def setUp(self):
self.marcxml = pkg_resources.resource_string('inspire.testsuite',
os.path.join(
'fixtures',
'test_hep_record.xml')
)
record = create_record(self.marcxml)
self.marcxml_to_json = hep.do(record)
self.json_to_marc = hep2marc.do(self.marcxml_to_json)
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))
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()
