diff --git a/swh/loader/package/gnu.py b/swh/loader/package/gnu.py index 6cf6d16..c4eea96 100644 --- a/swh/loader/package/gnu.py +++ b/swh/loader/package/gnu.py @@ -1,168 +1,170 @@ # Copyright (C) 2019 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 re from os import path from typing import Generator, Dict, Tuple, Sequence from swh.loader.package.loader import PackageLoader from swh.model.identifiers import normalize_timestamp # to recognize existing naming pattern extensions = [ 'zip', 'tar', 'gz', 'tgz', 'bz2', 'bzip2', 'lzma', 'lz', 'xz', 'Z', ] version_keywords = [ 'cygwin_me', 'w32', 'win32', 'nt', 'cygwin', 'mingw', 'latest', 'alpha', 'beta', 'release', 'stable', 'hppa', 'solaris', 'sunos', 'sun4u', 'sparc', 'sun', 'aix', 'ibm', 'rs6000', 'i386', 'i686', 'linux', 'redhat', 'linuxlibc', 'mips', 'powerpc', 'macos', 'apple', 'darwin', 'macosx', 'powermacintosh', 'unknown', 'netbsd', 'freebsd', 'sgi', 'irix', ] # Match a filename into components. # # We use Debian's release number heuristic: A release number starts # with a digit, and is followed by alphanumeric characters or any of # ., +, :, ~ and - # # We hardcode a list of possible extensions, as this release number # scheme would match them too... We match on any combination of those. # # Greedy matching is done right to left (we only match the extension # greedily with +, software_name and release_number are matched lazily # with +? and *?). pattern = r''' ^ (?: # We have a software name and a release number, separated with a # -, _ or dot. (?P.+?[-_.]) (?P(%(vkeywords)s|[0-9][0-9a-zA-Z_.+:~-]*?)+) | # We couldn't match a release number, put everything in the # software name. (?P.+?) ) (?P(?:\.(?:%(extensions)s))+) $ ''' % { 'extensions': '|'.join(extensions), 'vkeywords': '|'.join('%s[-]?' % k for k in version_keywords), } def get_version(url: str) -> str: """Extract branch name from tarball url Args: url (str): Tarball URL Returns: byte: Branch name Example: For url = https://ftp.gnu.org/gnu/8sync/8sync-0.2.0.tar.gz >>> get_version(url) '0.2.0' """ filename = path.split(url)[-1] m = re.match(pattern, filename, flags=re.VERBOSE | re.IGNORECASE) if m: d = m.groupdict() if d['software_name1'] and d['release_number']: return d['release_number'] if d['software_name2']: return d['software_name2'] return '' class GNULoader(PackageLoader): visit_type = 'gnu' SWH_PERSON = { 'name': b'Software Heritage', 'fullname': b'Software Heritage', 'email': b'robot@softwareheritage.org' } REVISION_MESSAGE = b'swh-loader-package: synthetic revision message' def __init__(self, package: str, package_url: str, tarballs: Sequence): """Loader constructor. For now, this is the lister's task output. Args: package: Package's name (unused) package_url: Origin url tarballs: List of dict with keys `date` (date) and `archive` (str) the url to retrieve one versioned archive """ super().__init__(url=package_url) self.tarballs = list(sorted(tarballs, key=lambda v: int(v['date']))) def get_versions(self) -> Sequence[str]: versions = [] for archive in self.tarballs: v = get_version(archive['archive']) if v: versions.append(v) return versions def get_default_release(self) -> str: # It's the most recent, so for this loader, it's the last one return get_version(self.tarballs[-1]['archive']) def get_artifacts(self, version: str) -> Generator[ Tuple[str, str, Dict], None, None]: for a_metadata in self.tarballs: url = a_metadata['archive'] filename = path.split(url)[-1] yield filename, url, a_metadata def build_revision( - self, a_metadata: Dict, a_uncompressed_path: str) -> Dict: - + self, a_metadata: Dict, a_uncompressed_path: str, + visit_date: str) -> Dict: normalized_date = normalize_timestamp(int(a_metadata['date'])) return { 'message': self.REVISION_MESSAGE, 'date': normalized_date, 'author': self.SWH_PERSON, 'committer': self.SWH_PERSON, 'committer_date': normalized_date, 'parents': [], 'metadata': { - 'package': { - 'date': a_metadata['date'], - 'archive': a_metadata['archive'], + 'intrinsic': {}, + 'extrinsic': { + 'provider': self.url, + 'when': visit_date, + 'raw': a_metadata, }, }, } diff --git a/swh/loader/package/loader.py b/swh/loader/package/loader.py index c6ca625..7686ab6 100644 --- a/swh/loader/package/loader.py +++ b/swh/loader/package/loader.py @@ -1,359 +1,353 @@ # Copyright (C) 2019 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 Generator, Dict, Tuple, Sequence, List, Optional from swh.core.tarball import uncompress from swh.core.config import SWHConfig from swh.model.from_disk import Directory 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.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 = '' - # Url providing the artifact information - provider_url = '' 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 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_artifacts(self, version: str) -> Generator[ Tuple[str, str, Dict], None, None]: """Given a release version of a package, retrieve the associated artifact information for such version. Args: version: Package version Returns: (artifact filename, artifact uri, raw artifact metadata) """ yield from {} def build_revision( self, a_metadata: Dict, a_uncompressed_path: str) -> Dict: """Build the revision dict Returns: SWH data dict """ return {} def get_default_release(self) -> str: """Retrieve the latest release version Returns: Latest version """ return '' def last_snapshot(self) -> Optional[Dict]: """Retrieve the last snapshot """ visit = self.storage.origin_visit_get_latest( self.url, require_snapshot=True) if visit: return snapshot_get_all_branches( self.storage, visit['snapshot']['id']) def known_artifacts(self, snapshot: 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 original_artifact = revision['metadata'].get('original_artifact') if original_artifact: ret[revision['id']] = original_artifact 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 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: Dict[str, List] = {} snapshot = None try: # Prepare origin and origin_visit origin = {'url': self.url} self.storage.origin_add([origin]) visit_date = datetime.datetime.now(tz=datetime.timezone.utc) visit_id = self.storage.origin_visit_add( origin=self.url, date=visit_date, type=self.visit_type)['visit'] 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 (the "latest" one) default_release = self.get_default_release() logger.debug('default release: %s', default_release) for version in self.get_versions(): # for each logger.debug('version: %s', version) tmp_revisions[version] = [] # `a_` stands for `artifact_` for a_filename, a_uri, a_metadata in self.get_artifacts( version): logger.debug('a_metadata: %s', a_metadata) revision_id = self.resolve_revision_from( known_artifacts, a_metadata) if revision_id is None: with tempfile.TemporaryDirectory() as tmpdir: try: # a_c_: archive_computed_ a_path, a_c_metadata = download( a_uri, dest=tmpdir) except Exception as e: logger.warning( 'Unable to retrieve %s. Reason: %s', a_uri, e) status_visit = 'partial' continue logger.debug('archive_path: %s', a_path) logger.debug('archive_computed_metadata: %s', a_c_metadata) uncompressed_path = os.path.join(tmpdir, 'src') uncompress(a_path, dest=uncompressed_path) 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 = objects['content'].values() logger.debug('Number of contents: %s', len(contents)) self.storage.content_add( map(content_for_storage, contents)) status_load = 'eventful' directories = objects['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( - a_metadata, uncompressed_path) + a_metadata, uncompressed_path, + visit_date.isoformat()) revision.update({ 'type': 'tar', 'synthetic': True, 'directory': directory.hash, }) revision['metadata'].update({ 