diff --git a/constitution.py b/constitution.py index 26ba130e885e0e69fb7874ca5c3f07f42100a150..71fc46b4c9a4d7a9ea7bf319860b0db1acc4a673 100644 --- a/constitution.py +++ b/constitution.py @@ -503,20 +503,22 @@ def _epochs_in_ledger() -> list[int]: def build_pairwise_prompt(side_a: dict, side_b: dict) -> str: - return f"""You are ranking contributions to an open source project. -Compare these two sides (each may be one or more commits). Decide which side contributed more. + return f"""You are ranking individual git commits to an open source project. +Compare these two commits. Decide which commit contributed more. Return ONLY a JSON object: {{"winner": "A" or "B", "ratio": "N:M", "explanation": "..."}} -Side A — commit messages: +Side A — contributor: {side_a.get('contributor', '?')} +Side A — commit message: {side_a['message']} -Side A — unified diffs (full patches): +Side A — unified diff (full patch): {side_a['diff']} -Side B — commit messages: +Side B — contributor: {side_b.get('contributor', '?')} +Side B — commit message: {side_b['message']} -Side B — unified diffs (full patches): +Side B — unified diff (full patch): {side_b['diff']}""" @@ -1518,16 +1520,28 @@ async def broadcast_js(js: str): await queue.put(js) -def _author_side_for_llm(author: str, author_commits: dict) -> dict: - cs = author_commits[author] +def _commit_side_for_llm(row: dict) -> dict: + oid = row["oid"] + short = oid.split(":", 1)[1][:8] if ":" in oid else oid[:8] return { - "message": "\n".join(f"[{c['sha']}] {c['message']}" for c in cs), - "diff": "\n\n".join(f"=== {c['sha']} ===\n{c['diff']}" for c in cs), - "commit_ids": [c["commit_id"] for c in cs], - "contributor": author, + "message": f"[{short}] {row['message']}", + "diff": row["patch"] or "", + "commit_id": commit_id_for_oid(oid), + "contributor": row["contributor"], + "oid": oid, } +def _rollup_contributor_scores( + ordered: list[dict], commit_scores: list[Decimal] +) -> dict[str, Decimal]: + totals: dict[str, Decimal] = {} + for row, score in zip(ordered, commit_scores): + contributor = row["contributor"] + totals[contributor] = totals.get(contributor, Decimal("0")) + score + return totals + + def _find_judgment(comparison_id: str, model_id: str) -> dict | None: for e in evidence_by_kind("llm.judgment"): p = e.payload @@ -1546,101 +1560,106 @@ def _find_ranking_models(ranking_run_id: str) -> list[str] | None: async def rank_commits(commits: list[dict], *, epoch: int = -1): + """Pairwise-rank every eligible commit; roll scores up to contributors.""" if not commits: return {}, [], {"ranking_run_id": "", "ranking_event_id": ""} - commit_ids = sorted(commit_id_for_oid(row["oid"]) for row in commits) + ordered = sorted(commits, key=lambda r: r["oid"]) + commit_ids = [commit_id_for_oid(row["oid"]) for row in ordered] ranking_run_id = _content_id("rank", { "epoch": epoch, - "commit_ids": commit_ids, + "commit_ids": sorted(commit_ids), }) - contributors = sorted(set(c["contributor"] for c in commits)) + contributors = sorted({c["contributor"] for c in ordered}) - if len(contributors) == 1: + # Nothing to compare: a single commit (not a single contributor). + if len(ordered) == 1: await append_evidence(epoch, "ranking.started", { "ranking_run_id": ranking_run_id, "commit_ids": commit_ids, "contributors": contributors, "models": [], - "summary": f"ranking epoch {epoch}: single contributor", + "summary": f"ranking epoch {epoch}: single commit", }) - ranking = {contributors[0]: Decimal("1")} + commit_ranking = {commit_ids[0]: "1"} + contributor_ranking = {ordered[0]["contributor"]: Decimal("1")} completed = await append_evidence(epoch, "ranking.completed", { "ranking_run_id": ranking_run_id, "models": [], - "ranking": {contributors[0]: "1"}, + "commit_ranking": commit_ranking, + "contributor_ranking": {ordered[0]["contributor"]: "1"}, + "ranking": {ordered[0]["contributor"]: "1"}, "judgment_ids": [], - "summary": f"Only {contributors[0]} is eligible; rank is 1.0", + "summary": f"Only one