The asyncio tools use structures from the _remote_debugging module.
In tests we currently reimplement these structures:
|
FrameInfo = namedtuple('FrameInfo', ['funcname', 'filename', 'lineno']) |
|
CoroInfo = namedtuple('CoroInfo', ['call_stack', 'task_name']) |
|
TaskInfo = namedtuple('TaskInfo', ['task_id', 'task_name', 'coroutine_stack', 'awaited_by']) |
|
AwaitedInfo = namedtuple('AwaitedInfo', ['thread_id', 'awaited_by']) |
|
|
In GH-142394, the upstream structures from _remote_debugging were changed, but the reimplementations in the tests were not updated to match. This caused a regression from GH-143952 to sneak in.
Let's use structures from _remote_debugging directly, as it is currently done in test_external_inspection, to prevent any further unnoticed discrepancies.
cc @pablogsal