diff --git a/requirements-swh.txt b/requirements-swh.txt index 77c3d83..4177d43 100644 --- a/requirements-swh.txt +++ b/requirements-swh.txt @@ -1,4 +1,4 @@ swh.core >= 0.0.75 -swh.model >= 0.0.18 +swh.model >= 0.0.54 swh.scheduler swh.storage >= 0.0.163 diff --git a/swh/loader/core/converters.py b/swh/loader/core/converters.py index cc1c30c..4a51ffb 100644 --- a/swh/loader/core/converters.py +++ b/swh/loader/core/converters.py @@ -1,46 +1,76 @@ -# Copyright (C) 2015-2017 The Software Heritage developers +# Copyright (C) 2015-2020 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 """Convert objects to dictionaries suitable for swh.storage""" +import logging + +from typing import Dict, Iterable, List, Optional, Tuple + from swh.model.hashutil import hash_to_hex +from swh.model.model import BaseContent, Content, SkippedContent + + +logger = logging.getLogger(__name__) + + +def prepare_contents( + contents: Iterable[Dict], max_content_size: Optional[int] = None, + origin_url: Optional[str] = None) -> Tuple[ + List[Dict], List[Dict]]: + """Prepare contents for storage from a list of contents + Returns + tuple of content iterable, skipped content iterable -def content_for_storage(content, log=None, max_content_size=None, - origin_url=None): + """ + present_contents: List[Dict] = [] + skipped_contents: List[Dict] = [] + for _content in contents: + content = content_for_storage( + _content, max_content_size=max_content_size, origin_url=origin_url) + if isinstance(content, SkippedContent): + skipped_contents.append(content.to_dict()) + else: + present_contents.append(content.to_dict()) + return present_contents, skipped_contents + + +def content_for_storage( + content: Dict, max_content_size: Optional[int] = None, + origin_url: Optional[str] = None) -> BaseContent: """Prepare content to be ready for storage Note: - 'data' is returned only if max_content_size is not reached. Returns: content with added data (or reason for being missing) """ ret = content.copy() + ret.pop('perms', None) if max_content_size and ret['length'] > max_content_size: - if log: - log.info('Skipping content %s, too large (%s > %s)' % - (hash_to_hex(content['sha1_git']), - ret['length'], - max_content_size)) + logger.info('Skipping content %s, too large (%s > %s)' % + (hash_to_hex(content['sha1_git']), + ret['length'], + max_content_size)) ret.pop('data', None) ret.update({'status': 'absent', 'reason': 'Content too large', 'origin': origin_url}) - return ret + return SkippedContent.from_dict(ret) if 'data' not in ret: ret['data'] = open(ret['path'], 'rb').read() # Extra keys added by swh.model.from_disk, that are not accepted # by swh-storage - ret.pop('perms', None) ret.pop('path', None) ret['status'] = 'visible' - return ret + return Content.from_dict(ret) diff --git a/swh/loader/core/loader.py b/swh/loader/core/loader.py index 067342d..f1bc7c9 100644 --- a/swh/loader/core/loader.py +++ b/swh/loader/core/loader.py @@ -1,399 +1,398 @@ # Copyright (C) 2015-2020 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 hashlib import logging import os from abc import ABCMeta, abstractmethod from typing import Any, Dict, Iterable, Optional, Tuple, Union from swh.core import config from swh.storage import get_storage -from swh.loader.core.converters import content_for_storage +from swh.loader.core.converters import prepare_contents class BaseLoader(config.SWHConfig, metaclass=ABCMeta): """Mixin base class for loader. To use this class, you must: - inherit from this class - and implement the @abstractmethod methods: - :func:`prepare`: First step executed by the loader to prepare some state needed by the `func`:load method. - :func:`get_origin`: Retrieve the origin that is currently being loaded. - :func:`fetch_data`: Fetch the data is actually the method to implement to compute data to inject in swh (through the store_data method) - :func:`store_data`: Store data fetched. - :func:`visit_status`: Explicit status of the visit ('partial' or 'full') - :func:`load_status`: Explicit status of the loading, for use by the scheduler (eventful/uneventful/temporary failure/permanent failure). - :func:`cleanup`: Last step executed by the loader. The entry point for the resulting loader is :func:`load`. You can take a look at some example classes: - :class:`BaseSvnLoader` """ CONFIG_BASE_FILENAME = None # type: Optional[str] DEFAULT_CONFIG = { 'storage': ('dict', { 'cls': 'remote', 'args': { 'url': 'http://localhost:5002/', } }), 'max_content_size': ('int', 100 * 1024 * 1024), 'save_data': ('bool', False), 'save_data_path': ('str', ''), } # type: Dict[str, Tuple[str, Any]] ADDITIONAL_CONFIG = {} # type: Dict[str, Tuple[str, Any]] def __init__(self, logging_class: Optional[str] = None, config: Dict[str, Any] = {}): if config: self.config = config else: self.config = self.parse_config_file( additional_configs=[self.ADDITIONAL_CONFIG]) self.storage = get_storage(**self.config['storage']) if logging_class is None: logging_class = '%s.%s' % (self.__class__.__module__, self.__class__.__name__) self.log = logging.getLogger(logging_class) _log = logging.getLogger('requests.packages.urllib3.connectionpool') _log.setLevel(logging.WARN) self.max_content_size = self.config['max_content_size'] # possibly overridden in self.prepare method self.visit_date: Optional[Union[str, datetime.datetime]] = None self.origin: Dict[str, Any] = {} if not hasattr(self, 'visit_type'): self.visit_type: Optional[str] = None self.origin_metadata: Dict[str, Any] = {} # Make sure the config is sane save_data = self.config.get('save_data') if save_data: path = self.config['save_data_path'] os.stat(path) if not os.access(path, os.R_OK | os.W_OK): raise PermissionError("Permission denied: %r" % path) def save_data(self) -> None: """Save the data associated to the current load""" raise NotImplementedError def get_save_data_path(self) -> str: """The path to which we archive the loader's raw data""" if not hasattr(self, '__save_data_path'): year = str(self.visit_date.year) # type: ignore url = self.origin['url'].encode('utf-8') origin_url_hash = hashlib.sha1(url).hexdigest() path = '%s/sha1:%s/%s/%s' % ( self.config['save_data_path'], origin_url_hash[0:2], origin_url_hash, year, ) os.makedirs(path, exist_ok=True) self.__save_data_path = path return self.__save_data_path def flush(self) -> None: """Flush any potential dangling data not sent to swh-storage. Bypass the maybe_load_* methods which awaits threshold reached signal. We actually want to store those as we are done loading. """ if hasattr(self.storage, 'flush'): self.storage.flush() @abstractmethod def cleanup(self) -> None: """Last step executed by the loader. """ pass @abstractmethod def prepare_origin_visit(self, *args, **kwargs) -> None: """First step executed by the loader to prepare origin and visit references. Set/update self.origin, and optionally self.origin_url, self.visit_date. """ pass def _store_origin_visit(self) -> None: """Store origin and visit references. Sets the self.origin_visit and self.visit references. """ origin = self.origin.copy() self.storage.origin_add_one(origin) if not self.visit_date: # now as default visit_date if not provided self.visit_date = datetime.datetime.now(tz=datetime.timezone.utc) self.origin_visit = self.storage.origin_visit_add( origin['url'], self.visit_date, self.visit_type) self.visit = self.origin_visit['visit'] @abstractmethod def prepare(self, *args, **kwargs) -> None: """Second step executed by the loader to prepare some state needed by the loader. """ pass def get_origin(self) -> Dict[str, Any]: """Get the origin that is currently being loaded. self.origin should be set in :func:`prepare_origin` Returns: dict: an origin ready to be sent to storage by :func:`origin_add_one`. """ return self.origin @abstractmethod def fetch_data(self) -> bool: """Fetch the data from the source the loader is currently loading (ex: git/hg/svn/... repository). Returns: a value that is interpreted as a boolean. If True, fetch_data needs to be called again to complete loading. """ pass @abstractmethod def store_data(self): """Store fetched data in the database. Should call the :func:`maybe_load_xyz` methods, which handle the bundles sent to storage, rather than send directly. """ pass def store_metadata(self) -> None: """Store fetched metadata in the database. For more information, see implementation in :class:`DepositLoader`. """ pass def load_status(self) -> Dict[str, str]: """Detailed loading status. Defaults to logging an eventful load. Returns: a dictionary that is eventually passed back as the task's result to the scheduler, allowing tuning of the task recurrence mechanism. """ return { 'status': 'eventful', } def post_load(self, success: bool = True) -> None: """Permit the loader to do some additional actions according to status after the loading is done. The flag success indicates the loading's status. Defaults to doing nothing. This is up to the implementer of this method to make sure this does not break. Args: success (bool): the success status of the loading """ pass def visit_status(self) -> str: """Detailed visit status. Defaults to logging a full visit. """ return 'full' def pre_cleanup(self) -> None: """As a first step, will try and check for dangling data to cleanup. This should do its best to avoid raising issues. """ pass def load(self, *args, **kwargs) -> Dict[str, str]: r"""Loading logic for the loader to follow: - 1. Call :meth:`prepare_origin_visit` to prepare the origin and visit we will associate loading data to - 2. Store the actual ``origin_visit`` to storage - 3. Call :meth:`prepare` to prepare any eventual state - 4. Call :meth:`get_origin` to get the origin we work with and store - while True: - 5. Call :meth:`fetch_data` to fetch the data to store - 6. Call :meth:`store_data` to store the data - 7. Call :meth:`cleanup` to clean up any eventual state put in place in :meth:`prepare` method. """ try: self.pre_cleanup() except Exception: msg = 'Cleaning up dangling data failed! Continue loading.' self.log.warning(msg) self.prepare_origin_visit(*args, **kwargs) self._store_origin_visit() try: self.prepare(*args, **kwargs) while True: more_data_to_fetch = self.fetch_data() self.store_data() if not more_data_to_fetch: break self.store_metadata() self.storage.origin_visit_update( self.origin['url'], self.visit, self.visit_status() ) self.post_load() except Exception: self.log.exception('Loading failure, updating to `partial` status', extra={ 'swh_task_args': args, 'swh_task_kwargs': kwargs, }) self.storage.origin_visit_update( self.origin['url'], self.visit, 'partial' ) self.post_load(success=False) return {'status': 'failed'} finally: self.flush() self.cleanup() return self.load_status() class DVCSLoader(BaseLoader): """This base class is a pattern for dvcs loaders (e.g. git, mercurial). Those loaders are able to load all the data in one go. For example, the loader defined in swh-loader-git :class:`BulkUpdater`. For other loaders (stateful one, (e.g :class:`SWHSvnLoader`), inherit directly from :class:`BaseLoader`. """ ADDITIONAL_CONFIG = {} # type: Dict[str, Tuple[str, Any]] def cleanup(self) -> None: """Clean up an eventual state installed for computations.""" pass def has_contents(self) -> bool: """Checks whether we need to load contents""" return True def get_contents(self) -> Iterable[Dict[str, Any]]: """Get the contents that need to be loaded""" raise NotImplementedError def has_directories(self) -> bool: """Checks whether we need to load directories""" return True def get_directories(self) -> Iterable[Dict[str, Any]]: """Get the directories that need to be loaded""" raise NotImplementedError def has_revisions(self) -> bool: """Checks whether we need to load revisions""" return True def get_revisions(self) -> Iterable[Dict[str, Any]]: """Get the revisions that need to be loaded""" raise NotImplementedError def has_releases(self) -> bool: """Checks whether we need to load releases""" return True def get_releases(self) -> Iterable[Dict[str, Any]]: """Get the releases that need to