add(hse): new model and transformation rules

cds_migrator_kit/rdm/records/transform/config.py

@@ -227,12 +227,18 @@
     "witch is433": "IS433",
     "isoltrap": "IS394",
     "is485 (rex-isolde)": "IS485",
+    "charm": "ea-irrad mixed-field",
 }
 
 # 693__e values that are keywords rather than real experiments (curated as
 # "mots clef"), mapped to subjects instead of cern:experiments
 EXPERIMENTS_AS_SUBJECTS = ["d3", "r104", "r105a"]
 
+# Legacy accelerator names remapped to vocabulary ids before lookup
+ACCELERATOR_ALIASES = {
+    "cern aa": "antiproton accumulator",
+}
+
 # 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.

cds_migrator_kit/rdm/records/transform/mappers/custom_fields.py

@@ -13,6 +13,7 @@
     UnexpectedValue,
 )
 from cds_migrator_kit.rdm.records.transform.config import (
+    ACCELERATOR_ALIASES,
     EXPERIMENT_ALIASES,
     EXPERIMENTS_AS_SUBJECTS,
 )
@@ -169,6 +170,9 @@ def apply(self, ctx):
                 "cern recognized expt.",
             ]:
                 continue
+            accelerator = ACCELERATOR_ALIASES.get(
+                accelerator.lower().strip(), accelerator
+            )
             result = search_vocabulary(accelerator, "accelerators")
             if result and result not in accelerators_out:
                 accelerators_out.append(result)

cds_migrator_kit/rdm/records/transform/models/hse.py

@@ -0,0 +1,47 @@
+# -*- 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.
+
+"""CDS-RDM HSE model."""
+from cds_migrator_kit.rdm.records.transform.models._config import IGNORE_SYSTEM_KEYS
+from cds_migrator_kit.rdm.records.transform.models.base_publication_record import (
+    rdm_base_publication_model,
+)
+from cds_migrator_kit.transform.overdo import CdsOverdo
+
+
+class HseModel(CdsOverdo):
+    """Translation model for HSE records."""
+
+    __query__ = """595__:CERN-HSE"""
+
+    __ignore_keys__ = IGNORE_SYSTEM_KEYS | {
+        "100__j",
+        "300__c",  # value only '9 p': 2712787, 2789695
+        "506__a",  # Access note restrictions, only 2 records have "Free" but they are restricted
+        "595__a",  # Unneeded: CERN-HSE or "To be check by ..."
+        "690C_a",  # Duplicate of 980__a resource type
+        # "6531_9",  # keyword scheme
+        "700__j",
+        "720__a",  # Author's duplicate
+        "7870_r",  # detailed description of record relation (2862345)
+        # Drop citation field https://cds.cern.ch/record/202927/
+        "913__t",  # Citation field
+        "913__y",  # Citation field
+        "913__v",  # Citation field
+        "913__c",  # Citation field
+    }
+
+    _default_fields = {
+        "custom_fields": {},
+        "record_restriction": "public",  # All HSE papers are public, but some of them are missing 963:PUBLIC
+    }
+
+
+hse_model = HseModel(
+    bases=(rdm_base_publication_model,),
+    entry_point_group="cds_migrator_kit.migrator.rules.hse",
+)

cds_migrator_kit/rdm/records/transform/models/radio_protection.py

@@ -0,0 +1,28 @@
+# -*- 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.
+
+"""CDS-RDM Radio Protection model."""
+from cds_migrator_kit.rdm.records.transform.models.hse import HseModel, hse_model
+
+
+class RadioProtectionModel(HseModel):
+    """Translation model for Radio Protection records."""
+
+    __query__ = """980__:RP_RESTRICTED"""
+
+    _default_fields = {
+        "resource_type": {
+            "id": "publication-preprint"
+        },  # Default (only 2 articles which the rules will update)
+        "custom_fields": {},
+    }
+
+
+radio_protection_model = RadioProtectionModel(
+    bases=(hse_model,),  # inherits the HSE rules
+    entry_point_group="cds_migrator_kit.migrator.rules.radio_protection",
+)

