feat(rdm.migration): Configurable auto-adding secondary community

cds_migrator_kit/errors.py

@@ -85,3 +85,9 @@ class RecordFlaggedCuration(CDSMigrationException):
     """Record statistics error."""
 
     description = "[Record needs to be curated]"
+
+
+class MissingConfiguration(CDSMigrationException):
+    """Missing configuration exception."""
+
+    description = "[Missing configuration]"

cds_migrator_kit/rdm/records/transform/config.py

@@ -158,6 +158,24 @@
 
 FILE_SUBFORMATS_TO_DROP = ["pdfa", "unstamped"]
 
+
+# Public research publication resource types that are auto-included in the CERN Research community.
+CERN_SCIENTIFIC_RESOURCE_TYPES = {
+    "publication-dissertation",  # Already included by the migrator for thesis records.
+    "publication-book",
+    "publication-section",
+    "publication-conferencepaper",
+    "publication-conferenceproceeding",
+    "publication-conferencenote",
+    "publication-journal",
+    "publication-article",
+    "publication-preprint",
+    "publication-report",
+    "publication-technicalnote",
+    "publication-note",
+    "publication",
+}
+
 # Legacy experiment names remapped to vocabulary ids before lookup
 EXPERIMENT_ALIASES = {
     "t2k": "re13",

cds_migrator_kit/rdm/records/transform/transform.py

@@ -6,7 +6,7 @@
 # the terms of the MIT License; see LICENSE file for more details.
 
 """CDS-RDM transform step module."""
-import datetime
+
 import logging
 from collections import OrderedDict
 from copy import deepcopy
@@ -32,6 +32,7 @@
 
 from cds_migrator_kit.errors import (
     ManualImportRequired,
+    MissingConfiguration,
     MissingRequiredField,
     MultipleModelsMatched,
     RecordFlaggedCuration,
@@ -43,6 +44,7 @@
     VOCABULARIES_NAMES_SCHEMES,
 )
 from cds_migrator_kit.rdm.records.transform.config import (
+    CERN_SCIENTIFIC_RESOURCE_TYPES,
     EXPERIMENT_ALIASES,
     FILE_SUBFORMATS_TO_DROP,
     IDENTIFIERS_SCHEMES_TO_DROP,
@@ -857,9 +859,42 @@ def __init__(
         self.db_state = {"affiliations": CDSMigrationAffiliationMapping}
         super().__init__(workers, throw)
 
+    def _should_add_scientific_community(self, entry, record):
+        """
+        Determine if the scientific community should be added to the record.
+
+        The scientific community is added if the following conditions are met:
+        - The record is public
+        - The files are public
+        - The record has a resource type in the CERN Scientific resource types
+        """
+        if self.restricted or record.get("access") != "public":
+            return False
+        if any(file.get("status") for file in entry.get("files", [])):
+            return False
+        resource_type_id = (
+            record.get("json", {})
+            .get("metadata", {})
+            .get("resource_type", {})
+            .get("id")
+        )
+        return resource_type_id in CERN_SCIENTIFIC_RESOURCE_TYPES
+
     def _communities_ids(self, entry, record):
         communities = record.get("communities", [])
         communities = self.communities_ids + [slug for slug in communities]
+
+        scientific_community = current_app.config.get(
+            "CDS_CERN_SCIENTIFIC_COMMUNITY_ID"
+        )
+        if not scientific_community:
+            raise MissingConfiguration(
+                "CDS_CERN_SCIENTIFIC_COMMUNITY_ID is not configured"
+            )
+        if self._should_add_scientific_community(entry, record):
+            if scientific_community not in communities:
+                communities.append(scientific_community)
+
         if communities:
             return {"ids": communities, "default": self.communities_ids[0]}
         return {}
@@ -913,6 +948,7 @@ def _transform(self, entry):
             ManualImportRequired,
             MissingRequiredField,
             MultipleModelsMatched,
+            MissingConfiguration,
         ) as e:
             migration_logger.add_log(e, record=entry)
 

tests/cds-rdm/conftest.py

@@ -278,7 +278,7 @@ def running_app(
 
 
 @pytest.fixture
-def test_app(running_app):
+def test_app(running_app, cern_scientific_community):
     """Get current app."""
     running_app.app.config["RDM_PERSISTENT_IDENTIFIERS"]["doi"]["required"] = False
     running_app.app.config["RDM_PARENT_PERSISTENT_IDENTIFIERS"]["doi"][
@@ -1703,6 +1703,18 @@ def community(running_app, db):
     return comm
 
 
+@pytest.fixture()
+def cern_scientific_community(running_app, db, app):
+    """A CERN Research community fixture."""
+    comm = Community.create({})
+    comm.slug = "cern-research"
+    comm.metadata = {"title": "CERN Research"}
+    comm.commit()
+    db.session.commit()
+    app.config["CDS_CERN_SCIENTIFIC_COMMUNITY_ID"] = str(comm.id)
+    return comm
+
+
 # @pytest.fixture()
 # def users(app, db):
 #     """Create example user."""

