Export every recorded input, not only the ones that vary

`export runs` dropped a `param.*` column whose value was constant across the
exported runs, so a downstream filter broke depending on which runs the
selection happened to hold. Every input the selection recorded is a column
now; `--params` still narrows it to a sweep's axis. The metrics default is
unchanged.

Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_019Hra4ndWMCLU5F3KjUuVAc

backend/fluksio/api/routes/runs.py

@@ -422,25 +422,6 @@ def _dig(record: dict[str, Any], path: str) -> Any:
     return value
 
 
-def _varying(runs: list[Run]) -> list[str]:
-    """The inputs that differ across these runs — the axis of a sweep.
-
-    What a reader comparing arms wants as columns, compared leaf by leaf: two
-    configurations differing in one field give that field as a column rather
-    than two blobs that are not the same. Under two runs nothing can differ,
-    and a table of one run with none of its inputs in it is not worth reading,
-    so all of them are kept.
-    """
-    keys = sorted({path for run in runs for path, _ in _leaves(run.params)})
-    if len(runs) < 2:
-        return keys
-    return [
-        key
-        for key in keys
-        if len({json.dumps(_dig(run.params, key), sort_keys=True) for run in runs}) > 1
-    ]
-
-
 def _scored(runs: list[Run]) -> list[str]:
     """A run's final numbers: every number its declared outputs carry, however
     deep it sits. A flag is not a number, and neither is a label."""
@@ -557,9 +538,9 @@ def export_runs(
     """One row per run: what it was given, what it scored, what code it ran.
 
     The arm-comparison table. Inputs are columns rather than one JSON blob —
-    by default the ones that vary across the selection, which is the sweep
-    axis; ``params`` names them instead. ``metrics`` narrows the final numbers
-    to a few of a run's declared outputs.
+    every input the selection recorded, so the schema is the same whichever
+    runs are asked for; ``params`` narrows it. ``metrics`` narrows the final
+    numbers to a few of a run's declared outputs.
 
     Both take dotted paths into a record a node returned:
     ``metrics=final_metrics.train_loss,test_metrics.known.perfect`` selects
@@ -567,7 +548,9 @@ def export_runs(
     depth.
     """
     runs = _selected(session, flow, status, group, ids, since, until)
-    inputs = [part for part in params.split(",") if part] or _varying(runs)
+    inputs = [part for part in params.split(",") if part] or sorted(
+        {path for run in runs for path, _ in _leaves(run.params)}
+    )
     scores = [part for part in metrics.split(",") if part] or _scored(runs)
     columns = [
         *RUN_COLUMNS,

backend/fluksio/sdk/cli.py

@@ -1394,7 +1394,7 @@ def add_parsers(subparsers: Any) -> None:
         metavar="A,B",
         help=(
             "the inputs to put in columns, dotted into a record "
-            "(default: the ones that vary)"
+            "(default: every input the runs recorded)"
         ),
     )
     sub.add_argument(

backend/fluksio/sdk/client.py

@@ -428,9 +428,9 @@ class Client:
     ) -> list[dict[str, Any]]:
         """One row per run: its inputs as columns, its final numbers, its code.
 
-        The arm-comparison table. The inputs kept are the ones that vary
-        across the selection unless ``params`` names them, which is the axis a
-        sweep is read along.
+        The arm-comparison table. Every input the selection recorded is a
+        column, so the schema does not depend on which runs were asked for;
+        ``params`` narrows it to the axis a sweep is read along.
         """
         query: dict[str, Any] = {}
         if params:

backend/tests/api/routes/test_runs.py

@@ -873,10 +873,10 @@ def test_an_export_strides_each_series_and_names_its_run(
     assert all(steps == [0, 2] for steps in curves.values())
 
 
-def test_an_exported_run_row_carries_the_inputs_that_vary(
+def test_an_exported_run_row_carries_every_recorded_input(
     client, superuser_token_headers, exported
 ):
-    """The sweep axis becomes columns; what every run shares stays out of them."""
+    """Every input is a column, so the schema does not move with the selection."""
 
     def export(**extra):
         answer = client.get(
@@ -890,12 +890,15 @@ def test_an_exported_run_row_carries_the_inputs_that_vary(
     rows = _lines(export(format="jsonl"))
     assert [row["id"] for row in rows] == ["exp-1", "exp-0"]
     assert {row["param.lr"] for row in rows} == {0.1, 0.01}
-    # `epochs` is the same on both runs, so it is not what they differ by, and
-    # neither is the model inside the config — but the depth beside it is.
-    assert "param.epochs" not in rows[0]
-    assert "param.config.model" not in rows[0]
+    # `epochs` is the same on both runs and stays a column anyway: which runs
+    # were asked for is not something a downstream filter should have to know.
+    assert rows[0]["param.epochs"] == 10
+    assert rows[0]["param.config.model"] == "mlp"
     assert {row["param.config.depth"] for row in rows} == {1, 2}
-    assert "param.epochs" in _lines(export(format="jsonl", params="epochs"))[0]
+
+    narrowed = _lines(export(format="jsonl", params="epochs"))[0]
+    assert "param.epochs" in narrowed
+    assert "param.lr" not in narrowed
 
     # A number inside a record is a column of its own, however deep; a string
     # is not one of the run's numbers wherever it sits.
@@ -912,7 +915,7 @@ def test_an_exported_run_row_carries_the_inputs_that_vary(
     header = export().text.splitlines()[0]
     assert header.startswith("id,flow,status,")
     assert header.endswith(
-        "param.config.depth,param.lr,"
+        "param.config.depth,param.config.model,param.epochs,param.lr,"
         "metric.acc,metric.final_metrics.train_loss,"
         "metric.test_metrics.known.perfect"
     )