github: Ignore archive_url and issues_url; use custom codemeta:issueTracker

The codemeta crosswalk maps archive_url and issues_url to codemeta
properties, but they are not actually useable because they are
link templates to API endpoints.

swh/indexer/metadata_dictionary/github.py

@@ -8,7 +8,7 @@ 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, FORGEFED, SCHEMA
+from swh.indexer.namespaces import ACTIVITYSTREAMS, CODEMETA, FORGEFED, SCHEMA
 
 from .base import BaseExtrinsicMapping, JsonMapping, produce_terms
 from .utils import prettyprint_graph  # noqa
@@ -24,9 +24,7 @@ class GitHubMapping(BaseExtrinsicMapping, JsonMapping):
         "clone_url": SCHEMA.codeRepository,
     }
     uri_fields = [
-        "archive_url",
         "clone_url",
-        "issues_url",
     ]
     date_fields = [
         "created_at",
@@ -46,6 +44,15 @@ class GitHubMapping(BaseExtrinsicMapping, JsonMapping):
         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"),
+                )
+            )
+
     def get_root_uri(self, content_dict: dict) -> URIRef:
         if isinstance(content_dict.get("html_url"), str):
             return URIRef(content_dict["html_url"])

swh/indexer/tests/metadata_dictionary/test_github.py

@@ -63,6 +63,8 @@ def test_compute_metadata_github():
   "created_at": "2017-01-31T13:05:39Z",
   "updated_at": "2022-06-22T08:02:20Z",
   "pushed_at": "2022-06-29T09:01:08Z",
+  "archive_url": "https://api.github.com/repos/SoftwareHeritage/swh-indexer/{archive_format}{/ref}",
+  "issues_url": "https://api.github.com/repos/SoftwareHeritage/swh-indexer/issues{/number}",
   "git_url": "git://github.com/SoftwareHeritage/swh-indexer.git",
   "ssh_url": "git@github.com:SoftwareHeritage/swh-indexer.git",
   "clone_url": "https://github.com/SoftwareHeritage/swh-indexer.git",
@@ -114,7 +116,7 @@ def test_compute_metadata_github():
   "subscribers_count": 6
 }
 
-    """
+    """  # noqa
     result = MAPPINGS["GitHubMapping"]().translate(content)
     assert result == {
         "@context": CONTEXT,
@@ -158,3 +160,19 @@ def test_github_topics():
         "type": "forge:Repository",
         "id": "https://github.com/SoftwareHeritage/swh-indexer",
     }
+
+
+def test_github_issues():
+    content = b"""
+{
+  "html_url": "https://github.com/SoftwareHeritage/swh-indexer",
+  "has_issues": true
+}
+    """
+    result = MAPPINGS["GitHubMapping"]().translate(content)
+    assert result == {
+        "@context": CONTEXT,
+        "type": "forge:Repository",
+        "id": "https://github.com/SoftwareHeritage/swh-indexer",
+        "issueTracker": "https://github.com/SoftwareHeritage/swh-indexer/issues",
+    }