workflows: fix author submissions not erroring

also some small config fixes for easy testing with docker

* ref: cern-sis/issues-inspire#643

Makefile

@@ -9,6 +9,7 @@ run: services start-inspirehep sleep setup-inspirehep start-backoffice sleep set
 run-inspirehep: services start-inspirehep sleep setup-inspirehep
 run-backoffice: services start-backoffice sleep setup-backoffice
 
+start: services start-inspirehep start-backoffice
 
 start-inspirehep:
 	echo -e "\033[0;32m Starting HEP. \033[0m"

backend/inspirehep/config.py

@@ -31,7 +31,7 @@
 FEATURE_FLAG_ENABLE_ORCID_PUSH = False
 FEATURE_FLAG_ENABLE_REDIRECTION_OF_PIDS = False
 FEATURE_FLAG_ENABLE_ASSIGN_AUTHOR_PAPERS = False
-FEATURE_FLAG_ENABLE_SEND_TO_BACKOFFICE = False
+FEATURE_FLAG_ENABLE_SEND_TO_BACKOFFICE = True
 # Only push to ORCIDs that match this regex.
 # Examples:
 #   any ORCID -> ".*"
@@ -51,8 +51,8 @@
 INSPIRE_NEXT_URL = "http://localhost:5000"
 LEGACY_BASE_URL = "https://old.inspirehep.net"
 LEGACY_RECORD_URL_PATTERN = "http://inspirehep.net/record/{recid}"
-INSPIRE_BACKOFFICE_URL = "https://backoffice.dev.inspirebeta.net"
-AUTHENTICATION_TOKEN_BACKOFFICE = "CHANGE_ME"
+INSPIRE_BACKOFFICE_URL = "http://host.docker.internal:8001"
+AUTHENTICATION_TOKEN_BACKOFFICE = "2e04111a61e8f5ba6ecec52af21bbb9e81732085"
 MAX_API_RESULTS = 10000
 REST_MIMETYPE_QUERY_ARG_NAME = "format"
 

workflows/dags/author/author_create/author_create_init.py

@@ -21,13 +21,14 @@
     params={
         "workflow_id": Param(type="string", default=""),
         "data": Param(type="object", default={}),
+        "collection": Param(type="string", default=AUTHORS),
     },
     start_date=datetime.datetime(2024, 5, 5),
     schedule=None,
     catchup=False,
     # TODO: what if callback fails? Data in backoffice not up to date!
     on_failure_callback=set_workflow_status_to_error,
-    tags=["authors"],
+    tags=[AUTHORS],
 )
 def author_create_initialization_dag():
     """

workflows/plugins/hooks/backoffice/workflow_management_hook.py

@@ -50,7 +50,7 @@ def update_workflow(self, workflow_id: str, workflow_data: dict) -> Response:
         return self.run_with_advanced_retry(
             _retry_args=self.tenacity_retry_kwargs,
             method="PUT",
-            data=workflow_data,
+            json=workflow_data,
             endpoint=endpoint,
         )
 
@@ -61,6 +61,6 @@ def partial_update_workflow(
         return self.run_with_advanced_retry(
             _retry_args=self.tenacity_retry_kwargs,
             method="PATCH",
-            data=workflow_partial_update_data,
+            json=workflow_partial_update_data,
             endpoint=endpoint,
         )

workflows/plugins/include/utils/set_workflow_status.py

@@ -1,6 +1,6 @@
 import logging
 
-from hooks.backoffice.workflow_management_hook import AUTHORS, WorkflowManagementHook
+from hooks.backoffice.workflow_management_hook import WorkflowManagementHook
 from requests import Response
 
 logger = logging.getLogger(__name__)
@@ -36,10 +36,10 @@ def set_workflow_status_to_error(context: dict) -> None:
         workflow_id (str): The identifier for the workflow.
     """
     logger.info("Setting workflow status to error")
-    response = WorkflowManagementHook().set_workflow_status(
-        status_name="error",
-        workflow_id=context["params"]["workflow_id"],
-        collection=AUTHORS,
+    response = WorkflowManagementHook(
+        collection=context["params"]["collection"]
+    ).set_workflow_status(
+        status_name="error", workflow_id=context["params"]["workflow_id"]
     )
     try:
         response.raise_for_status()