feat: read git history with go-git, not the git binary (#15)

bin/tools/test_gitgraph.py

@@ -9,12 +9,6 @@ sys.path.insert(0, str(Path(__file__).resolve().parent))
 import kblib  # noqa: E402
 import gitimport  # noqa: E402
 
-SAMPLE = (
-    "\x1e" + "a1b2c3d" + "\x1f" + "Ada Lovelace" + "\x1f" + "ada@example.com"
-    + "\x1f" + "2026-08-10T12:00:00+01:00" + "\x1f" + "feat: first commit"
-    + "\n\nREADME.md\nsrc/main.c\n"
-)
-
 COMMIT_PERSON_SCHEMA = (
     "CREATE NODE TABLE IF NOT EXISTS Commit (id STRING, repo STRING, subject STRING, "
     "author STRING, email STRING, date STRING, PRIMARY KEY(id))"
@@ -26,6 +20,17 @@ HAS_VERSION_SCHEMA = "CREATE REL TABLE IF NOT EXISTS HAS_VERSION (FROM File TO C
 AUTHORED_SCHEMA = "CREATE REL TABLE IF NOT EXISTS AUTHORED (FROM Commit TO Person)"
 
 
+def sample_commit() -> gitimport.Commit:
+    return gitimport.Commit(
+        sha="a1b2c3d",
+        author="Ada Lovelace",
+        email="ada@example.com",
+        date="2026-08-10T12:00:00+01:00",
+        subject="feat: first commit",
+        files=["README.md", "src/main.c"],
+    )
+
+
 class GitGraphTest(unittest.TestCase):
     def setUp(self):
         self.dir = tempfile.mkdtemp()
@@ -42,14 +47,12 @@ class GitGraphTest(unittest.TestCase):
         self.db.close()
 
     def test_index_commits_creates_nodes_and_edges(self):
-        cs = gitimport.parse_log(SAMPLE)
-        gitimport.index_commits(self.conn, cs, "sample-repo")
+        gitimport.index_commits(self.conn, [sample_commit()], "sample-repo")
         rp = self.conn.execute("MATCH (p:Person) RETURN p.name, p.email").get_all()
         self.assertEqual([tuple(r) for r in rp], [("Ada Lovelace", "ada@example.com")])
         rc = self.conn.execute("MATCH (c:Commit) RETURN c.id, c.repo").get_all()
         self.assertEqual(len(rc), 1)
         self.assertEqual(rc[0][1], "sample-repo")
-        # File -[:HAS_VERSION]-> Commit -[:AUTHORED]-> Person
         rf = self.conn.execute(
             "MATCH (f:File)-[:HAS_VERSION]->(c:Commit)-[:AUTHORED]->(p:Person) "
             "RETURN f.path, c.id, p.email").get_all()
@@ -58,7 +61,7 @@ class GitGraphTest(unittest.TestCase):
         self.assertTrue(all(r[2] == "ada@example.com" for r in rf))
 
     def test_index_commits_idempotent(self):
-        cs = gitimport.parse_log(SAMPLE)
+        cs = [sample_commit()]
         gitimport.index_commits(self.conn, cs, "sample-repo")
         gitimport.index_commits(self.conn, cs, "sample-repo")
         n = self.conn.execute("MATCH (c:Commit) RETURN count(*)").get_all()[0][0]
@@ -68,4 +71,4 @@ class GitGraphTest(unittest.TestCase):
 
 
 if __name__ == "__main__":
-    unittest.main()
+    unittest.main()