Exemplo n.º 1
0
def marcxml_filter(record):
    from inspire.dojson.hep import hep2marc
    from inspire.dojson.hepnames import hepnames2marc
    from inspire.dojson.utils import legacy_export_as_marc

    collections = [collection["primary"] for collection in record["collections"]]

    if "HEP" in collections:
        return legacy_export_as_marc(hep2marc.do(record))
    elif "HEPNAMES" in collections:
        return legacy_export_as_marc(hepnames2marc.do(record))
Exemplo n.º 2
0
 def _create_marcxml_record(obj, eng):
     from inspire.dojson.hepnames import hepnames2marc
     from inspire.dojson.utils import legacy_export_as_marc
     obj.extra_data["marcxml"] = legacy_export_as_marc(
         hepnames2marc.do(obj.data)
     )
     obj.log.info("Produced MarcXML: \n {}".format(
         obj.extra_data["marcxml"])
     )
Exemplo n.º 3
0
 def _finalize_sip(obj, eng):
     from inspire.dojson.utils import legacy_export_as_marc
     model = eng.workflow_definition.model(obj)
     sip = model.get_latest_sip()
     sip.package = legacy_export_as_marc(processor.do(sip.metadata))
     model.update()
def do_upgrade():
    """Implement your upgrades here."""
    from invenio_workflows.models import (
        BibWorkflowObject,
        Workflow,
        ObjectVersion
    )
    from invenio_workflows.registry import workflows
    from invenio_deposit.models import Deposition

    from inspire.dojson.utils import legacy_export_as_marc
    from inspire.dojson.hep import hep2marc
    from inspire.modules.workflows.dojson import bibfield
    from inspire.modules.workflows.models import Payload

    def rename_object_action(obj):
        if obj.get_action() == "arxiv_approval":
            obj.set_action("hep_approval", obj.get_action_message())

    def reset_workflow_object_states(obj):
        """Fix workflow positions and states.

        Old states from Prod/QA:
        {(), (0,), (5, 3, 14), (5, 3, 14, 0), (5, 3, 15), (5, 3, 15, 1)}

        {(),
         (0,),
         (5,),
         (5, 3, 1),
         (5, 3, 10),
         (5, 3, 11),
         (5, 3, 12),
         (5, 3, 14),
         (5, 3, 14, 0),
         (6, 3, 4)}

        OLD -> NEW
        5, 3, 14 -> 0 end
        5, 3, 10 -> 14, 0 halted
        """
        pos = obj.get_current_task()
        if obj.version == ObjectVersion.COMPLETED:
            obj.save(task_counter=[len(workflows.get(obj.workflow.name).workflow) - 1])
            return
        elif obj.version == ObjectVersion.RUNNING:
            # Running? Nah that cannot be.
            obj.version = ObjectVersion.ERROR
        try:
            obj.get_current_task_info()
        except IndexError:
            # The current task counter is Invalid
            obj.version = ObjectVersion.ERROR

        if obj.workflow.name == "process_record_arxiv":
            if tuple(pos) in [
                    (5,), (5, 3, 14), (5, 3, 14, 0), (5, 3, 15), (5, 3, 15, 1)]:
                pos = [len(workflows.get(obj.workflow.name).workflow) - 1]  # finished
            elif tuple(pos) in [(5, 3, 10), (5, 3, 11), (5, 3, 12)]:
                pos = [14, 0]  # halted
            elif len(pos) > 1 and pos[0] == 6:
                # We need to update pos from 6 to start of pre_processing part
                pos = [7]
            else:
                pos = [0]  # Nothing here, we go to start
            obj.save(task_counter=pos)

    # Special submission handling
    for deposit in BibWorkflowObject.query.filter(
            BibWorkflowObject.module_name == "webdeposit"):
        reset_workflow_object_states(deposit)
        d = Deposition(deposit)
        sip = d.get_latest_sip()
        if sip:
            sip.metadata = bibfield.do(sip.metadata)
            sip.package = legacy_export_as_marc(hep2marc.do(sip.metadata))
            d.update()
        deposit.save()

    # Special workflow handling
    workflows_to_fix = ["process_record_arxiv"]
    workflow_objects = []
    for workflow_name in workflows_to_fix:
        workflow_objects += BibWorkflowObject.query.join(
            BibWorkflowObject.workflow).filter(Workflow.name == workflow_name).all()

    for obj in workflow_objects:
        metadata = obj.get_data()
        reset_workflow_object_states(obj)
        rename_object_action(obj)
        if isinstance(metadata, six.string_types):
            # Ignore records that have string as data
            continue
        if 'drafts' in metadata:
            # New data model detected
            continue
        if hasattr(metadata, 'dumps'):
            metadata = metadata.dumps(clean=True)
        obj.data = bibfield.do(metadata)
        payload = Payload.create(
            type=obj.workflow.name,
            workflow_object=obj
        )
        payload.save()