'original_artifact': a_c_metadata, - 'extrinsic': { - 'provider': self.provider_url, - 'when': visit_date, - 'raw': a_metadata, - }, }) revision['id'] = revision_id = identifier_to_bytes( revision_identifier(revision)) logger.debug('Revision: %s', revision) self.storage.revision_add([revision]) tmp_revisions[version].append({ 'filename': a_filename, 'target': revision_id, }) # Build and load the snapshot branches = {} for version, v_branches in tmp_revisions.items(): if len(v_branches) == 1: branch_name = ('releases/%s' % version).encode('utf-8') if version == default_release: branches[b'HEAD'] = { 'target_type': 'alias', 'target': branch_name, } branches[branch_name] = { 'target_type': 'revision', 'target': v_branches[0]['target'], } else: for x in v_branches: branch_name = ('releases/%s/%s' % ( version, v_branches['filename'])).encode('utf-8') branches[branch_name] = { 'target_type': 'revision', 'target': x['target'], } snapshot = { 'branches': branches } 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 as e: logger.warning('Fail to load %s. Reason: %s' % (self.url, e)) status_visit = 'partial' finally: self.storage.origin_visit_update( origin=self.url, visit_id=visit_id, status=status_visit, snapshot=snapshot) return {'status': status_load} diff --git a/swh/loader/package/npm.py b/swh/loader/package/npm.py index 8c3680f..77762d9 100644 --- a/swh/loader/package/npm.py +++ b/swh/loader/package/npm.py @@ -1,286 +1,292 @@ # Copyright (C) 2019 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 json import logging import os import re from codecs import BOM_UTF8 from typing import Generator, Dict, Tuple, Sequence, Optional import chardet import iso8601 from swh.model.identifiers import normalize_timestamp from swh.loader.package.loader import PackageLoader from swh.loader.package.utils import api_info logger = logging.getLogger(__name__) _EMPTY_AUTHOR = {'fullname': b'', 'name': None, 'email': None} # https://github.com/jonschlinkert/author-regex _author_regexp = r'([^<(]+?)?[ \t]*(?:<([^>(]+?)>)?[ \t]*(?:\(([^)]+?)\)|$)' def parse_npm_package_author(author_str): """ Parse npm package author string. It works with a flexible range of formats, as detailed below:: name name (url) name (url) name (url) name(url) name (url) name (url) name(url) name(url) name (url) name(url) name name (url) (url) (url) (url) (url) Args: author_str (str): input author string Returns: dict: A dict that may contain the following keys: * name * email * url """ author = {} matches = re.findall(_author_regexp, author_str.replace('<>', '').replace('()', ''), re.M) for match in matches: if match[0].strip(): author['name'] = match[0].strip() if match[1].strip(): author['email'] = match[1].strip() if match[2].strip(): author['url'] = match[2].strip() return author def extract_npm_package_author(package_json): """ Extract package author from a ``package.json`` file content and return it in swh format. Args: package_json (dict): Dict holding the content of parsed ``package.json`` file Returns: dict: A dict with the following keys: * fullname * name * email """ def _author_str(author_data): if type(author_data) is dict: author_str = '' if 'name' in author_data: author_str += author_data['name'] if 'email' in author_data: author_str += ' <%s>' % author_data['email'] return author_str elif type(author_data) is list: return _author_str(author_data[0]) if len(author_data) > 0 else '' else: return author_data author_data = {} for author_key in ('author', 'authors'): if author_key in package_json: author_str = _author_str(package_json[author_key]) author_data = parse_npm_package_author(author_str) name = author_data.get('name') email = author_data.get('email') fullname = None if name and email: fullname = '%s <%s>' % (name, email) elif name: fullname = name if not fullname: return _EMPTY_AUTHOR if fullname: fullname = fullname.encode('utf-8') if name: name = name.encode('utf-8') if email: email = email.encode('utf-8') return {'fullname': fullname, 'name': name, 'email': email} def _lstrip_bom(s, bom=BOM_UTF8): if s.startswith(bom): return s[len(bom):] else: return s def load_json(json_bytes): """ Try to load JSON from bytes and return a dictionary. First try to decode from utf-8. If the decoding failed, try to detect the encoding and decode again with replace error handling. If JSON is malformed, an empty dictionary will be returned. Args: json_bytes (bytes): binary content of a JSON file Returns: dict: JSON data loaded in a dictionary """ json_data = {} try: json_str = _lstrip_bom(json_bytes).decode('utf-8') except UnicodeDecodeError: encoding = chardet.detect(json_bytes)['encoding'] if encoding: json_str = json_bytes.decode(encoding, 'replace') try: json_data = json.loads(json_str) except json.decoder.JSONDecodeError: pass return json_data def extract_intrinsic_metadata(dir_path: str) -> Dict: """Given an uncompressed path holding the pkginfo file, returns a pkginfo parsed structure as a dict. The release artifact contains at their root one folder. For example: $ tar tvf zprint-0.0.6.tar.gz drwxr-xr-x root/root 0 2018-08-22 11:01 zprint-0.0.6/ ... Args: dir_path (str): Path to the uncompressed directory representing a release artifact from npm. Returns: the pkginfo parsed structure as a dict if any or None if none was present. """ # Retrieve the root folder of the archive if not os.path.exists(dir_path): return {} lst = os.listdir(dir_path) if len(lst) == 0: return {} project_dirname = lst[0] package_json_path = os.path.join(dir_path, project_dirname, 'package.json') if not os.path.exists(package_json_path): return {} with open(package_json_path, 'rb') as package_json_file: package_json_bytes = package_json_file.read() return load_json(package_json_bytes) class NpmLoader(PackageLoader): visit_type = 'npm' def __init__(self, package_name, package_url, package_metadata_url): super().__init__(url=package_url) self.provider_url = package_metadata_url self._info = None self._versions = None # if package_url is None: # package_url = 'https://www.npmjs.com/package/%s' % package_name # if package_metadata_url is None: # package_metadata_url = 'https://replicate.npmjs.com/%s/' %\ # quote(package_name, safe='') @property def info(self) -> Dict: """Return the project metadata information (fetched from npm registry) """ if not self._info: self._info = api_info(self.provider_url) return self._info def get_versions(self) -> Sequence[str]: return sorted(self.info['versions'].keys()) def get_default_release(self) -> str: return self.info['dist-tags'].get('latest', '') def get_artifacts(self, version: str) -> Generator[ Tuple[str, str, Dict], None, None]: meta = self.info['versions'][version] url = meta['dist']['tarball'] filename = os.path.basename(url) yield filename, url, meta def resolve_revision_from( self, known_artifacts: Dict, artifact_metadata: Dict) \ -> Optional[bytes]: shasum = artifact_metadata['dist']['shasum'] for rev_id, known_artifact in known_artifacts.items(): if shasum == known_artifact['checksums']['sha1']: return rev_id def build_revision( - self, a_metadata: Dict, a_uncompressed_path: str) -> Dict: + self, a_metadata: Dict, a_uncompressed_path: str, + visit_date: str) -> Dict: # Parse metadata (project, artifact metadata) i_metadata = extract_intrinsic_metadata(a_uncompressed_path) # from intrinsic metadata author = extract_npm_package_author(i_metadata) # extrinsic metadata version = i_metadata['version'] date = self.info['time'][version] date = iso8601.parse_date(date) date = normalize_timestamp(int(date.timestamp())) message = version.encode('ascii') return { 'author': author, 'date': date, 'committer': author, 'committer_date': date, 'message': message, + 'parents': [], 'metadata': { 'intrinsic': { 