fix: select PR CI changes from the merge base

.github/workflows/ci.yml

@@ -141,7 +141,8 @@ jobs:
           HEAD_SHA: ${{ github.event.pull_request.head.sha }}
         run: |
           set -euo pipefail
-          mapfile -t changed_files < <(git diff --name-only "$BASE_SHA".."$HEAD_SHA")
+          MERGE_BASE=$(git merge-base "$BASE_SHA" "$HEAD_SHA")
+          mapfile -t changed_files < <(git diff --name-only "$MERGE_BASE".."$HEAD_SHA")
           if [ "${#changed_files[@]}" -eq 0 ]; then
             echo "No changed files; skipping submission diff validation."
             exit 0
@@ -152,7 +153,7 @@ jobs:
               exit 0
             fi
           done
-          lake exe lean-eval validate-submission --base "$BASE_SHA" --head "$HEAD_SHA"
+          lake exe lean-eval validate-submission --base "$MERGE_BASE" --head "$HEAD_SHA"
 
       - name: Run Lean unit tests
         run: |

scripts/select_ci_problems.py

@@ -137,9 +137,21 @@ def parse_git_changes(output: bytes) -> tuple[Change, ...]:
     return tuple(changes)
 
 
-def git_changes(root: pathlib.Path, base: str, head: str) -> tuple[Change, ...]:
+def git_changes(root: pathlib.Path, base: str, head: str, event: str) -> tuple[Change, ...]:
     if base == ZERO_SHA:
         return (Change("A", ("<initial-push>",)),)
+    if event == "pull_request":
+        # The base branch may have advanced after the contributor branched.
+        # Comparing its tip directly to the PR head treats base-only changes
+        # as reversions by the PR, triggering unrelated catalog checks.
+        ancestor = subprocess.run(
+            ["git", "merge-base", base, head],
+            cwd=root,
+            check=True,
+            stdout=subprocess.PIPE,
+            text=True,
+        )
+        base = ancestor.stdout.strip()
     result = subprocess.run(
         ["git", "diff", "--name-status", "-z", "--find-renames", f"{base}..{head}"],
         cwd=root,
@@ -359,7 +371,7 @@ def main(argv: Sequence[str] | None = None) -> int:
         parser.error("--shards must be positive")
     root = args.root.resolve()
     problems = load_problems(root)
-    changes = git_changes(root, args.base, args.head)
+    changes = git_changes(root, args.base, args.head, args.event)
     graph = load_import_graph(args.import_graph)
     selection = select(root, args.event, changes, problems, graph)
     matrix = make_matrix(selection, problems, args.shards)

tests/python/test_select_ci_problems.py

@@ -2,7 +2,9 @@
 
 import importlib.util
 import json
+import os
 import pathlib
+import subprocess
 import sys
 import tempfile
 import unittest
@@ -170,6 +172,41 @@ def test_nul_git_changes_reject_truncated_records(self):
         with self.assertRaisesRegex(ValueError, "unexpected git diff record"):
             SELECTOR.parse_git_changes(b"R100\0LeanEval/Old.lean\0")
 
+    def test_pr_changes_exclude_updates_only_on_base_branch(self):
+        with tempfile.TemporaryDirectory() as directory:
+            root = pathlib.Path(directory)
+            env = {**os.environ, "GIT_CONFIG_GLOBAL": os.devnull, "GIT_CONFIG_NOSYSTEM": "1"}
+
+            def git(*args: str) -> str:
+                return subprocess.check_output(
+                    ["git", *args], cwd=root, text=True, env=env
+                ).strip()
+
+            git("init", "-q")
+            git("config", "user.name", "CI test")
+            git("config", "user.email", "ci@example.test")
+            (root / "README.md").write_text("base\n", encoding="utf-8")
+            git("add", "README.md")
+            git("commit", "-qm", "base")
+            git("branch", "-M", "main")
+            git("switch", "-qc", "feature")
+            (root / "LeanEval").mkdir()
+            (root / "LeanEval" / "New.lean").write_text("-- new\n", encoding="utf-8")
+            git("add", "LeanEval/New.lean")
+            git("commit", "-qm", "new problem")
+            head = git("rev-parse", "HEAD")
+            git("switch", "-q", "main")
+            (root / "generated").mkdir()
+            (root / "generated" / "index.json").write_text("{}\n", encoding="utf-8")
+            git("add", "generated/index.json")
+            git("commit", "-qm", "regenerate main")
+            base = git("rev-parse", "HEAD")
+
+            pr_changes = SELECTOR.git_changes(root, base, head, "pull_request")
+            self.assertEqual(pr_changes, (Change("A", ("LeanEval/New.lean",)),))
+            push_changes = SELECTOR.git_changes(root, base, head, "push")
+            self.assertIn(Change("D", ("generated/index.json",)), push_changes)
+
     def test_loads_graph_emitted_by_lean(self):
         payload = [
             {