': len(self.type_to_ids[b'tree']), 'revisions': len(self.type_to_ids[b'commit']), 'releases': len(self.type_to_ids[b'tag']), } def save_data(self): """We already have the data locally, no need to save it""" pass def load_status(self): """The load was eventful if the current occurrences are different to the ones we retrieved at the beginning of the run""" eventful = False if self.previous_snapshot: eventful = self.snapshot['id'] != self.previous_snapshot['id'] else: eventful = bool(self.snapshot['branches']) return {'status': ('eventful' if eventful else 'uneventful')} class GitLoaderFromArchive(GitLoader): """Load a git repository from an archive. """ def project_name_from_archive(self, archive_path): """Compute the project name from the archive's path. """ return os.path.basename(os.path.dirname(archive_path)) def prepare(self, origin_url, archive_path, visit_date): """1. Uncompress the archive in temporary location. 2. Prepare as the GitLoader does 3. Load as GitLoader does """ project_name = self.project_name_from_archive(archive_path) self.temp_dir, self.repo_path = utils.init_git_repo_from_archive( project_name, archive_path) self.log.info('Project %s - Uncompressing archive %s at %s' % ( origin_url, os.path.basename(archive_path), self.repo_path)) super().prepare(origin_url, self.repo_path, visit_date) def cleanup(self): """Cleanup the temporary location (if it exists). """ if self.temp_dir and os.path.exists(self.temp_dir): shutil.rmtree(self.temp_dir) self.log.info('Project %s - Done injecting %s' % ( self.origin_url, self.repo_path)) if __name__ == '__main__': import click import logging logging.basicConfig( level=logging.DEBUG, format='%(asctime)s %(process)d %(message)s' ) @click.command() @click.option('--origin-url', help='origin url') @click.option('--git-directory', help='Path to git repository to load') @click.option('--visit-date', default=None, help='Visit date') def main(origin_url, git_directory, visit_date): if not visit_date: visit_date = datetime.datetime.now(tz=datetime.timezone.utc) return GitLoader().load(origin_url, git_directory, visit_date) main() diff --git a/swh/loader/git/reader.py b/swh/loader/git/reader.py index aea8c7a..2491530 100644 --- a/swh/loader/git/reader.py +++ b/swh/loader/git/reader.py @@ -1,258 +1,258 @@ # Copyright (C) 2016-2018 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 from collections import defaultdict import logging import pprint import click from swh.core import utils from swh.model import hashutil from .updater import BulkUpdater, SWHRepoRepresentation from . import converters class SWHRepoFullRepresentation(SWHRepoRepresentation): """Overridden representation of a swh repository to permit to read completely the remote repository. """ def __init__(self, storage, origin_id, occurrences=None): self.storage = storage self._parents_cache = {} self._type_cache = {} self.heads = set() def determine_wants(self, refs): """Filter the remote references to figure out which ones Software Heritage needs. In this particular context, we want to know everything. """ if not refs: return [] for target in refs.values(): self.heads.add(target) return self.filter_unwanted_refs(refs).values() def find_remote_ref_types_in_swh(self, remote_refs): """Find the known swh remote. In that particular context, we know nothing. """ return {} class DummyGraphWalker(object): """Dummy graph walker which claims that the client doesn’t have any objects. """ def ack(self, sha): pass def next(self): pass def __next__(self): pass class BaseGitRemoteReader(BulkUpdater): CONFIG_BASE_FILENAME = 'loader/git-remote-reader' ADDITIONAL_CONFIG = { 'pack_size_bytes': ('int', 4 * 1024 * 1024 * 1024), 'pack_storage_base': ('str', ''), # don't want to store packs so empty 'next_task': ( 'dict', { 'queue': 'swh.storage.archiver.tasks.SWHArchiverToBackendTask', 'batch_size': 100, 'destination': 'azure' } ) } def __init__(self): super().