'tool': 'package.json', 'raw': i_metadata, - } + }, + 'extrinsic': { + 'provider': self.provider_url, + 'when': visit_date, + 'raw': a_metadata, + }, }, - 'parents': [], } diff --git a/swh/loader/package/pypi.py b/swh/loader/package/pypi.py index c5a540b..d3d22ce 100644 --- a/swh/loader/package/pypi.py +++ b/swh/loader/package/pypi.py @@ -1,173 +1,179 @@ # Copyright (C) 2019 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 from typing import Generator, Dict, Tuple, Sequence, Optional from urllib.parse import urlparse from pkginfo import UnpackedSDist import iso8601 from swh.model.identifiers import normalize_timestamp from swh.loader.package.loader import PackageLoader from swh.loader.package.utils import api_info def pypi_api_url(url: str) -> str: """Compute api url from a project url Args: url (str): PyPI instance's url (e.g: https://pypi.org/project/requests) This deals with correctly transforming the project's api url (e.g https://pypi.org/pypi/requests/json) Returns: api url """ p_url = urlparse(url) project_name = p_url.path.split('/')[-1] url = '%s://%s/pypi/%s/json' % (p_url.scheme, p_url.netloc, project_name) return url def extract_intrinsic_metadata(dir_path: str) -> Dict: """Given an uncompressed path holding the pkginfo file, returns a pkginfo parsed structure as a dict. The release artifact contains at their root one folder. For example: $ tar tvf zprint-0.0.6.tar.gz drwxr-xr-x root/root 0 2018-08-22 11:01 zprint-0.0.6/ ... Args: dir_path (str): Path to the uncompressed directory representing a release artifact from pypi. Returns: the pkginfo parsed structure as a dict if any or None if none was present. """ # Retrieve the root folder of the archive if not os.path.exists(dir_path): return {} lst = os.listdir(dir_path) if len(lst) != 1: return {} project_dirname = lst[0] pkginfo_path = os.path.join(dir_path, project_dirname, 'PKG-INFO') if not os.path.exists(pkginfo_path): return {} pkginfo = UnpackedSDist(pkginfo_path) raw = pkginfo.__dict__ raw.pop('filename') # this gets added with the ondisk location return raw def author(data: Dict) -> Dict: """Given a dict of project/release artifact information (coming from PyPI), returns an author subset. Args: data (dict): Representing either artifact information or release information. Returns: swh-model dict representing a person. """ name = data.get('author') email = data.get('author_email') if email: fullname = '%s <%s>' % (name, email) else: fullname = name if not fullname: return {'fullname': b'', 'name': None, 'email': None} fullname = fullname.encode('utf-8') if name is not None: name = name.encode('utf-8') if email is not None: email = email.encode('utf-8') return {'fullname': fullname, 'name': name, 'email': email} class PyPILoader(PackageLoader): """Load pypi origin's artifact releases into swh archive. """ visit_type = 'pypi' def __init__(self, url): super().__init__(url=url) self._info = None self.provider_url = pypi_api_url(self.url) @property def info(self) -> Dict: """Return the project metadata information (fetched from pypi registry) """ if not self._info: self._info = api_info(self.provider_url) return self._info def get_versions(self) -> Sequence[str]: return self.info['releases'].keys() def get_default_release(self) -> str: return self.info['info']['version'] def get_artifacts(self, version: str) -> Generator[ Tuple[str, str, Dict], None, None]: for meta in self.info['releases'][version]: yield meta['filename'], meta['url'], meta def resolve_revision_from( self, known_artifacts: Dict, artifact_metadata: Dict) \ -> Optional[bytes]: sha256 = artifact_metadata['digests']['sha256'] for rev_id, known_artifact in known_artifacts.items(): if sha256 == known_artifact['checksums']['sha256']: return rev_id def build_revision( - self, a_metadata: Dict, a_uncompressed_path: str) -> Dict: + self, a_metadata: Dict, a_uncompressed_path: str, + visit_date: str) -> Dict: # Parse metadata (project, artifact metadata) metadata = extract_intrinsic_metadata(a_uncompressed_path) # from intrinsic metadata name = metadata['version'] _author = author(metadata) # from extrinsic metadata message = a_metadata.get('comment_text', '') message = '%s: %s' % (name, message) if message else name date = normalize_timestamp( int(iso8601.parse_date(a_metadata['upload_time']).timestamp())) return { 'message': message.encode('utf-8'), 'author': _author, 'date': date, 'committer': _author, 'committer_date': date, 'parents': [], 'metadata': { 'intrinsic': { 'tool': 'PKG-INFO', 'raw': metadata, - } + }, + 'extrinsic': { + 'provider': self.provider_url, + 'when': visit_date, + 'raw': a_metadata, + }, } } diff --git a/swh/loader/package/tests/common.py b/swh/loader/package/tests/common.py index 81d105b..a46cd58 100644 --- a/swh/loader/package/tests/common.py +++ b/swh/loader/package/tests/common.py @@ -1,57 +1,101 @@ # Copyright (C) 2019 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 os import path import logging +from typing import Dict, List, Tuple + from swh.model.hashutil import hash_to_bytes, hash_to_hex logger = logging.getLogger(__file__) + DATADIR = path.join(path.abspath(path.dirname(__file__)), 'resources') def decode_target(target): """Test helper to ease readability in test """ if not target: return target target_type = target['target_type'] if target_type == 'alias': decoded_target = target['target'].decode('utf-8') else: decoded_target = hash_to_hex(target['target']) return { 'target': decoded_target, 'target_type': target_type } def check_snapshot(expected_snapshot, storage): """Check for snapshot match. Provide the hashes as hexadecimal, the conversion is done within the method. Args: expected_snapshot (dict): full snapshot with hex ids storage (Storage): expected storage """ expected_snapshot_id = expected_snapshot['id'] expected_branches = expected_snapshot['branches'] snap = storage.snapshot_get(hash_to_bytes(expected_snapshot_id)) assert snap is not None branches = { branch.decode('utf-8'): decode_target(target) for branch, target in snap['branches'].items() } assert expected_branches == branches + + +def check_metadata(metadata: Dict, key_path: str, raw_type: str): + """Given a metadata dict, ensure the associated key_path value is of type + raw_type. + + Args: + metadata: Dict to check + key_path: Path to check + raw_type: Type to check the path with + + Raises: + Assertion error in case of mismatch + + """ + data = metadata + keys = key_path.split('.') + for k in keys: + try: + data = data[k] + except (TypeError, KeyError) as e: + # KeyError: because path too long + # TypeError: data is not a dict + raise AssertionError(e) + assert isinstance(data, raw_type) + + +def check_metadata_paths(metadata: Dict, paths: List[Tuple[str, str]]): + """Given a metadata dict, ensure the keys are of expected types + + Args: + metadata: Dict to check + key_path: Path to check + raw_type: Type to check the path with + + Raises: + Assertion error in case of mismatch + + """ + for key_path, raw_type in paths: + check_metadata(metadata, key_path, raw_type) diff --git a/swh/loader/package/tests/test_common.py b/swh/loader/package/tests/test_common.py index 3f49c05..95a5a9a 100644 --- a/swh/loader/package/tests/test_common.py +++ b/swh/loader/package/tests/test_common.py @@ -1,101 +1,175 @@ # Copyright (C) 2019 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 pytest from swh.model.hashutil import hash_to_bytes -from swh.loader.package.tests.common import decode_target, check_snapshot +from swh.loader.package.tests.common import ( + decode_target, check_snapshot, check_metadata, check_metadata_paths +) from swh.storage import get_storage hash_hex = '43e45d56f88993aae6a0198013efa80716fd8920' def test_decode_target_edge(): assert