tests/cds-rdm/test_bulletin_issue.py

@@ -80,8 +80,3 @@ def test_bulletin_issue(
         )
         if record["legacy_recid"] == "2234683":
             parent_related_identifier(loaded_rec)
-
-
-# FAILED tests/cds-rdm/test_access_permissions.py::test_access_permissions - assert 2 == 3
-# FAILED tests/cds-rdm/test_bulletin_issue.py::test_bulletin_issue - AssertionError: assert [{'identifier...heme': 'cds'}] == [{'identifier...heme': 'cds'}]
-# FAILED tests/cds-rdm/test_full_migration.py::test_full_migration_stream - assert 0 == 2

tests/cds-rdm/test_ep_approval_entry.py

@@ -14,11 +14,11 @@
 import pytest
 
 from cds_migrator_kit.errors import UnexpectedValue
-from cds_migrator_kit.rdm.migration_config import CDS_CERN_SCIENTIFIC_COMMUNITY_ID
 from cds_migrator_kit.rdm.records.load.ep_approval_entry import (
     EPPHAPP_FILE_TYPE,
     PublicEntry,
     RestrictedEntry,
+    _cern_scientific_community_id,
 )
 
 RECID = "12345"
@@ -526,22 +526,22 @@ def test_public_adds_cern_scientific_community(self, app):
             entry, _make_approval_request(), _make_migration_logger()
         ).build()
 
-        assert CDS_CERN_SCIENTIFIC_COMMUNITY_ID in (
+        assert _cern_scientific_community_id() in (
             result["parent"]["json"]["communities"]["ids"]
         )
 
     def test_public_does_not_duplicate_community(self, app):
         entry = _make_entry(_versions_with_epphapp())
         entry["parent"]["json"]["communities"]["ids"] = [
             "example-community",
-            CDS_CERN_SCIENTIFIC_COMMUNITY_ID,
+            _cern_scientific_community_id(),
         ]
         result = PublicEntry(
             entry, _make_approval_request(), _make_migration_logger()
         ).build()
 
         community_ids = result["parent"]["json"]["communities"]["ids"]
-        assert community_ids.count(CDS_CERN_SCIENTIFIC_COMMUNITY_ID) == 1
+        assert community_ids.count(_cern_scientific_community_id()) == 1
 
 
 class TestEntryImmutability:

tests/cds-rdm/test_full_migration.py

@@ -471,6 +471,7 @@ def test_full_migration_stream(
     superuser_identity,
     orcid_name_data,
     community,
+    cern_scientific_community,  # Fixture for the CERN Scientific community, so that it gets auto included for PUBLIC-ations.
     mocker,
     groups,
 ):
@@ -515,6 +516,8 @@ def test_full_migration_stream(
         if record["legacy_recid"] == "2783104":
             file_restricted(loaded_rec)
             parent_access_fields(loaded_rec)
+            # Skip scientific community inclusion check since it has restricted files
+            continue
         if record["legacy_recid"] == "2046076":
             irregular_exp_field(loaded_rec)
         if record["legacy_recid"] == "2041388":
@@ -526,6 +529,8 @@ def test_full_migration_stream(
         if record["legacy_recid"] == "2294138":
             author_with_inspire(loaded_rec)
 
+        # Check if the record is also included in the CERN Scientific community since it is a publication report
+        scientific_community_inclusion(loaded_rec, cern_scientific_community.id)
     # Check if remote account has the correct metadata
     # Check if user profile has the correct metadata
     user_metadata()
@@ -595,3 +600,9 @@ def user_metadata():
         "person_id": "11115",
         "department": "IT",
     }
+
+
+def scientific_community_inclusion(record, cern_scientific_community_uuid):
+    """Checks if the record is included in the CERN Scientific community."""
+    assert str(cern_scientific_community_uuid) in record._record.parent.communities.ids
+    assert cern_scientific_community_uuid != record._record.parent.communities.default

