diff --git a/swh/loader/package/gnu.py b/swh/loader/package/gnu.py index d7bd940..12d765d 100644 --- a/swh/loader/package/gnu.py +++ b/swh/loader/package/gnu.py @@ -1,169 +1,188 @@ # 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 logging import re from os import path -from typing import Generator, Dict, Tuple, Sequence +from typing import Dict, Generator, Optional, Sequence, Tuple from swh.loader.package.loader import PackageLoader from swh.model.identifiers import normalize_timestamp +logger = logging.getLogger(__name__) + + # 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']))) + self.tarballs = list(sorted(tarballs, key=lambda v: v['time'])) 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 + artifact_version = get_version(url) + if version == artifact_version: + filename = path.split(url)[-1] + yield filename, url, a_metadata + + def resolve_revision_from( + self, known_artifacts: Dict, artifact_metadata: Dict) \ + -> Optional[bytes]: + def pk(d): + return [d.get(k) for k in ['time', 'archive', 'length']] + + artifact_pk = pk(artifact_metadata) + for rev_id, known_artifact in known_artifacts.items(): + logging.debug('known_artifact: %s', known_artifact) + known_pk = pk(known_artifact['extrinsic']['raw']) + if artifact_pk == known_pk: + return rev_id def build_revision( self, a_metadata: Dict, a_uncompressed_path: str) -> Dict: - normalized_date = normalize_timestamp(int(a_metadata['date'])) + normalized_date = normalize_timestamp(int(a_metadata['time'])) return { 'message': self.REVISION_MESSAGE, 'date': normalized_date, 'author': self.SWH_PERSON, 'committer': self.SWH_PERSON, 'committer_date': normalized_date, 'parents': [], 'metadata': { 'intrinsic': {}, 'extrinsic': { 'provider': self.url, 'when': self.visit_date.isoformat(), 'raw': a_metadata, }, }, } diff --git a/swh/loader/package/loader.py b/swh/loader/package/loader.py index bf9a61a..d4ebe83 100644 --- a/swh/loader/package/loader.py +++ b/swh/loader/package/loader.py @@ -1,361 +1,359 @@ # 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 = '' 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) 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 + 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 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_id = self.storage.origin_visit_add( origin=self.url, date=self.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, filename=a_filename) except Exception: logger.exception('Unable to retrieve %s', a_uri) 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) revision.update({ 'type': 'tar', 'synthetic': True, 'directory': directory.hash, }) revision['metadata'].update({ 'original_artifact': a_c_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 = ( version if version == 'HEAD' else '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 } 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) result = { 'status': status_load, } if snapshot: result['snapshot_id'] = snapshot['id'] return result diff --git a/swh/loader/package/npm.py b/swh/loader/package/npm.py index 48b2cb7..4aca83e 100644 --- a/swh/loader/package/npm.py +++ b/swh/loader/package/npm.py @@ -1,291 +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']: + original_artifact = known_artifact['original_artifact'] + if shasum == original_artifact['checksums']['sha1']: return rev_id def build_revision( self, a_metadata: Dict, a_uncompressed_path: 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': self.visit_date.isoformat(), 'raw': a_metadata, }, }, } diff --git a/swh/loader/package/pypi.py b/swh/loader/package/pypi.py index 7c4ccc1..0ecc10d 100644 --- a/swh/loader/package/pypi.py +++ b/swh/loader/package/pypi.py @@ -1,178 +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']: + original_artifact = known_artifact['original_artifact'] + if sha256 == original_artifact['checksums']['sha256']: return rev_id def build_revision( self, a_metadata: Dict, a_uncompressed_path: 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': self.visit_date.isoformat(), 'raw': a_metadata, }, } } diff --git