not decode_target(None) def test_decode_target(): actual_alias_decode_target = decode_target({ 'target_type': 'alias', 'target': b'something', }) assert actual_alias_decode_target == { 'target_type': 'alias', 'target': 'something', } actual_decode_target = decode_target({ 'target_type': 'revision', 'target': hash_to_bytes(hash_hex), }) assert actual_decode_target == { 'target_type': 'revision', 'target': hash_hex, } def test_check_snapshot(): storage = get_storage(cls='memory', args={}) snap_id = '2498dbf535f882bc7f9a18fb16c9ad27fda7bab7' snapshot = { 'id': hash_to_bytes(snap_id), 'branches': { b'master': { 'target': hash_to_bytes(hash_hex), 'target_type': 'revision', }, }, } s = storage.snapshot_add([snapshot]) assert s == { 'snapshot:add': 1, } expected_snapshot = { 'id': snap_id, 'branches': { 'master': { 'target': hash_hex, 'target_type': 'revision', } } } check_snapshot(expected_snapshot, storage) def test_check_snapshot_failure(): storage = get_storage(cls='memory', args={}) snapshot = { 'id': hash_to_bytes('2498dbf535f882bc7f9a18fb16c9ad27fda7bab7'), 'branches': { b'master': { 'target': hash_to_bytes(hash_hex), 'target_type': 'revision', }, }, } s = storage.snapshot_add([snapshot]) assert s == { 'snapshot:add': 1, } unexpected_snapshot = { 'id': '2498dbf535f882bc7f9a18fb16c9ad27fda7bab7', 'branches': { 'master': { 'target': hash_hex, 'target_type': 'release', # wrong value } } } with pytest.raises(AssertionError): check_snapshot(unexpected_snapshot, storage) + + +def test_check_metadata(): + metadata = { + 'a': { + 'raw': { + 'time': 'something', + }, + }, + 'b': [], + 'c': 1, + } + + for raw_path, raw_type in [ + ('a.raw', dict), + ('a.raw.time', str), + ('b', list), + ('c', int), + ]: + check_metadata(metadata, raw_path, raw_type) + + +def test_check_metadata_ko(): + metadata = { + 'a': { + 'raw': 'hello', + }, + 'b': [], + 'c': 1, + } + + for raw_path, raw_type in [ + ('a.b', dict), + ('a.raw.time', str), + ]: + with pytest.raises(AssertionError): + check_metadata(metadata, raw_path, raw_type) + + +def test_check_metadata_paths(): + metadata = { + 'a': { + 'raw': { + 'time': 'something', + }, + }, + 'b': [], + 'c': 1, + } + + check_metadata_paths(metadata, [ + ('a.raw', dict), + ('a.raw.time', str), + ('b', list), + ('c', int), + ]) + + +def test_check_metadata_paths_ko(): + metadata = { + 'a': { + 'raw': 'hello', + }, + 'b': [], + 'c': 1, + } + + with pytest.raises(AssertionError): + check_metadata_paths(metadata, [ + ('a.b', dict), + ('a.raw.time', str), + ]) diff --git a/swh/loader/package/tests/test_gnu.py b/swh/loader/package/tests/test_gnu.py index 55af941..181e35c 100644 --- a/swh/loader/package/tests/test_gnu.py +++ b/swh/loader/package/tests/test_gnu.py @@ -1,214 +1,251 @@ # Copyright (C) 2019 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 re from swh.model.hashutil import hash_to_bytes from swh.loader.package.gnu import GNULoader, get_version -from swh.loader.package.tests.common import check_snapshot +from swh.loader.package.tests.common import ( + check_snapshot, check_metadata_paths +) def test_get_version(): """From url to branch name should yield something relevant """ for url, expected_branchname in [ ('https://gnu.org/sthg/info-2.1.0.tar.gz', '2.1.0'), ('https://gnu.org/sthg/info-2.1.2.zip', '2.1.2'), ('https://sthg.org/gnu/sthg.tar.gz', 'sthg'), ('https://sthg.org/gnu/DLDF-1.1.4.tar.gz', '1.1.4'), ('https://sthg.org/gnu/anubis-latest.tar.bz2', 'latest'), ('https://ftp.org/gnu/aris-w32.zip', 'w32'), ('https://ftp.org/gnu/aris-w32-2.2.zip', 'w32-2.2'), ('https://ftp.org/gnu/autogen.info.tar.gz', 'autogen.info'), ('https://ftp.org/gnu/crypto-build-demo.tar.gz', 'crypto-build-demo'), ('https://ftp.org/gnu/clue+clio+xit.clisp.tar.gz', 'clue+clio+xit.clisp'), ('https://ftp.org/gnu/clue+clio.for-pcl.tar.gz', 'clue+clio.for-pcl'), ('https://ftp.org/gnu/clisp-hppa2.0-hp-hpux10.20.tar.gz', 'hppa2.0-hp-hpux10.20'), ('clisp-i386-solaris2.6.tar.gz', 'i386-solaris2.6'), ('clisp-mips-sgi-irix6.5.tar.gz', 'mips-sgi-irix6.5'), ('clisp-powerpc-apple-macos.tar.gz', 'powerpc-apple-macos'), ('clisp-powerpc-unknown-linuxlibc6.tar.gz', 'powerpc-unknown-linuxlibc6'), ('clisp-rs6000-ibm-aix3.2.5.tar.gz', 'rs6000-ibm-aix3.2.5'), ('clisp-sparc-redhat51-linux.tar.gz', 'sparc-redhat51-linux'), ('clisp-sparc-sun-solaris2.4.tar.gz', 'sparc-sun-solaris2.4'), ('clisp-sparc-sun-sunos4.1.3_U1.tar.gz', 'sparc-sun-sunos4.1.3_U1'), ('clisp-2.25.1-powerpc-apple-MacOSX.tar.gz', '2.25.1-powerpc-apple-MacOSX'), ('clisp-2.27-PowerMacintosh-powerpc-Darwin-1.3.7.tar.gz', '2.27-PowerMacintosh-powerpc-Darwin-1.3.7'), ('clisp-2.27-i686-unknown-Linux-2.2.19.tar.gz', '2.27-i686-unknown-Linux-2.2.19'), ('clisp-2.28-i386-i386-freebsd-4.3-RELEASE.tar.gz', '2.28-i386-i386-freebsd-4.3-RELEASE'), ('clisp-2.28-i686-unknown-cygwin_me-4.90-1.3.10.tar.gz', '2.28-i686-unknown-cygwin_me-4.90-1.3.10'), ('clisp-2.29-i386-i386-freebsd-4.6-STABLE.tar.gz', '2.29-i386-i386-freebsd-4.6-STABLE'), ('clisp-2.29-i686-unknown-cygwin_nt-5.0-1.3.12.tar.gz', '2.29-i686-unknown-cygwin_nt-5.0-1.3.12'), ('gcl-2.5.3-ansi-japi-xdr.20030701_mingw32.zip', '2.5.3-ansi-japi-xdr.20030701_mingw32'), ('gettext-runtime-0.13.1.bin.woe32.zip', '0.13.1.bin.woe32'), ('sather-logo_images.tar.gz', 'sather-logo_images'), ('sather-specification-000328.html.tar.gz', '000328.html') ]: actual_branchname = get_version(url) assert actual_branchname == expected_branchname _expected_new_contents_first_visit = [ 'e9258d81faf5881a2f96a77ba609396f82cb97ad', '1170cf105b04b7e2822a0e09d2acf71da7b9a130', 'fbd27c3f41f2668624ffc80b7ba5db9b92ff27ac', '0057bec9b5422aff9256af240b177ac0e3ac2608', '2b8d0d0b43a1078fc708930c8ddc2956a86c566e', '27de3b3bc6545d2a797aeeb4657c0e215a0c2e55', '2e6db43f5cd764e677f416ff0d0c78c7a82ef19b', 'ae9be03bd2a06ed8f4f118d3fe76330bb1d77f62', 'edeb33282b2bffa0e608e9d2fd960fd08093c0ea', 'd64e64d4c73679323f8d4cde2643331ba6c20af9', '7a756602914be889c0a2d3952c710144b3e64cb0', '84fb589b554fcb7f32b806951dcf19518d67b08f', '8624bcdae55baeef00cd11d5dfcfa60f68710a02', 'e08441aeab02704cfbd435d6445f7c072f8f524e', 'f67935bc3a83a67259cda4b2d43373bd56703844', '809788434b433eb2e3cfabd5d591c9a659d5e3d8', '7d7c6c8c5ebaeff879f61f37083a3854184f6c41', 'b99fec102eb24bffd53ab61fc30d59e810f116a2', '7d149b28eaa228b3871c91f0d5a95a2fa7cb0c68', 'f0c97052e567948adf03e641301e9983c478ccff', '7fb724242e2b62b85ca64190c31dcae5303e19b3', '4f9709e64a9134fe8aefb36fd827b84d8b617ab5', '7350628ccf194c2c3afba4ac588c33e3f3ac778d', '0bb892d9391aa706dc2c3b1906567df43cbe06a2', '49d4c0ce1a16601f1e265d446b6c5ea6b512f27c', '6b5cc594ac466351450f7f64a0b79fdaf4435ad3', '3046e5d1f70297e2a507b98224b6222c9688d610', '1572607d456d7f633bc6065a2b3048496d679a31', ] _expected_new_directories_first_visit = [ 'daabc65ec75d487b1335ffc101c0ac11c803f8fc', '263be23b4a8101d3ad0d9831319a3e0f2b065f36', '7f6e63ba6eb3e2236f65892cd822041f1a01dd5c', '4db0a3ecbc976083e2dac01a62f93729698429a3', 'dfef1c80e1098dd5deda664bb44a9ab1f738af13', 'eca971d346ea54d95a6e19d5051f900237fafdaa', '3aebc29ed1fccc4a6f2f2010fb8e57882406b528', ] _expected_new_revisions_first_visit = { '44183488c0774ce3c957fa19ba695cf18a4a42b3': '3aebc29ed1fccc4a6f2f2010fb8e57882406b528' } _expected_branches_first_visit = { 'HEAD': { 'target_type': 'alias', 'target': 'releases/0.1.0', }, 'releases/0.1.0': { 'target_type': 'revision', 'target': '44183488c0774ce3c957fa19ba695cf18a4a42b3', }, } # hash is different then before as we changed the snapshot # gnu used to use `release/` (singular) instead of plural _expected_new_snapshot_first_visit_id = 'c419397fd912039825ebdbea378bc6283f006bf5' # noqa def test_release_artifact_not_found(swh_config, requests_mock): package = '8sync' package_url = 'https://ftp.gnu.org/gnu/8sync/' tarballs = [{ 'date': '944729610', 'archive': 'https://ftp.gnu.org/gnu/8sync/8sync-0.1.0.tar.gz', }] loader = GNULoader(package, package_url, tarballs) requests_mock.get(re.compile('https://'), status_code=404) actual_load_status = loader.load() assert actual_load_status == {'status': 'uneventful'} stats = loader.storage.stat_counters() assert { 'content': 0, 'directory': 0, 'origin': 1, 'origin_visit': 1, 'person': 0, 'release': 0, 'revision': 0, 'skipped_content': 0, 'snapshot': 1, } == stats origin_visit = next(loader.storage.origin_visit_get(package_url)) assert origin_visit['status'] == 'partial' +def test_revision_metadata_structure(swh_config, local_get): + package = '8sync' + package_url = 'https://ftp.gnu.org/gnu/8sync/' + tarballs = [{ + 'date': '944729610', + 'archive': 'https://ftp.gnu.org/gnu/8sync/8sync-0.1.0.tar.gz', + }] + + loader = GNULoader(package, package_url, tarballs) + + actual_load_status = loader.load() + assert actual_load_status == {'status': 'eventful'} + + expected_revision_id = hash_to_bytes( + '44183488c0774ce3c957fa19ba695cf18a4a42b3') + revision = list(loader.storage.revision_get([expected_revision_id]))[0] + + assert revision is not None + + assert isinstance(revision['metadata'], dict) + assert isinstance(revision['metadata']['intrinsic'], dict) + assert isinstance(revision['metadata']['extrinsic'], dict) + assert isinstance(revision['metadata']['original_artifact'], dict) + + check_metadata_paths(revision['metadata'], paths=[ + ('intrinsic', dict), + ('extrinsic.provider', str), + ('extrinsic.when', str), + ('extrinsic.raw', dict), + ('original_artifact.filename', str), + ('original_artifact.length', int), + ('original_artifact.checksums', dict), + ]) + + def test_release_artifact_no_prior_visit(swh_config, local_get): """With no prior visit, load a gnu project ends up with 1 snapshot """ assert 'SWH_CONFIG_FILENAME' in os.environ # cf. tox.ini package = '8sync' package_url = 'https://ftp.gnu.org/gnu/8sync/' tarballs = [{ 'date': '944729610', 'archive': 'https://ftp.gnu.org/gnu/8sync/8sync-0.1.0.tar.gz', }] loader = GNULoader(package, package_url, tarballs) actual_load_status = loader.load() assert actual_load_status == {'status': 'eventful'} stats = loader.storage.stat_counters() assert { 'content': len(_expected_new_contents_first_visit), 'directory': len(_expected_new_directories_first_visit), 'origin': 1, 'origin_visit': 1, 'person': 1, 'release': 0, 'revision': len(_expected_new_revisions_first_visit), 'skipped_content': 0, 'snapshot': 1 } == stats expected_contents = map(hash_to_bytes, _expected_new_contents_first_visit) assert list(loader.storage.content_missing_per_sha1(expected_contents)) \ == [] expected_dirs = map(hash_to_bytes, _expected_new_directories_first_visit) assert list(loader.storage.directory_missing(expected_dirs)) == [] expected_revs = map(hash_to_bytes, _expected_new_revisions_first_visit) assert list(loader.storage.revision_missing(expected_revs)) == [] expected_snapshot = { 'id': _expected_new_snapshot_first_visit_id, 'branches': _expected_branches_first_visit, } check_snapshot(expected_snapshot, loader.storage) diff --git a/swh/loader/package/tests/test_npm.py b/swh/loader/package/tests/test_npm.py index d1b95cf..6ccf13f 100644 --- a/swh/loader/package/tests/test_npm.py +++ b/swh/loader/package/tests/test_npm.py @@ -1,496 +1,530 @@ # Copyright (C) 2019 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 json import os from swh.model.hashutil import hash_to_bytes from swh.loader.package.npm import ( parse_npm_package_author, extract_npm_package_author ) -from swh.loader.package.tests.common import DATADIR, check_snapshot +from swh.loader.package.tests.common import ( + DATADIR, check_snapshot, check_metadata_paths +) from swh.loader.package.npm import NpmLoader def _parse_author_string_test(author_str, expected_result): assert parse_npm_package_author(author_str) == expected_result assert parse_npm_package_author(' %s' % author_str) == expected_result assert parse_npm_package_author('%s ' % author_str) == expected_result def test_parse_npm_package_author(): _parse_author_string_test( 'John Doe', { 'name': 'John Doe' } ) _parse_author_string_test( '', { 'email': 'john.doe@foo.bar' } ) _parse_author_string_test( '(https://john.doe)', { 'url': 'https://john.doe' } ) _parse_author_string_test( 'John Doe ', { 'name': 'John Doe', 'email': 'john.doe@foo.bar' } ) _parse_author_string_test( 'John Doe', { 'name': 'John Doe', 'email': 'john.doe@foo.bar' } ) _parse_author_string_test( 'John Doe (https://john.doe)', { 'name': 'John Doe', 'url': 'https://john.doe' } ) _parse_author_string_test( 'John Doe(https://john.doe)', { 'name': 'John Doe', 'url': 'https://john.doe' } ) _parse_author_string_test( ' (https://john.doe)', { 'email': 'john.doe@foo.bar', 'url': 'https://john.doe' } ) _parse_author_string_test( '(https://john.doe) ', { 'email': 'john.doe@foo.bar', 'url': 'https://john.doe' } ) _parse_author_string_test( 'John Doe (https://john.doe)', { 'name': 'John Doe', 'email': 'john.doe@foo.bar', 'url': 'https://john.doe' } ) _parse_author_string_test( 'John Doe (https://john.doe) ', { 'name': 'John Doe', 'email': 'john.doe@foo.bar', 'url': 'https://john.doe' } ) _parse_author_string_test( 'John Doe (https://john.doe)', { 'name': 'John Doe', 'email': 'john.doe@foo.bar', 'url': 'https://john.doe' } ) _parse_author_string_test( 'John Doe(https://john.doe)', { 'name': 'John Doe', 'email': 'john.doe@foo.bar', 'url': 'https://john.doe' } ) _parse_author_string_test('', {}) _parse_author_string_test('<>', {}) _parse_author_string_test(' <>', {}) _parse_author_string_test('<>()', {}) _parse_author_string_test('<> ()', {}) _parse_author_string_test('()', {}) _parse_author_string_test(' ()', {}) _parse_author_string_test( 'John Doe <> ()', { 'name': 'John Doe' } ) _parse_author_string_test( 'John Doe <>', { 'name': 'John Doe' } ) _parse_author_string_test( 'John Doe ()', { 'name': 'John Doe' } ) def test_extract_npm_package_author(): package_metadata_filepath = os.path.join( DATADIR, 'replicate.npmjs.com', 'org_visit1') with open(package_metadata_filepath) as json_file: package_metadata = json.load(json_file) extract_npm_package_author(package_metadata['versions']['0.0.2']) == \ { 'fullname': b'mooz ', 'name': b'mooz', 'email': b'stillpedant@gmail.com' } assert ( extract_npm_package_author(package_metadata['versions']['0.0.3']) == { 'fullname': b'Masafumi Oyamada ', 'name': b'Masafumi Oyamada', 'email': b'stillpedant@gmail.com' } ) package_json = json.loads(''' { "name": "highlightjs-line-numbers.js", "version": "2.7.0", "description": "Highlight.js line numbers plugin.", "main": "src/highlightjs-line-numbers.js", "dependencies": {}, "devDependencies": { "gulp": "^4.0.0", "gulp-rename": "^1.4.0", "gulp-replace": "^0.6.1", "gulp-uglify": "^1.2.0" }, "repository": { "type": "git", "url": "https://github.com/wcoder/highlightjs-line-numbers.js.git" }, "author": "Yauheni Pakala ", "license": "MIT", "bugs": { "url": "https://github.com/wcoder/highlightjs-line-numbers.js/issues" }, "homepage": "http://wcoder.github.io/highlightjs-line-numbers.js/" }''') # noqa assert extract_npm_package_author(package_json) == \ { 'fullname': b'Yauheni Pakala ', 'name': b'Yauheni Pakala', 'email': b'evgeniy.pakalo@gmail.com' } package_json = json.loads(''' { "name": "3-way-diff", "version": "0.0.1", "description": "3-way diffing of JavaScript objects", "main": "index.js", "authors": [ { "name": "Shawn Walsh", "url": "https://github.com/shawnpwalsh" }, { "name": "Markham F Rollins IV", "url": "https://github.com/mrollinsiv" } ], "keywords": [ "3-way diff", "3 way diff", "three-way diff", "three way