PR#369: Add celery task ID for regenerated composes - odcs

odcs

#369 Add celery task ID for regenerated composes

Merged 4 years ago by lsedlar. Opened 4 years ago by onosek.

onosek/odcs add_task_id into master

Add celery task ID for regenerated composes

Ondrej Nosek • 4 years ago

650d67e

server/odcs/server/views.py

file modified

+6 -2

		`@@ -234,9 +234,13 @@`

		`if CELERY_AVAILABLE and conf.celery_broker_url:`
		`if compose.source_type == PungiSourceType.PULP:`
		`- generate_pulp_compose.delay(compose.id)`
		`+ result = generate_pulp_compose.delay(compose.id)`
		`else:`
		`- generate_pungi_compose.delay(compose.id)`
		`+ result = generate_pungi_compose.delay(compose.id)`
		`+`
		`+ compose.celery_task_id = result.id`
		`+ db.session.add(compose)`
		`+ db.session.commit()`

		`return jsonify(compose.json()), 200`
		`else:`

onosek commented 4 years ago

JIRA: RHELCMP-107

Signed-off-by: Ondrej Nosek onosek@redhat.com

rebased onto 81b8e1ad1237b73baa4ab705cdddbfc8b9a91ce2

4 years ago

rebased onto 1048f4493c84495ad27a485c9634cbf0c636fba5

4 years ago

onosek commented 4 years ago

I was curious about the doubled session.add() construction. First I thought that it could generate new (incremented) compose.id for every call. Then I found:
https://docs.sqlalchemy.org/en/13/orm/session_api.html
as per this page "Repeated calls to add() will be ignored."

I did't verified it yet, because I do not have fully working testing environment.

rebased onto 650d67e

4 years ago

onosek commented 4 years ago

OK, my doubts weren't confirmed during testing, so I reverted changes back.