diff --git a/docs/specs/spec-meta-deposit.rst b/docs/specs/spec-meta-deposit.rst
--- a/docs/specs/spec-meta-deposit.rst
+++ b/docs/specs/spec-meta-deposit.rst
@@ -109,12 +109,7 @@
-
+
diff --git a/swh/deposit/parsers.py b/swh/deposit/parsers.py
--- a/swh/deposit/parsers.py
+++ b/swh/deposit/parsers.py
@@ -8,6 +8,7 @@
"""
+from typing import Dict, Optional, Union
from xml.parsers.expat import ExpatError
from django.conf import settings
@@ -15,6 +16,7 @@
import xmltodict
from swh.deposit.errors import ParserError
+from swh.model.identifiers import SWHID, parse_swhid
class SWHFileUploadZipParser(FileUploadParser):
@@ -51,6 +53,7 @@
"http://purl.org/dc/terms/": None,
"https://doi.org/10.5063/SCHEMA/CODEMETA-2.0": "codemeta",
"http://purl.org/net/sword/": "sword",
+ "https://www.softwareheritage.org/schema/2018/deposit": "swh",
}
data = xmltodict.parse(
@@ -101,3 +104,52 @@
return SWHXMLParser().parse(raw_content)
except ExpatError as e:
raise ParserError(str(e))
+
+
+def parse_swh_reference(metadata: Dict) -> Optional[Union[str, SWHID]]:
+ """Parse swh reference within the metadata dict (or origin) reference if found, None
+ otherwise.
+
+
+
+
+
+
+
+ or:
+
+
+
+
+
+
+ Raises:
+ ValidationError in case the swhid referenced (if any) is invalid
+
+ Returns:
+ Either swhid or origin reference if any. None otherwise.
+
+ """ # noqa
+ 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
+ return parse_swhid(swhid)
diff --git a/swh/deposit/tests/api/test_parser.py b/swh/deposit/tests/api/test_parser.py
--- a/swh/deposit/tests/api/test_parser.py
+++ b/swh/deposit/tests/api/test_parser.py
@@ -1,4 +1,4 @@
-# Copyright (C) 2018-2019 The Software Heritage developers
+# 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
@@ -6,7 +6,11 @@
from collections import OrderedDict
import io
-from swh.deposit.parsers import SWHXMLParser
+import pytest
+
+from swh.deposit.parsers import SWHXMLParser, parse_swh_reference, parse_xml
+from swh.model.exceptions import ValidationError
+from swh.model.identifiers import parse_swhid
def test_parsing_without_duplicates():
@@ -127,3 +131,102 @@
]
)
assert expected_dict == actual_result
+
+
+@pytest.fixture
+def xml_with_origin_reference():
+ xml_data = """
+
+
+
+
+
+
+
+ """
+ return xml_data.strip()
+
+
+def test_parse_swh_reference_origin(xml_with_origin_reference):
+ url = "https://url"
+ xml_data = xml_with_origin_reference.format(url=url)
+ metadata = parse_xml(xml_data)
+
+ actual_origin = parse_swh_reference(metadata)
+ assert actual_origin == url
+
+
+@pytest.fixture
+def xml_with_empty_reference():
+ xml_data = """
+
+
+ {swh_reference}
+
+
+ """
+ return xml_data.strip()
+
+
+@pytest.mark.parametrize(
+ "xml_ref",
+ [
+ "",
+ "",
+ "",
+ """""",
+ ],
+)
+def test_parse_swh_reference_empty(xml_with_empty_reference, xml_ref):
+ xml_body = xml_with_empty_reference.format(swh_reference=xml_ref)
+ metadata = parse_xml(xml_body)
+
+ assert parse_swh_reference(metadata) is None
+
+
+@pytest.fixture
+def xml_with_swhid():
+ xml_data = """
+
+
+
+
+
+
+
+ """
+ return xml_data.strip()
+
+
+@pytest.mark.parametrize(
+ "swhid",
+ [
+ "swh:1:dir:31b5c8cc985d190b5a7ef4878128ebfdc2358f49;origin=https://hal.archives-ouvertes.fr/hal-01243573;visit=swh:1:snp:4fc1e36fca86b2070204bedd51106014a614f321;anchor=swh:1:rev:9c5de20cfb54682370a398fcc733e829903c8cba;path=/moranegg-AffectationRO-df7f68b/", # noqa
+ "swh:1:dir:31b5c8cc985d190b5a7ef4878128ebfdc2358f49",
+ ],
+)
+def test_parse_swh_reference_swhid(swhid, xml_with_swhid):
+ xml_data = xml_with_swhid.format(swhid=swhid)
+ metadata = parse_xml(xml_data)
+
+ actual_swhid = parse_swh_reference(metadata)
+ assert actual_swhid is not None
+
+ expected_swhid = parse_swhid(swhid)
+ assert actual_swhid == expected_swhid
+
+
+def test_parse_swh_reference_invalid_swhid(xml_with_swhid):
+ """Unparsable swhid should raise
+
+ """
+ invalid_swhid = "swh:1:dir:31b5c8cc985d190b5a7ef4878128ebfdc235"
+ xml_invalid_swhid = xml_with_swhid.format(swhid=invalid_swhid)
+ metadata = parse_xml(xml_invalid_swhid)
+
+ with pytest.raises(ValidationError, match="Unexpected length"):
+ parse_swh_reference(metadata)