diff" ], "devDependencies": { "babel-core": "^6.20.0", "babel-preset-es2015": "^6.18.0", "mocha": "^3.0.2" }, "dependencies": { "lodash": "^4.15.0" } }''') assert extract_npm_package_author(package_json) == \ { 'fullname': b'Shawn Walsh', 'name': b'Shawn Walsh', 'email': None } package_json = json.loads(''' { "name": "yfe-ynpm", "version": "1.0.0", "homepage": "http://gitlab.ywwl.com/yfe/yfe-ynpm", "repository": { "type": "git", "url": "git@gitlab.ywwl.com:yfe/yfe-ynpm.git" }, "author": [ "fengmk2 (https://fengmk2.com)", "xufuzi (https://7993.org)" ], "license": "MIT" }''') assert extract_npm_package_author(package_json) == \ { 'fullname': b'fengmk2 ', 'name': b'fengmk2', 'email': b'fengmk2@gmail.com' } package_json = json.loads(''' { "name": "umi-plugin-whale", "version": "0.0.8", "description": "Internal contract component", "authors": { "name": "xiaohuoni", "email": "448627663@qq.com" }, "repository": "alitajs/whale", "devDependencies": { "np": "^3.0.4", "umi-tools": "*" }, "license": "MIT" }''') assert extract_npm_package_author(package_json) == \ { 'fullname': b'xiaohuoni <448627663@qq.com>', 'name': b'xiaohuoni', 'email': b'448627663@qq.com' } def normalize_hashes(hashes): if isinstance(hashes, str): return hash_to_bytes(hashes) if isinstance(hashes, list): return [hash_to_bytes(x) for x in hashes] return {hash_to_bytes(k): hash_to_bytes(v) for k, v in hashes.items()} _expected_new_contents_first_visit = normalize_hashes([ '4ce3058e16ab3d7e077f65aabf855c34895bf17c', '858c3ceee84c8311adc808f8cdb30d233ddc9d18', '0fa33b4f5a4e0496da6843a38ff1af8b61541996', '85a410f8ef8eb8920f2c384a9555566ad4a2e21b', '9163ac8025923d5a45aaac482262893955c9b37b', '692cf623b8dd2c5df2c2998fd95ae4ec99882fb4', '18c03aac6d3e910efb20039c15d70ab5e0297101', '41265c42446aac17ca769e67d1704f99e5a1394d', '783ff33f5882813dca9239452c4a7cadd4dba778', 'b029cfb85107aee4590c2434a3329bfcf36f8fa1', '112d1900b4c2e3e9351050d1b542c9744f9793f3', '5439bbc4bd9a996f1a38244e6892b71850bc98fd', 'd83097a2f994b503185adf4e719d154123150159', 'd0939b4898e83090ee55fd9d8a60e312cfadfbaf', 'b3523a26f7147e4af40d9d462adaae6d49eda13e', 'cd065fb435d6fb204a8871bcd623d0d0e673088c', '2854a40855ad839a54f4b08f5cff0cf52fca4399', 'b8a53bbaac34ebb8c6169d11a4b9f13b05c583fe', '0f73d56e1cf480bded8a1ecf20ec6fc53c574713', '0d9882b2dfafdce31f4e77fe307d41a44a74cefe', '585fc5caab9ead178a327d3660d35851db713df1', 'e8cd41a48d79101977e3036a87aeb1aac730686f', '5414efaef33cceb9f3c9eb5c4cc1682cd62d14f7', '9c3cc2763bf9e9e37067d3607302c4776502df98', '3649a68410e354c83cd4a38b66bd314de4c8f5c9', 'e96ed0c091de1ebdf587104eaf63400d1974a1fe', '078ca03d2f99e4e6eab16f7b75fbb7afb699c86c', '38de737da99514de6559ff163c988198bc91367a', ]) _expected_new_directories_first_visit = normalize_hashes([ '3370d20d6f96dc1c9e50f083e2134881db110f4f', '42753c0c2ab00c4501b552ac4671c68f3cf5aece', 'd7895533ef5edbcffdea3f057d9fef3a1ef845ce', '80579be563e2ef3e385226fe7a3f079b377f142c', '3b0ddc6a9e58b4b53c222da4e27b280b6cda591c', 'bcad03ce58ac136f26f000990fc9064e559fe1c0', '5fc7e82a1bc72e074665c6078c6d3fad2f13d7ca', 'e3cd26beba9b1e02f6762ef54bd9ac80cc5f25fd', '584b5b4b6cf7f038095e820b99386a9c232de931', '184c8d6d0d242f2b1792ef9d3bf396a5434b7f7a', 'bb5f4ee143c970367eb409f2e4c1104898048b9d', '1b95491047add1103db0dfdfa84a9735dcb11e88', 'a00c6de13471a2d66e64aca140ddb21ef5521e62', '5ce6c1cd5cda2d546db513aaad8c72a44c7771e2', 'c337091e349b6ac10d38a49cdf8c2401ef9bb0f2', '202fafcd7c0f8230e89d5496ad7f44ab12b807bf', '775cc516543be86c15c1dc172f49c0d4e6e78235', 'ff3d1ead85a14f891e8b3fa3a89de39db1b8de2e', ]) _expected_new_revisions_first_visit = normalize_hashes({ 'd8a1c7474d2956ac598a19f0f27d52f7015f117e': '42753c0c2ab00c4501b552ac4671c68f3cf5aece', '5f9eb78af37ffd12949f235e86fac04898f9f72a': '3370d20d6f96dc1c9e50f083e2134881db110f4f', 'ba019b192bdb94bd0b5bd68b3a5f92b5acc2239a': 'd7895533ef5edbcffdea3f057d9fef3a1ef845ce'} ) _expected_new_snapshot_first_visit_id = normalize_hashes( 'd0587e1195aed5a8800411a008f2f2d627f18e2d') _expected_branches_first_visit = { 'HEAD': { 'target': 'releases/0.0.4', 'target_type': 'alias' }, 'releases/0.0.2': { 'target': 'd8a1c7474d2956ac598a19f0f27d52f7015f117e', 'target_type': 'revision' }, 'releases/0.0.3': { 'target': '5f9eb78af37ffd12949f235e86fac04898f9f72a', 'target_type': 'revision' }, 'releases/0.0.4': { 'target': 'ba019b192bdb94bd0b5bd68b3a5f92b5acc2239a', 'target_type': 'revision' } } def package_url(package): return 'https://www.npmjs.com/package/%s' % package def package_metadata_url(package): return 'https://replicate.npmjs.com/%s/' % package +def test_revision_metadata_structure(swh_config, local_get): + package = 'org' + loader = NpmLoader(package, + package_url(package), + package_metadata_url(package)) + + actual_load_status = loader.load() + assert actual_load_status == {'status': 'eventful'} + + expected_revision_id = hash_to_bytes( + 'd8a1c7474d2956ac598a19f0f27d52f7015f117e') + revision = list(loader.storage.revision_get([expected_revision_id]))[0] + + assert revision is not None + + assert isinstance(revision['metadata'], dict) + assert isinstance(revision['metadata']['intrinsic'], dict) + assert isinstance(revision['metadata']['extrinsic'], dict) + assert isinstance(revision['metadata']['original_artifact'], dict) + + check_metadata_paths(revision['metadata'], paths=[ + ('intrinsic.tool', str), + ('intrinsic.raw', dict), + ('extrinsic.provider', str), + ('extrinsic.when', str), + ('extrinsic.raw', dict), + ('original_artifact.filename', str), + ('original_artifact.length', int), + ('original_artifact.checksums', dict), + ]) + + def test_npm_loader_first_visit(swh_config, local_get): package = 'org' loader = NpmLoader(package, package_url(package), package_metadata_url(package)) actual_load_status = loader.load() assert actual_load_status == {'status': 'eventful'} stats = loader.storage.stat_counters() assert { 'content': len(_expected_new_contents_first_visit), 'directory': len(_expected_new_directories_first_visit), 'origin': 1, 'origin_visit': 1, 'person': 2, 'release': 0, 'revision': len(_expected_new_revisions_first_visit), 'skipped_content': 0, 'snapshot': 1, } == stats assert len(list(loader.storage.content_get( _expected_new_contents_first_visit))) == len( _expected_new_contents_first_visit) assert list(loader.storage.directory_missing( _expected_new_directories_first_visit)) == [] assert list(loader.storage.revision_missing( _expected_new_revisions_first_visit)) == [] expected_snapshot = { 'id': _expected_new_snapshot_first_visit_id, 'branches': _expected_branches_first_visit, } check_snapshot(expected_snapshot, loader.storage) def test_npm_loader_incremental_visit(swh_config, local_get_visits): package = 'org' url = package_url(package) metadata_url = package_metadata_url(package) loader = NpmLoader(package, url, metadata_url) actual_load_status = loader.load() assert actual_load_status == {'status': 'eventful'} origin_visit = list(loader.storage.origin_visit_get(url))[-1] assert origin_visit['status'] == 'full' stats = loader.storage.stat_counters() assert { 'content': len(_expected_new_contents_first_visit), 'directory': len(_expected_new_directories_first_visit), 'origin': 1, 'origin_visit': 1, 'person': 2, 'release': 0, 'revision': len(_expected_new_revisions_first_visit), 'skipped_content': 0, 'snapshot': 1, } == stats loader._info = None # reset loader internal state actual_load_status2 = loader.load() assert actual_load_status2 == {'status': 'eventful'} origin_visit2 = list(loader.storage.origin_visit_get(url))[-1] assert origin_visit2['status'] == 'full' stats = loader.storage.stat_counters() assert { # 3 new releases artifacts 'content': len(_expected_new_contents_first_visit) + 14, 'directory': len(_expected_new_directories_first_visit) + 15, 'origin': 1, 'origin_visit': 2, 'person': 