diff --git a/.pre-commit-config.yaml b/.pre-commit-config.yaml index 3e223ca..a459ab9 100644 --- a/.pre-commit-config.yaml +++ b/.pre-commit-config.yaml @@ -1,41 +1,41 @@ repos: - repo: https://github.com/pre-commit/pre-commit-hooks - rev: v4.1.0 + rev: v4.3.0 hooks: - id: trailing-whitespace - id: check-json - id: check-yaml - - repo: https://gitlab.com/pycqa/flake8 - rev: 4.0.1 + - repo: https://github.com/pycqa/flake8 + rev: 5.0.4 hooks: - id: flake8 - additional_dependencies: [flake8-bugbear==22.3.23] + additional_dependencies: [flake8-bugbear==22.9.23] - repo: https://github.com/codespell-project/codespell - rev: v2.1.0 + rev: v2.2.2 hooks: - id: codespell name: Check source code spelling exclude: ^(swh/indexer/data/codemeta/crosswalk.csv)$ stages: [commit] - repo: local hooks: - id: mypy name: mypy entry: mypy args: [swh] pass_filenames: false language: system types: [python] - repo: https://github.com/PyCQA/isort rev: 5.10.1 hooks: - id: isort - repo: https://github.com/python/black - rev: 22.3.0 + rev: 22.10.0 hooks: - id: black diff --git a/PKG-INFO b/PKG-INFO index 362f6ba..4ec4e91 100644 --- a/PKG-INFO +++ b/PKG-INFO @@ -1,71 +1,71 @@ Metadata-Version: 2.1 Name: swh.indexer -Version: 2.7.1 +Version: 2.7.2 Summary: Software Heritage Content Indexer Home-page: https://forge.softwareheritage.org/diffusion/78/ Author: Software Heritage developers Author-email: swh-devel@inria.fr Project-URL: Bug Reports, https://forge.softwareheritage.org/maniphest Project-URL: Funding, https://www.softwareheritage.org/donate Project-URL: Source, https://forge.softwareheritage.org/source/swh-indexer Project-URL: Documentation, https://docs.softwareheritage.org/devel/swh-indexer/ Classifier: Programming Language :: Python :: 3 Classifier: Intended Audience :: Developers Classifier: License :: OSI Approved :: GNU General Public License v3 (GPLv3) Classifier: Operating System :: OS Independent Classifier: Development Status :: 5 - Production/Stable Requires-Python: >=3.7 Description-Content-Type: text/markdown Provides-Extra: testing License-File: LICENSE License-File: AUTHORS swh-indexer ============ Tools to compute multiple indexes on SWH's raw contents: - content: - mimetype - ctags - language - fossology-license - metadata - revision: - metadata An indexer is in charge of: - looking up objects - extracting information from those objects - store those information in the swh-indexer db There are multiple indexers working on different object types: - content indexer: works with content sha1 hashes - revision indexer: works with revision sha1 hashes - origin indexer: works with origin identifiers Indexation procedure: - receive batch of ids - retrieve the associated data depending on object type - compute for that object some index - store the result to swh's storage Current content indexers: - mimetype (queue swh_indexer_content_mimetype): detect the encoding and mimetype - language (queue swh_indexer_content_language): detect the programming language - ctags (queue swh_indexer_content_ctags): compute tags information - fossology-license (queue swh_indexer_fossology_license): compute the license - metadata: translate file into translated_metadata dict Current revision indexers: - metadata: detects files containing metadata and retrieves translated_metadata in content_metadata table in storage or run content indexer to translate files. diff --git a/mypy.ini b/mypy.ini index d63e789..28c26fb 100644 --- a/mypy.ini +++ b/mypy.ini @@ -1,33 +1,36 @@ [mypy] namespace_packages = True warn_unused_ignores = True # 3rd party libraries without stubs (yet) [mypy-celery.*] ignore_missing_imports = True [mypy-confluent_kafka.*] ignore_missing_imports = True +[mypy-iso8601.*] +ignore_missing_imports = True + [mypy-magic.*] ignore_missing_imports = True [mypy-pkg_resources.*] ignore_missing_imports = True [mypy-psycopg2.*] ignore_missing_imports = True [mypy-pyld.*] ignore_missing_imports = True [mypy-pytest.*] ignore_missing_imports = True [mypy-rdflib.*] ignore_missing_imports = True [mypy-xmltodict.*] ignore_missing_imports = True diff --git a/requirements.txt b/requirements.txt index 4dd61a2..1cfc8ea 100644 --- a/requirements.txt +++ b/requirements.txt @@ -1,11 +1,12 @@ python-magic >= 0.4.13 click # frozendict: dependency of pyld # the version 2.1.2 is causing segmentation faults # cf https://forge.softwareheritage.org/T3815 frozendict != 2.1.2 +iso8601 pyld rdflib sentry-sdk typing-extensions xmltodict diff --git a/swh.indexer.egg-info/PKG-INFO b/swh.indexer.egg-info/PKG-INFO index 362f6ba..4ec4e91 100644 --- a/swh.indexer.egg-info/PKG-INFO +++ b/swh.indexer.egg-info/PKG-INFO @@ -1,71 +1,71 @@ Metadata-Version: 2.1 Name: swh.indexer -Version: 2.7.1 +Version: 2.7.2 Summary: Software Heritage Content Indexer Home-page: https://forge.softwareheritage.org/diffusion/78/ Author: Software Heritage developers Author-email: swh-devel@inria.fr Project-URL: Bug Reports, https://forge.softwareheritage.org/maniphest Project-URL: Funding, https://www.softwareheritage.org/donate Project-URL: Source, https://forge.softwareheritage.org/source/swh-indexer Project-URL: Documentation, https://docs.softwareheritage.org/devel/swh-indexer/ Classifier: Programming Language :: Python :: 3 Classifier: Intended Audience :: Developers Classifier: License :: OSI Approved :: GNU General Public License v3 (GPLv3) Classifier: Operating System :: OS Independent Classifier: Development Status :: 5 - Production/Stable Requires-Python: >=3.7 Description-Content-Type: text/markdown Provides-Extra: testing License-File: LICENSE License-File: AUTHORS swh-indexer ============ Tools to compute multiple indexes on SWH's raw contents: - content: - mimetype - ctags - language - fossology-license - metadata - revision: - metadata An indexer is in charge of: - looking up objects - extracting information from those objects - store those information in the swh-indexer db There are multiple indexers working on different object types: - content indexer: works with content sha1 hashes - revision indexer: works with revision sha1 hashes - origin indexer: works with origin identifiers Indexation procedure: - receive batch of ids - retrieve the associated data depending on object type - compute for that object some index - store the result to swh's storage Current content indexers: - mimetype (queue swh_indexer_content_mimetype): detect the encoding and mimetype - language (queue swh_indexer_content_language): detect the programming language - ctags (queue swh_indexer_content_ctags): compute tags information - fossology-license (queue swh_indexer_fossology_license): compute the license - metadata: translate file into translated_metadata dict Current revision indexers: - metadata: detects files containing metadata and retrieves translated_metadata in content_metadata table in storage or run content indexer to translate files. diff --git a/swh.indexer.egg-info/requires.txt b/swh.indexer.egg-info/requires.txt index e45d096..2b28940 100644 --- a/swh.indexer.egg-info/requires.txt +++ b/swh.indexer.egg-info/requires.txt @@ -1,24 +1,25 @@ python-magic>=0.4.13 click frozendict!=2.1.2 +iso8601 pyld rdflib sentry-sdk typing-extensions xmltodict swh.core[db,http]>=2.14.1 swh.model>=0.0.15 swh.objstorage>=0.2.2 swh.scheduler>=0.5.2 swh.storage>=0.22.0 swh.journal>=0.1.0 [testing] confluent-kafka hypothesis>=3.11.0 pytest pytest-mock swh.scheduler[testing]>=0.5.0 swh.storage[testing]>=0.10.0 types-click types-pyyaml diff --git a/swh/indexer/indexer.py b/swh/indexer/indexer.py index 65f730c..fbc0e1f 100644 --- a/swh/indexer/indexer.py +++ b/swh/indexer/indexer.py @@ -1,721 +1,731 @@ # Copyright (C) 2016-2022 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 abc from contextlib import contextmanager import logging import os import shutil import tempfile from typing import ( Any, Dict, Generic, Iterable, Iterator, List, Optional, Set, Tuple, TypeVar, Union, ) import warnings import sentry_sdk from typing_extensions import TypedDict from swh.core import utils from swh.core.config import load_from_envvar, merge_configs from swh.indexer.storage import INDEXER_CFG_KEY, Sha1, get_indexer_storage from swh.indexer.storage.interface import IndexerStorageInterface from swh.model import hashutil from swh.model.model import Directory, Origin, Sha1Git from swh.objstorage.exc import ObjNotFoundError from swh.objstorage.factory import get_objstorage from swh.scheduler import CONFIG as SWH_CONFIG from swh.storage import get_storage from swh.storage.interface import StorageInterface class ObjectsDict(TypedDict, total=False): """Typed objects whose keys are names of Kafka topics and values are list of values of messages in that topic.""" content: List[Dict] directory: List[Dict] origin: List[Dict] origin_visit_status: List[Dict] raw_extrinsic_metadata: List[Dict] @contextmanager def write_to_temp(filename: str, data: bytes, working_directory: str) -> Iterator[str]: """Write the sha1's content in a temporary file. Args: filename: one of sha1's many filenames data: the sha1's content to write in temporary file working_directory: the directory into which the file is written Returns: The path to the temporary file created. That file is filled in with the raw content's data. """ os.makedirs(working_directory, exist_ok=True) temp_dir = tempfile.mkdtemp(dir=working_directory) content_path = os.path.join(temp_dir, filename) with open(content_path, "wb") as f: f.write(data) yield content_path shutil.rmtree(temp_dir) DEFAULT_CONFIG = { INDEXER_CFG_KEY: {"cls": "memory"}, "storage": {"cls": "memory"}, "objstorage": {"cls": "memory"}, } TId = TypeVar("TId") """type of the ids of index()ed objects.""" TData = TypeVar("TData") """type of the objects passed to index().""" TResult = TypeVar("TResult") """return type of index()""" class BaseIndexer(Generic[TId, TData, TResult], metaclass=abc.ABCMeta): """Base class for indexers to inherit from. The main entry point is the :func:`run` function which is in charge of triggering the computations on the batch dict/ids received. Indexers can: - filter out ids whose data has already been indexed. - retrieve ids data from storage or objstorage - index this data depending on the object and store the result in storage. To implement a new object type indexer, inherit from the BaseIndexer and implement indexing: :meth:`~BaseIndexer.run`: object_ids are different depending on object. For example: sha1 for content, sha1_git for revision, directory, release, and id for origin To implement a new concrete indexer, inherit from the object level classes: :class:`ContentIndexer`, :class:`DirectoryIndexer`, :class:`OriginIndexer`. Then you need to implement the following functions: :meth:`~BaseIndexer.filter`: filter out data already indexed (in storage). :meth:`~BaseIndexer.index_object`: compute index on id with data (retrieved from the storage or the objstorage by the id key) and return the resulting index computation. :meth:`~BaseIndexer.persist_index_computations`: persist the results of multiple index computations in the storage. The new indexer implementation can also override the following functions: :meth:`~BaseIndexer.prepare`: Configuration preparation for the indexer. When overriding, this must call the `super().prepare()` instruction. :meth:`~BaseIndexer.check`: Configuration check for the indexer. When overriding, this must call the `super().check()` instruction. :meth:`~BaseIndexer.register_tools`: This should return a dict of the tool(s) to use when indexing or filtering. """ results: List[TResult] USE_TOOLS = True catch_exceptions = True """Prevents exceptions in `index()` from raising too high. Set to False in tests to properly catch all exceptions.""" scheduler: Any storage: StorageInterface objstorage: Any idx_storage: IndexerStorageInterface def __init__(self, config=None, **kw) -> None: """Prepare and check that the indexer is ready to run.""" super().