Source code for swh.deposit.utils

# Copyright (C) 2018-2020 The Software Heritage developers
# See the AUTHORS file at the top-level directory of this distribution
# License: GNU General Public License version 3, or any later version
# See top-level LICENSE file for more information

import logging
from types import GeneratorType
from typing import Any, Dict, Optional, Tuple, Union

import iso8601
import xmltodict

from swh.model.exceptions import ValidationError
from swh.model.identifiers import (
from swh.model.model import MetadataTargetType

logger = logging.getLogger(__name__)

[docs]def parse_xml(stream, encoding="utf-8"): namespaces = { "": "atom", "": "app", "": "dc", "": "codemeta", "": "sword", "": "swh", } data = xmltodict.parse( stream, encoding=encoding, namespaces=namespaces, process_namespaces=True, dict_constructor=dict, ) if "atom:entry" in data: data = data["atom:entry"] return data
[docs]def merge(*dicts): """Given an iterator of dicts, merge them losing no information. Args: *dicts: arguments are all supposed to be dict to merge into one Returns: dict merged without losing information """ def _extend(existing_val, value): """Given an existing value and a value (as potential lists), merge them together without repetition. """ if isinstance(value, (list, map, GeneratorType)): vals = value else: vals = [value] for v in vals: if v in existing_val: continue existing_val.append(v) return existing_val d = {} for data in dicts: if not isinstance(data, dict): raise ValueError("dicts is supposed to be a variable arguments of dict") for key, value in data.items(): existing_val = d.get(key) if not existing_val: d[key] = value continue if isinstance(existing_val, (list, map, GeneratorType)): new_val = _extend(existing_val, value) elif isinstance(existing_val, dict): if isinstance(value, dict): new_val = merge(existing_val, value) else: new_val = _extend([existing_val], value) else: new_val = _extend([existing_val], value) d[key] = new_val return d
[docs]def normalize_date(date): """Normalize date fields as expected by swh workers. If date is a list, elect arbitrarily the first element of that list If date is (then) a string, parse it through dateutil.parser.parse to extract a datetime. Then normalize it through swh.model.identifiers.normalize_timestamp. Returns The swh date object """ if isinstance(date, list): date = date[0] if isinstance(date, str): date = iso8601.parse_date(date) return normalize_timestamp(date)
[docs]def compute_metadata_context( swhid_reference: Union[SWHID, str] ) -> Tuple[MetadataTargetType, Dict[str, Any]]: """Given a SWHID object, determine the context as a dict. The parse_swhid calls within are not expected to raise (because they should have been caught early on). """ metadata_context: Dict[str, Any] = {"origin": None} if isinstance(swhid_reference, SWHID): object_type = MetadataTargetType(swhid_reference.object_type) assert object_type != MetadataTargetType.ORIGIN if swhid_reference.metadata: path = swhid_reference.metadata.get("path") metadata_context = { "origin": swhid_reference.metadata.get("origin"), "path": path.encode() if path else None, } snapshot = swhid_reference.metadata.get("visit") if snapshot: metadata_context["snapshot"] = parse_swhid(snapshot) anchor = swhid_reference.metadata.get("anchor") if anchor: anchor_swhid = parse_swhid(anchor) metadata_context[anchor_swhid.object_type] = anchor_swhid else: object_type = MetadataTargetType.ORIGIN return object_type, metadata_context
[docs]def parse_swh_reference(metadata: Dict) -> Optional[Union[str, SWHID]]: """Parse swh reference within the metadata dict (or origin) reference if found, None otherwise. <swh:deposit> <swh:reference> <swh:origin url=''/> </swh:reference> </swh:deposit> or: <swh:deposit> <swh:reference> <swh:object swhid="swh:1:dir:31b5c8cc985d190b5a7ef4878128ebfdc2358f49;origin=;visit=swh:1:snp:4fc1e36fca86b2070204bedd51106014a614f321;anchor=swh:1:rev:9c5de20cfb54682370a398fcc733e829903c8cba;path=/moranegg-AffectationRO-df7f68b/" /> </swh:deposit> Raises: ValidationError in case the swhid referenced (if any) is invalid Returns: Either swhid or origin reference if any. None otherwise. """ # noqa visit_swhid = None anchor_swhid = None swh_deposit = metadata.get("swh:deposit") if not swh_deposit: return None swh_reference = swh_deposit.get("swh:reference") if not swh_reference: return None swh_origin = swh_reference.get("swh:origin") if swh_origin: url = swh_origin.get("@url") if url: return url swh_object = swh_reference.get("swh:object") if not swh_object: return None swhid = swh_object.get("@swhid") if not swhid: return None swhid_reference = parse_swhid(swhid) if swhid_reference.metadata: anchor = swhid_reference.metadata.get("anchor") if anchor: anchor_swhid = parse_swhid(anchor) if anchor_swhid.object_type not in ALLOWED_QUALIFIERS_NODE_TYPE: error_msg = ( "anchor qualifier should be a core SWHID with type one of " f" {', '.join(ALLOWED_QUALIFIERS_NODE_TYPE)}" ) raise ValidationError(error_msg) visit = swhid_reference.metadata.get("visit") if visit: visit_swhid = parse_swhid(visit) if visit_swhid.object_type != SNAPSHOT: raise ValidationError( f"visit qualifier should be a core SWHID with type {SNAPSHOT}" ) if ( visit_swhid and anchor_swhid and visit_swhid.object_type == SNAPSHOT and anchor_swhid.object_type == SNAPSHOT ): logger.warn( "SWHID use of both anchor and visit targeting " f"a snapshot: {swhid_reference}" ) raise ValidationError( "'anchor=swh:1:snp:' is not supported when 'visit' is also provided." ) return swhid_reference