tests/cds-rdm/test_scientific_community.py

@@ -0,0 +1,175 @@
+# -*- coding: utf-8 -*-
+#
+# Copyright (C) 2026 CERN.
+#
+# CDS-RDM is free software; you can redistribute it and/or modify it under
+# the terms of the MIT License; see LICENSE file for more details.
+
+"""Tests for auto-inclusion in the CERN Research community."""
+
+from unittest.mock import MagicMock
+
+import pytest
+
+from cds_migrator_kit.errors import MissingConfiguration
+from cds_migrator_kit.rdm.records.transform.config import (
+    CERN_SCIENTIFIC_RESOURCE_TYPES,
+)
+from cds_migrator_kit.rdm.records.transform.transform import CDSToRDMRecordTransform
+
+
+def _test_record(
+    access="public",
+    resource_type="publication-preprint",
+    communities=[],
+    recid="123456",
+):
+    """Build a minimal CDSToRDMRecordEntry.transform() output for community tests."""
+    metadata = {
+        "title": "Test record",
+        "publication_date": "2020-01-01",
+    }
+    if resource_type is not None:
+        metadata["resource_type"] = {"id": resource_type}
+
+    return {
+        "recid": recid,
+        "access": access,
+        "communities": communities,
+        "json": {
+            "files": {"enabled": False},
+            "metadata": metadata,
+        },
+    }
+
+
+def _test_entry(files_restricted=False):
+    """Build a minimal raw dump entry for community tests."""
+    status = "restricted" if files_restricted else ""
+    return {"files": [{"status": status}]}
+
+
+@pytest.fixture
+def transform(tmp_path, community):
+    """Transform instance with a collection community configured."""
+    return CDSToRDMRecordTransform(
+        files_dump_dir=tmp_path,
+        missing_users=tmp_path,
+        communities_ids=[str(community.id)],
+        migration_logger=MagicMock(),
+    )
+
+
+class TestCommunitiesIds:
+    """Test CDSToRDMRecordTransform._communities_ids()."""
+
+    def test_adds_scientific_community_for_public_research_test_record(
+        self, transform, community, cern_scientific_community
+    ):
+        """Public research records are included in the CERN Scientific community."""
+        record = _test_record()
+        record["json"]["files"] = {"enabled": True}
+        result = transform._communities_ids(
+            _test_entry(),
+            record,
+        )
+
+        assert result == {
+            "ids": [str(community.id), str(cern_scientific_community.id)],
+            "default": str(community.id),
+        }
+
+    def test_keep_collection_community_as_default(
+        self, transform, community, cern_scientific_community
+    ):
+        """Collection community remains the default when CERN Scientific community is added."""
+        result = transform._communities_ids(
+            _test_entry(),
+            _test_record(communities=["test-community"]),
+        )
+
+        assert result["default"] == str(community.id)
+        assert result["ids"] == [
+            str(community.id),
+            "test-community",
+            str(cern_scientific_community.id),
+        ]
+
+    @pytest.mark.parametrize("resource_type", CERN_SCIENTIFIC_RESOURCE_TYPES)
+    def test_research_resource_types(
+        self, transform, cern_scientific_community, resource_type
+    ):
+        """All configured public research resource types trigger inclusion."""
+        result = transform._communities_ids(
+            _test_entry(), _test_record(resource_type=resource_type)
+        )
+
+        assert str(cern_scientific_community.id) in result["ids"]
+        assert cern_scientific_community.id != result["default"]
+
+    def test_skip_restricted_test_record(
+        self, transform, community, cern_scientific_community
+    ):
+        """Restricted records are not included in the CERN Research community."""
+        result = transform._communities_ids(
+            _test_entry(), _test_record(access="restricted")
+        )
+
+        assert result == {
+            "ids": [str(community.id)],
+            "default": str(community.id),
+        }
+
+    def test_skip_restricted_files(
+        self, transform, community, cern_scientific_community
+    ):
+        """Records with restricted files are not included in the CERN Scientific community."""
+        result = transform._communities_ids(
+            _test_entry(files_restricted=True), _test_record()
+        )
+
+        assert result == {
+            "ids": [str(community.id)],
+            "default": str(community.id),
+        }
+
+    def test_skip_non_research_resource_type(
+        self, transform, community, cern_scientific_community
+    ):
+        """Non-research resource types are not included in the CERN Scientific community."""
+        result = transform._communities_ids(
+            _test_entry(), _test_record(resource_type="other")
+        )
+
+        assert result == {
+            "ids": [str(community.id)],
+            "default": str(community.id),
+        }
+
+    def test_skip_when_stream_is_restricted(
+        self, tmp_path, community, cern_scientific_community
+    ):
+        """Records on restricted migration streams are not included in the CERN Scientific community."""
+        transform = CDSToRDMRecordTransform(
+            files_dump_dir=tmp_path,
+            missing_users=tmp_path,
+            communities_ids=[str(community.id)],
+            restricted=True,
+            migration_logger=MagicMock(),
+        )
+
+        result = transform._communities_ids(_test_entry(), _test_record())
+
+        assert result == {
+            "ids": [str(community.id)],
+            "default": str(community.id),
+        }
+
+    def test_raise_when_cern_scientific_community_not_configured(
+        self, test_app, transform, community, monkeypatch
+    ):
+        """No CERN Scientific community is added when config is unset."""
+        monkeypatch.setitem(test_app.config, "CDS_CERN_SCIENTIFIC_COMMUNITY_ID", None)
+
+        with pytest.raises(MissingConfiguration):
+            transform._communities_ids(_test_entry(), _test_record())