__init__() if config is not None: self.config = config elif SWH_CONFIG: self.config = SWH_CONFIG.copy() else: self.config = load_from_envvar() self.config = merge_configs(DEFAULT_CONFIG, self.config) self.prepare() self.check() self.log.debug("%s: config=%s", self, self.config) def prepare(self) -> None: """Prepare the indexer's needed runtime configuration. Without this step, the indexer cannot possibly run. """ config_storage = self.config.get("storage") if config_storage: self.storage = get_storage(**config_storage) self.objstorage = get_objstorage(**self.config["objstorage"]) idx_storage = self.config[INDEXER_CFG_KEY] self.idx_storage = get_indexer_storage(**idx_storage) _log = logging.getLogger("requests.packages.urllib3.connectionpool") _log.setLevel(logging.WARN) self.log = logging.getLogger("swh.indexer") if self.USE_TOOLS: self.tools = list(self.register_tools(self.config.get("tools", []))) self.results = [] @property def tool(self) -> Dict: return self.tools[0] def check(self) -> None: """Check the indexer's configuration is ok before proceeding. If ok, does nothing. If not raise error. """ if self.USE_TOOLS and not self.tools: raise ValueError("Tools %s is unknown, cannot continue" % self.tools) def _prepare_tool(self, tool: Dict[str, Any]) -> Dict[str, Any]: """Prepare the tool dict to be compliant with the storage api.""" return {"tool_%s" % key: value for key, value in tool.items()} def register_tools( self, tools: Union[Dict[str, Any], List[Dict[str, Any]]] ) -> List[Dict[str, Any]]: """Permit to register tools to the storage. Add a sensible default which can be overridden if not sufficient. (For now, all indexers use only one tool) Expects the self.config['tools'] property to be set with one or more tools. Args: tools: Either a dict or a list of dict. Returns: list: List of dicts with additional id key. Raises: ValueError: if not a list nor a dict. """ if isinstance(tools, list): tools = list(map(self._prepare_tool, tools)) elif isinstance(tools, dict): tools = [self._prepare_tool(tools)] else: raise ValueError("Configuration tool(s) must be a dict or list!") if tools: return self.idx_storage.indexer_configuration_add(tools) else: return [] def index(self, id: TId, data: Optional[TData], **kwargs) -> List[TResult]: """Index computation for the id and associated raw data. Args: id: identifier or Dict object data: id's data from storage or objstorage depending on object type Returns: dict: a dict that makes sense for the :meth:`.persist_index_computations` method. """ raise NotImplementedError() def filter(self, ids: List[TId]) -> Iterator[TId]: """Filter missing ids for that particular indexer. Args: ids: list of ids Yields: iterator of missing ids """ yield from ids @abc.abstractmethod def persist_index_computations(self, results: List[TResult]) -> Dict[str, int]: """Persist the computation resulting from the index. Args: results: List of results. One result is the result of the index function. Returns: a summary dict of what has been inserted in the storage """ return {} def process_journal_objects(self, objects: ObjectsDict) -> Dict: """Read swh message objects (content, origin, ...) from the journal to: - retrieve the associated objects from the storage backend (e.g. storage, objstorage...) - execute the associated indexing computations - store the results in the indexer storage """ raise NotImplementedError() class ContentIndexer(BaseIndexer[Sha1, bytes, TResult], Generic[TResult]): """A content indexer working on the journal (method `process_journal_objects`) or on a list of ids directly (method `run`). To work on indexer partition, use the :class:`ContentPartitionIndexer` instead. Note: :class:`ContentIndexer` is not an instantiable object. To use it, one should inherit from this class and override the methods mentioned in the :class:`BaseIndexer` class. """ def process_journal_objects(self, objects: ObjectsDict) -> Dict: """Read content objects from the journal, retrieve their raw content and compute content indexing (e.g. mimetype, fossology license, ...). Note that once this is deployed, this supersedes the main ContentIndexer.run method call and the class ContentPartitionIndexer. """ summary: Dict[str, Any] = {"status": "uneventful"} try: results = [] contents = objects.get("content", []) # FIXME: with swh.objstorage > v2.0: self.objstorage.get_batch(contents) content_data = self.objstorage.get_batch(c["sha1"] for c in contents) for item, raw_content in zip(contents, content_data): id_ = item["sha1"] sentry_sdk.set_tag( "swh-indexer-content-sha1", hashutil.hash_to_hex(id_) ) if not raw_content: self.log.warning( "Content %s not found in objstorage", hashutil.hash_to_hex(id_), ) continue results.extend(self.index(id_, data=raw_content)) except Exception: if not self.catch_exceptions: raise self.log.exception("Problem when reading contents metadata.") sentry_sdk.capture_exception() summary["status"] = "failed" return summary + else: + # Reset tag after we finished processing the given content + sentry_sdk.set_tag("swh-indexer-content-sha1", "") summary_persist = self.persist_index_computations(results) self.results = results if summary_persist: for value in summary_persist.values(): if value > 0: summary["status"] = "eventful" summary.update(summary_persist) return summary def run(self, ids: List[Sha1], **kwargs) -> Dict: """Given a list of ids: - retrieve the content from the storage - execute the indexing computations - store the results Args: ids (Iterable[Union[bytes, str]]): sha1's identifier list **kwargs: passed to the `index` method Returns: A summary Dict of the task's status """ if "policy_update" in kwargs: warnings.warn( "'policy_update' argument is deprecated and ignored.", DeprecationWarning, ) del kwargs["policy_update"] sha1s = [ hashutil.hash_to_bytes(id_) if isinstance(id_, str) else id_ for id_ in ids ] results = [] summary: Dict = {"status": "uneventful"} try: for sha1 in sha1s: sentry_sdk.set_tag( "swh-indexer-content-sha1", hashutil.hash_to_hex(sha1) ) try: raw_content = self.objstorage.get(sha1) except ObjNotFoundError: self.log.warning( "Content %s not found in objstorage" % hashutil.hash_to_hex(sha1) ) continue res = self.index(sha1, raw_content, **kwargs) if res: # If no results, skip it results.extend(res) summary["status"] = "eventful" summary = self.persist_index_computations(results) self.results = results except Exception: if not self.catch_exceptions: raise self.log.exception("Problem when reading contents metadata.") sentry_sdk.capture_exception() summary["status"] = "failed" + else: + # Reset tag after we finished processing the given content + sentry_sdk.set_tag("swh-indexer-content-sha1", "") return summary class ContentPartitionIndexer(BaseIndexer[Sha1, bytes, TResult], Generic[TResult]): """A content partition indexer. This expects as input a partition_id and a nb_partitions. This will then index the contents within that partition. To work on a list of ids, use the :class:`ContentIndexer` instead. Note: :class:`ContentPartitionIndexer` is not an instantiable object. To use it, one should inherit from this class and override the methods mentioned in the :class:`BaseIndexer` class. """ @abc.abstractmethod def indexed_contents_in_partition( self, partition_id: int, nb_partitions: int ) -> Iterable[Sha1]: """Retrieve indexed contents within range [start, end]. Args: partition_id: Index of the partition to fetch nb_partitions: Total number of partitions to split into page_token: opaque token used for pagination """ pass def _list_contents_to_index( self, partition_id: int, nb_partitions: int, indexed: Set[Sha1] ) -> Iterable[Sha1]: """Compute from storage the new contents to index in the partition_id . The already indexed contents are skipped. Args: partition_id: Index of the partition to fetch data from nb_partitions: Total number of partition indexed: Set of content already indexed. Yields: Sha1 id (bytes) of contents to index """ if not isinstance(partition_id, int) or not isinstance(nb_partitions, int): raise TypeError( f"identifiers must be int, not {partition_id!r} and {nb_partitions!r}." ) next_page_token = None while True: result = self.storage.content_get_partition( partition_id, nb_partitions, page_token=next_page_token ) contents = result.results for c in contents: _id = hashutil.hash_to_bytes(c.sha1) if _id in indexed: continue yield _id next_page_token = result.next_page_token if next_page_token is None: break def _index_contents( self, partition_id: int, nb_partitions: int, indexed: Set[Sha1], **kwargs: Any ) -> Iterator[TResult]: """Index the contents within the partition_id. Args: start: Starting bound from range identifier end: End range identifier indexed: Set of content already indexed. Yields: indexing result as dict to persist in the indexer backend """ for sha1 in self._list_contents_to_index(partition_id, nb_partitions, indexed): try: raw_content = self.objstorage.get(sha1) except ObjNotFoundError: self.log.warning(f"Content {sha1.hex()} not found in objstorage") continue sentry_sdk.set_tag("swh-indexer-content-sha1", sha1) yield from self.index(sha1, raw_content, **kwargs) + sentry_sdk.set_tag("swh-indexer-content-sha1", "") def _index_with_skipping_already_done( self, partition_id: int, nb_partitions: int ) -> Iterator[TResult]: """Index not already indexed contents within the partition partition_id Args: partition_id: Index of the partition to fetch nb_partitions: Total number of partitions to split into Yields: indexing result as dict to persist in the indexer backend """ already_indexed_contents = set( self.indexed_contents_in_partition(partition_id, nb_partitions) ) return self._index_contents( partition_id, nb_partitions, already_indexed_contents ) def run( self, partition_id: int, nb_partitions: int, skip_existing: bool = True, **kwargs, ) -> Dict: """Given a partition of content ids, index the contents within. Either the indexer is incremental (filter out existing computed data) or it computes everything from scratch. Args: partition_id: Index of the partition to fetch nb_partitions: Total number of partitions to split into skip_existing: Skip existing indexed data (default) or not **kwargs: passed to the `index` method Returns: dict with the indexing task status """ summary: Dict[str, Any] = {"status": "uneventful"} count = 0 try: if skip_existing: gen = self._index_with_skipping_already_done( partition_id, nb_partitions ) else: gen = self._index_contents(partition_id, nb_partitions, indexed=set([])) count_object_added_key: Optional[str] = None for contents in utils.grouper(gen, n=self.config["write_batch_size"]): res = self.persist_index_computations(list(contents)) if not count_object_added_key: count_object_added_key = list(res.keys())[0] count += res[count_object_added_key] if count > 0: summary["status"] = "eventful" except Exception: if not self.catch_exceptions: raise self.log.exception("Problem when computing metadata.") sentry_sdk.capture_exception() summary["status"] = "failed" if count > 0 and count_object_added_key: summary[count_object_added_key] = count return summary class OriginIndexer(BaseIndexer[str, None, TResult], Generic[TResult]): """An object type indexer, inherits from the :class:`BaseIndexer` and implements Origin indexing using the run method Note: the :class:`OriginIndexer` is not an instantiable object. To use it in another context one should inherit from this class and override the methods mentioned in the :class:`BaseIndexer` class. """ def run(self, origin_urls: List[str], **kwargs) -> Dict: """Given a list of origin urls: - retrieve origins from storage - execute the indexing computations - store the results Args: origin_urls: list of origin urls. **kwargs: passed to the `index` method """ if "policy_update" in kwargs: warnings.warn( "'policy_update' argument is deprecated and ignored.", DeprecationWarning, ) del kwargs["policy_update"] origins = [{"url": url} for url in origin_urls] return self.process_journal_objects({"origin": origins}) def process_journal_objects(self, objects: ObjectsDict) -> Dict: """Worker function for ``JournalClient``.""" origins = [ Origin(url=status["origin"]) for status in objects.get("origin_visit_status", []) if status["status"] == "full" ] + [Origin(url=origin["url"]) for origin in objects.get("origin", [])] summary: Dict[str, Any] = {"status": "uneventful"} try: results = self.index_list( origins, # no need to check they exist, as we just received either an origin # or visit status; which cannot be created by swh-storage unless # the origin already exists check_origin_known=False, ) except Exception: if not self.catch_exceptions: raise self.log.exception("Problem when processing origins") sentry_sdk.capture_exception() summary["status"] = "failed" return summary summary_persist = self.persist_index_computations(results) self.results = results if summary_persist: for value in summary_persist.values(): if value > 0: summary["status"] = "eventful" summary.update(summary_persist) return summary def index_list(self, origins: List[Origin], **kwargs) -> List[TResult]: results = [] for origin in origins: sentry_sdk.set_tag("swh-indexer-origin-url", origin.url) results.extend(self.index(origin.url, **kwargs)) + sentry_sdk.set_tag("swh-indexer-origin-url", "") return results class DirectoryIndexer(BaseIndexer[Sha1Git, Directory, TResult], Generic[TResult]): """An object type indexer, inherits from the :class:`BaseIndexer` and implements Directory indexing using the run method Note: the :class:`DirectoryIndexer` is not an instantiable object. To use it in another context one should inherit from this class and override the methods mentioned in the :class:`BaseIndexer` class. """ def run(self, ids: List[Sha1Git], **kwargs) -> Dict: """Given a list of sha1_gits: - retrieve directories from storage - execute the indexing computations - store the results Args: ids: sha1_git's identifier list """ if "policy_update" in kwargs: warnings.warn( "'policy_update' argument is deprecated and ignored.", DeprecationWarning, ) del kwargs["policy_update"] directory_ids = [ hashutil.hash_to_bytes(id_) if isinstance(id_, str) else id_ for id_ in ids ] return self._process_directories([(dir_id, None) for dir_id in directory_ids]) def process_journal_objects(self, objects: ObjectsDict) -> Dict: """Worker function for ``JournalClient``.""" return self._process_directories( [ (dir_["id"], Directory.from_dict(dir_)) for dir_ in objects.get("directory", []) ] ) def _process_directories( self, directories: Union[List[Tuple[Sha1Git, Directory]], List[Tuple[Sha1Git, None]]], ) -> Dict: summary: Dict[str, Any] = {"status": "uneventful"} results = [] # TODO: fetch raw_manifest when useful? for (dir_id, dir_) in directories: swhid = f"swh:1:dir:{hashutil.hash_to_hex(dir_id)}" sentry_sdk.set_tag("swh-indexer-directory-swhid", swhid) try: results.extend(self.index(dir_id, dir_)) except Exception: if not self.catch_exceptions: raise self.log.exception("Problem when processing directory") sentry_sdk.capture_exception() summary["status"] = "failed" + else: + sentry_sdk.set_tag("swh-indexer-directory-swhid", "") summary_persist = self.persist_index_computations(results) if summary_persist: for value in summary_persist.values(): if value > 0: summary["status"] = "eventful" summary.update(summary_persist) self.results = results return summary diff --git a/swh/indexer/metadata.py b/swh/indexer/metadata.py index bcc31d3..f1e1e6a 100644 --- a/swh/indexer/metadata.py +++ b/swh/indexer/metadata.py @@ -1,561 +1,559 @@ # Copyright (C) 2017-2022 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 copy import deepcopy import itertools import logging import time from typing import ( Any, Callable, Dict, Iterable, Iterator, List, Optional, Tuple, TypeVar, cast, ) from urllib.parse import urlparse +import pkg_resources import sentry_sdk from swh.core.config import merge_configs from swh.core.utils import grouper from swh.indexer.codemeta import merge_documents from swh.indexer.indexer import ( BaseIndexer, ContentIndexer, DirectoryIndexer, ObjectsDict, OriginIndexer, ) from swh.indexer.metadata_detector import detect_metadata from swh.indexer.metadata_dictionary import EXTRINSIC_MAPPINGS, INTRINSIC_MAPPINGS from swh.indexer.metadata_dictionary.base import DirectoryLsEntry from swh.indexer.origin_head import get_head_swhid from swh.indexer.storage import INDEXER_CFG_KEY, Sha1 from swh.indexer.storage.model import ( ContentMetadataRow, DirectoryIntrinsicMetadataRow, OriginExtrinsicMetadataRow, OriginIntrinsicMetadataRow, ) from swh.model import hashutil from swh.model.model import Directory, MetadataAuthorityType from swh.model.model import ObjectType as ModelObjectType from swh.model.model import Origin, RawExtrinsicMetadata, Sha1Git from swh.model.swhids import CoreSWHID, ExtendedObjectType, ObjectType REVISION_GET_BATCH_SIZE = 10 RELEASE_GET_BATCH_SIZE = 10 ORIGIN_GET_BATCH_SIZE = 10 T1 = TypeVar("T1") T2 = TypeVar("T2") logger = logging.getLogger(__name__) def call_with_batches( f: Callable[[List[T1]], Iterable[T2]], args: List[T1], batch_size: int, ) -> Iterator[T2]: """Calls a function with batches of args, and concatenates the results.""" groups = grouper(args, batch_size) for group in groups: yield from f(list(group)) class ExtrinsicMetadataIndexer( BaseIndexer[Sha1Git, RawExtrinsicMetadata, OriginExtrinsicMetadataRow] ): def process_journal_objects(self, objects: ObjectsDict) -> Dict: summary: Dict[str, Any] = {"status": "uneventful"} try: results = {} for item in objects.get("raw_extrinsic_metadata", []): remd = RawExtrinsicMetadata.from_dict(item) sentry_sdk.set_tag("swh-indexer-remd-swhid", str(remd.swhid())) results[remd.target] = self.index(remd.id, data=remd) except Exception: if not self.catch_exceptions: raise summary["status"] = "failed" return summary self.results = list(itertools.chain.from_iterable(results.values())) summary_persist = self.persist_index_computations(self.results) if summary_persist: for value in summary_persist.values(): if value > 0: summary["status"] = "eventful" summary.update(summary_persist) return summary def index( self, id: Sha1Git, data: Optional[RawExtrinsicMetadata], **kwargs, ) -> List[OriginExtrinsicMetadataRow]: if data is None: raise NotImplementedError( "ExtrinsicMetadataIndexer.index() without RawExtrinsicMetadata data" ) if data.target.object_type != ExtendedObjectType.ORIGIN: # other types are not supported yet return [] if data.authority.type == MetadataAuthorityType.REGISTRY: # metadata provided by a third-party; don't trust it # (technically this could be handled below, but we check it here # to return early; sparing a translation and origin lookup) # TODO: add ways to define trusted authorities return [] metadata_items = [] mappings: List[str] = [] for mapping_cls in EXTRINSIC_MAPPINGS.values(): if data.format in mapping_cls.extrinsic_metadata_formats(): mapping = mapping_cls() metadata_item = mapping.translate(data.metadata) if metadata_item is not None: metadata_items.append(metadata_item) mappings.append(mapping.name) if not metadata_items: # Don't have any mapping to parse it, ignore return [] # TODO: batch requests to origin_get_by_sha1() for _ in range(6): origins = self.storage.origin_get_by_sha1([data.target.object_id]) try: (origin,) = origins if origin is not None: break except ValueError: pass # The origin does not exist. This may be due to some replication lag # between the loader's DB/journal and the DB we are consuming from. # Wait a bit and try again logger.debug("Origin %s not found, sleeping for 10s.", data.target) time.sleep(10) else: # Does not exist, or replication lag > 60s. raise ValueError(f"Unknown origin {data.target}") from None if urlparse(data.authority.url).netloc != urlparse(origin["url"]).netloc: # metadata provided by a third-party; don't trust it # TODO: add ways to define trusted authorities return [] metadata = merge_documents(metadata_items) return [ OriginExtrinsicMetadataRow( id=origin["url"], indexer_configuration_id=self.tool["id"], from_remd_id=data.id, mappings=mappings, metadata=metadata, ) ] def persist_index_computations( self, results: List[OriginExtrinsicMetadataRow] ) -> Dict[str, int]: """Persist the results in storage.""" return self.idx_storage.origin_extrinsic_metadata_add(results) class ContentMetadataIndexer(ContentIndexer[ContentMetadataRow]): """Content-level indexer This indexer is in charge of: - filtering out content already indexed in content_metadata - reading content from objstorage with the content's id sha1 - computing metadata by given context - using the metadata_dictionary as the 'swh-metadata-translator' tool - store result in content_metadata table """ def filter(self, ids): """Filter out known sha1s and return only missing ones.""" yield from self.idx_storage.content_metadata_missing( ( { "id": sha1, "indexer_configuration_id": self.tool["id"], } for sha1 in ids ) ) def index( self, id: Sha1, data: Optional[bytes] = None, log_suffix="unknown directory", **kwargs, ) -> List[ContentMetadataRow]: """Index sha1s' content and store result. Args: id: content's identifier data: raw content in bytes Returns: dict: dictionary representing a content_metadata. If the translation wasn't successful the metadata keys will be returned as None """ assert isinstance(id, bytes) assert data is not None metadata = None try: mapping_name = self.tool["tool_configuration"]["context"] log_suffix += ", content_id=%s" % hashutil.hash_to_hex(id) metadata = INTRINSIC_MAPPINGS[mapping_name](log_suffix).translate(data) except Exception: self.log.exception( "Problem during metadata translation " "for content %s" % hashutil.hash_to_hex(id) ) sentry_sdk.capture_exception() if metadata is None: return [] return [ ContentMetadataRow( id=id, indexer_configuration_id=self.tool["id"], metadata=metadata, ) ] def persist_index_computations( self, results: List[ContentMetadataRow] ) -> Dict[str, int]: """Persist the results in storage.""" return self.idx_storage.content_metadata_add(results) DEFAULT_CONFIG: Dict[str, Any] = { "tools": { - "name": "swh-metadata-detector", - "version": "0.0.2", + "name": "swh.indexer.metadata", + "version": pkg_resources.get_distribution("swh.indexer").version, "configuration": {}, }, } class DirectoryMetadataIndexer(DirectoryIndexer[DirectoryIntrinsicMetadataRow]): """Directory-level indexer This indexer is in charge of: - filtering directories already indexed in directory_intrinsic_metadata table with defined computation tool - retrieve all entry_files in directory - use metadata_detector for file_names containing metadata - compute metadata translation if necessary and possible (depends on tool) - send sha1s to content indexing if possible - store the results for directory """ def __init__(self, *args, **kwargs): super().__init__(*args, **kwargs) self.config = merge_configs(DEFAULT_CONFIG, self.config) def filter(self, sha1_gits): """Filter out known sha1s and return only missing ones.""" yield from self.idx_storage.directory_intrinsic_metadata_missing( ( { "id": sha1_git, "indexer_configuration_id": self.tool["id"], } for sha1_git in sha1_gits ) ) def index( self, id: Sha1Git, data: Optional[Directory] = None, **kwargs ) -> List[DirectoryIntrinsicMetadataRow]: """Index directory by processing it and organizing result. use metadata_detector to iterate on filenames, passes them to the content indexers, then merges (if more than one) Args: id: sha1_git of the directory data: should always be None Returns: dict: dictionary representing a directory_intrinsic_metadata, with keys: - id: directory's identifier (sha1_git) - indexer_configuration_id (bytes): tool used - metadata: dict of retrieved metadata """ dir_: List[DirectoryLsEntry] assert data is None, "Unexpected directory object" dir_ = cast( List[DirectoryLsEntry], list(self.storage.directory_ls(id, recursive=False)), ) try: if [entry["type"] for entry in dir_] == ["dir"]: # If the root is just a single directory, recurse into it # eg. PyPI packages, GNU tarballs subdir = dir_[0]["target"] dir_ = cast( List[DirectoryLsEntry], list(self.storage.directory_ls(subdir, recursive=False)), ) files = [entry for entry in dir_ if entry["type"] == "file"] (mappings, metadata) = self.translate_directory_intrinsic_metadata( files, log_suffix="directory=%s" % hashutil.hash_to_hex(id), ) except Exception as e: self.log.exception("Problem when indexing dir: %r", e) sentry_sdk.capture_exception() return [] return [ DirectoryIntrinsicMetadataRow( id=id, indexer_configuration_id=self.tool["id"], mappings=mappings, metadata=metadata, ) ] def persist_index_computations( self, results: List[DirectoryIntrinsicMetadataRow] ) -> Dict[str, int]: """Persist the results in storage.""" # TODO: add functions in storage to keep data in # directory_intrinsic_metadata return self.idx_storage.directory_intrinsic_metadata_add(results) def translate_directory_intrinsic_metadata( self, files: List[DirectoryLsEntry], log_suffix: str ) -> Tuple[List[Any], Any]: """ Determine plan of action to translate metadata in the given root directory Args: files: list of file entries, as returned by :meth:`swh.storage.interface.StorageInterface.directory_ls` Returns: (List[str], dict): list of mappings used and dict with translated metadata according to the CodeMeta vocabulary """ metadata = [] - tool = { - "name": "swh-metadata-translator", - "version": "0.0.2", - "configuration": {}, - } # TODO: iterate on each context, on each file # -> get raw_contents # -> translate each content - config = {k: self.config[k] for k in [INDEXER_CFG_KEY, "objstorage", "storage"]} - config["tools"] = [tool] + config = { + k: self.config[k] + for k in [INDEXER_CFG_KEY, "objstorage", "storage", "tools"] + } all_detected_files = detect_metadata(files) used_mappings = [ INTRINSIC_MAPPINGS[context].name for context in all_detected_files ] for (mapping_name, detected_files) in all_detected_files.items(): cfg = deepcopy(config) - cfg["tools"][0]["configuration"]["context"] = mapping_name + cfg["tools"]["configuration"]["context"] = mapping_name c_metadata_indexer = ContentMetadataIndexer(config=cfg) # sha1s that are in content_metadata table sha1s_in_storage = [] metadata_generator = self.idx_storage.content_metadata_get(detected_files) for c in metadata_generator: # extracting metadata sha1 = c.id sha1s_in_storage.append(sha1) local_metadata = c.metadata # local metadata is aggregated if local_metadata: metadata.append(local_metadata) sha1s_filtered = [ item for item in detected_files if item not in sha1s_in_storage ] if sha1s_filtered: # content indexing try: c_metadata_indexer.run( sha1s_filtered, log_suffix=log_suffix, ) # on the fly possibility: for result in c_metadata_indexer.results: local_metadata = result.metadata metadata.append(local_metadata) except Exception: self.log.exception("Exception while indexing metadata on contents") sentry_sdk.capture_exception() metadata = merge_documents(metadata) return (used_mappings, metadata) class OriginMetadataIndexer( OriginIndexer[Tuple[OriginIntrinsicMetadataRow, DirectoryIntrinsicMetadataRow]] ): USE_TOOLS = False def __init__(self, config=None, **kwargs) -> None: super().__init__(config=config, **kwargs) self.directory_metadata_indexer = DirectoryMetadataIndexer(config=config) def index_list( self, origins: List[Origin], *, check_origin_known: bool = True, **kwargs, ) -> List[Tuple[OriginIntrinsicMetadataRow, DirectoryIntrinsicMetadataRow]]: head_rev_ids = [] head_rel_ids = [] origin_heads: Dict[Origin, CoreSWHID] = {} # Filter out origins not in the storage if check_origin_known: known_origins = list( call_with_batches( self.storage.origin_get, [origin.url for origin in origins], ORIGIN_GET_BATCH_SIZE, ) ) else: known_origins = list(origins) for origin in known_origins: if origin is None: continue head_swhid = get_head_swhid(self.storage, origin.url) if head_swhid: origin_heads[origin] = head_swhid if head_swhid.object_type == ObjectType.REVISION: head_rev_ids.append(head_swhid.object_id) elif head_swhid.object_type == ObjectType.RELEASE: head_rel_ids.append(head_swhid.object_id) else: assert False, head_swhid head_revs = dict( zip( head_rev_ids, call_with_batches( self.storage.revision_get, head_rev_ids, REVISION_GET_BATCH_SIZE ), ) ) head_rels = dict( zip( head_rel_ids, call_with_batches( self.storage.release_get, head_rel_ids, RELEASE_GET_BATCH_SIZE ), ) ) results = [] for (origin, head_swhid) in origin_heads.items(): sentry_sdk.set_tag("swh-indexer-origin-url", origin.url) sentry_sdk.set_tag("swh-indexer-origin-head-swhid", str(head_swhid)) if head_swhid.object_type == ObjectType.REVISION: rev = head_revs[head_swhid.object_id] if not rev: self.log.warning( "Missing head object %s of origin %r", head_swhid, origin.url ) continue directory_id = rev.directory elif head_swhid.object_type == ObjectType.RELEASE: rel = head_rels[head_swhid.object_id] if not rel: self.log.warning( "Missing head object %s of origin %r", head_swhid, origin.url ) continue if rel.target_type != ModelObjectType.DIRECTORY: # TODO self.log.warning( "Head release %s of %r has unexpected target type %s", head_swhid, origin.url, rel.target_type, ) continue assert rel.target, rel directory_id = rel.target else: assert False, head_swhid for dir_metadata in self.directory_metadata_indexer.index(directory_id): # There is at most one dir_metadata orig_metadata = OriginIntrinsicMetadataRow( from_directory=dir_metadata.id, id=origin.url, metadata=dir_metadata.metadata, mappings=dir_metadata.mappings, indexer_configuration_id=dir_metadata.indexer_configuration_id, ) results.append((orig_metadata, dir_metadata)) return results def persist_index_computations( self, results: List[Tuple[OriginIntrinsicMetadataRow, DirectoryIntrinsicMetadataRow]], ) -> Dict[str, int]: # Deduplicate directories dir_metadata: Dict[bytes, DirectoryIntrinsicMetadataRow] = {} orig_metadata: Dict[str, OriginIntrinsicMetadataRow] = {} summary: Dict = {} for (orig_item, dir_item) in results: assert dir_item.metadata == orig_item.metadata if dir_item.metadata and not (dir_item.metadata.keys() <= {"@context"}): # Only store non-empty metadata sets if dir_item.id not in dir_metadata: dir_metadata[dir_item.id] = dir_item if orig_item.id not in orig_metadata: orig_metadata[orig_item.id] = orig_item if dir_metadata: summary_dir = self.idx_storage.directory_intrinsic_metadata_add( list(dir_metadata.values()) ) summary.update(summary_dir) if orig_metadata: summary_ori = self.idx_storage.origin_intrinsic_metadata_add( list(orig_metadata.values()) ) summary.update(summary_ori) return summary diff --git a/swh/indexer/metadata_dictionary/base.py b/swh/indexer/metadata_dictionary/base.py index b4e781a..7d63e0e 100644 --- a/swh/indexer/metadata_dictionary/base.py +++ b/swh/indexer/metadata_dictionary/base.py @@ -1,405 +1,404 @@ # Copyright (C) 2017-2022 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 from typing import Any, Callable, Dict, List, Optional, Tuple, TypeVar -import urllib.parse import uuid import xml.parsers.expat from pyld import jsonld import rdflib from typing_extensions import TypedDict import xmltodict import yaml from swh.indexer.codemeta import _document_loader, compact from swh.indexer.namespaces import RDF, SCHEMA from swh.indexer.storage.interface import Sha1 +from .utils import add_url_if_valid + TMP_ROOT_URI_PREFIX = "https://www.softwareheritage.org/schema/2022/indexer/tmp-node/" """Prefix used to generate temporary URIs for root nodes being translated.""" class DirectoryLsEntry(TypedDict): target: Sha1 sha1: Sha1 name: bytes type: str TTranslateCallable = TypeVar( "TTranslateCallable", bound=Callable[[Any, rdflib.Graph, rdflib.term.BNode, Any], None], ) def produce_terms(*uris: str) -> Callable[[TTranslateCallable], TTranslateCallable]: """Returns a decorator that marks the decorated function as adding the given terms to the ``translated_metadata`` dict""" def decorator(f: TTranslateCallable) -> TTranslateCallable: if not hasattr(f, "produced_terms"): f.produced_terms = [] # type: ignore f.produced_terms.extend(uris) # type: ignore return f return decorator class BaseMapping: """Base class for :class:`BaseExtrinsicMapping` and :class:`BaseIntrinsicMapping`, not to be inherited directly.""" def __init__(self, log_suffix=""): self.log_suffix = log_suffix self.log = logging.getLogger( "%s.%s" % (self.__class__.__module__, self.__class__.__name__) ) @property def name(self): """A name of this mapping, used as an identifier in the indexer storage.""" raise NotImplementedError(f"{self.__class__.__name__}.name") def translate(self, raw_content: bytes) -> Optional[Dict]: """ Translates content by parsing content from a bytestring containing mapping-specific data and translating with the appropriate mapping to JSON-LD using the Codemeta and ForgeFed vocabularies. Args: raw_content: raw content to translate Returns: translated metadata in JSON friendly form needed for the content if parseable, :const:`None` otherwise. """ raise NotImplementedError(f"{self.__class__.__name__}.translate") def normalize_translation(self, metadata: Dict[str, Any]) -> Dict[str, Any]: raise NotImplementedError(f"{self.__class__.__name__}.normalize_translation") class BaseExtrinsicMapping(BaseMapping): """Base class for extrinsic_metadata mappings to inherit from To implement a new mapping: - inherit this class - override translate function """ @classmethod def extrinsic_metadata_formats(cls) -> Tuple[str, ...]: """ Returns the list of extrinsic metadata formats which can be translated by this mapping """ raise NotImplementedError(f"{cls.__name__}.extrinsic_metadata_formats") def normalize_translation(self, metadata: Dict[str, Any]) -> Dict[str, Any]: return compact(metadata, forgefed=True) class BaseIntrinsicMapping(BaseMapping): """Base class for intrinsic-metadata mappings to inherit from To implement a new mapping: - inherit this class - override translate function """ @classmethod def detect_metadata_files(cls, file_entries: List[DirectoryLsEntry]) -> List[Sha1]: """ Returns the sha1 hashes of files which can be translated by this mapping """ raise NotImplementedError(f"{cls.__name__}.detect_metadata_files") def normalize_translation(self, metadata: Dict[str, Any]) -> Dict[str, Any]: return compact(metadata, forgefed=False) class SingleFileIntrinsicMapping(BaseIntrinsicMapping): """Base class for all intrinsic metadata mappings that use a single file as input.""" @property def filename(self): """The .json file to extract metadata from.""" raise NotImplementedError(f"{self.__class__.__name__}.filename") @classmethod def detect_metadata_files(cls, file_entries: List[DirectoryLsEntry]) -> List[Sha1]: for entry in file_entries: if entry["name"].lower() == cls.filename: return [entry["sha1"]] return [] class DictMapping(BaseMapping): """Base class for mappings that take as input a file that is mostly a key-value store (eg. a shallow JSON dict).""" string_fields: List[str] = [] """List of fields that are simple strings, and don't need any normalization.""" date_fields: List[str] = [] """List of fields that are strings that should be typed as http://schema.org/Date """ uri_fields: List[str] = [] """List of fields that are simple URIs, and don't need any normalization.""" @property def mapping(self): """A translation dict to map dict keys into a canonical name.""" raise NotImplementedError(f"{self.__class__.__name__}.mapping") @staticmethod def _normalize_method_name(name: str) -> str: return name.replace("-", "_") @classmethod def supported_terms(cls): # one-to-one mapping from the original key to a CodeMeta term simple_terms = { str(term) for (key, term) in cls.mapping.items() if key in cls.string_fields + cls.date_fields + cls.uri_fields or hasattr(cls, "normalize_" + cls._normalize_method_name(key)) } # more complex mapping from the original key to JSON-LD complex_terms = { str(term) for meth_name in dir(cls) if meth_name.startswith("translate_") for term in getattr(getattr(cls, meth_name), "produced_terms", []) } return simple_terms | complex_terms def get_root_uri(self, content_dict: Dict) -> rdflib.URIRef: """Returns an URI for the SoftwareSourceCode or Repository being described. The default implementation uses a temporary URI that is stripped before normalization by :meth:`_translate_dict`. """ # The main object being described (the SoftwareSourceCode) does not necessarily # may or may not have an id. # If it does, it will need to be set by a subclass. # If it doesn't we temporarily use this URI to identify it. Unfortunately, # we cannot use a blank node as we need to use it for JSON-LD framing later, # and blank nodes cannot be used for framing in JSON-LD >= 1.1 root_id = TMP_ROOT_URI_PREFIX + str(uuid.uuid4()) return rdflib.URIRef(root_id) def _translate_dict(self, content_dict: Dict) -> Dict[str, Any]: """ Translates content by parsing content from a dict object and translating with the appropriate mapping Args: content_dict (dict): content dict to translate Returns: dict: translated metadata in json-friendly form needed for the indexer """ graph = rdflib.Graph() root = self.get_root_uri(content_dict) self._translate_to_graph(graph, root, content_dict) self.sanitize(graph) # Convert from rdflib's internal graph representation to JSON s = graph.serialize(format="application/ld+json") # Load from JSON to a list of Python objects jsonld_graph = json.loads(s) # Use JSON-LD framing to turn the graph into a rooted tree # frame = {"@type": str(SCHEMA.SoftwareSourceCode)} translated_metadata = jsonld.frame( jsonld_graph, {"@id": str(root)}, options={ "documentLoader": _document_loader, "processingMode": "json-ld-1.1", }, ) # Remove the temporary id we added at the beginning assert isinstance(translated_metadata["@id"], str) if translated_metadata["@id"].startswith(TMP_ROOT_URI_PREFIX): del translated_metadata["@id"] return self.normalize_translation(translated_metadata) def _translate_to_graph( self, graph: rdflib.Graph, root: rdflib.term.Identifier, content_dict: Dict ) -> None: """ Translates content by parsing content from a dict object and translating with the appropriate mapping to the graph passed as parameter Args: content_dict (dict): content dict to translate """ graph.add((root, RDF.type, SCHEMA.SoftwareSourceCode)) for k, v in content_dict.items(): # First, check if there is a specific translation # method for this key translation_method = getattr( self, "translate_" + self._normalize_method_name(k), None ) if translation_method: translation_method(graph, root, v) elif k in self.mapping: # if there is no method, but the key is known from the # crosswalk table codemeta_key = self.mapping[k] # if there is a normalization method, use it on the value, # and add its results to the triples normalization_method = getattr( self, "normalize_" + self._normalize_method_name(k), None ) if normalization_method: v = normalization_method(v) if v is None: pass elif isinstance(v, list): for item in reversed(v): - graph.add((root, codemeta_key, item)) + if isinstance(item, rdflib.URIRef): + add_url_if_valid(graph, root, codemeta_key, str(item)) + else: + graph.add((root, codemeta_key, item)) else: - graph.add((root, codemeta_key, v)) + if isinstance(v, rdflib.URIRef): + add_url_if_valid(graph, root, codemeta_key, str(v)) + else: + graph.add((root, codemeta_key, v)) elif k in self.string_fields and isinstance(v, str): graph.add((root, codemeta_key, rdflib.Literal(v))) elif k in self.string_fields and isinstance(v, list): for item in v: graph.add((root, codemeta_key, rdflib.Literal(item))) elif k in self.date_fields and isinstance(v, str): typed_v = rdflib.Literal(v, datatype=SCHEMA.Date) graph.add((root, codemeta_key, typed_v)) elif k in self.date_fields and isinstance(v, list): for item in v: if isinstance(item, str): typed_item = rdflib.Literal(item, datatype=SCHEMA.Date) graph.add((root, codemeta_key, typed_item)) elif k in self.uri_fields and isinstance(v, str): - # Workaround for https://github.com/digitalbazaar/pyld/issues/91 : drop - # URLs that are blatantly invalid early, so PyLD does not crash. - parsed_url = urllib.parse.urlparse(v) - if parsed_url.netloc: - graph.add((root, codemeta_key, rdflib.URIRef(v))) + add_url_if_valid(graph, root, codemeta_key, v) elif k in self.uri_fields and isinstance(v, list): for item in v: - if isinstance(item, str): - # ditto - parsed_url = urllib.parse.urlparse(item) - if parsed_url.netloc: - graph.add((root, codemeta_key, rdflib.URIRef(item))) + add_url_if_valid(graph, root, codemeta_key, item) else: continue self.extra_translation(graph, root, content_dict) def sanitize(self, graph: rdflib.Graph) -> None: # Remove triples that make PyLD crash for (subject, predicate, _) in graph.triples((None, None, rdflib.URIRef(""))): graph.remove((subject, predicate, rdflib.URIRef(""))) # Should not happen, but we's better check as this may lead to incorrect data invalid = False for triple in graph.triples((rdflib.URIRef(""), None, None)): invalid = True logging.error("Empty triple subject URI: %r", triple) if invalid: raise ValueError("Empty triple subject(s)") def extra_translation( self, graph: rdflib.Graph, root: rdflib.term.Node, d: Dict[str, Any] ) -> None: """Called at the end of the translation process, and may add arbitrary triples to ``graph`` based on the input dictionary (passed as ``d``). """ pass class JsonMapping(DictMapping): """Base class for all mappings that use JSON data as input.""" def translate(self, raw_content: bytes) -> Optional[Dict]: try: raw_content_string: str = raw_content.decode() except UnicodeDecodeError: self.log.warning("Error unidecoding from %s", self.log_suffix) return None try: content_dict = json.loads(raw_content_string) except json.JSONDecodeError: self.log.warning("Error unjsoning from %s", self.log_suffix) return None if isinstance(content_dict, dict): return self._translate_dict(content_dict) return None class XmlMapping(DictMapping): """Base class for all mappings that use XML data as input.""" def translate(self, raw_content: bytes) -> Optional[Dict]: try: d = xmltodict.parse(raw_content) except xml.parsers.expat.ExpatError: self.log.warning("Error parsing XML from %s", self.log_suffix) return None except UnicodeDecodeError: self.log.warning("Error unidecoding XML from %s", self.log_suffix) return None except (LookupError, ValueError): # unknown encoding or multi-byte encoding self.log.warning("Error detecting XML encoding from %s", self.log_suffix) return None if not isinstance(d, dict): self.log.warning("Skipping ill-formed XML content: %s", raw_content) return None return self._translate_dict(d) class SafeLoader(yaml.SafeLoader): yaml_implicit_resolvers = { k: [r for r in v if r[0] != "tag:yaml.org,2002:timestamp"] for k, v in yaml.SafeLoader.yaml_implicit_resolvers.items() } class YamlMapping(DictMapping, SingleFileIntrinsicMapping): """Base class for all mappings that use Yaml data as input.""" def