eligible commit; {ordered[0]['contributor']} rank 1.0", }) await broadcast_audit( "ranking", - f"Only {contributors[0]} is eligible; rank is 1.0", + f"Only one eligible commit; {ordered[0]['contributor']} rank 1.0", progress=90, phase="finalizing", evidence_event_id=completed.event_id, evidence_url=_evidence_url("event", completed.event_id), links={"epoch": _evidence_url("epoch", str(epoch))}, ) - return ranking, [], { + return contributor_ranking, [], { "ranking_run_id": ranking_run_id, "ranking_event_id": completed.event_id, } if not (OPENROUTER_API_KEY or "").strip(): raise RuntimeError( - "OPENROUTER_API_KEY is required when multiple contributors need ranking" + "OPENROUTER_API_KEY is required when multiple commits need ranking" ) models = _find_ranking_models(ranking_run_id) if models is None: models = await fetch_top_models(n=3) if not models: - raise RuntimeError("no council models available for contributor ranking") + raise RuntimeError("no council models available for commit ranking") await append_evidence(epoch, "ranking.started", { "ranking_run_id": ranking_run_id, "commit_ids": commit_ids, "contributors": contributors, "models": models, - "summary": f"Council selected: {', '.join(models)}", + "summary": ( + f"Council selected: {', '.join(models)} — " + f"{len(ordered)} commits" + ), }) await broadcast_audit( "council", - f"Council selected: {', '.join(models)}", + f"Council selected: {', '.join(models)} — ranking {len(ordered)} commits", progress=35, phase="ranking", ) await broadcast_js(exec_event(Three[Selector("#emission-log")][PREPEND][ - ["div.log-council", f"Council: {', '.join(models)} — {len(commits)} commits"] + ["div.log-council", + f"Council: {', '.join(models)} — {len(ordered)} commits"] ])) - authors = contributors - author_commits = {a: [] for a in authors} - for row in sorted(commits, key=lambda r: r["oid"]): - author_commits[row["contributor"]].append({ - "message": row["message"], - "sha": row["oid"].split(":", 1)[1][:8], - "diff": row["patch"], - "commit_id": commit_id_for_oid(row["oid"]), - }) - + sides = [_commit_side_for_llm(row) for row in ordered] judgment_ids: list[str] = [] async def compare_fn(i, j): - a1, a2 = authors[i], authors[j] - side_a = _author_side_for_llm(a1, author_commits) - side_b = _author_side_for_llm(a2, author_commits) + side_a, side_b = sides[i], sides[j] + label_a = f"{side_a['commit_id'][:16]} ({side_a['contributor']})" + label_b = f"{side_b['commit_id'][:16]} ({side_b['contributor']})" prompt = build_pairwise_prompt(side_a, side_b) comparison_material = { "ranking_run_id": ranking_run_id, "side_a": { - "contributor": a1, - "commit_ids": side_a["commit_ids"], + "contributor": side_a["contributor"], + "commit_id": side_a["commit_id"], + "commit_ids": [side_a["commit_id"]], + "oid": side_a["oid"], "message": _bytes_blob(side_a["message"]), "diff": _bytes_blob(side_a["diff"]), }, "side_b": { - "contributor": a2, - "commit_ids": side_b["commit_ids"], + "contributor": side_b["contributor"], + "commit_id": side_b["commit_id"], + "commit_ids": [side_b["commit_id"]], + "oid": side_b["oid"], "message": _bytes_blob(side_b["message"]), "diff": _bytes_blob(side_b["diff"]), }, @@ -1650,22 +1669,24 @@ async def rank_commits(commits: list[dict], *, epoch: int = -1): comparison_material = { **comparison_material, "comparison_id": comparison_id, - "summary": f"Comparing {a1} with {a2}", + "summary": f"Comparing {label_a} with {label_b}", } cmp_ev = await append_evidence(epoch, "comparison.input", comparison_material) await broadcast_audit( "comparison", - f"Comparing {a1} with {a2}", + f"Comparing commits {label_a} vs {label_b}", phase="ranking", evidence_event_id=cmp_ev.event_id, evidence_url=_evidence_url("comparison", comparison_id), links={ "comparison": _evidence_url("comparison", comparison_id), + "commit_a": _evidence_url("commit", side_a["commit_id"]), + "commit_b": _evidence_url("commit", side_b["commit_id"]), "epoch": _evidence_url("epoch", str(epoch)), }, ) await broadcast_js(exec_event(Three[Selector("#emission-status")][MORPH][ - ["div#emission-status", f"Comparing {a1} vs {a2}…"] + ["div#emission-status", f"Comparing {label_a} vs {label_b}…"] ])) results = [] for model in models: @@ -1697,9 +1718,15 @@ async def rank_commits(commits: list[dict], *, epoch: int = -1): jud_id = (existing or _find_judgment(comparison_id, model) or {}).get( "judgment_id" ) + win_label = ( + f"{sides[w]['commit_id'][:16]} ({sides[w]['contributor']})" + ) + lose_label = ( + f"{sides[l]['commit_id'][:16]} ({sides[l]['contributor']})" + ) await broadcast_audit( "vote", - f"{model}: {authors[w]} over {authors[l]} ({result['ratio']})", + f"{model}: {win_label} over {lose_label} ({result['ratio']})", phase="ranking", evidence_url=( _evidence_url("judgment", jud_id) if jud_id else None @@ -1714,8 +1741,8 @@ async def rank_commits(commits: list[dict], *, epoch: int = -1): await broadcast_js(exec_event(Three[Selector("#emission-log")][PREPEND][ ["div.log-vote", ["span.model", model], " — ", - ["span.winner", authors[w]], f" beat ", - ["span.loser", authors[l]], f" ({result['ratio']}) ", + ["span.winner", win_label], f" beat ", + ["span.loser", lose_label], f" ({result['ratio']}) ", ["span.explanation", result["explanation"]], ] ])) @@ -1745,24 +1772,46 @@ async def rank_commits(commits: list[dict], *, epoch: int = -1): ["div#emission-status", label] ])) - pairs = await pairwise_rank(len(authors), compare_fn, progress_fn) + pairs = await pairwise_rank(len(ordered), compare_fn, progress_fn) if not pairs: - ranking = {authors[0]: Decimal("1")} if authors else {} + commit_score_list = [Decimal("1")] else: scores = rank_centrality(pairs) - ranking = {authors[i]: Decimal(str(scores[i])) for i in range(len(authors))} - ranking_rows = sorted(ranking.items(), key=lambda x: x[1], reverse=True) + commit_score_list = [Decimal(str(scores[i])) for i in range(len(ordered))] + + commit_ranking = { + commit_ids[i]: str(commit_score_list[i]) for i in range(len(ordered)) + } + contributor_totals = _rollup_contributor_scores(ordered, commit_score_list) + contrib_rows = sorted( + contributor_totals.items(), key=lambda x: x[1], reverse=True + ) + commit_rows = sorted( + ((commit_ids[i], commit_score_list[i], ordered[i]["contributor"]) + for i in range(len(ordered))), + key=lambda x: x[1], + reverse=True, + ) completed = await append_evidence(epoch, "ranking.completed", { "ranking_run_id": ranking_run_id, "models": models, - "ranking": {a: str(s) for a, s in ranking_rows}, + "commit_ranking": commit_ranking, + "contributor_ranking": {a: str(s) for a, s in contrib_rows}, + "ranking": {a: str(s) for a, s in contrib_rows}, "judgment_ids": judgment_ids, - "summary": "Ranking: " + ", ".join(f"{a} {s:.4f}" for a, s in ranking_rows), + "summary": ( + "Commit ranking: " + + ", ".join( + f"{cid[:16]}={float(s):.4f}" for cid, s, _ in commit_rows[:12] + ) + + ("…" if len(commit_rows) > 12 else "") + ), }) await broadcast_audit( "ranking", - "Ranking: " + ", ".join(f"{a} {s:.4f}" for a, s in ranking_rows), + "Contributor rollup: " + + ", ".join(f"{a} {s:.4f}" for a, s in contrib_rows), progress=90, phase="finalizing", evidence_event_id=completed.event_id, @@ -1772,10 +1821,10 @@ async def rank_commits(commits: list[dict], *, epoch: int = -1): await broadcast_js(exec_event(Three[Selector("#emission-log")][PREPEND][ ["div.log-ranking", ["b", "Ranking: "], - *[["span.rank-entry", f"{a} {float(s):.3f} "] for a, s in ranking_rows], + *[["span.rank-entry", f"{a} {float(s):.3f} "] for a, s in contrib_rows], ] ])) - return ranking, models, { + return contributor_totals, models, { "ranking_run_id": ranking_run_id, "ranking_event_id": completed.event_id, } @@ -2352,6 +2401,23 @@ async def epoch_detail(epoch: int): ranking_nodes: list = [] if ranking_completed: + commit_ranking = ranking_completed.payload.get("commit_ranking") or {} + contrib_ranking = ( + ranking_completed.payload.get("contributor_ranking") + or ranking_completed.payload.get("ranking") + or {} + ) + commit_rank_links = [ + ( + f"{cid[:20]} = {score}", + _evidence_path("commit", cid), + ) + for cid, score in sorted( + commit_ranking.items(), + key=lambda kv: Decimal(str(kv[1])), + reverse=True, + ) + ] ranking_nodes = [ ["p", ranking_completed.payload.get("summary") or "ranking completed"], _dl_rows([ @@ -2362,10 +2428,10 @@ async def epoch_detail(epoch: int): ranking_completed.event_id, )), ]), - ["pre.blob", json.dumps( - ranking_completed.payload.get("ranking") or {}, - indent=2, sort_keys=True, - )], + ["h3", "commit ranking"], + _link_list(commit_rank_links) if commit_rank_links else ["p.note", "(none)"], + ["h3", "contributor rollup"], + ["pre.blob", json.dumps(contrib_ranking, indent=2, sort_keys=True)], ] elif ranking_started: ranking_nodes = [["p.note", f"Ranking started: {ranking_started.event_id}"]] @@ -2386,14 +2452,9 @@ async def epoch_detail(epoch: int): else: emission_node = ["p.note", "No emission for this epoch."] - contributors = { - e.payload.get("contributor") - for e in commit_evs - if e.payload.get("contributor") - } no_comparisons_note = "No comparisons." - if len(contributors) <= 1: - no_comparisons_note += " Single-contributor — no LLM judgments." + if len(commit_evs) <= 1: + no_comparisons_note += " Fewer than two eligible commits — nothing to pairwise-rank." body = [ _evidence_nav(), @@ -2511,9 +2572,14 @@ async def comparison_detail(comparison_id: str): j.payload.get("summary") or jid, _evidence_path("judgment", jid), )) - commit_links = [] - for cid in (side_a.get("commit_ids") or []) + (side_b.get("commit_ids") or []): - commit_links.append((cid, _evidence_path("commit", cid))) + commit_ids = [] + for side in (side_a, side_b): + cid = side.get("commit_id") + if cid: + commit_ids.append(cid) + else: + commit_ids.extend(side.get("commit_ids") or []) + commit_links = [(cid, _evidence_path("commit", cid)) for cid in commit_ids] return _evidence_page(f"comparison {comparison_id[:24]}", [ _evidence_nav(_a(_evidence_path("epoch", str(ev.epoch)), f"epoch {ev.epoch}")), ["div.eyebrow", "comparison"], @@ -2522,8 +2588,8 @@ async def comparison_detail(comparison_id: str): ("summary", p.get("summary")), ("ranking_run_id", p.get("ranking_run_id")), ("evidence_event", _a(_evidence_path("event", ev.event_id), ev.event_id)), - ("side_a", side_a.get("contributor")), - ("side_b", side_b.get("contributor")), + ("side_a", f"{side_a.get('commit_id', '?')} ({side_a.get('contributor', '?')})"), + ("side_b", f"{side_b.get('commit_id', '?')} ({side_b.get('contributor', '?')})"), ]), ["p", _a(f"/comparisons/{comparison_id}/prompt", "download prompt")], ["h2", "commits"], @@ -2534,12 +2600,12 @@ async def comparison_detail(comparison_id: str): _link_list(judgment_links), ["h2", "prompt"], _pre_blob(_blob_text(p.get("prompt"))), - ["h2", f"side A — {side_a.get('contributor', '?')}"], + ["h2", f"side A — {side_a.get('commit_id', side_a.get('contributor', '?'))}"], ["h3", "message"], _pre_blob(_blob_text(side_a.get("message"))), ["h3", "diff"], _pre_blob(_blob_text(side_a.get("diff"))), - ["h2", f"side B — {side_b.get('contributor', '?')}"], + ["h2", f"side B — {side_b.get('commit_id', side_b.get('contributor', '?'))}"], ["h3", "message"], _pre_blob(_blob_text(side_b.get("message"))), ["h3", "diff"], @@ -3156,9 +3222,9 @@ async def watch(): ], ["p.note", ( - "Pairwise council voting is ready." + "Pairwise council ranks every eligible commit." if key_ok else - "Single-contributor epochs can finalize, but contested rankings require OPENROUTER_API_KEY." + "Epochs with two or more eligible commits require OPENROUTER_API_KEY." ) ], ], diff --git a/tests/integration.clj b/tests/integration.clj index 11140e95a8cd86768a661a30e6b29a3ddfc95fcc..39b2476cb5ddf80733769f61343c81ba7287a974 100644 --- a/tests/integration.clj +++ b/tests/integration.clj @@ -507,7 +507,7 @@ (bind or-state @(:state or-mock)) (assert! (pos? (:model-requests or-state)) "OpenRouter /models was called") - (assert! (>= (:compare-requests