BDMS 227: permission & well pump type transfer fix

transfers/permissions_transfer.py

@@ -1,5 +1,6 @@
 from sqlalchemy.orm import Session
 from datetime import datetime
+from pandas import isna
 
 from db import Thing, PermissionHistory
 from transfers.util import read_csv, logger, replace_nans
@@ -15,8 +16,10 @@
 
 def transfer_permissions(session: Session):
     """
-    The transferred wells and contacts need to be queried to know who gave
-    permission to which well since contact_id is required for PermissionHistory
+    The transferred wells and contacts need to be transferred first
+    - to access the auto-generated well IDs
+    - to know who gave permission to which well since contact_id is required for
+        PermissionHistory
     """
     wdf = read_csv("WellData", dtype={"OSEWelltagID": str})
     wdf = replace_nans(wdf)
@@ -38,7 +41,11 @@ def transfer_permissions(session: Session):
         allow_water_level_samples = wdf.loc[
             wdf["PointID"] == well.name, "MonitorOK"
         ].values
-        if len(allow_water_level_samples) > 0 and allow_water_level_samples is not None:
+        if len(allow_water_level_samples) == 0:
+            pass
+        elif isna(allow_water_level_samples[0]):
+            pass
+        else:
             try:
                 permission_allowed = bool(allow_water_level_samples[0])
                 permission = PermissionHistory(
@@ -61,10 +68,11 @@ def transfer_permissions(session: Session):
         allow_water_chemistry_samples = wdf.loc[
             wdf["PointID"] == well.name, "SampleOK"
         ].values
-        if (
-            len(allow_water_chemistry_samples) > 0
-            and allow_water_chemistry_samples is not None
-        ):
+        if len(allow_water_chemistry_samples) == 0:
+            pass
+        elif isna(allow_water_chemistry_samples[0]):
+            pass
+        else:
             try:
                 permission_allowed = bool(allow_water_chemistry_samples[0])
                 permission = PermissionHistory(

transfers/well_transfer.py

@@ -16,7 +16,7 @@
 import json
 import time
 from datetime import datetime, UTC
-
+import re
 import pandas as pd
 from pandas import isna
 from pydantic import ValidationError
@@ -122,19 +122,26 @@ def _extract_casing_materials(row) -> list[str]:
     return materials
 
 
+pattern = re.compile(
+    r"\b(?P<term>jet|hand|submersible)\b|\b(?P<phrase>line[-\s]+shaft)\b", re.IGNORECASE
+)
+
+
+def first_matched_term(text: str):
+    m = pattern.search(text)
+    if not m:
+        return None
+    return m.group("term") or m.group("phrase")
+
+
+PUMP_MAPPING = {"jet": "Jet", "hand": "Hand", "submersible": "Submersible"}
+
+
 def _extract_well_pump_type(row) -> str | None:
+    if isna(row.ConstructionNotes):
+        return None
     construction_notes = row.ConstructionNotes.lower()
-    if "pump" in construction_notes:
-        if "submersible" in construction_notes:
-            return "Submersible"
-        elif "jet" in construction_notes:
-            return "Jet"
-        elif "line shaft" in construction_notes or "lineshaft" in construction_notes:
-            return "Line Shaft"
-        elif "hand" in construction_notes:
-            return "Hand"
-        else:
-            return None
+    return PUMP_MAPPING.get(first_matched_term(construction_notes), None)
 
 
 # Parse aquifer codes
@@ -357,9 +364,7 @@ def transfer_wells(session: Session, flags: dict = None, limit: int = 0) -> None
             well_casing_materials = (
                 [] if isna(row.CasingDescription) else _extract_casing_materials(row)
             )
-            well_pump_type = (
-                _extract_well_pump_type(row) if row.ConstructionNotes else None
-            )
+            well_pump_type = _extract_well_pump_type(row)
 
             # manually add the well rather than add_well from services/thing_helper.py
             # so that effective_start can be set on the location assocation