be loaded""" raise NotImplementedError def get_snapshot(self) -> Dict[str, Any]: """Get the snapshot that needs to be loaded""" raise NotImplementedError def eventful(self) -> bool: """Whether the load was eventful""" raise NotImplementedError def store_data(self) -> None: if self.config['save_data']: self.save_data() if self.has_contents(): - self.storage.content_add([ - content_for_storage( - c, max_content_size=self.max_content_size, - origin_url=self.origin['url']) - for c in self.get_contents() - ]) + contents, skipped_contents = prepare_contents( + self.get_contents(), max_content_size=self.max_content_size, + origin_url=self.origin['url']) + self.storage.skipped_content_add(skipped_contents) + self.storage.content_add(contents) if self.has_directories(): self.storage.directory_add(self.get_directories()) if self.has_revisions(): self.storage.revision_add(self.get_revisions()) if self.has_releases(): self.storage.release_add(self.get_releases()) self.flush() # to ensure the snapshot targets existing objects snapshot = self.get_snapshot() self.storage.snapshot_add([snapshot]) self.storage.origin_visit_update( self.origin['url'], self.visit, snapshot=snapshot['id']) self.flush() diff --git a/swh/loader/core/tests/test_converters.py b/swh/loader/core/tests/test_converters.py index ae7a04d..ffe9f9d 100644 --- a/swh/loader/core/tests/test_converters.py +++ b/swh/loader/core/tests/test_converters.py @@ -1,102 +1,101 @@ -# Copyright (C) 2015-2017 The Software Heritage developers +# Copyright (C) 2015-2020 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 os import tempfile import unittest -from unittest.mock import Mock from swh.loader.core import converters -from swh.model.from_disk import Content +from swh.model import from_disk +from swh.model.model import Content, SkippedContent def tmpfile_with_content(fromdir, contentfile): """Create a temporary file with content contentfile in directory fromdir. """ tmpfilepath = tempfile.mktemp( suffix='.swh', prefix='tmp-file-for-test', dir=fromdir) with open(tmpfilepath, 'wb') as f: f.write(contentfile) return tmpfilepath class TestContentForStorage(unittest.TestCase): maxDiff = None def setUp(self): super().setUpClass() self.tmpdir = tempfile.TemporaryDirectory( prefix='test-swh-loader-core.' ) def tearDown(self): self.tmpdir.cleanup() def test_content_for_storage_path(self): # given data = b'temp file for testing content storage conversion' tmpfile = tmpfile_with_content(self.tmpdir.name, data) - obj = Content.from_file(path=os.fsdecode(tmpfile), - save_path=True).get_data() + obj = from_disk.Content.from_file(path=os.fsdecode(tmpfile), + save_path=True).get_data() expected_content = obj.copy() expected_content['data'] = data expected_content['status'] = 'visible' del expected_content['path'] del expected_content['perms'] + expected_content = Content.from_dict(expected_content) # when content = converters.content_for_storage(obj) # then self.assertEqual(content, expected_content) def test_content_for_storage_data(self): # given data = b'temp file for testing content storage conversion' - obj = Content.from_bytes(data=data, mode=0o100644).get_data() + obj = from_disk.Content.from_bytes(data=data, mode=0o100644).get_data() expected_content = obj.copy() expected_content['status'] = 'visible' del expected_content['perms'] + expected_content = Content.from_dict(expected_content) # when content = converters.content_for_storage(obj) # then self.assertEqual(content, expected_content) def test_content_for_storage_too_long(self): # given data = b'temp file for testing content storage conversion' - obj = Content.from_bytes(data=data, mode=0o100644).get_data() - - log = Mock() + obj = from_disk.Content.from_bytes(data=data, mode=0o100644).get_data() + del obj['perms'] expected_content = obj.copy() expected_content.pop('data') expected_content['status'] = 'absent' expected_content['origin'] = 'http://example.org/' expected_content['reason'] = 'Content too large' + expected_content = SkippedContent.from_dict(expected_content) # when content = converters.content_for_storage( - obj, log, max_content_size=len(data) - 1, - origin_url=expected_content['origin'], + obj, max_content_size=len(data) - 1, + origin_url=expected_content.origin, ) # then self.assertEqual(content, expected_content) - self.assertTrue(log.info.called) - self.assertIn('Skipping content', log.info.call_args[0][0]) - self.assertIn('too large', log.info.call_args[0][0]) diff --git a/swh/loader/package/loader.py b/swh/loader/package/loader.py index b52811d..5ad6661 100644 --- a/swh/loader/package/loader.py +++ b/swh/loader/package/loader.py @@ -1,406 +1,411 @@ -# Copyright (C) 2019 The Software Heritage developers +# Copyright (C) 2019-2020 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 logging import tempfile import os from typing import ( Any, Dict, Generator, List, Mapping, Optional, Sequence, Tuple ) from swh.core.tarball import uncompress from swh.core.config import SWHConfig from swh.model.from_disk import Directory from swh.model.hashutil import hash_to_hex from swh.model.identifiers import ( revision_identifier, snapshot_identifier, identifier_to_bytes ) from swh.storage import get_storage from swh.storage.algos.snapshot import snapshot_get_all_branches -from swh.loader.core.converters import content_for_storage +from swh.loader.core.converters import prepare_contents from swh.loader.package.utils import download logger = logging.getLogger(__name__) # Not implemented yet: # - clean up disk routines from previous killed workers (when OOMkilled) # -> separation of concern would like this to be abstracted from the code # -> experience tells us it's complicated to do as such (T903, T964, T982, # etc...) # # - model: swh.model.merkle.from_disk should output swh.model.model.* objects # to avoid this layer's conversion routine call # -> Take this up within swh.model's current implementation class PackageLoader: # Origin visit type (str) set by the loader visit_type = '' def __init__(self, url): """Loader's constructor. This raises exception if the minimal required configuration is missing (cf. fn:`check` method). Args: url (str): Origin url to load data from """ # This expects to use the environment variable SWH_CONFIG_FILENAME self.config = SWHConfig.parse_config_file() self._check_configuration() self.storage = get_storage(**self.config['storage']) self.url = url self.visit_date = datetime.datetime.now(tz=datetime.timezone.utc) + self.max_content_size = self.config['max_content_size'] def _check_configuration(self): """Checks the minimal configuration required is set for the loader. If some required configuration is missing, exception detailing the issue is raised. """ if 'storage' not in self.config: raise ValueError( 'Misconfiguration, at least the storage key should be set') def get_versions(self) -> Sequence[str]: """Return the list of all published package versions. Returns: Sequence of published versions """ return [] def get_package_info(self, version: str) -> Generator[ Tuple[str, Mapping[str, Any]], None, None]: """Given a release version of a package, retrieve the associated package information for such version. Args: version: Package version Returns: (branch name, package metadata) """ yield from {} def build_revision( self, a_metadata: Dict, uncompressed_path: str) -> Dict: """Build the revision dict from the archive metadata (extrinsic artifact metadata) and the intrinsic metadata. Args: a_metadata: Artifact metadata uncompressed_path: Artifact uncompressed path on disk Returns: SWH data dict """ return {} def get_default_version(self) -> str: """Retrieve the latest release version if any. Returns: Latest version """ return '' def last_snapshot(self) -> Optional[Dict]: """Retrieve the last snapshot """ snapshot = None visit = self.storage.origin_visit_get_latest( self.url, require_snapshot=True) if visit: snapshot = snapshot_get_all_branches( self.storage, visit['snapshot']) return snapshot def known_artifacts(self, snapshot: Optional[Dict]) -> Dict: """Retrieve the known releases/artifact for the origin. Args snapshot: snapshot for the visit Returns: Dict of keys revision id (bytes), values a metadata Dict. """ if not snapshot or 'branches' not in snapshot: return {} # retrieve only revisions (e.g the alias we do not want here) revs = [rev['target'] for rev in snapshot['branches'].values() if rev and rev['target_type'] == 'revision'] known_revisions = self.storage.revision_get(revs) ret = {} for revision in known_revisions: if not revision: # revision_get can return None continue ret[revision['id']] = revision['metadata'] return ret def resolve_revision_from( self, known_artifacts: Dict, artifact_metadata: Dict) \ -> Optional[bytes]: """Resolve the revision from a snapshot and an artifact metadata dict. If the artifact has already been downloaded, this will return the existing revision targeting that uncompressed artifact directory. Otherwise, this returns None. Args: snapshot: Snapshot artifact_metadata: Information dict Returns: None or revision identifier """ return None def download_package(self, p_info: Mapping[str, Any], tmpdir: str) -> List[Tuple[str, Mapping]]: """Download artifacts for a specific package. All downloads happen in in the tmpdir folder. Default implementation expects the artifacts package info to be about one artifact per package. Note that most implementation have 1 artifact per package. But some implementation have multiple artifacts per package (debian), some have none, the package is the artifact (gnu). Args: artifacts_package_info: Information on the package artifacts to download (url, filename, etc...) tmpdir: Location to retrieve such artifacts Returns: List of (path, computed hashes) """ a_uri = p_info['url'] filename = p_info.get('filename') return [download(a_uri, dest=tmpdir, filename=filename)] def uncompress(self, dl_artifacts: List[Tuple[str, Mapping[str, Any]]], dest: str) -> str: """Uncompress the artifact(s) in the destination folder dest. Optionally, this could need to use the p_info dict for some more information (debian). """ uncompressed_path = os.path.join(dest, 'src') for a_path, _ in dl_artifacts: uncompress(a_path, dest=uncompressed_path) return uncompressed_path def load(self) -> Dict: """Load for a specific origin the associated contents. for each package version of the origin 1. Fetch the files for one package version By default, this can be implemented as a simple HTTP request. Loaders with more specific requirements can override this, e.g.: the PyPI loader checks the integrity of the downloaded files; the Debian loader has to download and check several files for one package version. 2. Extract the downloaded files By default, this would be a universal archive/tarball extraction. Loaders for specific formats can override this method (for instance, the Debian loader uses dpkg-source -x). 3. Convert the extracted directory to a set of Software Heritage objects Using swh.model.from_disk. 4. Extract the metadata from the unpacked directories This would only be applicable for "smart" loaders like npm (parsing the package.json), PyPI (parsing the PKG-INFO file) or Debian (parsing debian/changelog and debian/control). On "minimal-metadata" sources such as the GNU archive, the lister should provide the minimal set of metadata needed to populate the revision/release objects (authors, dates) as an argument to the task. 5. Generate the revision/release objects for the given version. From the data generated at steps 3 and 4. end for each 6. Generate and load the snapshot for the visit Using the revisions/releases collected at step 5., and the branch information from step 0., generate a snapshot and load it into the Software Heritage archive """ status_load = 'uneventful' # either: eventful, uneventful, failed status_visit = 'full' # either: partial, full tmp_revisions = {} # type: Dict[str, List] snapshot = None # Prepare origin and origin_visit origin = {'url': self.url} try: self.storage.origin_add_one(origin) visit_id = self.storage.origin_visit_add( origin=self.url, date=self.visit_date, type=self.visit_type)['visit'] except Exception as e: logger.error( 'Failed to create origin/origin_visit. Reason: %s', e) return {'status': 'failed'} try: last_snapshot = self.last_snapshot() logger.debug('last snapshot: %s', last_snapshot) known_artifacts = self.known_artifacts(last_snapshot) logger.debug('known artifacts: %s', known_artifacts) # Retrieve the default release version (the "latest" one) default_version = self.get_default_version() logger.debug('default version: %s', default_version) for version in self.get_versions(): # for each logger.debug('version: %s', version) tmp_revisions[version] = [] # `p_` stands for `package_` for branch_name, p_info in self.get_package_info(version): logger.debug('package_info: %s', p_info) revision_id = self.resolve_revision_from( known_artifacts, p_info['raw']) if revision_id is None: with tempfile.TemporaryDirectory() as tmpdir: try: dl_artifacts = self.download_package( p_info, tmpdir) except Exception: logger.exception('Unable to retrieve %s', p_info) status_visit = 'partial' continue uncompressed_path = self.uncompress( dl_artifacts, dest=tmpdir) logger.debug('uncompressed_path: %s', uncompressed_path) directory = Directory.from_disk( path=uncompressed_path.encode('utf-8'), data=True) # noqa # FIXME: Try not to load the full raw content in # memory objects = directory.collect() - contents = list( - objects.get('content', {}).values()) + contents, skipped_contents = prepare_contents( + objects.get('content', {}).values(), + max_content_size=self.max_content_size, + origin_url=origin['url']) + self.storage.skipped_content_add(skipped_contents) + logger.debug('Number of skipped contents: %s', + len(skipped_contents)) + self.storage.content_add(contents) logger.debug('Number of contents: %s', len(contents)) - self.storage.content_add( - [content_for_storage(x) for x in contents]) status_load = 'eventful' directories = list( objects.get('directory', {}).values()) logger.debug('Number of directories: %s', len(directories)) self.storage.directory_add(directories) # FIXME: This should be release. cf. D409 revision = self.build_revision( p_info['raw'], uncompressed_path) if not revision: # Some artifacts are missing intrinsic metadata # skipping those continue revision.update({ 'synthetic': True, 'directory': directory.hash, }) revision['metadata'].update({ 'original_artifact': [ hashes for _, hashes in dl_artifacts ], }) revision['id'] = revision_id = identifier_to_bytes( revision_identifier(revision)) logger.debug('Revision: %s', revision) self.storage.revision_add([revision]) tmp_revisions[version].append((branch_name, revision_id)) logger.debug('tmp_revisions: %s', tmp_revisions) # Build and load the snapshot branches = {} # type: Dict[bytes, Mapping[str, Any]] for version, branch_name_revisions in tmp_revisions.items(): if version == default_version and \ len(branch_name_revisions) == 1: # only 1 branch (no ambiguity), we can create an alias # branch 'HEAD' branch_name, _ = branch_name_revisions[0] # except for some corner case (deposit) if branch_name != 'HEAD': branches[b'HEAD'] = { 'target_type': 'alias', 'target': branch_name.encode('utf-8'), } for branch_name, target in branch_name_revisions: branches[branch_name.encode('utf-8')] = { 'target_type': 'revision', 'target': target, } snapshot = { 'branches': branches } logger.debug('snapshot: %s', snapshot) snapshot['id'] = identifier_to_bytes( snapshot_identifier(snapshot)) logger.debug('snapshot: %s', snapshot) self.storage.snapshot_add([snapshot]) if hasattr(self.storage, 'flush'): self.storage.flush() except Exception: logger.exception('Fail to load %s' % self.url) status_visit = 'partial' status_load = 'failed' finally: self.storage.origin_visit_update( origin=self.url, visit_id=visit_id, status=status_visit, snapshot=snapshot and snapshot['id']) result = { 'status': status_load, } # type: Dict[str, Any] if snapshot: result['snapshot_id'] = hash_to_hex(snapshot['id']) return result