a/swh/loader/package/tests/resources/ftp.gnu.org/gnu_8sync_8sync-0.2.0.tar.gz b/swh/loader/package/tests/resources/ftp.gnu.org/gnu_8sync_8sync-0.2.0.tar.gz new file mode 100644 index 0000000..ad9cbfa Binary files /dev/null and b/swh/loader/package/tests/resources/ftp.gnu.org/gnu_8sync_8sync-0.2.0.tar.gz differ diff --git a/swh/loader/package/tests/test_gnu.py b/swh/loader/package/tests/test_gnu.py index e00cdf3..13e5f06 100644 --- a/swh/loader/package/tests/test_gnu.py +++ b/swh/loader/package/tests/test_gnu.py @@ -1,244 +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 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, 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): +def test_visit_with_no_artifact_found(swh_config, requests_mock): package = '8sync' package_url = 'https://ftp.gnu.org/gnu/8sync/' tarballs = [{ - 'date': '944729610', + 'time': '944729610', 'archive': 'https://ftp.gnu.org/gnu/8sync/8sync-0.1.0.tar.gz', + 'length': 221837, }] 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): +def test_check_revision_metadata_structure(swh_config, local_get): package = '8sync' package_url = 'https://ftp.gnu.org/gnu/8sync/' tarballs = [{ - 'date': '944729610', + 'time': '944729610', 'archive': 'https://ftp.gnu.org/gnu/8sync/8sync-0.1.0.tar.gz', + 'length': 221837, }] 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 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): +def test_visit_with_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', + 'time': 944729610, 'archive': 'https://ftp.gnu.org/gnu/8sync/8sync-0.1.0.tar.gz', + 'length': 221837, }] 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) + + +def test_2_visits_without_change(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' + url = 'https://ftp.gnu.org/gnu/8sync/' + tarballs = [{ + 'time': 944729610, + 'archive': 'https://ftp.gnu.org/gnu/8sync/8sync-0.1.0.tar.gz', + 'length': 221837, + }] + + loader = GNULoader(package, url, tarballs) + 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' + + actual_load_status2 = loader.load() + assert actual_load_status2['status'] == 'uneventful' + origin_visit2 = list(loader.storage.origin_visit_get(url))[-1] + assert origin_visit2['status'] == 'full' + + urls = [ + m.url for m in local_get.request_history + if m.url.startswith('https://ftp.gnu.org') + ] + assert len(urls) == 1 + + +def test_2_visits_with_new_artifact(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' + url = 'https://ftp.gnu.org/gnu/8sync/' + tarball1 = { + 'time': 944729610, + 'archive': 'https://ftp.gnu.org/gnu/8sync/8sync-0.1.0.tar.gz', + 'length': 221837, + } + + loader = GNULoader(package, url, [tarball1]) + 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': 1, + 'release': 0, + 'revision': len(_expected_new_revisions_first_visit), + 'skipped_content': 0, + 'snapshot': 1 + } == stats + + urls = [ + m.url for m in local_get.request_history + if m.url.startswith('https://ftp.gnu.org') + ] + assert len(urls) == 1 + + tarball2 = { + 'time': 1480991830, + 'archive': 'https://ftp.gnu.org/gnu/8sync/8sync-0.2.0.tar.gz', + 'length': 238466, + } + loader2 = GNULoader(package, url, [tarball1, tarball2]) + # implementation detail: share the storage in between visits + loader2.storage = loader.storage + stats2 = loader2.storage.stat_counters() + assert stats == stats2 # ensure we share the storage + + actual_load_status2 = loader2.load() + assert actual_load_status2['status'] == 'eventful' + + stats2 = loader.storage.stat_counters() + assert { + 'content': len(_expected_new_contents_first_visit) + 14, + 'directory': len(_expected_new_directories_first_visit) + 8, + 'origin': 1, + 'origin_visit': 1 + 1, + 'person': 1, + 'release': 0, + 'revision': len(_expected_new_revisions_first_visit) + 1, + 'skipped_content': 0, + 'snapshot': 1 + 1, + } == stats2 + + origin_visit2 = list(loader.storage.origin_visit_get(url))[-1] + assert origin_visit2['status'] == 'full' + + urls = [ + m.url for m in local_get.request_history + if m.url.startswith('https://ftp.gnu.org') + ] + # 1 artifact (2nd time no modification) + 1 new artifact + assert len(urls) == 2