or-state) 3) "at least 3 pairwise LLM calls (2 authors × 3 models)") + (assert! (>= (:compare-requests or-state) 3) "at least 3 pairwise LLM calls (2 commits × 3 models)") (bind ledger2 (get-json base-url "/api/ledger")) (assert! (>= (count ledger2) 3) @@ -543,6 +543,8 @@ "epoch page links comparisons") (assert! (str/includes? epoch-html "/judgments/") "epoch page links judgments") + (assert! (str/includes? epoch-html "commit ranking") + "epoch page shows per-commit ranking") (bind commit-href (second (re-find #"/commits/(c_[a-f0-9]+)" epoch-html))) (assert! (some? commit-href) "found a commit id on epoch page") diff --git a/tests/test_git_discovery.py b/tests/test_git_discovery.py index 00d26bfa85737b178c8822804e278211974e0efe..0f002a58bd9a122d41c5a85e32c16ed2b4404d2f 100644 --- a/tests/test_git_discovery.py +++ b/tests/test_git_discovery.py @@ -434,7 +434,7 @@ def test_emission_distribution_sums_exactly_to_total( assert entry.discovery_snapshot_id == "ranked-snapshot" -def test_single_contributor_ranking_is_total_and_uses_no_pairwise_votes( +def test_single_commit_ranking_skips_pairwise( discovery_config, monkeypatch, ): monkeypatch.setattr(c, "store", c.JsonlStore(discovery_config / "ledger.jsonl")) @@ -454,6 +454,65 @@ def test_single_contributor_ranking_is_total_and_uses_no_pairwise_votes( assert info["ranking_event_id"] +def test_same_contributor_multiple_commits_runs_pairwise( + discovery_config, monkeypatch, +): + monkeypatch.setattr(c, "store", c.JsonlStore(discovery_config / "ledger.jsonl")) + calls = {"n": 0} + + async def models(n=3): + return ["m1", "m2", "m3"] + + async def compare(model_id, side_a, side_b, **kwargs): + calls["n"] += 1 + assert "commit_id" in side_a and "commit_id" in side_b + if kwargs.get("persist"): + attempt_id = c.attempt_id_for(kwargs["comparison_id"], model_id, 1) + await c.append_evidence(kwargs["epoch"], "llm.judgment", { + "judgment_id": c.judgment_id_for({ + "attempt_id": attempt_id, + "comparison_id": kwargs["comparison_id"], + "model_id": model_id, + "winner": "A", + "ratio": "2:1", + "explanation": "ok", + }), + "attempt_id": attempt_id, + "comparison_id": kwargs["comparison_id"], + "model_id": model_id, + "winner": "A", + "ratio": "2:1", + "explanation": "ok", + "summary": "ok", + }) + return {"winner": "A", "ratio": "2:1", "explanation": "ok"} + + monkeypatch.setattr(c, "fetch_top_models", models) + monkeypatch.setattr(c, "llm_pairwise_compare", compare) + monkeypatch.setattr(c, "OPENROUTER_API_KEY", "test-key") + commits = [ + { + "contributor": "alice", + "oid": "sha1:" + char * 40, + "message": f"msg-{char}", + "patch": f"patch-{char}", + } + for char in ("a", "b", "c") + ] + ranking, used, info = asyncio.run(c.rank_commits(commits, epoch=0)) + assert set(ranking) == {"alice"} + assert ranking["alice"] > 0 + assert used == ["m1", "m2", "m3"] + assert calls["n"] >= 3 + completed = next( + e for e in c.store.read() + if isinstance(e, c.Evidence) and e.kind == "ranking.completed" + ) + assert len(completed.payload["commit_ranking"]) == 3 + assert "alice" in completed.payload["contributor_ranking"] + assert info["ranking_event_id"] + + def test_any_council_failure_aborts_ranking(discovery_config, monkeypatch): monkeypatch.setattr(c, "store", c.JsonlStore(discovery_config / "ledger.jsonl")) @@ -479,16 +538,16 @@ def test_any_council_failure_aborts_ranking(discovery_config, monkeypatch): asyncio.run(c.rank_commits(commits, epoch=0)) -def test_contested_ranking_requires_openrouter_key(monkeypatch): +def test_multi_commit_ranking_requires_openrouter_key(monkeypatch): monkeypatch.setattr(c, "OPENROUTER_API_KEY", "") commits = [ { - "contributor": contributor, + "contributor": "alice", "oid": "sha1:" + char * 40, - "message": contributor, + "message": char, "patch": "patch", } - for contributor, char in [("alice", "a"), ("bob", "b")] + for char in ("a", "b") ] with pytest.raises(RuntimeError, match="OPENROUTER_API_KEY"): asyncio.run(c.rank_commits(commits))