__init__(SWHRepoFullRepresentation) self.next_task = self.config['next_task'] self.batch_size = self.next_task['batch_size'] self.task_destination = self.next_task['queue'] self.destination = self.next_task['destination'] def graph_walker(self): return DummyGraphWalker() - def prepare_origin(self, origin_url, base_url=None): + def prepare_origin_visit(self, origin_url, base_url=None): self.origin = converters.origin_url_to_origin(origin_url) self.origin_id = 0 def prepare(self, origin_url, base_url=None): """Only retrieve information about the origin, set everything else to empty. """ self.base_occurrences = [] self.base_origin_id = 0 def keep_object(self, obj): """Do we want to keep this object or not?""" raise NotImplementedError('Please implement keep_object') def get_id_and_data(self, obj): """Get the id, type and data of the given object""" raise NotImplementedError('Please implement get_id_and_data') def list_pack(self, pack_data, pack_size): """Override list_pack to only keep contents' sha1. Returns: id_to_type (dict): keys are sha1, values are their associated type type_to_ids (dict): keys are types, values are list of associated data (sha1 for blobs) """ self.data = {} id_to_type = {} type_to_ids = defaultdict(set) inflater = self.get_inflater() for obj in inflater: if not self.keep_object(obj): continue object_id, type, data = self.get_id_and_data(obj) id_to_type[object_id] = type type_to_ids[type].add(object_id) self.data[object_id] = data return id_to_type, type_to_ids def load(self, *args, **kwargs): """Override the loading part which simply reads the repository's contents' sha1. Returns: Returns the list of discovered sha1s for that origin. """ self.prepare(*args, **kwargs) self.fetch_data() class GitSha1RemoteReader(BaseGitRemoteReader): """Read sha1 git from a remote repository and dump only repository's content sha1 as list. """ def keep_object(self, obj): """Only keep blobs""" return obj.type_name == b'blob' def get_id_and_data(self, obj): """We want to store only object identifiers""" # compute the sha1 (obj.id is the sha1_git) data = obj.as_raw_string() hashes = hashutil.hash_data(data, {'sha1'}) oid = hashes['sha1'] return (oid, b'blob', oid) class GitSha1RemoteReaderAndSendToQueue(GitSha1RemoteReader): """Read sha1 git from a remote repository and dump only repository's content sha1 as list and send batch of those sha1s to a celery queue for consumption. """ def load(self, *args, **kwargs): """Retrieve the list of sha1s for a particular origin and send those sha1s as group of sha1s to a specific queue. """ super().load(*args, **kwargs) data = self.type_to_ids[b'blob'] from swh.scheduler.celery_backend.config import app try: # optional dependency from swh.storage.archiver import tasks # noqa except ImportError: pass from celery import group task_destination = app.tasks[self.task_destination] groups = [] for ids in utils.grouper(data, self.batch_size): sig_ids = task_destination.s(destination=self.destination, batch=list(ids)) groups.append(sig_ids) group(groups).delay() return data class GitCommitRemoteReader(BaseGitRemoteReader): def keep_object(self, obj): return obj.type_name == b'commit' def get_id_and_data(self, obj): return obj.id, b'commit', converters.dulwich_commit_to_revision(obj) def load(self, *args, **kwargs): super().load(*args, **kwargs) return self.data @click.group() @click.option('--origin-url', help='Origin url') @click.pass_context def main(ctx, origin_url): logging.basicConfig( level=logging.DEBUG, format='%(asctime)s %(process)d %(message)s' ) ctx.obj['origin_url'] = origin_url @main.command() @click.option('--send/--nosend', default=False, help='Origin\'s url') @click.pass_context def blobs(ctx, send): origin_url = ctx.obj['origin_url'] if send: loader = GitSha1RemoteReaderAndSendToQueue() ids = loader.load(origin_url) print('%s sha1s were sent to queue' % len(ids)) return loader = GitSha1RemoteReader() ids = loader.load(origin_url) if ids: for oid in ids: print(hashutil.hash_to_hex(oid)) @main.command() @click.option('--ids-only', is_flag=True, help='print ids only') @click.pass_context def commits(ctx, ids_only): origin_url = ctx.obj['origin_url'] reader = GitCommitRemoteReader() commits = reader.load(origin_url) for commit_id, commit in commits.items(): if ids_only: print(commit_id.decode()) else: pprint.pprint(commit) if __name__ == '__main__': main(obj={}) diff --git a/swh/loader/git/updater.py b/swh/loader/git/updater.py index