translate(self, raw_content: bytes) -> Optional[Dict[str, str]]: raw_content_string: str = raw_content.decode() try: content_dict = yaml.load(raw_content_string, Loader=SafeLoader) except yaml.scanner.ScannerError: return None if isinstance(content_dict, dict): return self._translate_dict(content_dict) return None diff --git a/swh/indexer/metadata_dictionary/cff.py b/swh/indexer/metadata_dictionary/cff.py index 8ced23e..78ba661 100644 --- a/swh/indexer/metadata_dictionary/cff.py +++ b/swh/indexer/metadata_dictionary/cff.py @@ -1,60 +1,65 @@ # Copyright (C) 2021-2022 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 typing import List +import urllib.parse from rdflib import BNode, Graph, Literal, URIRef import rdflib.term from swh.indexer.codemeta import CROSSWALK_TABLE from swh.indexer.namespaces import RDF, SCHEMA from .base import YamlMapping from .utils import add_map DOI = URIRef("https://doi.org/") SPDX = URIRef("https://spdx.org/licenses/") class CffMapping(YamlMapping): """Dedicated class for Citation (CITATION.cff) mapping and translation""" name = "cff" filename = b"CITATION.cff" mapping = CROSSWALK_TABLE["Citation File Format Core (CFF-Core) 1.0.2"] string_fields = ["keywords", "license", "abstract", "version", "doi"] date_fields = ["date-released"] uri_fields = ["repository-code"] def _translate_author(self, graph: Graph, author: dict) -> rdflib.term.Node: node: rdflib.term.Node - if "orcid" in author and isinstance(author["orcid"], str): + if ( + "orcid" in author + and isinstance(author["orcid"], str) + and urllib.parse.urlparse(author["orcid"]).netloc + ): node = URIRef(author["orcid"]) else: node = BNode() graph.add((node, RDF.type, SCHEMA.Person)) if "affiliation" in author and isinstance(author["affiliation"], str): affiliation = BNode() graph.add((node, SCHEMA.affiliation, affiliation)) graph.add((affiliation, RDF.type, SCHEMA.Organization)) graph.add((affiliation, SCHEMA.name, Literal(author["affiliation"]))) if "family-names" in author and isinstance(author["family-names"], str): graph.add((node, SCHEMA.familyName, Literal(author["family-names"]))) if "given-names" in author and isinstance(author["given-names"], str): graph.add((node, SCHEMA.givenName, Literal(author["given-names"]))) return node def translate_authors( self, graph: Graph, root: URIRef, authors: List[dict] ) -> None: add_map(graph, root, SCHEMA.author, self._translate_author, authors) def normalize_doi(self, s: str) -> URIRef: if isinstance(s, str): return DOI + s def normalize_license(self, s: str) -> URIRef: if isinstance(s, str): return SPDX + s diff --git a/swh/indexer/metadata_dictionary/codemeta.py b/swh/indexer/metadata_dictionary/codemeta.py index 2ceee00..7472123 100644 --- a/swh/indexer/metadata_dictionary/codemeta.py +++ b/swh/indexer/metadata_dictionary/codemeta.py @@ -1,156 +1,168 @@ # Copyright (C) 2018-2022 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 collections import json import re -from typing import Any, Dict, List, Optional, Tuple +from typing import Any, Dict, List, Optional, Tuple, Union import xml.etree.ElementTree as ET +import iso8601 import xmltodict from swh.indexer.codemeta import CODEMETA_CONTEXT_URL, CODEMETA_TERMS, compact, expand from .base import BaseExtrinsicMapping, SingleFileIntrinsicMapping ATOM_URI = "http://www.w3.org/2005/Atom" _TAG_RE = re.compile(r"\{(?P.*?)\}(?P.*)") _IGNORED_NAMESPACES = ("http://www.w3.org/2005/Atom",) +_DATE_RE = re.compile("^[0-9]{4}-[0-9]{1,2}-[0-9]{1,2}$") class CodemetaMapping(SingleFileIntrinsicMapping): """ dedicated class for CodeMeta (codemeta.json) mapping and translation """ name = "codemeta" filename = b"codemeta.json" string_fields = None @classmethod def supported_terms(cls) -> List[str]: return [term for term in CODEMETA_TERMS if not term.startswith("@")] def translate(self, content: bytes) -> Optional[Dict[str, Any]]: try: return self.normalize_translation(expand(json.loads(content.decode()))) except Exception: return None class SwordCodemetaMapping(BaseExtrinsicMapping): """ dedicated class for mapping and translation from JSON-LD statements embedded in SWORD documents, optionally using Codemeta contexts, as described in the :ref:`deposit-protocol`. """ name = "sword-codemeta" @classmethod def extrinsic_metadata_formats(cls) -> Tuple[str, ...]: return ( "sword-v2-atom-codemeta", "sword-v2-atom-codemeta-v2", ) @classmethod def supported_terms(cls) -> List[str]: return [term for term in CODEMETA_TERMS if not term.startswith("@")] - def xml_to_jsonld(self, e: ET.Element) -> Dict[str, Any]: - doc: Dict[str, List[Dict[str, Any]]] = collections.defaultdict(list) + def xml_to_jsonld(self, e: ET.Element) -> Union[str, Dict[str, Any]]: + doc: Dict[str, List[Union[str, Dict[str, Any]]]] = collections.defaultdict(list) for child in e: m = _TAG_RE.match(child.tag) assert m, f"Tag with no namespace: {child}" namespace = m.group("namespace") localname = m.group("localname") if namespace == ATOM_URI and localname in ("title", "name"): # Convert Atom to Codemeta name; in case codemeta:name # is not provided or different doc["name"].append(self.xml_to_jsonld(child)) elif namespace == ATOM_URI and localname in ("author", "email"): # ditto for these author properties (note that author email is also # covered by the previous test) doc[localname].append(self.xml_to_jsonld(child)) elif namespace in _IGNORED_NAMESPACES: # SWORD-specific namespace that is not interesting to translate pass elif namespace.lower() == CODEMETA_CONTEXT_URL: # It is a term defined by the context; write is as-is and JSON-LD # expansion will convert it to a full URI based on # "@context": CODEMETA_CONTEXT_URL jsonld_child = self.xml_to_jsonld(child) - if localname == "type" and isinstance(jsonld_child, dict): - # With a codemeta context, this is later translated to a JSON-LD - # @type, which must be either an array of strings or a string. - if set(jsonld_child) != {"@value"}: - raise ValueError(f'Unexpected value for "type": {jsonld_child}') - jsonld_child = jsonld_child["@value"] + if ( + localname + in ( + "dateCreated", + "dateModified", + "datePublished", + ) + and isinstance(jsonld_child, str) + and _DATE_RE.match(jsonld_child) + ): + # Dates missing a leading zero for their day/month, used + # to be allowed by the deposit; so we need to reformat them + # to be valid ISO8601. + jsonld_child = iso8601.parse_date(jsonld_child).date().isoformat() doc[localname].append(jsonld_child) else: # Otherwise, we already know the URI doc[f"{namespace}{localname}"].append(self.xml_to_jsonld(child)) # The above needed doc values to be list to work; now we allow any type # of value as key "@value" cannot have a list as value. doc_: Dict[str, Any] = doc text = e.text.strip() if e.text else None if text: # TODO: check doc is empty, and raise mixed-content error otherwise? - doc_["@value"] = text + return text return doc_ def translate(self, content: bytes) -> Optional[Dict[str, Any]]: # Parse XML root = ET.fromstring(content) # Transform to JSON-LD document doc = self.xml_to_jsonld(root) + assert isinstance(doc, dict), f"Root object is not a dict: {doc}" + # Add @context to JSON-LD expansion replaces the "codemeta:" prefix # hash (which uses the context URL as namespace URI for historical # reasons) into properties in `http://schema.org/` and # `https://codemeta.github.io/terms/` namespaces doc["@context"] = CODEMETA_CONTEXT_URL # Normalize as a Codemeta document return self.normalize_translation(expand(doc)) def normalize_translation(self, metadata: Dict[str, Any]) -> Dict[str, Any]: return compact(metadata, forgefed=False) class JsonSwordCodemetaMapping(SwordCodemetaMapping): """ Variant of :class:`SwordCodemetaMapping` that reads the legacy ``sword-v2-atom-codemeta-v2-in-json`` format and converts it back to ``sword-v2-atom-codemeta-v2`` XML """ name = "json-sword-codemeta" @classmethod def extrinsic_metadata_formats(cls) -> Tuple[str, ...]: return ("sword-v2-atom-codemeta-v2-in-json",) def translate(self, content: bytes) -> Optional[Dict[str, Any]]: # ``content`` was generated by calling ``xmltodict.parse()`` on a XML document, # so ``xmltodict.unparse()`` is guaranteed to return a document that is # semantically equivalent to the original and pass it to SwordCodemetaMapping. json_doc = json.loads(content) if json_doc.get("@xmlns") != ATOM_URI: # Technically, non-default XMLNS were allowed, but it does not seem like # anyone used them, so they do not need to be implemented here. raise NotImplementedError(f"Unexpected XMLNS set: {json_doc}") # Root tag was stripped by swh-deposit json_doc = {"entry": json_doc} return super().translate(xmltodict.unparse(json_doc)) diff --git a/swh/indexer/metadata_dictionary/github.py b/swh/indexer/metadata_dictionary/github.py index 25f6eff..0435c41 100644 --- a/swh/indexer/metadata_dictionary/github.py +++ b/swh/indexer/metadata_dictionary/github.py @@ -1,137 +1,136 @@ # Copyright (C) 2022 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 typing import Any, Tuple from rdflib import RDF, BNode, Graph, Literal, URIRef from swh.indexer.codemeta import CROSSWALK_TABLE from swh.indexer.namespaces import ACTIVITYSTREAMS, CODEMETA, FORGEFED, SCHEMA from .base import BaseExtrinsicMapping, JsonMapping, produce_terms -from .utils import prettyprint_graph # noqa +from .utils import add_url_if_valid, prettyprint_graph # noqa SPDX = URIRef("https://spdx.org/licenses/") class GitHubMapping(BaseExtrinsicMapping, JsonMapping): name = "github" mapping = { **CROSSWALK_TABLE["GitHub"], "topics": SCHEMA.keywords, # TODO: submit this to the official crosswalk "clone_url": SCHEMA.codeRepository, } uri_fields = [ "clone_url", ] date_fields = [ "created_at", "updated_at", ] string_fields = [ "description", "full_name", "topics", ] @classmethod def extrinsic_metadata_formats(cls) -> Tuple[str, ...]: return ("application/vnd.github.v3+json",) def extra_translation(self, graph, root, content_dict): graph.remove((root, RDF.type, SCHEMA.SoftwareSourceCode)) graph.add((root, RDF.type, FORGEFED.Repository)) if content_dict.get("has_issues"): - graph.add( - ( - root, - CODEMETA.issueTracker, - URIRef(content_dict["html_url"] + "/issues"), - ) + add_url_if_valid( + graph, + root, + CODEMETA.issueTracker, + URIRef(content_dict["html_url"] + "/issues"), ) def get_root_uri(self, content_dict: dict) -> URIRef: if isinstance(content_dict.get("html_url"), str): return URIRef(content_dict["html_url"]) else: raise ValueError( f"GitHub metadata has missing/invalid html_url: {content_dict}" ) @produce_terms(FORGEFED.forks, ACTIVITYSTREAMS.totalItems) def translate_forks_count(self, graph: Graph, root: BNode, v: Any) -> None: """ >>> graph = Graph() >>> root = URIRef("http://example.org/test-software") >>> GitHubMapping().translate_forks_count(graph, root, 42) >>> prettyprint_graph(graph, root) { "@id": ..., "https://forgefed.org/ns#forks": { "@type": "https://www.w3.org/ns/activitystreams#OrderedCollection", "https://www.w3.org/ns/activitystreams#totalItems": 42 } } """ if isinstance(v, int): collection = BNode() graph.add((root, FORGEFED.forks, collection)) graph.add((collection, RDF.type, ACTIVITYSTREAMS.OrderedCollection)) graph.add((collection, ACTIVITYSTREAMS.totalItems, Literal(v))) @produce_terms(ACTIVITYSTREAMS.likes, ACTIVITYSTREAMS.totalItems) def translate_stargazers_count(self, graph: Graph, root: BNode, v: Any) -> None: """ >>> graph = Graph() >>> root = URIRef("http://example.org/test-software") >>> GitHubMapping().translate_stargazers_count(graph, root, 42) >>> prettyprint_graph(graph, root) { "@id": ..., "https://www.w3.org/ns/activitystreams#likes": { "@type": "https://www.w3.org/ns/activitystreams#Collection", "https://www.w3.org/ns/activitystreams#totalItems": 42 } } """ if isinstance(v, int): collection = BNode() graph.add((root, ACTIVITYSTREAMS.likes, collection)) graph.add((collection, RDF.type, ACTIVITYSTREAMS.Collection)) graph.add((collection, ACTIVITYSTREAMS.totalItems, Literal(v))) @produce_terms(ACTIVITYSTREAMS.followers, ACTIVITYSTREAMS.totalItems) def translate_watchers_count(self, graph: Graph, root: BNode, v: Any) -> None: """ >>> graph = Graph() >>> root = URIRef("http://example.org/test-software") >>> GitHubMapping().translate_watchers_count(graph, root, 42) >>> prettyprint_graph(graph, root) { "@id": ..., "https://www.w3.org/ns/activitystreams#followers": { "@type": "https://www.w3.org/ns/activitystreams#Collection", "https://www.w3.org/ns/activitystreams#totalItems": 42 } } """ if isinstance(v, int): collection = BNode() graph.add((root, ACTIVITYSTREAMS.followers, collection)) graph.add((collection, RDF.type, ACTIVITYSTREAMS.Collection)) graph.add((collection, ACTIVITYSTREAMS.totalItems, Literal(v))) def normalize_license(self, d): """ >>> GitHubMapping().normalize_license({'spdx_id': 'MIT'}) rdflib.term.URIRef('https://spdx.org/licenses/MIT') """ if isinstance(d, dict) and isinstance(d.get("spdx_id"), str): return SPDX + d["spdx_id"] diff --git a/swh/indexer/metadata_dictionary/maven.py b/swh/indexer/metadata_dictionary/maven.py index 8b3e48d..5575ba9 100644 --- a/swh/indexer/metadata_dictionary/maven.py +++ b/swh/indexer/metadata_dictionary/maven.py @@ -1,162 +1,163 @@ # Copyright (C) 2018-2022 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 Any, Dict -from rdflib import Graph, Literal, URIRef +from rdflib import Graph, Literal from swh.indexer.codemeta import CROSSWALK_TABLE from swh.indexer.namespaces import SCHEMA from .base import SingleFileIntrinsicMapping, XmlMapping -from .utils import prettyprint_graph # noqa +from .utils import add_url_if_valid, prettyprint_graph # noqa class MavenMapping(XmlMapping, SingleFileIntrinsicMapping): """ dedicated class for Maven (pom.xml) mapping and