Repository navigation
Conversation
boring-cyborg
Bot
added
area:API
c56e618 to
59eae68
Compare
888cf42 to
c79cd1f
Compare
| ("stop", State.SUCCESS, 1, True, False, [0]), | ||
| ("continue", State.SUCCESS, 1, True, True, [0]), | ||
| (None, State.FAILED, 1, True, False, [0]), | ||
| ("continue", State.SUCCESS, 3, False, False, [1, 2]), |
There was a problem hiding this comment.
With this row now [1, 2], the if mapped and passes == [0, 1]: block further down (L2584) no longer matches any row, so the check that the next pass's mapped task starts as a -1 placeholder and then expands silently stopped running. It needs passes == [1, 2] and parent_region_index == 2 there.
| ) == [0, 1] | ||
| ) == [1, 2] | ||
|
|
||
| def test_loop_gate_completion_uses_its_pinned_limit(self, client, session, running_loop_gate, dag_maker): |
There was a problem hiding this comment.
The ti.region_index == 1 filter in this test now picks up the first pass (gate and terminal) instead of the successor, so len(successors) == 2 and the version check pass whatever the server does with the pinned limit. Should be == 2.
| The loop iteration and mapped position are separate "coordinates". In this | ||
| example, each iteration has mapped instances 0 and 1, so mapped instance 1 in | ||
| iteration 0 is distinct from mapped instance 1 in iteration 1. | ||
| The loop iteration and mapped position are separate "coordinates", and they do |
There was a problem hiding this comment.
"Rerun part of an iteration" further down (L389 and L396) still counts from zero: "iterations 0 through 4" and "Iterations 0 and 1 remain untouched". With this change that becomes "1 through 4" and "Iteration 1 remains untouched".
| assert client.head(url, params=params).headers["Content-Range"] == "map_indexes 1" | ||
|
|
||
|
|
||
| def test_previous_iteration_of_the_first_pass_resolves_to_nothing(client, loop_xcoms, authenticate_as): |
There was a problem hiding this comment.
This also passes with the old < 0 guard: pass 1 then looks up pass 0, which never exists, so it 404s either way. To pin "stops before looking for one", assert the producer query never runs (capture_orm_selects, as elsewhere in this file), or trim the docstring to the 404.
| run_id: Mapped[str] = mapped_column(StringID(), nullable=False) | ||
| # The slot this instance fills in its region, so its meaning follows the region's node: a loop | ||
| # region numbers its iterations from one, a mapped expansion numbers its slots from zero, and -1 | ||
| # is the placeholder for a mapped task that has not been expanded yet. Code that reads the column |
There was a problem hiding this comment.
Nit: plain tasks outside any region also store -1 (they get (-1,) in dagrun.py), and that's the most common case. Can we mention it here too?
c79cd1f to
3258cc3
Compare
A loop iteration is a number people read: it shows up in the loop filter, on the Gantt tooltip, in the task instance list and in the `loop.index` a Dag author branches on. Starting it at zero meant each of those readers had to translate between the number Airflow stored and the number a person would say, and that translation was easy to apply in one place and forget in another. Numbering from one in storage removes the translation instead of hiding it behind a display-time adjustment, so the stored index, the API field and the UI agree without anyone adding to it. Mapped expansion indexes share the column and stay zero-based: a map index is a position, not a count, so code reading the column has to know whether the region executes a loop or an expansion. Task loops have not shipped in a release, so no stored run and no client depends on the old numbering, and the execution API version that introduces the loop context is still unreleased. Co-Authored-By: Claude <noreply@anthropic.com>
3258cc3 to
e9da72d
Compare
This branch has not been deployed
3 participants
Task Loop Iterations were starting at 0 which was confusing in the UI. Instead of remembering a UI-only "+1" change that we could easily forget to add. This PR changes it so that all region_indexes that are for loop iterations start at 1.
Was generative AI tooling used to co-author this PR?
Claude Opus 5
Generated-by: Claude Opus 5 following the guidelines
{pr_number}.significant.rst, in airflow-core/newsfragments. You can add this file in a follow-up commit after the PR is created so you know the PR number.