b307aef..83c48d1 100644 --- a/swh/loader/git/updater.py +++ b/swh/loader/git/updater.py @@ -1,473 +1,462 @@ # Copyright (C) 2016-2018 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 import datetime import dulwich.client import logging import os import pickle import sys from collections import defaultdict from io import BytesIO from dulwich.object_store import ObjectStoreGraphWalker from dulwich.pack import PackData, PackInflater from swh.model import hashutil from swh.loader.core.loader import SWHStatelessLoader from . import converters class SWHRepoRepresentation: """Repository representation for a Software Heritage origin.""" def __init__(self, storage, origin_id, base_snapshot=None): self.storage = storage self._parents_cache = {} self._type_cache = {} if origin_id: self.heads = set(self._cache_heads(origin_id, base_snapshot)) else: self.heads = set() def _fill_parents_cache(self, commits): """When querying for a commit's parents, we fill the cache to a depth of 1000 commits.""" root_revs = self._encode_for_storage(commits) for rev, parents in self.storage.revision_shortlog(root_revs, 1000): rev_id = hashutil.hash_to_bytehex(rev) if rev_id not in self._parents_cache: self._parents_cache[rev_id] = [ hashutil.hash_to_bytehex(parent) for parent in parents ] for rev in commits: if rev not in self._parents_cache: self._parents_cache[rev] = [] def _cache_heads(self, origin_id, base_snapshot): """Return all the known head commits for `origin_id`""" if not base_snapshot: base_snapshot = self.storage.snapshot_get_latest(origin_id) if base_snapshot: return self._decode_from_storage( target['target'] for target in base_snapshot['branches'].values() ) else: return [] def get_parents(self, commit): """get the parent commits for `commit`""" # Prime the parents cache if not self._parents_cache and self.heads: self._fill_parents_cache(self.heads) if commit not in self._parents_cache: self._fill_parents_cache([commit]) return self._parents_cache[commit] def get_heads(self): return self.heads @staticmethod def _encode_for_storage(objects): return [hashutil.bytehex_to_hash(object) for object in objects] @staticmethod def _decode_from_storage(objects): return set(hashutil.hash_to_bytehex(object) for object in objects) def graph_walker(self): return ObjectStoreGraphWalker(self.get_heads(), self.get_parents) @staticmethod def filter_unwanted_refs(refs): """Filter the unwanted references from refs""" ret = {} for ref, val in refs.items(): if ref.endswith(b'^{}'): # Peeled refs make the git protocol explode continue elif ref.startswith(b'refs/pull/') and ref.endswith(b'/merge'): # We filter-out auto-merged GitHub pull requests continue else: ret[ref] = val return ret def determine_wants(self, refs): """Filter the remote references to figure out which ones Software Heritage needs. """ if not refs: return [] # Find what objects Software Heritage has refs = self.find_remote_ref_types_in_swh(refs) # Cache the objects found in swh as existing heads for target in refs.values(): if target['target_type'] is not None: self.heads.add(target['target']) ret = set() for target in self.filter_unwanted_refs(refs).values(): if target['target_type'] is None: # The target doesn't exist in Software Heritage, let's retrieve # it. ret.add(target['target']) return list(ret) def get_stored_objects(self, objects): return self.storage.object_find_by_sha1_git( self._encode_for_storage(objects)) def find_remote_ref_types_in_swh(self, remote_refs): """Parse the remote refs information and list the objects that exist in Software Heritage. """ all_objs = set(remote_refs.values()) - set(self._type_cache) type_by_id = {} for id, objs in self.get_stored_objects(all_objs).items(): id = hashutil.hash_to_bytehex(id) if objs: type_by_id[id] = objs[0]['type'] self._type_cache.update(type_by_id) ret = {} for ref, id in remote_refs.items(): ret[ref] = { 'target': id, 'target_type': self._type_cache.get(id), } return ret class BulkUpdater(SWHStatelessLoader): """A bulk loader for a git repository""" CONFIG_BASE_FILENAME = 'loader/git-updater' ADDITIONAL_CONFIG = { 'pack_size_bytes': ('int', 4 * 1024 * 1024 * 1024), } def __init__(self, repo_representation=SWHRepoRepresentation, config=None): """Initialize the bulk updater. Args: repo_representation: swh's repository representation which is in charge of filtering between known and remote data. """ super().