translation """ name = "maven" filename = b"pom.xml" mapping = CROSSWALK_TABLE["Java (Maven)"] string_fields = ["name", "version", "description", "email"] _default_repository = {"url": "https://repo.maven.apache.org/maven2/"} def _translate_dict(self, d: Dict[str, Any]) -> Dict[str, Any]: return super()._translate_dict(d.get("project") or {}) def extra_translation(self, graph: Graph, root, d): self.parse_repositories(graph, root, d) def parse_repositories(self, graph: Graph, root, d): """https://maven.apache.org/pom.html#Repositories >>> import rdflib >>> import xmltodict >>> from pprint import pprint >>> d = xmltodict.parse(''' ... ... ... codehausSnapshots ... Codehaus Snapshots ... http://snapshots.maven.codehaus.org/maven2 ... default ... ... ... ''') >>> MavenMapping().parse_repositories(rdflib.Graph(), rdflib.BNode(), d) """ repositories = d.get("repositories") if not repositories: self.parse_repository(graph, root, d, self._default_repository) elif isinstance(repositories, dict): repositories = repositories.get("repository") or [] if not isinstance(repositories, list): repositories = [repositories] for repo in repositories: self.parse_repository(graph, root, d, repo) def parse_repository(self, graph: Graph, root, d, repo): if not isinstance(repo, dict): return if repo.get("layout", "default") != "default": return # TODO ? url = repo.get("url") group_id = d.get("groupId") artifact_id = d.get("artifactId") if ( isinstance(url, str) and isinstance(group_id, str) and isinstance(artifact_id, str) ): repo = os.path.join(url, *group_id.split("."), artifact_id) if "${" in repo: # Often use as templating in pom.xml files collected from VCSs return - graph.add((root, SCHEMA.codeRepository, URIRef(repo))) + add_url_if_valid(graph, root, SCHEMA.codeRepository, repo) def normalize_groupId(self, id_): """https://maven.apache.org/pom.html#Maven_Coordinates >>> MavenMapping().normalize_groupId('org.example') rdflib.term.Literal('org.example') """ if isinstance(id_, str): return Literal(id_) def translate_licenses(self, graph, root, licenses): """https://maven.apache.org/pom.html#Licenses >>> import xmltodict >>> import json + >>> from rdflib import URIRef >>> d = xmltodict.parse(''' ... ... ... Apache License, Version 2.0 ... https://www.apache.org/licenses/LICENSE-2.0.txt ... ... ... ''') >>> print(json.dumps(d, indent=4)) { "licenses": { "license": { "name": "Apache License, Version 2.0", "url": "https://www.apache.org/licenses/LICENSE-2.0.txt" } } } >>> graph = Graph() >>> root = URIRef("http://example.org/test-software") >>> MavenMapping().translate_licenses(graph, root, d["licenses"]) >>> prettyprint_graph(graph, root) { "@id": ..., "http://schema.org/license": { "@id": "https://www.apache.org/licenses/LICENSE-2.0.txt" } } or, if there are more than one license: >>> import xmltodict >>> from pprint import pprint >>> d = xmltodict.parse(''' ... ... ... Apache License, Version 2.0 ... https://www.apache.org/licenses/LICENSE-2.0.txt ... ... ... MIT License ... https://opensource.org/licenses/MIT ... ... ... ''') >>> graph = Graph() >>> root = URIRef("http://example.org/test-software") >>> MavenMapping().translate_licenses(graph, root, d["licenses"]) >>> pprint(set(graph.triples((root, URIRef("http://schema.org/license"), None)))) {(rdflib.term.URIRef('http://example.org/test-software'), rdflib.term.URIRef('http://schema.org/license'), rdflib.term.URIRef('https://opensource.org/licenses/MIT')), (rdflib.term.URIRef('http://example.org/test-software'), rdflib.term.URIRef('http://schema.org/license'), rdflib.term.URIRef('https://www.apache.org/licenses/LICENSE-2.0.txt'))} """ if not isinstance(licenses, dict): return licenses = licenses.get("license") if isinstance(licenses, dict): licenses = [licenses] elif not isinstance(licenses, list): return for license in licenses: - if isinstance(license, dict) and isinstance(license.get("url"), str): - graph.add((root, SCHEMA.license, URIRef(license["url"]))) + if isinstance(license, dict): + add_url_if_valid(graph, root, SCHEMA.license, license.get("url")) diff --git a/swh/indexer/metadata_dictionary/npm.py b/swh/indexer/metadata_dictionary/npm.py index fd627b7..b838e5a 100644 --- a/swh/indexer/metadata_dictionary/npm.py +++ b/swh/indexer/metadata_dictionary/npm.py @@ -1,298 +1,288 @@ # Copyright (C) 2018-2022 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 -import urllib.parse from rdflib import RDF, BNode, Graph, Literal, URIRef from swh.indexer.codemeta import CROSSWALK_TABLE from swh.indexer.namespaces import SCHEMA from .base import JsonMapping, SingleFileIntrinsicMapping -from .utils import add_list, prettyprint_graph # noqa +from .utils import add_list, add_url_if_valid, prettyprint_graph # noqa SPDX = URIRef("https://spdx.org/licenses/") class NpmMapping(JsonMapping, SingleFileIntrinsicMapping): """ dedicated class for NPM (package.json) mapping and translation """ name = "npm" mapping = CROSSWALK_TABLE["NodeJS"] filename = b"package.json" string_fields = ["name", "version", "description", "email"] uri_fields = ["homepage"] _schema_shortcuts = { "github": "git+https://github.com/%s.git", "gist": "git+https://gist.github.com/%s.git", "gitlab": "git+https://gitlab.com/%s.git", # Bitbucket supports both hg and git, and the shortcut does not # tell which one to use. # 'bitbucket': 'https://bitbucket.org/', } def normalize_repository(self, d): """https://docs.npmjs.com/files/package.json#repository >>> NpmMapping().normalize_repository({ ... 'type': 'git', ... 'url': 'https://example.org/foo.git' ... }) rdflib.term.URIRef('git+https://example.org/foo.git') >>> NpmMapping().normalize_repository( ... 'gitlab:foo/bar') rdflib.term.URIRef('git+https://gitlab.com/foo/bar.git') >>> NpmMapping().normalize_repository( ... 'foo/bar') rdflib.term.URIRef('git+https://github.com/foo/bar.git') """ if ( isinstance(d, dict) and isinstance(d.get("type"), str) and isinstance(d.get("url"), str) ): url = "{type}+{url}".format(**d) elif isinstance(d, str): if "://" in d: url = d elif ":" in d: (schema, rest) = d.split(":", 1) if schema in self._schema_shortcuts: url = self._schema_shortcuts[schema] % rest else: return None else: url = self._schema_shortcuts["github"] % d else: return None return URIRef(url) def normalize_bugs(self, d): """https://docs.npmjs.com/files/package.json#bugs >>> NpmMapping().normalize_bugs({ ... 'url': 'https://example.org/bugs/', ... 'email': 'bugs@example.org' ... }) rdflib.term.URIRef('https://example.org/bugs/') >>> NpmMapping().normalize_bugs( ... 'https://example.org/bugs/') rdflib.term.URIRef('https://example.org/bugs/') """ if isinstance(d, dict) and isinstance(d.get("url"), str): url = d["url"] elif isinstance(d, str): url = d else: url = "" - parsed_url = urllib.parse.urlparse(url) - if parsed_url.netloc: - return URIRef(url) - else: - return None + return URIRef(url) _parse_author = re.compile( r"^ *" r"(?P.*?)" r"( +<(?P.*)>)?" r"( +\((?P.*)\))?" r" *$" ) def translate_author(self, graph: Graph, root, d): r"""https://docs.npmjs.com/files/package.json#people-fields-author-contributors' >>> from pprint import pprint >>> root = URIRef("http://example.org/test-software") >>> graph = Graph() >>> NpmMapping().translate_author(graph, root, { ... 'name': 'John Doe', ... 'email': 'john.doe@example.org', ... 'url': 'https://example.org/~john.doe', ... }) >>> prettyprint_graph(graph, root) { "@id": ..., "http://schema.org/author": { "@list": [ { "@type": "http://schema.org/Person", "http://schema.org/email": "john.doe@example.org", "http://schema.org/name": "John Doe", "http://schema.org/url": { "@id": "https://example.org/~john.doe" } } ] } } >>> graph = Graph() >>> NpmMapping().translate_author(graph, root, ... 'John Doe (https://example.org/~john.doe)' ... ) >>> prettyprint_graph(graph, root) { "@id": ..., "http://schema.org/author": { "@list": [ { "@type": "http://schema.org/Person", "http://schema.org/email": "john.doe@example.org", "http://schema.org/name": "John Doe", "http://schema.org/url": { "@id": "https://example.org/~john.doe" } } ] } } >>> graph = Graph() >>> NpmMapping().translate_author(graph, root, { ... 'name': 'John Doe', ... 'email': 'john.doe@example.org', ... 'url': 'https:\\\\example.invalid/~john.doe', ... }) >>> prettyprint_graph(graph, root) { "@id": ..., "http://schema.org/author": { "@list": [ { "@type": "http://schema.org/Person", "http://schema.org/email": "john.doe@example.org", "http://schema.org/name": "John Doe" } ] } } """ # noqa author = BNode() graph.add((author, RDF.type, SCHEMA.Person)) if isinstance(d, dict): name = d.get("name", None) email = d.get("email", None) url = d.get("url", None) elif isinstance(d, str): match = self._parse_author.match(d) if not match: return None name = match.group("name") email = match.group("email") url = match.group("url") else: return None if name and isinstance(name, str): graph.add((author, SCHEMA.name, Literal(name))) if email and isinstance(email, str): graph.add((author, SCHEMA.email, Literal(email))) - if url and isinstance(url, str): - # Workaround for https://github.com/digitalbazaar/pyld/issues/91 : drop - # URLs that are blatantly invalid early, so PyLD does not crash. - parsed_url = urllib.parse.urlparse(url) - if parsed_url.netloc: - graph.add((author, SCHEMA.url, URIRef(url))) + add_url_if_valid(graph, author, SCHEMA.url, url) add_list(graph, root, SCHEMA.author, [author]) def normalize_description(self, description): r"""Try to re-decode ``description`` as UTF-16, as this is a somewhat common mistake that causes issues in the database because of null bytes in JSON. >>> NpmMapping().normalize_description("foo bar") rdflib.term.Literal('foo bar') >>> NpmMapping().normalize_description( ... "\ufffd\ufffd#\x00 \x00f\x00o\x00o\x00 \x00b\x00a\x00r\x00\r\x00 \x00" ... ) rdflib.term.Literal('foo bar') >>> NpmMapping().normalize_description( ... "\ufffd\ufffd\x00#\x00 \x00f\x00o\x00o\x00 \x00b\x00a\x00r\x00\r\x00 " ... ) rdflib.term.Literal('foo bar') >>> NpmMapping().normalize_description( ... # invalid UTF-16 and meaningless UTF-8: ... "\ufffd\ufffd\x00#\x00\x00\x00 \x00\x00\x00\x00f\x00\x00\x00\x00" ... ) is None True >>> NpmMapping().normalize_description( ... # ditto (ut looks like little-endian at first) ... "\ufffd\ufffd#\x00\x00\x00 \x00\x00\x00\x00f\x00\x00\x00\x00\x00" ... ) is None True >>> NpmMapping().normalize_description(None) is None True """ if not isinstance(description, str): return None # XXX: if this function ever need to support more cases, consider # switching to https://pypi.org/project/ftfy/ instead of adding more hacks if description.startswith("\ufffd\ufffd") and "\x00" in description: # 2 unicode replacement characters followed by '# ' encoded as UTF-16 # is a common mistake, which indicates a README.md was saved as UTF-16, # and some NPM tool opened it as UTF-8 and used the first line as # description. description_bytes = description.encode() # Strip the the two unicode replacement characters assert description_bytes.startswith(b"\xef\xbf\xbd\xef\xbf\xbd") description_bytes = description_bytes[6:] # If the following attempts fail to recover the description, discard it # entirely because the current indexer storage backend (postgresql) cannot # store zero bytes in JSON columns. description = None if not description_bytes.startswith(b"\x00"): # try UTF-16 little-endian (the most common) first try: description = description_bytes.decode("utf-16le") except UnicodeDecodeError: pass if description is None: # if it fails, try UTF-16 big-endian try: description = description_bytes.decode("utf-16be") except UnicodeDecodeError: pass if description: if description.startswith("# "): description = description[2:] return Literal(description.rstrip()) else: return None return Literal(description) def normalize_license(self, s): """https://docs.npmjs.com/files/package.json#license >>> NpmMapping().normalize_license('MIT') rdflib.term.URIRef('https://spdx.org/licenses/MIT') """ if isinstance(s, str): if s.startswith("SEE LICENSE IN "): # Very common pattern, because it is an example in the specification. # It is followed by the filename; and the indexer architecture currently # does not allow accessing that from metadata mappings. # (Plus, an hypothetical license mapping would eventually pick it up) return if " " in s: # Either an SPDX expression, or unusable data # TODO: handle it return return SPDX + s def normalize_keywords(self, lst): """https://docs.npmjs.com/files/package.json#homepage >>> NpmMapping().normalize_keywords(['foo', 'bar']) [rdflib.term.Literal('foo'), rdflib.term.Literal('bar')] """ if isinstance(lst, list): return [Literal(x) for x in lst if isinstance(x, str)] diff --git a/swh/indexer/metadata_dictionary/nuget.py b/swh/indexer/metadata_dictionary/nuget.py index 087ec0e..b22100c 100644 --- a/swh/indexer/metadata_dictionary/nuget.py +++ b/swh/indexer/metadata_dictionary/nuget.py @@ -1,95 +1,95 @@ # Copyright (C) 2022 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.path import re from typing import Any, Dict, List from rdflib import RDF, BNode, Graph, Literal, URIRef from swh.indexer.codemeta import _DATA_DIR, read_crosstable from swh.indexer.namespaces import SCHEMA from swh.indexer.storage.interface import Sha1 from .base import BaseIntrinsicMapping, DirectoryLsEntry, XmlMapping -from .utils import add_list +from .utils import add_list, add_url_if_valid NUGET_TABLE_PATH = os.path.join(_DATA_DIR, "nuget.csv") with open(NUGET_TABLE_PATH) as fd: (CODEMETA_TERMS, NUGET_TABLE) = read_crosstable(fd) SPDX = URIRef("https://spdx.org/licenses/") class NuGetMapping(XmlMapping, BaseIntrinsicMapping): """ dedicated class for NuGet (.nuspec) mapping and translation """ name = "nuget" mapping = NUGET_TABLE["NuGet"] mapping["copyright"] = URIRef("http://schema.org/copyrightNotice") mapping["language"] = URIRef("http://schema.org/inLanguage") string_fields = [ "description", "version", "name", "tags", "license", "summary", "copyright", "language", ] uri_fields = ["projectUrl", "licenseUrl"] @classmethod def detect_metadata_files(cls, file_entries: List[DirectoryLsEntry]) -> List[Sha1]: for entry in file_entries: if entry["name"].endswith(b".nuspec"): return [entry["sha1"]] return [] def _translate_dict(self, d: Dict[str, Any]) -> Dict[str, Any]: return super()._translate_dict(d.get("package", {}).get("metadata", {})) def translate_repository(self, graph, root, v): if isinstance(v, dict) and isinstance(v["@url"], str): codemeta_key = URIRef(self.mapping["repository.url"]) - graph.add((root, codemeta_key, URIRef(v["@url"]))) + add_url_if_valid(graph, root, codemeta_key, v["@url"]) def normalize_license(self, v): if isinstance(v, dict) and v["@type"] == "expression": license_string = v["#text"] if not bool( re.search(r" with |\(|\)| and ", license_string, re.IGNORECASE) ): return [ SPDX + license_type.strip() for license_type in re.split( r" or ", license_string, flags=re.IGNORECASE ) ] else: return None def translate_authors(self, graph: Graph, root, s): if isinstance(s, str): authors = [] for author_name in s.split(","): author_name = author_name.strip() author = BNode() graph.add((author, RDF.type, SCHEMA.Person)) graph.add((author, SCHEMA.name, Literal(author_name))) authors.append(author) add_list(graph, root, SCHEMA.author, authors) def translate_releaseNotes(self, graph: Graph, root, s): if isinstance(s, str): graph.add((root, SCHEMA.releaseNotes, Literal(s))) def normalize_tags(self, s): if isinstance(s, str): return [Literal(tag) for tag in s.split(" ")] diff --git a/swh/indexer/metadata_dictionary/utils.py b/swh/indexer/metadata_dictionary/utils.py index 173b146..8a5fdb9 100644 --- a/swh/indexer/metadata_dictionary/utils.py +++ b/swh/indexer/metadata_dictionary/utils.py @@ -1,72 +1,112 @@ # Copyright (C) 2022 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 -from typing import Callable, Iterable, Optional, Sequence, TypeVar +from typing import Any, Callable, Iterable, Optional, Sequence, TypeVar +import urllib.parse from pyld import jsonld from rdflib import RDF, Graph, URIRef import rdflib.term from swh.indexer.codemeta import _document_loader def prettyprint_graph(graph: Graph, root: URIRef): s = graph.serialize(format="application/ld+json") jsonld_graph = json.loads(s) translated_metadata = jsonld.frame( jsonld_graph, {"@id": str(root)}, options={ "documentLoader": _document_loader, "processingMode": "json-ld-1.1", }, ) print(json.dumps(translated_metadata, indent=4)) def add_list( graph: Graph, subject: rdflib.term.Node, predicate: rdflib.term.Identifier, objects: Sequence[rdflib.term.Node], ) -> None: """Adds triples to the ``graph`` so that they are equivalent to this JSON-LD object:: { "@id": subject, predicate: {"@list": objects} } This is a naive implementation of https://json-ld.org/spec/latest/json-ld-api/#list-to-rdf-conversion """ # JSON-LD's @list is syntactic sugar for a linked list / chain in the RDF graph, # which is what we are going to construct, starting from the end: last_link: rdflib.term.Node last_link = RDF.nil for item in reversed(objects): link = rdflib.BNode() graph.add((link, RDF.first, item)) graph.add((link, RDF.rest, last_link)) last_link = link graph.add((subject, predicate, last_link)) TValue = TypeVar("TValue") def add_map( graph: Graph, subject: rdflib.term.Node, predicate: rdflib.term.Identifier, f: Callable[[Graph, TValue], Optional[rdflib.term.Node]], values: Iterable[TValue], ) -> None: """Helper for :func:`add_list` that takes a mapper function ``f``.""" nodes = [f(graph, value) for value in values] add_list(graph, subject, predicate, [node for node in nodes if node]) + + +def add_url_if_valid( + graph: Graph, + subject: rdflib.term.Node, + predicate: rdflib.term.Identifier, + url: Any, +) -> None: + """Adds ``(subject, predicate, url)`` to the graph if ``url`` is well-formed. + + This is meant as a workaround for https://github.com/digitalbazaar/pyld/issues/91 + to drop URLs that are blatantly invalid early, so PyLD does not crash. + + >>> from pprint import pprint + >>> graph = Graph() + >>> subject = rdflib.term.URIRef("http://example.org/test-software") + >>> predicate = rdflib.term.URIRef("http://schema.org/license") + >>> add_url_if_valid( + ... graph, subject, predicate, "https//www.apache.org/licenses/LICENSE-2.0.txt" + ... ) + >>> add_url_if_valid( + ... graph, subject, predicate, "http:s//www.apache.org/licenses/LICENSE-2.0.txt" + ... ) + >>> add_url_if_valid( + ... graph, subject, predicate, "https://www.apache.org/licenses/LICENSE-2.0.txt" + ... ) + >>> add_url_if_valid( + ... graph, subject, predicate, 42 + ... ) + >>> pprint(set(graph.triples((subject, predicate, None)))) + {(rdflib.term.URIRef('http://example.org/test-software'), + rdflib.term.URIRef('http://schema.org/license'), + rdflib.term.URIRef('https://www.apache.org/licenses/LICENSE-2.0.txt'))} + """ + if not isinstance(url, str): + return + if " " in url or not urllib.parse.urlparse(url).netloc: + return + graph.add((subject, predicate, rdflib.term.URIRef(url))) diff --git a/swh/indexer/tests/metadata_dictionary/test_codemeta.py b/swh/indexer/tests/metadata_dictionary/test_codemeta.py index 586c4e9..bc08b25 100644 --- a/swh/indexer/tests/metadata_dictionary/test_codemeta.py +++ b/swh/indexer/tests/metadata_dictionary/test_codemeta.py @@ -1,393 +1,422 @@ # Copyright (C) 2017-2022 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 from hypothesis import HealthCheck, given, settings from swh.indexer.codemeta import CODEMETA_TERMS from swh.indexer.metadata_detector import detect_metadata from swh.indexer.metadata_dictionary import MAPPINGS from ..utils import json_document_strategy def test_compute_metadata_valid_codemeta(): raw_content = b"""{ "@context": "https://doi.org/10.5063/schema/codemeta-2.0", "@type": "SoftwareSourceCode", "identifier": "CodeMeta", "description": "CodeMeta is a concept vocabulary that can be used to standardize the exchange of software metadata across repositories and organizations.", "name": "CodeMeta: Minimal metadata schemas for science software and code, in JSON-LD", "codeRepository": "https://github.com/codemeta/codemeta", "issueTracker": "https://github.com/codemeta/codemeta/issues", "license": "https://spdx.org/licenses/Apache-2.0", "version": "2.0", "author": [ { "@type": "Person", "givenName": "Carl", "familyName": "Boettiger", "email": "cboettig@gmail.com", "@id": "http://orcid.org/0000-0002-1642-628X" }, { "@type": "Person", "givenName": "Matthew B.", "familyName": "Jones", "email": "jones@nceas.ucsb.edu", "@id": "http://orcid.org/0000-0003-0077-4738" } ], "maintainer": { "@type": "Person", "givenName": "Carl", "familyName": "Boettiger", "email": "cboettig@gmail.com", "@id": "http://orcid.org/0000-0002-1642-628X" }, "contIntegration": "https://travis-ci.org/codemeta/codemeta", "developmentStatus": "active", "downloadUrl": "https://github.com/codemeta/codemeta/archive/2.0.zip", "funder": { "@id": "https://doi.org/10.13039/100000001", "@type": "Organization", "name": "National Science Foundation" }, "funding":"1549758; Codemeta: A Rosetta Stone for Metadata in Scientific Software", "keywords": [ "metadata", "software" ], "version":"2.0", "dateCreated":"2017-06-05", "datePublished":"2017-06-05", "programmingLanguage": "JSON-LD" }""" # noqa expected_result = { "@context": "https://doi.org/10.5063/schema/codemeta-2.0", "type": "SoftwareSourceCode", "identifier": "CodeMeta", "description": "CodeMeta is a concept vocabulary that can " "be used to standardize the exchange of software metadata " "across repositories and organizations.", "name": "CodeMeta: Minimal metadata schemas for science " "software and code, in JSON-LD", "codeRepository": "https://github.com/codemeta/codemeta", "issueTracker": "https://github.com/codemeta/codemeta/issues", "license": "https://spdx.org/licenses/Apache-2.0", "version": "2.0", "author": [ { "type": "Person", "givenName": "Carl", "familyName": "Boettiger", "email": "cboettig@gmail.com", "id": "http://orcid.org/0000-0002-1642-628X", }, { "type": "Person", "givenName": "Matthew B.", "familyName": "Jones", "email": "jones@nceas.ucsb.edu", "id": "http://orcid.org/0000-0003-0077-4738", }, ], "maintainer": { "type": "Person", "givenName": "Carl", "familyName": "Boettiger", "email": "cboettig@gmail.com", "id": "http://orcid.org/0000-0002-1642-628X", }, "contIntegration": "https://travis-ci.org/codemeta/codemeta", "developmentStatus": "active", "downloadUrl": "https://github.com/codemeta/codemeta/archive/2.0.zip", "funder": { "id": "https://doi.org/10.13039/100000001", "type": "Organization", "name": "National Science Foundation", }, "funding": "1549758; Codemeta: A Rosetta Stone for Metadata " "in Scientific Software", "keywords": ["metadata", "software"], "version": "2.0", "dateCreated": "2017-06-05", "datePublished": "2017-06-05", "programmingLanguage": "JSON-LD", } result = MAPPINGS["CodemetaMapping"]().translate(raw_content) assert result == expected_result def test_compute_metadata_codemeta_alternate_context(): raw_content = b"""{ "@context": "https://raw.githubusercontent.com/codemeta/codemeta/master/codemeta.jsonld", "@type": "SoftwareSourceCode", "identifier": "CodeMeta" }""" # noqa expected_result = { "@context": "https://doi.org/10.5063/schema/codemeta-2.0", "type": "SoftwareSourceCode", "identifier": "CodeMeta", } result = MAPPINGS["CodemetaMapping"]().translate(raw_content) assert result == expected_result @settings(suppress_health_check=[HealthCheck.too_slow]) @given(json_document_strategy(keys=CODEMETA_TERMS)) def test_codemeta_adversarial(doc): raw = json.dumps(doc).encode() MAPPINGS["CodemetaMapping"]().translate(raw) def test_detect_metadata_codemeta_json_uppercase(): df = [ { "sha1_git": b"abc", "name": b"index.html", "target": b"abc", "length": 897, "status": "visible", "type": "file", "perms": 33188, "dir_id": b"dir_a", "sha1": b"bcd", }, { "sha1_git": b"aab", "name": b"CODEMETA.json", "target": b"aab", "length": 712, "status": "visible", "type": "file", "perms": 33188, "dir_id": b"dir_a", "sha1": b"bcd", }, ] results = detect_metadata(df) expected_results = {"CodemetaMapping": [b"bcd"]} assert expected_results == results def test_sword_default_xmlns(): content = """ My Software Author 1 foo@example.org Author 2 """ result = MAPPINGS["SwordCodemetaMapping"]().translate(content) assert result == { "@context": "https://doi.org/10.5063/schema/codemeta-2.0", "name": "My Software", "author": [ {"name": "Author 1", "email": "foo@example.org"}, {"name": "Author 2"}, ], } def test_sword_basics(): content = """ My Software Author 1 foo@example.org Author 2 + 2022-10-26 Author 3 bar@example.org """ result = MAPPINGS["SwordCodemetaMapping"]().translate(content) assert result == { "@context": "https://doi.org/10.5063/schema/codemeta-2.0", "name": "My Software", "author": [ {"name": "Author 1", "email": "foo@example.org"}, {"name": "Author 2"}, {"name": "Author 3", "email": "bar@example.org"}, ], + "dateCreated": "2022-10-26", } def test_sword_mixed(): content = """ My Software blah 1.2.3 blih """ result = MAPPINGS["SwordCodemetaMapping"]().translate(content) assert result == { "@context": "https://doi.org/10.5063/schema/codemeta-2.0", "name": "My Software", "version": "1.2.3", } def test_sword_schemaorg_in_codemeta(): content = """ My Software 1.2.3 """ result = MAPPINGS["SwordCodemetaMapping"]().translate(content) assert result == { "@context": "https://doi.org/10.5063/schema/codemeta-2.0", "name": "My Software", "version": "1.2.3", } def test_sword_schemaorg_in_codemeta_constrained(): """Resulting property has the compact URI 'schema:url' instead of just the term 'url', because term 'url' is defined by the Codemeta schema - has having type '@id'.""" + has having type '@id'. + Ditto for dates (with type http://schema.org/Date).""" content = """ My Software http://example.org/my-software + foo + 2022-10-26 """ result = MAPPINGS["SwordCodemetaMapping"]().translate(content) assert result == { "@context": "https://doi.org/10.5063/schema/codemeta-2.0", "name": "My Software", "schema:url": "http://example.org/my-software", + "schema:dateCreated": "foo", + "schema:dateModified": "2022-10-26", } def test_sword_schemaorg_not_in_codemeta(): content = """ My Software http://example.org/my-software """ result = MAPPINGS["SwordCodemetaMapping"]().translate(content) assert result == { "@context": "https://doi.org/10.5063/schema/codemeta-2.0", "name": "My Software", "schema:sameAs": "http://example.org/my-software", } def test_sword_atom_name(): content = """ My Software """ result = MAPPINGS["SwordCodemetaMapping"]().translate(content) assert result == { "@context": "https://doi.org/10.5063/schema/codemeta-2.0", "name": "My Software", } def test_sword_multiple_names(): content = """ Atom Name 1 Atom Name 2 Atom Title 1 Atom Title 2 Codemeta Name 1 Codemeta Name 2 """ result = MAPPINGS["SwordCodemetaMapping"]().translate(content) assert result == { "@context": "https://doi.org/10.5063/schema/codemeta-2.0", "name": [ "Atom Name 1", "Atom Name 2", "Atom Title 1", "Atom Title 2", "Codemeta Name 1", "Codemeta Name 2", ], } def test_sword_propertyvalue(): content = """ Name schema:PropertyValue HAL-ID hal-03780423 """ result = MAPPINGS["SwordCodemetaMapping"]().translate(content) assert result == { "@context": "https://doi.org/10.5063/schema/codemeta-2.0", "name": "Name", "identifier": { "schema:propertyID": "HAL-ID", "schema:value": "hal-03780423", "type": "schema:PropertyValue", }, } +def