Scan all eligible CodeQL benchmark pairs

codeql_compare.py

@@ -32,15 +32,17 @@ def rows(path):
 
 
 def selected_pairs(args):
-    cohort = json.loads(Path(args.cohort).read_text())
+    cohort = None if args.all_supported else json.loads(Path(args.cohort).read_text())
     selected = {}
     for domain in ("bug", "security"):
-        repos = set(cohort[domain])
+        repos = set(cohort[domain]) if cohort else None
         selected[domain] = [
             pair for pair in rows(Path(args.data) / domain / "benchmark.jsonl")
-            if pair["repository"] in repos
+            if (repos is None or pair["repository"] in repos)
             and pair["language"].lower() in LANGUAGES
-            and all(pair[role].get("revision") for role in ("before", "after"))
+            and all(isinstance(pair[role].get("revision"), str)
+                    and re.fullmatch(r"[0-9a-f]{40}", pair[role]["revision"])
+                    for role in ("before", "after"))
         ]
     return selected
 
@@ -390,6 +392,8 @@ def main():
     parser.add_argument("--data", required=True, help="directory with bug/security/benchmark.jsonl")
     parser.add_argument("--predictions", required=True, help="directory with bug/security/jev.jsonl")
     parser.add_argument("--cohort", default=str(Path(__file__).with_name("codeql-cohort.json")))
+    parser.add_argument("--all-supported", action="store_true",
+                        help="scan every pair in supported languages with two Git revisions")
     parser.add_argument("--codeql", help="unpacked CodeQL CLI directory")
     parser.add_argument("--rebench-parquet", help="SWE-rebench-V2 source Parquet with gold patches")
     parser.add_argument("--image", default="node:22-bookworm")

test_codeql_compare.py

@@ -3,11 +3,33 @@ import sys
 import tempfile
 import unittest
 from pathlib import Path
+from types import SimpleNamespace
 
-from codeql_compare import codeql_alert, method_span, metrics, run
+from codeql_compare import codeql_alert, method_span, metrics, run, selected_pairs
 
 
 class CodeQLComparisonTests(unittest.TestCase):
+    def test_all_supported_selects_every_eligible_repository(self):
+        revision = "a" * 40
+        eligible = {"pair_id": "eligible", "repository": "other/repo", "language": "python",
+                    "before": {"revision": revision}, "after": {"revision": revision}}
+        unsupported = {**eligible, "pair_id": "unsupported", "language": "php"}
+        missing_revision = {**eligible, "pair_id": "missing",
+                            "after": {"revision": None}}
+        with tempfile.TemporaryDirectory() as directory:
+            root = Path(directory)
+            for domain in ("bug", "security"):
+                path = root / domain / "benchmark.jsonl"
+                path.parent.mkdir()
+                path.write_text("".join(json.dumps(row) + "\n" for row in
+                                        (eligible, unsupported, missing_revision)))
+            cohort = root / "cohort.json"
+            cohort.write_text(json.dumps({"bug": ["selected/repo"],
+                                          "security": ["selected/repo"]}))
+            args = SimpleNamespace(data=str(root), cohort=str(cohort), all_supported=True)
+            self.assertEqual([row["pair_id"] for row in selected_pairs(args)["bug"]],
+                             ["eligible"])
+
     def test_alert_requires_primary_location_inside_method_and_relevant_tag(self):
         side = {
             "path": "src/example.py",