__init__(logging_class='swh.loader.git.BulkLoader', config=config) self.repo_representation = repo_representation def fetch_pack_from_origin(self, origin_url, base_origin_id, base_snapshot, do_activity): """Fetch a pack from the origin""" pack_buffer = BytesIO() base_repo = self.repo_representation(self.storage, base_origin_id, base_snapshot) client, path = dulwich.client.get_transport_and_path(origin_url, thin_packs=False) size_limit = self.config['pack_size_bytes'] def do_pack(data, pack_buffer=pack_buffer, limit=size_limit, origin_url=origin_url): cur_size = pack_buffer.tell() would_write = len(data) if cur_size + would_write > limit: raise IOError('Pack file too big for repository %s, ' 'limit is %d bytes, current size is %d, ' 'would write %d' % (origin_url, limit, cur_size, would_write)) pack_buffer.write(data) remote_refs = client.fetch_pack(path, base_repo.determine_wants, base_repo.graph_walker(), do_pack, progress=do_activity) if remote_refs: local_refs = base_repo.find_remote_ref_types_in_swh(remote_refs) else: local_refs = remote_refs = {} pack_buffer.flush() pack_size = pack_buffer.tell() pack_buffer.seek(0) return { 'remote_refs': base_repo.filter_unwanted_refs(remote_refs), 'local_refs': local_refs, 'pack_buffer': pack_buffer, 'pack_size': pack_size, } def list_pack(self, pack_data, pack_size): id_to_type = {} type_to_ids = defaultdict(set) inflater = self.get_inflater() for obj in inflater: type, id = obj.type_name, obj.id id_to_type[id] = type type_to_ids[type].add(id) return id_to_type, type_to_ids - def prepare_origin(self, origin_url, base_url=None): - origin = converters.origin_url_to_origin(origin_url) - base_origin = converters.origin_url_to_origin(base_url) - - prev_snapshot = {} - base_origin_id = origin_id = None - - db_origin = self.storage.origin_get(origin) - if db_origin: - base_origin_id = origin_id = db_origin['id'] + def prepare_origin_visit(self, origin_url, base_url=None): + self.visit_date = datetime.datetime.now(tz=datetime.timezone.utc) + self.origin = converters.origin_url_to_origin(origin_url) - if origin_id: - prev_snapshot = self.storage.snapshot_get_latest(origin_id) + def prepare(self, origin_url, base_url=None): + base_origin_id = origin_id = self.origin_id + prev_snapshot = self.storage.snapshot_get_latest(origin_id) if base_url and not prev_snapshot: + base_origin = converters.origin_url_to_origin(base_url) base_origin = self.storage.origin_get(base_origin) if base_origin: base_origin_id = base_origin['id'] prev_snapshot = self.storage.snapshot_get_latest( base_origin_id ) self.base_snapshot = prev_snapshot self.base_origin_id = base_origin_id - self.origin = origin - - return super().prepare_origin(origin_url, base_url=None) - - def prepare(self, origin_url, base_url=None): - self.visit_date = datetime.datetime.now(tz=datetime.timezone.utc) def fetch_data(self): def do_progress(msg): sys.stderr.buffer.write(msg) sys.stderr.flush() fetch_info = self.fetch_pack_from_origin( self.origin['url'], self.base_origin_id, self.base_snapshot, do_progress) self.pack_buffer = fetch_info['pack_buffer'] self.pack_size = fetch_info['pack_size'] self.remote_refs = fetch_info['remote_refs'] self.local_refs = fetch_info['local_refs'] origin_url = self.origin['url'] self.log.info('Listed %d refs for repo %s' % ( len(self.remote_refs), origin_url), extra={ 'swh_type': 'git_repo_list_refs', 'swh_repo': origin_url, 'swh_num_refs': len(self.remote_refs), }) # We want to load the repository, walk all the objects id_to_type, type_to_ids = self.list_pack(self.pack_buffer, self.pack_size) self.id_to_type = id_to_type self.type_to_ids = type_to_ids def save_data(self): """Store a pack for archival""" write_size = 8192 pack_dir = self.get_save_data_path() pack_name = "%s.pack" % self.visit_date.isoformat() refs_name = "%s.refs" % self.visit_date.isoformat() with open(os.path.join(pack_dir, pack_name), 'xb') as f: while True: r = self.pack_buffer.read(write_size) if not r: break f.write(r) self.pack_buffer.seek(0) with