2, 'release': 0, 'revision': len(_expected_new_revisions_first_visit) + 3, 'skipped_content': 0, 'snapshot': 2, } == stats urls = [ m.url for m in local_get_visits.request_history if m.url.startswith('https://registry.npmjs.org') ] assert len(urls) == len(set(urls)) # we visited each artifact once across diff --git a/swh/loader/package/tests/test_pypi.py b/swh/loader/package/tests/test_pypi.py index e625690..2b3b6b1 100644 --- a/swh/loader/package/tests/test_pypi.py +++ b/swh/loader/package/tests/test_pypi.py @@ -1,541 +1,571 @@ # Copyright (C) 2019 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 from os import path import pytest from unittest.mock import patch from swh.core.tarball import uncompress from swh.model.hashutil import hash_to_bytes from swh.loader.package.pypi import ( PyPILoader, pypi_api_url, author, extract_intrinsic_metadata ) from swh.loader.package.tests.common import ( - check_snapshot, DATADIR + check_snapshot, DATADIR, check_metadata_paths ) from swh.loader.package.tests.conftest import local_get_factory def test_author_basic(): data = { 'author': "i-am-groot", 'author_email': 'iam@groot.org', } actual_author = author(data) expected_author = { 'fullname': b'i-am-groot ', 'name': b'i-am-groot', 'email': b'iam@groot.org', } assert actual_author == expected_author def test_author_empty_email(): data = { 'author': 'i-am-groot', 'author_email': '', } actual_author = author(data) expected_author = { 'fullname': b'i-am-groot', 'name': b'i-am-groot', 'email': b'', } assert actual_author == expected_author def test_author_empty_name(): data = { 'author': "", 'author_email': 'iam@groot.org', } actual_author = author(data) expected_author = { 'fullname': b' ', 'name': b'', 'email': b'iam@groot.org', } assert actual_author == expected_author def test_author_malformed(): data = { 'author': "['pierre', 'paul', 'jacques']", 'author_email': None, } actual_author = author(data) expected_author = { 'fullname': b"['pierre', 'paul', 'jacques']", 'name': b"['pierre', 'paul', 'jacques']", 'email': None, } assert actual_author == expected_author def test_author_malformed_2(): data = { 'author': '[marie, jeanne]', 'author_email': '[marie@some, jeanne@thing]', } actual_author = author(data) expected_author = { 'fullname': b'[marie, jeanne] <[marie@some, jeanne@thing]>', 'name': b'[marie, jeanne]', 'email': b'[marie@some, jeanne@thing]', } assert actual_author == expected_author def test_author_malformed_3(): data = { 'author': '[marie, jeanne, pierre]', 'author_email': '[marie@somewhere.org, jeanne@somewhere.org]', } actual_author = author(data) expected_author = { 'fullname': b'[marie, jeanne, pierre] <[marie@somewhere.org, jeanne@somewhere.org]>', # noqa 'name': b'[marie, jeanne, pierre]', 'email': b'[marie@somewhere.org, jeanne@somewhere.org]', } actual_author == expected_author # configuration error # def test_badly_configured_loader_raise(monkeypatch): """Badly configured loader should raise""" monkeypatch.delenv('SWH_CONFIG_FILENAME', raising=False) with pytest.raises(ValueError) as e: PyPILoader(url='some-url') assert 'Misconfiguration' in e.value.args[0] def test_pypi_api_url(): """Compute pypi api url from the pypi project url should be ok""" url = pypi_api_url('https://pypi.org/project/requests') assert url == 'https://pypi.org/pypi/requests/json' @pytest.mark.fs def test_extract_intrinsic_metadata(tmp_path): """Parsing existing archive's PKG-INFO should yield results""" uncompressed_archive_path = str(tmp_path) archive_path = path.join( DATADIR, 'files.pythonhosted.org', '0805nexter-1.1.0.zip') uncompress(archive_path, dest=uncompressed_archive_path) actual_metadata = extract_intrinsic_metadata(uncompressed_archive_path) expected_metadata = { 'metadata_version': '1.0', 'name': '0805nexter', 'version': '1.1.0', 'summary': 'a simple printer of nested lest', 'home_page': 'http://www.hp.com', 'author': 'hgtkpython', 'author_email': '2868989685@qq.com', 'platforms': ['UNKNOWN'], } assert actual_metadata == expected_metadata @pytest.mark.fs def test_extract_intrinsic_metadata_failures(tmp_path): """Parsing inexistant path/archive/PKG-INFO yield None""" # inexistant first level path assert extract_intrinsic_metadata('/something-inexistant') == {} # inexistant second level path (as expected by pypi archives) assert extract_intrinsic_metadata(tmp_path) == {} # inexistant PKG-INFO within second level path existing_path_no_pkginfo = str(tmp_path / 'something') os.mkdir(existing_path_no_pkginfo) assert extract_intrinsic_metadata(tmp_path) == {} # LOADER SCENARIO # # "edge" cases (for the same origin) # # no release artifact: # {visit full, status: uneventful, no contents, etc...} local_get_missing_all = local_get_factory(ignore_urls=[ 'https://files.pythonhosted.org/packages/ec/65/c0116953c9a3f47de89e71964d6c7b0c783b01f29fa3390584dbf3046b4d/0805nexter-1.1.0.zip', # noqa 'https://files.pythonhosted.org/packages/c4/a0/4562cda161dc4ecbbe9e2a11eb365400c0461845c5be70d73869786809c4/0805nexter-1.2.0.zip', # noqa ]) def test_no_release_artifact(swh_config, local_get_missing_all): """Load a pypi project with all artifacts missing ends up with no snapshot """ url = 'https://pypi.org/project/0805nexter' loader = PyPILoader(url) actual_load_status = loader.load() assert actual_load_status == {'status': 'uneventful'} stats = loader.storage.stat_counters() assert { 'content': 0, 'directory': 0, 'origin': 1, 'origin_visit': 1, 'person': 0, 'release': 0, 'revision': 0, 'skipped_content': 0, 'snapshot': 1, } == stats origin_visit = next(loader.storage.origin_visit_get(url)) assert origin_visit['status'] == 'partial' # problem during loading: # {visit: partial, status: uneventful, no snapshot} def test_release_with_traceback(swh_config): url = 'https://pypi.org/project/0805nexter' with patch('swh.loader.package.pypi.PyPILoader.get_default_release', side_effect=ValueError('Problem')): loader = PyPILoader(url) actual_load_status = loader.load() assert actual_load_status == {'status': 'uneventful'} stats = loader.storage.stat_counters() assert { 'content': 0, 'directory': 0, 'origin': 1, 'origin_visit': 1, 'person': 0, 'release': 0, 'revision': 0, 'skipped_content': 0, 'snapshot': 0, } == stats origin_visit = next(loader.storage.origin_visit_get(url)) assert origin_visit['status'] == 'partial' # problem during loading: failure early enough in between swh contents... # some contents (contents, directories, etc...) have been written in storage # {visit: partial, status: eventful, no snapshot} # problem during loading: failure late enough we can have snapshots (some # revisions are written in storage already) # {visit: partial, status: eventful, snapshot} # "normal" cases (for the same origin) # local_get_missing_one = local_get_factory(ignore_urls=[ 'https://files.pythonhosted.org/packages/ec/65/c0116953c9a3f47de89e71964d6c7b0c783b01f29fa3390584dbf3046b4d/0805nexter-1.1.0.zip', # noqa ]) # some missing release artifacts: # {visit partial, status: eventful, 1 snapshot} +def test_revision_metadata_structure(swh_config, local_get): + url = 'https://pypi.org/project/0805nexter' + loader = PyPILoader(url) + + actual_load_status = loader.load() + assert actual_load_status == {'status': 'eventful'} + + expected_revision_id = hash_to_bytes( + 'e445da4da22b31bfebb6ffc4383dbf839a074d21') + revision = list(loader.storage.revision_get([expected_revision_id]))[0] + + assert revision is not None + + assert isinstance(revision['metadata'], dict) + assert isinstance(revision['metadata']['intrinsic'], dict) + assert isinstance(revision['metadata']['extrinsic'], dict) + assert isinstance(revision['metadata']['original_artifact'], dict) + + check_metadata_paths(revision['metadata'], paths=[ + ('intrinsic.tool', str), + ('intrinsic.raw', dict), + ('extrinsic.provider', str), + ('extrinsic.when', str), + ('extrinsic.raw', dict), + ('original_artifact.filename', str), + ('original_artifact.length', int), + ('original_artifact.checksums', dict), + ]) + + def test_release_with_missing_artifact(swh_config, local_get_missing_one): """Load a pypi project with some missing artifacts ends up with 1 snapshot """ url = 'https://pypi.org/project/0805nexter' loader = PyPILoader(url) actual_load_status = loader.load() assert actual_load_status == {'status': 'eventful'} stats = loader.storage.stat_counters() assert { 'content': 3, 'directory': 2, 'origin': 1, 'origin_visit': 1, 'person': 1, 'release': 0, 'revision': 1, 'skipped_content': 0, 'snapshot': 1 } == stats expected_contents = map(hash_to_bytes, [ '405859113963cb7a797642b45f171d6360425d16', 'e5686aa568fdb1d19d7f1329267082fe40482d31', '83ecf6ec1114fd260ca7a833a2d165e71258c338', ]) assert list(loader.storage.content_missing_per_sha1(expected_contents))\ == [] expected_dirs = map(hash_to_bytes, [ 'b178b66bd22383d5f16f4f5c923d39ca798861b4', 'c3a58f8b57433a4b56caaa5033ae2e0931405338', ]) assert list(loader.storage.directory_missing(expected_dirs)) == [] # {revision hash: directory hash} expected_revs = { hash_to_bytes('e445da4da22b31bfebb6ffc4383dbf839a074d21'): hash_to_bytes('b178b66bd22383d5f16f4f5c923d39ca798861b4'), # noqa } assert list(loader.storage.revision_missing(expected_revs)) == [] expected_branches = { 'releases/1.2.0': { 'target': 'e445da4da22b31bfebb6ffc4383dbf839a074d21', 'target_type': 'revision', }, 'HEAD': { 'target': 'releases/1.2.0', 'target_type': 'alias', }, } expected_snapshot = { 'id': 'dd0e4201a232b1c104433741dbf45895b8ac9355', 'branches': expected_branches, } check_snapshot(expected_snapshot, storage=loader.storage) origin_visit = next(loader.storage.origin_visit_get(url)) assert origin_visit['status'] == 'partial' def test_release_artifact_no_prior_visit(swh_config, local_get): """With no prior visit, load a pypi project ends up with 1 snapshot """ url = 'https://pypi.org/project/0805nexter' loader = PyPILoader(url) actual_load_status = loader.load() assert actual_load_status == {'status': 'eventful'} stats = loader.storage.stat_counters() assert { 'content': 6, 'directory': 4, 'origin': 1, 'origin_visit': 1, 'person': 1, 'release': 0, 'revision': 2, 'skipped_content': 0, 'snapshot': 1 } == stats expected_contents = map(hash_to_bytes, [ 'a61e24cdfdab3bb7817f6be85d37a3e666b34566', '938c33483285fd8ad57f15497f538320df82aeb8', 'a27576d60e08c94a05006d2e6d540c0fdb5f38c8', '405859113963cb7a797642b45f171d6360425d16', 'e5686aa568fdb1d19d7f1329267082fe40482d31', '83ecf6ec1114fd260ca7a833a2d165e71258c338', ]) assert list(loader.storage.content_missing_per_sha1(expected_contents))\ == [] expected_dirs = map(hash_to_bytes, [ '05219ba38bc542d4345d5638af1ed56c7d43ca7d', 'cf019eb456cf6f78d8c4674596f1c9a97ece8f44', 'b178b66bd22383d5f16f4f5c923d39ca798861b4', 'c3a58f8b57433a4b56caaa5033ae2e0931405338', ]) assert list(loader.storage.directory_missing(expected_dirs)) == [] # {revision hash: directory hash} expected_revs = { hash_to_bytes('4c99891f93b81450385777235a37b5e966dd1571'): hash_to_bytes('05219ba38bc542d4345d5638af1ed56c7d43ca7d'), # noqa hash_to_bytes('e445da4da22b31bfebb6ffc4383dbf839a074d21'): hash_to_bytes('b178b66bd22383d5f16f4f5c923d39ca798861b4'), # noqa } assert list(loader.storage.revision_missing(expected_revs)) == [] expected_branches = { 'releases/1.1.0': { 'target': '4c99891f93b81450385777235a37b5e966dd1571', 'target_type': 'revision', }, 'releases/1.2.0': { 'target': 'e445da4da22b31bfebb6ffc4383dbf839a074d21', 'target_type': 'revision', }, 'HEAD': { 'target': 'releases/1.2.0', 'target_type': 'alias', }, } expected_snapshot = { 'id': 'ba6e158ada75d0b3cfb209ffdf6daa4ed34a227a', 'branches': expected_branches, } check_snapshot(expected_snapshot, loader.storage) origin_visit = next(loader.storage.origin_visit_get(url)) assert origin_visit['status'] == 'full' # release artifact, new artifact # {visit full, status full, new snapshot with shared history as prior snapshot} def test_release_artifact_incremental_visit(swh_config, local_get_visits): """With prior visit, 2nd load will result with a different snapshot with some shared history """ url = 'https://pypi.org/project/0805nexter' loader = PyPILoader(url) visit1_actual_load_status = loader.load() visit1_stats = loader.storage.stat_counters() assert visit1_actual_load_status == {'status': 'eventful'} origin_visit1 = next(loader.storage.origin_visit_get(url)) assert origin_visit1['status'] == 'full' assert { 'content': 6, 'directory': 4, 'origin': 1, 'origin_visit': 1, 'person': 1, 'release': 0, 'revision': 2, 'skipped_content': 0, 'snapshot': 1 } == visit1_stats # Reset internal state loader._info = None visit2_actual_load_status = loader.load() visit2_stats = loader.storage.stat_counters() assert visit2_actual_load_status == {'status': 'eventful'} visits = list(loader.storage.origin_visit_get(url)) assert len(visits) == 2 assert visits[1]['status'] == 'full' assert { 'content': 6 + 1, # 1 more content 'directory': 4 + 2, # 2 more directories 'origin': 1, 'origin_visit': 1 + 1, 'person': 1, 'release': 0, 'revision': 2 + 1, # 1 more revision 'skipped_content': 0, 'snapshot': 1 + 1, # 1 more snapshot } == visit2_stats expected_contents = map(hash_to_bytes, [ 'a61e24cdfdab3bb7817f6be85d37a3e666b34566', '938c33483285fd8ad57f15497f538320df82aeb8', 'a27576d60e08c94a05006d2e6d540c0fdb5f38c8', '405859113963cb7a797642b45f171d6360425d16', 'e5686aa568fdb1d19d7f1329267082fe40482d31', '83ecf6ec1114fd260ca7a833a2d165e71258c338', '92689fa2b7fb4d4fc6fb195bf73a50c87c030639' ]) assert list(loader.storage.content_missing_per_sha1(expected_contents))\ == [] expected_dirs = map(hash_to_bytes, [ '05219ba38bc542d4345d5638af1ed56c7d43ca7d', 'cf019eb456cf6f78d8c4674596f1c9a97ece8f44', 'b178b66bd22383d5f16f4f5c923d39ca798861b4', 'c3a58f8b57433a4b56caaa5033ae2e0931405338', 'e226e7e4ad03b4fc1403d69a18ebdd6f2edd2b3a', '52604d46843b898f5a43208045d09fcf8731631b', ]) assert list(loader.storage.directory_missing(expected_dirs)) == [] # {revision hash: directory hash} expected_revs = { hash_to_bytes('4c99891f93b81450385777235a37b5e966dd1571'): hash_to_bytes('05219ba38bc542d4345d5638af1ed56c7d43ca7d'), # noqa hash_to_bytes('e445da4da22b31bfebb6ffc4383dbf839a074d21'): hash_to_bytes('b178b66bd22383d5f16f4f5c923d39ca798861b4'), # noqa hash_to_bytes('51247143b01445c9348afa9edfae31bf7c5d86b1'): hash_to_bytes('e226e7e4ad03b4fc1403d69a18ebdd6f2edd2b3a'), # noqa } assert list(loader.storage.revision_missing(expected_revs)) == [] expected_branches = { 'releases/1.1.0': { 'target': '4c99891f93b81450385777235a37b5e966dd1571', 'target_type': 'revision', }, 'releases/1.2.0': { 'target': 'e445da4da22b31bfebb6ffc4383dbf839a074d21', 'target_type': 'revision', }, 'releases/1.3.0': { 'target': '51247143b01445c9348afa9edfae31bf7c5d86b1', 'target_type': 'revision', }, 'HEAD': { 'target': 'releases/1.3.0', 'target_type': 'alias', }, } expected_snapshot = { 'id': '2e5149a7b0725d18231a37b342e9b7c4e121f283', 'branches': expected_branches, } check_snapshot(expected_snapshot, loader.storage) origin_visit = list(loader.storage.origin_visit_get(url))[-1] assert origin_visit['status'] == 'full' urls = [ m.url for m in local_get_visits.request_history if m.url.startswith('https://files.pythonhosted.org') ] # visited each artifact once across 2 visits assert len(urls) == len(set(urls)) # release artifact, no new artifact # {visit full, status uneventful, same snapshot as before} # release artifact, old artifact with different checksums # {visit full, status full, new snapshot with shared history and some new # different history}