cds_migrator_kit/rdm/records/transform/xml_processing/rules/hse.py

@@ -0,0 +1,96 @@
+# -*- 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.
+
+"""CDS-RDM HSE rules."""
+
+from dojson.errors import IgnoreKey
+
+from cds_migrator_kit.errors import UnexpectedValue
+from cds_migrator_kit.transform.xml_processing.quality.decorators import (
+    for_each_value,
+)
+
+from ...models.hse import hse_model as model
+
+# from .base import copyrights as base_copyright
+# from .base import licenses as base_licenses
+from .base import note as base_note
+from .hr import corpo_author
+from .it import imprint
+from .research import resource_type as research_resource_type
+
+# from dojson.utils import filter_values, force_list
+
+
+# from cds_migrator_kit.transform.xml_processing.rules.base import (
+#     description as base_description,
+# )
+
+
+model.over("creators", "(^110__)")(corpo_author)
+model.over("imprint_info", "(^250__)")(imprint)  # Only one record: 156663
+
+
+@model.over("physical_copies", "^964__", override=True)
+@for_each_value
+def physical_copies(self, key, value):
+    """Translates physical copies field."""
+    number_of_copies = value.get("a").strip()
+    if number_of_copies and number_of_copies not in [
+        "0002",
+        "0001",
+        "0004",
+        "0003",
+        "0005",
+    ]:
+        raise UnexpectedValue(subfield="a", value=number_of_copies, field=key)
+    raise IgnoreKey("physical_copies")
+
+
+@model.over("internal_notes", "^595__")
+@for_each_value
+def note_hse(self, key, value):
+    """Translates notes. HSE Articles query is 595__:CERN-HSE."""
+    _note = (value.get("a") or "").strip()
+    if _note in ("CERN-HSE", "To be check by Salome"):
+        raise IgnoreKey("internal_notes")
+    else:
+        return base_note(self, key, value)
+
+
+# Reuse the generic 980__/697C_ resource type rule. HSE papers have no default,
+# a record without a match is rejected by ``ResourceTypeMapper``.
+model.over("resource_type", "(^980__)|(^697C_)", override=True)(research_resource_type)
+
+
+# @model.over("rights", "^540__", override=True)
+# @for_each_value
+# @filter_values
+# def licenses(self, key, value):
+#     """Translates rights."""
+#     material = value.get("3")
+#     if material and material.lower() not in ["publication"]:
+#         raise UnexpectedValue(subfield="3", value=material, field=key)
+#     license = base_licenses(self, key, value)
+#     return license[0]
+
+
+# @model.over("copyright", "^542__", override=True)
+# def copyrights(self, key, value):
+#     """Translates copyright."""
+#     material = value.get("3")
+#     if material and material.lower() not in ["publication"]:
+#         raise UnexpectedValue(subfield="3", value=material, field=key)
+#     return base_copyright(self, key, value)
+# @model.over("description", "^520__", override=True)
+# def description(self, key, value):
+#     """Translates description."""
+#     abstract_number = value.get("9")
+#     # TODO: can we ignore?
+#     if abstract_number and abstract_number not in ["JACoW", "submitter"]:
+#         raise UnexpectedValue(subfield="9", value=abstract_number, field=key)
+#     return base_description(self, key, value)

cds_migrator_kit/rdm/records/transform/xml_processing/rules/radio_protection.py

@@ -0,0 +1,35 @@
+# -*- 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.
+
+"""CDS-RDM Radio Protection rules."""
+
+from ...models.radio_protection import radio_protection_model as model
+from .research import resource_type as research_resource_type
+
+# The rules (including the 980__/697C_ resource type) are inherited from the HSE
+# model, only the model defaults differ (resource_type: publication-preprint).
+
+# Radio Protection also has the document type in the 490__a series. Clean the
+# value to the legacy document types known by the generic rule, it shares the
+# rank with 980__ so the highest priority value wins. Replaces the inherited
+# series rule.
+SERIES_TO_DOCUMENT_TYPE = {
+    "internal report": "report",
+    "rp annual report": "report",
+    "divisionnal report": "report",
+    "technical inspection": "report",
+    "technical notes": "technical note",
+    "technical memorandum": "technical note",
+}
+
+
+@model.over("resource_type", "^490__", override_tag=True)
+def series_resource_type(self, key, value):
+    """Translates resource_type from the 490__a series."""
+    series = (value.get("a") or "").strip().lower()
+    cleaned = {"a": SERIES_TO_DOCUMENT_TYPE.get(series, series)}
+    return research_resource_type(self, key, cleaned)

migration_cli/collections/hse.yaml

@@ -0,0 +1,6 @@
+name: hse
+queries:
+  - q: "595__:CERN-HSE"
+    dump_file_prefix: hse
+communities:
+    - hse

migration_cli/collections/radio_protection.yaml

@@ -0,0 +1,8 @@
+name: radio_protection
+queries:
+  - q: "980__:RP_RESTRICTED"
+    dump_file_prefix: radio_protection
+flags:
+  restricted: true
+communities:
+    - radio-protection

setup.cfg

@@ -94,6 +94,8 @@ cds_migrator_kit.migrator.models =
     it = cds_migrator_kit.rdm.records.transform.models.it:it_model
     it_meetings = cds_migrator_kit.rdm.records.transform.models.it_meetings:it_meetings_model
     hr = cds_migrator_kit.rdm.records.transform.models.hr:hr_model
+    hse = cds_migrator_kit.rdm.records.transform.models.hse:hse_model
+    radio_protection = cds_migrator_kit.rdm.records.transform.models.radio_protection:radio_protection_model
     sy = cds_migrator_kit.rdm.records.transform.models.sy:sy_model
     te = cds_migrator_kit.rdm.records.transform.models.te:te_model
     en = cds_migrator_kit.rdm.records.transform.models.en:en_model
@@ -183,6 +185,12 @@ cds_migrator_kit.migrator.rules.hr =
     base = cds_migrator_kit.transform.xml_processing.rules.base
     base_records = cds_migrator_kit.rdm.records.transform.xml_processing.rules.base
     hr = cds_migrator_kit.rdm.records.transform.xml_processing.rules.hr
+cds_migrator_kit.migrator.rules.hse =
+    base = cds_migrator_kit.transform.xml_processing.rules.base
+    base_records = cds_migrator_kit.rdm.records.transform.xml_processing.rules.base
+    hse = cds_migrator_kit.rdm.records.transform.xml_processing.rules.hse
+cds_migrator_kit.migrator.rules.radio_protection =
+    radio_protection = cds_migrator_kit.rdm.records.transform.xml_processing.rules.radio_protection
 cds_migrator_kit.migrator.rules.sy =
     base = cds_migrator_kit.transform.xml_processing.rules.base
     base_records = cds_migrator_kit.rdm.records.transform.xml_processing.rules.base