open(os.path.join(pack_dir, refs_name), 'xb') as f: pickle.dump(self.remote_refs, f) def get_inflater(self): """Reset the pack buffer and get an object inflater from it""" self.pack_buffer.seek(0) return PackInflater.for_pack_data( PackData.from_file(self.pack_buffer, self.pack_size)) def has_contents(self): return bool(self.type_to_ids[b'blob']) def get_content_ids(self): """Get the content identifiers from the git repository""" for raw_obj in self.get_inflater(): if raw_obj.type_name != b'blob': continue yield converters.dulwich_blob_to_content_id(raw_obj) def get_contents(self): """Format the blobs from the git repository as swh contents""" max_content_size = self.config['content_size_limit'] missing_contents = set(self.storage.content_missing( self.get_content_ids(), 'sha1_git')) for raw_obj in self.get_inflater(): if raw_obj.type_name != b'blob': continue if raw_obj.sha().digest() not in missing_contents: continue yield converters.dulwich_blob_to_content( raw_obj, log=self.log, max_content_size=max_content_size, origin_id=self.origin_id) def has_directories(self): return bool(self.type_to_ids[b'tree']) def get_directory_ids(self): """Get the directory identifiers from the git repository""" return (hashutil.hash_to_bytes(id.decode()) for id in self.type_to_ids[b'tree']) def get_directories(self): """Format the trees as swh directories""" missing_dirs = set(self.storage.directory_missing( sorted(self.get_directory_ids()))) for raw_obj in self.get_inflater(): if raw_obj.type_name != b'tree': continue if raw_obj.sha().digest() not in missing_dirs: continue yield converters.dulwich_tree_to_directory(raw_obj, log=self.log) def has_revisions(self): return bool(self.type_to_ids[b'commit']) def get_revision_ids(self): """Get the revision identifiers from the git repository""" return (hashutil.hash_to_bytes(id.decode()) for id in self.type_to_ids[b'commit']) def get_revisions(self): """Format commits as swh revisions""" missing_revs = set(self.storage.revision_missing( sorted(self.get_revision_ids()))) for raw_obj in self.get_inflater(): if raw_obj.type_name != b'commit': continue if raw_obj.sha().digest() not in missing_revs: continue yield converters.dulwich_commit_to_revision(raw_obj, log=self.log) def has_releases(self): return bool(self.type_to_ids[b'tag']) def get_release_ids(self): """Get the release identifiers from the git repository""" return (hashutil.hash_to_bytes(id.decode()) for id in self.type_to_ids[b'tag']) def get_releases(self): """Retrieve all the release objects from the git repository""" missing_rels = set(self.storage.release_missing( sorted(self.get_release_ids()))) for raw_obj in self.get_inflater(): if raw_obj.type_name != b'tag': continue if raw_obj.sha().digest() not in missing_rels: continue yield converters.dulwich_tag_to_release(raw_obj, log=self.log) def get_snapshot(self): branches = {} for ref in self.remote_refs: ret_ref = self.local_refs[ref].copy() if not ret_ref['target_type']: target_type = self.id_to_type[ret_ref['target']] ret_ref['target_type'] = converters.DULWICH_TYPES[target_type] ret_ref['target'] = hashutil.bytehex_to_hash(ret_ref['target']) branches[ref] = ret_ref self.snapshot = converters.branches_to_snapshot(branches) return self.snapshot def get_fetch_history_result(self): return { 'contents': len(self.type_to_ids[b'blob']), 'directories': len(self.type_to_ids[b'tree']), 'revisions': len(self.type_to_ids[b'commit']), 'releases': len(self.type_to_ids[b'tag']), } def load_status(self): """The load was eventful if the current snapshot is different to the one we retrieved at the beginning of the run""" eventful = False if self.base_snapshot: eventful = self.snapshot['id'] != self.base_snapshot['id'] else: eventful = bool(self.snapshot['branches']) return {'status': ('eventful' if eventful else 'uneventful')} if __name__ == '__main__': import click logging.basicConfig( level=logging.DEBUG, format='%(asctime)s %(process)d %(message)s' ) @click.command() @click.option('--origin-url', help='Origin url') @click.option('--base-url', default=None, help='Optional Base url') def main(origin_url, base_url): return BulkUpdater().load(origin_url, base_url=base_url) main()