bench(learn): harvest signals for the decision engine (decision-engine task 5)

`ousast plane harvest` (plane/harvest.py) writes two Runs over a label snapshot:
verify a/b + agree on the verify-family units (both sides of each labelled pair as
an opaque files-only Workspace, candidates outside the hunt's root, no declared
sites; v1 and dev-php pins with repo-facts first) and model roles on the candidate
files (pair excerpts grouped, one Task per population pin). Budgets sum under the
ceiling by construction. New roles task template; the generator re-pins it.

Spent by attribution: verify $2.54 (1961 calls, 150 units, 182 hunts per pass),
roles $1.40 (1034 calls, 799 files, 0 unclassified chunks); DeepSeek balance
19.63 -> 18.23. Engine on the pair corpus (benchmarks/learn/engine_pairs.py,
frozen e6ed785, one 3g container) is PARTIAL: 384 of 686 sides at commit, still
running; unfinished sides stay missing. benchmarks/learn/build_features.py built
1712 static+plane records into the memory store; counts-only record under
benchmarks/measurements/2026-10-01-decision-engine-harvest/.

source/roles parts are `none`, not `failed`, for a language without a declaration
pattern (C). ops/ax/README: Substrate golden-actor disk leak and the inline
Workspace limit (~20 KB).

Co-Authored-By: Claude Opus 5.5 <[email protected]>
Claude-Session: https://claude.ai/code/session_01EmM1j46qgKMh3zqD1QMwEo
This commit is contained in:
Marius Ciepluch
2026-10-01 14:04:28 +02:00
co-authored by Claude Opus 5.5
parent e6ed7854bb
commit 34da916030
21 changed files with 1739 additions and 137 deletions
+6 -3
View File
@@ -254,7 +254,9 @@ def function_span(lines: Sequence[str], language: str, function: str) -> int | N
def source_part(lines: Sequence[str] | None, language: str, function: str) -> Part:
if lines is None:
"""The function's span; ``none`` without the file or a declaration pattern for its language (C, say: no coverage,
not a failure), ``failed`` when a covered file does not declare the function."""
if lines is None or (function != GLOBAL and language not in DECLARATION):
return NONE
span = function_span(lines, language, function)
return Part("ran", {"facts.function_lines": span}) if span is not None else Part("failed")
@@ -294,8 +296,9 @@ def agree_part(final: str | None) -> Part:
def roles_part(function: Function | None, roles: RoleSet | None, language: str) -> Part:
"""The deterministic roles of the candidate's function (language-level entries plus wrapper inference; priors
off): ``none`` when inference did not run, ``failed`` when the function's body could not be found."""
if roles is None or "inferred_wrapper" not in roles.sources:
off): ``none`` when inference did not run or cannot read the language's declarations, ``failed`` when the
function's body could not be found."""
if roles is None or "inferred_wrapper" not in roles.sources or language not in DECLARATION:
return NONE
if function is None:
return Part("failed", version=VERSION)