test_sword_fix_date(): + content = """ + + Name + 2020-12-1 + 2020-12-2 + 2020-12-3 + + """ + + result = MAPPINGS["SwordCodemetaMapping"]().translate(content) + assert result == { + "@context": "https://doi.org/10.5063/schema/codemeta-2.0", + "name": "Name", + "dateModified": "2020-12-01", + "dateCreated": "2020-12-02", + "datePublished": "2020-12-03", + } + + def test_json_sword(): content = """{"id": "hal-01243573", "@xmlns": "http://www.w3.org/2005/Atom", "author": {"name": "Author 1", "email": "foo@example.org"}, "client": "hal", "codemeta:url": "http://example.org/", "codemeta:name": "The assignment problem", "@xmlns:codemeta": "https://doi.org/10.5063/SCHEMA/CODEMETA-2.0", "codemeta:author": {"codemeta:name": "Author 2"}, "codemeta:license": {"codemeta:name": "GNU General Public License v3.0 or later"}}""" # noqa result = MAPPINGS["JsonSwordCodemetaMapping"]().translate(content) assert result == { "@context": "https://doi.org/10.5063/schema/codemeta-2.0", "author": [ {"name": "Author 1", "email": "foo@example.org"}, {"name": "Author 2"}, ], "license": {"name": "GNU General Public License v3.0 or later"}, "name": "The assignment problem", - "schema:url": "http://example.org/", + "url": "http://example.org/", "name": "The assignment problem", } diff --git a/swh/indexer/tests/metadata_dictionary/test_npm.py b/swh/indexer/tests/metadata_dictionary/test_npm.py index 804ac64..9b52bfd 100644 --- a/swh/indexer/tests/metadata_dictionary/test_npm.py +++ b/swh/indexer/tests/metadata_dictionary/test_npm.py @@ -1,438 +1,449 @@ # Copyright (C) 2017-2022 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 from hypothesis import HealthCheck, given, settings import pytest from swh.indexer.metadata_detector import detect_metadata from swh.indexer.metadata_dictionary import MAPPINGS from swh.indexer.storage.model import ContentMetadataRow from ..test_metadata import TRANSLATOR_TOOL, ContentMetadataTestIndexer from ..utils import ( BASE_TEST_CONFIG, MAPPING_DESCRIPTION_CONTENT_SHA1, json_document_strategy, ) def test_compute_metadata_none(): """ testing content empty content is empty should return None """ content = b"" # None if no metadata was found or an error occurred declared_metadata = None result = MAPPINGS["NpmMapping"]().translate(content) assert declared_metadata == result def test_compute_metadata_npm(): """ testing only computation of metadata with hard_mapping_npm """ content = b""" { "name": "test_metadata", "version": "0.0.2", "description": "Simple package.json test for indexer", "repository": { "type": "git", "url": "https://github.com/moranegg/metadata_test" }, "author": { "email": "moranegg@example.com", "name": "Morane G" } } """ declared_metadata = { "@context": "https://doi.org/10.5063/schema/codemeta-2.0", "type": "SoftwareSourceCode", "name": "test_metadata", "version": "0.0.2", "description": "Simple package.json test for indexer", "codeRepository": "git+https://github.com/moranegg/metadata_test", "author": [ { "type": "Person", "name": "Morane G", "email": "moranegg@example.com", } ], } result = MAPPINGS["NpmMapping"]().translate(content) assert declared_metadata == result def test_compute_metadata_invalid_description_npm(): """ testing only computation of metadata with hard_mapping_npm """ content = b""" { "name": "test_metadata", "version": "0.0.2", "description": 1234 } """ declared_metadata = { "@context": "https://doi.org/10.5063/schema/codemeta-2.0", "type": "SoftwareSourceCode", "name": "test_metadata", "version": "0.0.2", } result = MAPPINGS["NpmMapping"]().translate(content) assert declared_metadata == result def test_index_content_metadata_npm(storage, obj_storage): """ testing NPM with package.json - one sha1 uses a file that can't be translated to metadata and should return None in the translated metadata """ sha1s = [ MAPPING_DESCRIPTION_CONTENT_SHA1["json:test-metadata-package.json"], MAPPING_DESCRIPTION_CONTENT_SHA1["json:npm-package.json"], MAPPING_DESCRIPTION_CONTENT_SHA1["python:code"], ] # this metadata indexer computes only metadata for package.json # in npm context with a hard mapping config = BASE_TEST_CONFIG.copy() config["tools"] = [TRANSLATOR_TOOL] metadata_indexer = ContentMetadataTestIndexer(config=config) metadata_indexer.run(sha1s, log_suffix="unknown content") results = list(metadata_indexer.idx_storage.content_metadata_get(sha1s)) expected_results = [ ContentMetadataRow( id=sha1s[0], tool=TRANSLATOR_TOOL, metadata={ "@context": "https://doi.org/10.5063/schema/codemeta-2.0", "type": "SoftwareSourceCode", "codeRepository": "git+https://github.com/moranegg/metadata_test", "description": "Simple package.json test for indexer", "name": "test_metadata", "version": "0.0.1", }, ), ContentMetadataRow( id=sha1s[1], tool=TRANSLATOR_TOOL, metadata={ "@context": "https://doi.org/10.5063/schema/codemeta-2.0", "type": "SoftwareSourceCode", "issueTracker": "https://github.com/npm/npm/issues", "author": [ { "type": "Person", "name": "Isaac Z. Schlueter", "email": "i@izs.me", "url": "http://blog.izs.me", } ], "codeRepository": "git+https://github.com/npm/npm", "description": "a package manager for JavaScript", "license": "https://spdx.org/licenses/Artistic-2.0", "version": "5.0.3", "name": "npm", "url": "https://docs.npmjs.com/", }, ), ] for result in results: del result.tool["id"] result.metadata.pop("keywords", None) # The assertion below returns False sometimes because of nested lists assert expected_results == results def test_npm_null_list_item_normalization(): package_json = b"""{ "name": "foo", "keywords": [ "foo", null ], "homepage": [ "http://example.org/", null ] }""" result = MAPPINGS["NpmMapping"]().translate(package_json) assert result == { "@context": "https://doi.org/10.5063/schema/codemeta-2.0", "name": "foo", "type": "SoftwareSourceCode", "url": "http://example.org/", "keywords": "foo", } def test_npm_bugs_normalization(): # valid dictionary package_json = b"""{ "name": "foo", "bugs": { "url": "https://github.com/owner/project/issues", "email": "foo@example.com" } }""" result = MAPPINGS["NpmMapping"]().translate(package_json) assert result == { "@context": "https://doi.org/10.5063/schema/codemeta-2.0", "name": "foo", "issueTracker": "https://github.com/owner/project/issues", "type": "SoftwareSourceCode", } # "invalid" dictionary package_json = b"""{ "name": "foo", "bugs": { "email": "foo@example.com" } }""" result = MAPPINGS["NpmMapping"]().translate(package_json) assert result == { "@context": "https://doi.org/10.5063/schema/codemeta-2.0", "name": "foo", "type": "SoftwareSourceCode", } # string package_json = b"""{ "name": "foo", "bugs": "https://github.com/owner/project/issues" }""" result = MAPPINGS["NpmMapping"]().translate(package_json) assert result == { "@context": "https://doi.org/10.5063/schema/codemeta-2.0", "name": "foo", "issueTracker": "https://github.com/owner/project/issues", "type": "SoftwareSourceCode", } def test_npm_repository_normalization(): # normal package_json = b"""{ "name": "foo", "repository": { "type" : "git", "url" : "https://github.com/npm/cli.git" } }""" result = MAPPINGS["NpmMapping"]().translate(package_json) assert result == { "@context": "https://doi.org/10.5063/schema/codemeta-2.0", "name": "foo", "codeRepository": "git+https://github.com/npm/cli.git", "type": "SoftwareSourceCode", } # missing url package_json = b"""{ "name": "foo", "repository": { "type" : "git" } }""" result = MAPPINGS["NpmMapping"]().translate(package_json) assert result == { "@context": "https://doi.org/10.5063/schema/codemeta-2.0", "name": "foo", "type": "SoftwareSourceCode", } # github shortcut package_json = b"""{ "name": "foo", "repository": "github:npm/cli" }""" result = MAPPINGS["NpmMapping"]().translate(package_json) expected_result = { "@context": "https://doi.org/10.5063/schema/codemeta-2.0", "name": "foo", "codeRepository": "git+https://github.com/npm/cli.git", "type": "SoftwareSourceCode", } assert result == expected_result # github shortshortcut package_json = b"""{ "name": "foo", "repository": "npm/cli" }""" result = MAPPINGS["NpmMapping"]().translate(package_json) assert result == expected_result # gitlab shortcut package_json = b"""{ "name": "foo", "repository": "gitlab:user/repo" }""" result = MAPPINGS["NpmMapping"]().translate(package_json) assert result == { "@context": "https://doi.org/10.5063/schema/codemeta-2.0", "name": "foo", "codeRepository": "git+https://gitlab.com/user/repo.git", "type": "SoftwareSourceCode", } def test_npm_author(): package_json = rb"""{ "version": "1.0.0", "author": "Foo Bar (@example)" }""" result = MAPPINGS["NpmMapping"]().translate(package_json) assert result == { "@context": "https://doi.org/10.5063/schema/codemeta-2.0", "type": "SoftwareSourceCode", "author": [{"name": "Foo Bar", "type": "Person"}], "version": "1.0.0", } def test_npm_invalid_uris(): package_json = rb"""{ "version": "1.0.0", "homepage": "", "author": { "name": "foo", "url": "http://example.org" } }""" result = MAPPINGS["NpmMapping"]().translate(package_json) assert result == { "@context": "https://doi.org/10.5063/schema/codemeta-2.0", "type": "SoftwareSourceCode", "author": [{"name": "foo", "type": "Person", "url": "http://example.org"}], "version": "1.0.0", } package_json = rb"""{ "version": "1.0.0", "homepage": "http://example.org", "author": { "name": "foo", "url": "" } }""" result = MAPPINGS["NpmMapping"]().translate(package_json) assert result == { "@context": "https://doi.org/10.5063/schema/codemeta-2.0", "type": "SoftwareSourceCode", "author": [{"name": "foo", "type": "Person"}], "url": "http://example.org", "version": "1.0.0", } package_json = rb"""{ "version": "1.0.0", "homepage": "", "author": { "name": "foo", "url": "" }, "bugs": "" }""" result = MAPPINGS["NpmMapping"]().translate(package_json) assert result == { "@context": "https://doi.org/10.5063/schema/codemeta-2.0", "type": "SoftwareSourceCode", "author": [{"name": "foo", "type": "Person"}], "version": "1.0.0", } package_json = rb"""{ "version": "1.0.0", "homepage": "http:example.org", "author": { "name": "foo", "url": "http:example.com" }, "bugs": { "url": "http:example.com" } }""" result = MAPPINGS["NpmMapping"]().translate(package_json) assert result == { "@context": "https://doi.org/10.5063/schema/codemeta-2.0", "type": "SoftwareSourceCode", "author": [{"name": "foo", "type": "Person"}], "version": "1.0.0", } + package_json = rb"""{ + "version": "1.0.0", + "repository": "git+https://g ithub.com/foo/bar.git" +}""" + result = MAPPINGS["NpmMapping"]().translate(package_json) + assert result == { + "@context": "https://doi.org/10.5063/schema/codemeta-2.0", + "type": "SoftwareSourceCode", + "version": "1.0.0", + } + def test_npm_invalid_licenses(): package_json = rb"""{ "version": "1.0.0", "license": "SEE LICENSE IN LICENSE.md", "author": { "name": "foo", "url": "http://example.org" } }""" result = MAPPINGS["NpmMapping"]().translate(package_json) assert result == { "@context": "https://doi.org/10.5063/schema/codemeta-2.0", "type": "SoftwareSourceCode", "author": [{"name": "foo", "type": "Person", "url": "http://example.org"}], "version": "1.0.0", } @settings(suppress_health_check=[HealthCheck.too_slow]) @given(json_document_strategy(keys=list(MAPPINGS["NpmMapping"].mapping))) # type: ignore def test_npm_adversarial(doc): raw = json.dumps(doc).encode() MAPPINGS["NpmMapping"]().translate(raw) @pytest.mark.parametrize( "filename", [b"package.json", b"Package.json", b"PACKAGE.json", b"PACKAGE.JSON"] ) def test_detect_metadata_package_json(filename): df = [ { "sha1_git": b"abc", "name": b"index.js", "target": b"abc", "length": 897, "status": "visible", "type": "file", "perms": 33188, "dir_id": b"dir_a", "sha1": b"bcd", }, { "sha1_git": b"aab", "name": filename, "target": b"aab", "length": 712, "status": "visible", "type": "file", "perms": 33188, "dir_id": b"dir_a", "sha1": b"cde", }, ] results = detect_metadata(df) expected_results = {"NpmMapping": [b"cde"]} assert expected_results == results diff --git a/tox.ini b/tox.ini index f0bda88..b9e5d2b 100644 --- a/tox.ini +++ b/tox.ini @@ -1,79 +1,80 @@ [tox] envlist=black,flake8,mypy,py3 [testenv] extras = testing deps = pytest-cov swh-scheduler[testing] >= 0.5.0 swh-storage[testing] >= 0.10.0 dev: pdbpp commands = pytest --doctest-modules \ !slow: --hypothesis-profile=fast \ slow: --hypothesis-profile=slow \ {envsitepackagesdir}/swh/indexer \ --cov={envsitepackagesdir}/swh/indexer \ --cov-branch {posargs} [testenv:black] skip_install = true deps = - black==22.3.0 + black==22.10.0 commands = {envpython} -m black --check swh [testenv:flake8] skip_install = true deps = - flake8==4.0.1 - flake8-bugbear==22.3.23 + flake8==5.0.4 + flake8-bugbear==22.9.23 + pycodestyle==2.9.1 commands = {envpython} -m flake8 [testenv:mypy] extras = testing deps = mypy==0.942 commands = mypy swh # build documentation outside swh-environment using the current # git HEAD of swh-docs, is executed on CI for each diff to prevent # breaking doc build [testenv:sphinx] whitelist_externals = make usedevelop = true extras = testing deps = # fetch and install swh-docs in develop mode -e git+https://forge.softwareheritage.org/source/swh-docs#egg=swh.docs setenv = SWH_PACKAGE_DOC_TOX_BUILD = 1 # turn warnings into errors SPHINXOPTS = -W commands = make -I ../.tox/sphinx/src/swh-docs/swh/ -C docs # build documentation only inside swh-environment using local state # of swh-docs package [testenv:sphinx-dev] whitelist_externals = make usedevelop = true extras = testing deps = # install swh-docs in develop mode -e ../swh-docs setenv = SWH_PACKAGE_DOC_TOX_BUILD = 1 # turn warnings into errors SPHINXOPTS = -W commands = make -I ../.tox/sphinx-dev/src/swh-docs/swh/ -C docs