⚡ Bolt: json 파일 읽기 성능 최적화

.jules/bolt.md

@@ -63,3 +63,7 @@
 ## 2024-07-30 - Avoid chained string replace when checking character sets
 **Learning:** Using chained `.replace(a, "").replace(b, "")` to check if a string consists entirely of specific characters requires intermediate string allocations for every call. In benchmarks, using `.strip("ab")` is ~30% faster and avoids multiple allocations in the hot path.
 **Action:** When checking if a string is solely composed of specific characters, use `.strip(chars)` instead of chained `.replace()` calls to improve performance.
+
+## 2024-07-31 - Bypass string allocation when reading JSON files
+**Learning:** Passing `path.read_bytes()` directly to `json.loads()` is faster than `path.read_text()` because it bypasses the intermediate string allocation and decoding overhead.
+**Action:** When reading JSON files, use `path.read_bytes()` for performance optimization.

src/newsdom_api/mineru_runner.py

@@ -262,7 +262,8 @@ def _execute_mineru(cmd: list[str]) -> subprocess.CompletedProcess[str]:
 def _read_mineru_json(path: Path, *, artifact: str) -> Any:
     """Read a MinerU JSON artifact with safe, differentiated failure messages."""
     try:
-        return json.loads(path.read_text(encoding="utf-8"))
+        # ⚡ Bolt: Read bytes directly to bypass intermediate string allocation and decoding overhead
+        return json.loads(path.read_bytes())
     except json.JSONDecodeError as exc:
         raise MineruIncompleteOutputError(f"{artifact} JSON was malformed") from exc
     except (OSError, UnicodeDecodeError) as exc:

tests/test_mineru_runner_paths.py

@@ -286,14 +286,14 @@ def test_parse_mineru_output_distinguishes_read_failures(
         json.dumps([{"layout_dets": []}]), encoding="utf-8"
     )
 
-    original_read_text = Path.read_text
+    original_read_bytes = Path.read_bytes
 
-    def fake_read_text(self, *args, **kwargs):
+    def fake_read_bytes(self, *args, **kwargs):
         if self.name == file_name:
             raise read_error
-        return original_read_text(self, *args, **kwargs)
+        return original_read_bytes(self, *args, **kwargs)
 
-    monkeypatch.setattr(Path, "read_text", fake_read_text)
+    monkeypatch.setattr(Path, "read_bytes", fake_read_bytes)
 
     with pytest.raises(MineruIncompleteOutputError, match=expected_detail) as exc_info:
         mineru_runner._parse_mineru_output(tmp_path, Path("sample.pdf"))
@@ -613,14 +613,14 @@ class Result:
 
         return Result()
 
-    original_read_text = Path.read_text
+    original_read_bytes = Path.read_bytes
 
-    def fake_read_text(self, *args, **kwargs):
+    def fake_read_bytes(self, *args, **kwargs):
         if self.name == file_name:
             raise read_error
-        return original_read_text(self, *args, **kwargs)
+        return original_read_bytes(self, *args, **kwargs)
 
-    monkeypatch.setattr(Path, "read_text", fake_read_text)
+    monkeypatch.setattr(Path, "read_bytes", fake_read_bytes)
     monkeypatch.setattr(mineru_runner.subprocess, "run", fake_run)
 
     with pytest.raises(Exception) as exc_info: