Clean up sys.path setup for modal_app.images imports

- Use existing sys/Path imports instead of aliased re-imports
- Remove duplicate sys.path block in pipeline.py (now handled once at top)
- Add sys.path fix to restage.py (also imports from modal_app)
- Consistent pattern across all modal_app/ entrypoints:
  sys.path gets /root/policyengine-us-data (baked image) and
  local repo root before any from modal_app.* imports

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

Author: baogorek

modal_app/data_build.py

@@ -11,14 +11,11 @@
 
 import modal
 
-import sys as _sys
-from pathlib import Path as _Path
-
 _baked = "/root/policyengine-us-data"
-_local = str(_Path(__file__).resolve().parent.parent)
+_local = str(Path(__file__).resolve().parent.parent)
 for _p in (_baked, _local):
-    if _p not in _sys.path:
-        _sys.path.insert(0, _p)
+    if _p not in sys.path:
+        sys.path.insert(0, _p)
 
 from modal_app.images import cpu_image as image
 

modal_app/local_area.py

@@ -13,18 +13,17 @@
 
 import os
 import subprocess
+import sys
 import json
 import modal
 from pathlib import Path
 from typing import List, Dict
 
-import sys as _sys
-
 _baked = "/root/policyengine-us-data"
 _local = str(Path(__file__).resolve().parent.parent)
 for _p in (_baked, _local):
-    if _p not in _sys.path:
-        _sys.path.insert(0, _p)
+    if _p not in sys.path:
+        sys.path.insert(0, _p)
 
 from modal_app.images import cpu_image as image
 

modal_app/pipeline.py

@@ -35,6 +35,7 @@
 import json
 import os
 import subprocess
+import sys
 import time
 import traceback
 from dataclasses import asdict, dataclass, field
@@ -44,13 +45,12 @@
 from typing import Optional
 
 import modal
-import sys as _sys
 
 _baked = "/root/policyengine-us-data"
 _local = str(Path(__file__).resolve().parent.parent)
 for _p in (_baked, _local):
-    if _p not in _sys.path:
-        _sys.path.insert(0, _p)
+    if _p not in sys.path:
+        sys.path.insert(0, _p)
 
 from modal_app.images import cpu_image as image
 
@@ -253,20 +253,7 @@ def _record_step(
 # app.include() merges functions from other apps into this one,
 # ensuring Modal mounts their files and registers their functions
 # (with their GPU/memory/volume configs) in the ephemeral run.
-#
-# Inside Modal containers the auto-mounted package root may not be
-# on sys.path when the module first loads; ensure it is importable.
-import sys
-
-_parent = str(Path(__file__).resolve().parent.parent)
-if _parent not in sys.path:
-    sys.path.insert(0, _parent)
-# The image bakes the repo at /root/policyengine-us-data, but Modal
-# auto-mounts the entrypoint elsewhere, so _parent may not contain
-# modal_app/.  Ensure the baked repo root is always importable.
-_baked = "/root/policyengine-us-data"
-if _baked not in sys.path:
-    sys.path.insert(0, _baked)
+# sys.path setup is handled at the top of this file.
 
 from modal_app.data_build import app as _data_build_app
 from modal_app.data_build import build_datasets

modal_app/remote_calibration_runner.py

@@ -1,15 +1,15 @@
 import os
 import subprocess
-import modal
+import sys
+from pathlib import Path
 
-import sys as _sys
-from pathlib import Path as _Path
+import modal
 
 _baked = "/root/policyengine-us-data"
-_local = str(_Path(__file__).resolve().parent.parent)
+_local = str(Path(__file__).resolve().parent.parent)
 for _p in (_baked, _local):
-    if _p not in _sys.path:
-        _sys.path.insert(0, _p)
+    if _p not in sys.path:
+        sys.path.insert(0, _p)
 
 from modal_app.images import gpu_image as image
 

restage.py

@@ -1,5 +1,14 @@
 """Re-upload files from Modal staging volume to HF staging."""
 
+import sys
+from pathlib import Path
+
+_baked = "/root/policyengine-us-data"
+_local = str(Path(__file__).resolve().parent)
+for _p in (_baked, _local):
+    if _p not in sys.path:
+        sys.path.insert(0, _p)
+
 from modal_app.local_area import app, validate_staging, upload_to_staging
 
 branch = "fix-would-file-blend-and-entity-weights"