add(faser-draft): new model and transformation rules

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

@@ -0,0 +1,71 @@
+# -*- 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 FASER drafts model."""
+
+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 FaserDraftsModel(CdsOverdo):
+    """Translation model for FASER draft records."""
+
+    __query__ = (
+        '693__:FASER AND 591__b:"Draft" AND ('
+        "980__.a:NOTE OR "
+        "(980__.a:CONFERENCEPAPER OR 980__.a:SLIDE OR 980__.a:CONFERENCENOTE) OR "
+        "((980__.a:ARTICLE OR 980__.a:PREPRINT) -980__.a:CONFERENCEPAPER)"
+        ")"
+    )
+
+    __ignore_keys__ = {
+        "0248_a",
+        "0248_p",
+        "0248_q",
+        "035__d",  # oai harvest tag
+        "035__h",  # oai harvest tag
+        "035__m",  # oai harvest tag
+        "035__t",  # oai harvest tag
+        "100__m",  # email of contributor
+        "300__a",  # number of pages
+        "500__9",
+        "520__9",
+        "540__3",
+        "542__3",
+        "700__m",  # email of contributor
+        "773__p",  # TODO: title of the related links: 2959848, 2949836. values: twiki, internal comments
+        "8564_8",
+        "8564_s",
+        "8564_x",
+        "8564_y",  # file description - done by files dump
+        "916__y",  # year of publication, redundant
+        "937__c",  # last modified by
+        "937__s",  # last modification date
+        "960__a",  # base number
+        "961__c",  # CDS modification tag # TODO
+        "961__h",  # CDS modification tag # TODO
+        "961__l",  # CDS modification tag # TODO
+        "961__x",  # CDS modification tag # TODO
+        "981__a",  # duplicate record id
+        "980__b",  # additional article tag
+    }
+
+    _default_fields = {
+        "custom_fields": {},
+        # TODO: some records are missing creators, is it okay if we add a default value?
+        "creators": [
+            {"person_or_org": {"type": "organizational", "name": "FASER Collaboration"}}
+        ],
+    }
+
+
+faser_drafts_model = FaserDraftsModel(
+    bases=(rdm_base_publication_model,),
+    entry_point_group="cds_migrator_kit.migrator.rules.faser_drafts",
+)

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

@@ -0,0 +1,86 @@
+# -*- 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 FASER publication model."""
+
+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 FaserPublicationModel(CdsOverdo):
+    """Translation model for FASER publication records."""
+
+    __query__ = (
+        '693__:FASER -591__b:"Draft" AND ('
+        "980__.a:NOTE OR "
+        "(980__.a:CONFERENCEPAPER OR 980__.a:SLIDE OR 980__.a:CONFERENCENOTE) OR "
+        "((980__.a:ARTICLE OR 980__.a:PREPRINT) -980__.a:CONFERENCEPAPER)"
+        ")"
+    )
+
+    __ignore_keys__ = {
+        "0248_a",
+        "0248_p",
+        "0248_q",
+        "035__d",  # oai harvest tag
+        "035__h",  # oai harvest tag
+        "035__m",  # oai harvest tag
+        "035__t",  # oai harvest tag
+        "035__u",  # oai harvest tag
+        "037__c",  # arXiv subject
+        "100__m",  # email of contributor
+        "100__v",  # explanation of the affiliation? TODO: can we ignore? 2816452, 2924565
+        "270__m",  # contact person email
+        "300__a",  # number of pages
+        "500__9",
+        "520__9",
+        "540__3",
+        "542__3",
+        "700__m",  # email of contributor
+        "700__v",  # explanation of the affiliation? TODO: can we ignore? 2816452, 2924565
+        "8564_8",
+        "8564_s",
+        "8564_x",
+        "8564_y",  # file description - done by files dump
+        "8564_z",  # TODO value:Stamped by WebSubmit. Only EP records: 2888582, 2917427, 2853245, 2766207
+        "903__s",  # public: 2651328, 2642351, 2702868
+        "905__m",  # email of spokesperson
+        "916__y",  # year of publication, redundant
+        "937__c",  # last modified by
+        "937__s",  # last modification date
+        "960__a",  # base number
+        "961__c",  # CDS modification tag # TODO
+        "961__h",  # CDS modification tag # TODO
+        "961__l",  # CDS modification tag # TODO
+        "961__x",  # CDS modification tag # TODO
+        "981__a",  # duplicate record id
+        #  TODO
+        "773__0",  # TODO Check what is this field
+        "773__z",  # TODO: check what is this field
+        "773__p",  # TODO: title of the related links: 2959848, 2949836. values: twiki, internal comments
+        "773__x",  # TODO: one record 2652275
+        "542__h",  # TODO: year is in the wrong field 2816452
+        "0247_9",  # TODO: 20 records has 2 and 9 subfields.
+        "906__a",  # TODO: one record 2702868
+        "906__p",
+        "906__m",
+        "595_Ds",  # # TODO: some note? 2798675, 2791032
+        "595_Dd",
+        "595_Da",
+    }
+
+    _default_fields = {
+        "custom_fields": {},
+    }
+
+
+faser_publication_model = FaserPublicationModel(
+    bases=(rdm_base_publication_model,),
+    entry_point_group="cds_migrator_kit.migrator.rules.faser_publication",
+)

cds_migrator_kit/rdm/records/transform/transform.py

@@ -518,7 +518,8 @@ def subjects(json_entry):
             if item in keys:
                 keys.remove(item)
 
-        forgotten_keys = [key for key in keys if key not in list(metadata.keys())]
+        used_keys = list(metadata.keys()) + ["ep_approval"]
+        forgotten_keys = [key for key in keys if key not in used_keys]
         if forgotten_keys:
             raise ManualImportRequired("Unassigned metadata key", value=forgotten_keys)
         return {k: v for k, v in metadata.items() if v}

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

@@ -0,0 +1,106 @@
+# -*- 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 FASER drafts rules."""
+
+from dojson.errors import IgnoreKey
+from dojson.utils import for_each_value
+
+from cds_migrator_kit.errors import UnexpectedValue
+from cds_migrator_kit.rdm.records.transform.xml_processing.rules.base import urls
+
+from ...models.faser_drafts import faser_drafts_model as model
+
+
+@model.over("collection", "^690C_", override=True)
+@for_each_value
+def collection(self, key, value):
+    """Translates collection field."""
+    collection = value.get("a").strip()
+    if collection.lower() not in ["cern", "faser", "preprint"]:
+        raise UnexpectedValue(subfield="a", field=key, value=value)
+    subjects = self.get("subjects", [])
+    subjects.append(
+        {
+            "subject": f"collection:{collection}",
+        }
+    )
+    self["subjects"] = subjects
+    raise IgnoreKey("collection")
+
+
+@model.over("resource_type", "^980__", override=True)
+def resource_type(self, key, value):
+    """Translates resource_type."""
+    value = value.get("a")
+    if value:
+        value = value.strip().lower()
+
+    map_ = {
+        "article": {"id": "publication-article"},
+        "preprint": {"id": "publication-preprint"},
+        "scicommpubllhcc": {"id": "publication-article"},
+        "bookchapter": {"id": "publication-section"},
+        "faser_papers": {"id": "publication-article"},
+        "note": {"id": "publication-technicalnote"},
+        "conferencepaper": {"id": "publication-conferenceproceeding"},
+        "slide": {"id": "presentation"},
+        # TODO: is this correct?
+        "conferencenote": {"id": "publication-technicalnote"},
+    }
+    try:
+        return map_[value]
+    except KeyError:
+        raise UnexpectedValue("Unknown resource type (FASER)", field=key, value=value)
+
+
+@model.over("status", "^591__")
+@for_each_value
+def status(self, key, value):
+    """Translates status field."""
+    status = value.get("b").strip().lower()
+    # TODO: there is no other value, can we ignore?
+    if status != "draft":
+        raise UnexpectedValue(subfield="b", field=key, value=value)
+    raise IgnoreKey("status")
+
+
+@model.over("restriction_access_notes", "^5061_")
+@for_each_value
+def restriction_access_notes(self, key, value):
+    """Translates status field."""
+    restriction_access_notes = value.get("a", "").strip().lower()
+    # TODO: how to handle these values?
+    if restriction_access_notes and restriction_access_notes not in [
+        "faser-preprint",
+        "faser-slide",
+        "faser-confnote",
+        "faser-confpaper",
+    ]:
+        raise UnexpectedValue(subfield="a", field=key, value=value)
+    raise IgnoreKey("restriction_access_notes")
+
+
+@model.over("related_ids", "(^773__)", override=True)
+@for_each_value
+def related_ids(self, key, value):
+    """Translated related links."""
+    related_link = value.get("u", "")
+    if not related_link:
+        raise UnexpectedValue(subfield="u", field=key, value=value)
+    # TODO: how to transform? https://cds.cern.ch/record/1452204/export/xm
+    # Transform like the base `urls` rule, we lose the title "p" it's okay?
+    rel_ids = urls(self, key, value)
+    if not rel_ids:
+        raise IgnoreKey("related_ids")
+    rel_id = rel_ids[0]
+    related_identifiers = self.get("related_identifiers", [])
+    if rel_id not in related_identifiers:
+        related_identifiers.append(rel_id)
+    self["related_identifiers"] = related_identifiers
+
+    raise IgnoreKey("related_ids")