fix issue with how negligent devs are identified; add property counts…

… for city owners

data/src/data_utils/negligent_devs.py

@@ -1,40 +1,114 @@
+import re
+
+import pandas as pd
+
+replacements = {
+    "STREET": "ST",
+    "AVENUE": "AVE",
+    "ROAD": "RD",
+    "BOULEVARD": "BLVD",
+    "PLACE": "PL",
+    "FLOOR": "FL",
+    "FLR": "FL",
+    "FIRST": "1ST",
+    "SECOND": "2ND",
+    "THIRD": "3RD",
+    "FOURTH": "4TH",
+    "FIFTH": "5TH",
+    "SIXTH": "6TH",
+    "SEVENTH": "7TH",
+    "EIGHTH": "8TH",
+    "NINTH": "9TH",
+    "NORTH": "N",
+    "SOUTH": "S",
+    "EAST": "E",
+    "WEST": "W",
+    "SUITE": "STE",
+    "LA": "LN",
+    "LANE": "LN",
+    "PARKWAY": "PKY",
+}
+
+
+def standardize_street(street):
+    if not isinstance(street, str):
+        return ""
+    for full, abbr in replacements.items():
+        street = re.sub(r"\b{}\b".format(full), abbr, street, flags=re.IGNORECASE)
+    return street
+
+
+def create_standardized_address(row):
+    parts = [
+        row["mailing_address_1"].strip(),
+        row["mailing_address_2"].strip(),
+        row["mailing_street"].strip(),
+        row["mailing_city_state"].strip(),
+        row["mailing_zip"].strip(),
+    ]
+    standardized_address = ", ".join(part for part in parts if part)
+    return standardized_address.lower()
+
+
 def negligent_devs(primary_featurelayer):
     devs = primary_featurelayer.gdf
+    city_owners = devs[~devs["city_owner_agency"].isna()]
+    non_city_owners = devs[devs["city_owner_agency"].isna()]
 
-    devs = devs[devs["city_owner_agency"].isna()]
-
-    devs["full_mailing_address"] = (
-        devs["mailing_address_1"].str.strip()
-        + ", "
-        + devs["mailing_street"].str.strip()
-        + ", "
-        + devs["mailing_city_state"].str.strip()
-        + ", "
-        + devs["mailing_zip"].str.strip()
+    non_city_owners["mailing_street"] = (
+        non_city_owners["mailing_street"].astype(str).apply(standardize_street)
+    )
+    for term in ["ST", "AVE", "RD", "BLVD"]:
+        non_city_owners["mailing_street"] = non_city_owners["mailing_street"].replace(
+            regex={f"{term}.*": term}
+        )
+    non_city_owners["mailing_address_1"] = non_city_owners["mailing_address_1"].fillna(
+        ""
     )
+    non_city_owners["mailing_address_2"] = non_city_owners["mailing_address_2"].fillna(
+        ""
+    )
+    non_city_owners["mailing_street"] = non_city_owners["mailing_street"].fillna("")
+    non_city_owners["mailing_city_state"] = non_city_owners[
+        "mailing_city_state"
+    ].fillna("")
+    non_city_owners["mailing_zip"] = non_city_owners["mailing_zip"].fillna("")
 
-    devs["standardized_address"] = devs["full_mailing_address"].str.lower().str.strip()
+    non_city_owners["standardized_address"] = non_city_owners.apply(
+        create_standardized_address, axis=1
+    )
 
     address_counts = (
-        devs.groupby("standardized_address").size().reset_index(name="property_count")
+        non_city_owners.groupby("standardized_address")
+        .size()
+        .reset_index(name="property_count")
     )
-
     sorted_address_counts = address_counts.sort_values(
         by="property_count", ascending=False
     )
 
-    devs = devs.merge(sorted_address_counts, on="standardized_address", how="left")
+    non_city_owners = non_city_owners.merge(
+        sorted_address_counts, on="standardized_address", how="left"
+    )
 
-    primary_featurelayer.gdf = primary_featurelayer.gdf.merge(
-        devs[["opa_id", "property_count"]], on="opa_id", how="left"
+    city_owner_counts = (
+        city_owners.groupby("city_owner_agency")
+        .size()
+        .reset_index(name="property_count")
+    )
+    city_owners = city_owners.merge(
+        city_owner_counts, on="city_owner_agency", how="left"
     )
 
+    devs_combined = pd.concat([city_owners, non_city_owners], axis=0)
+    primary_featurelayer.gdf = primary_featurelayer.gdf.merge(
+        devs_combined[["opa_id", "property_count"]], on="opa_id", how="left"
+    )
     primary_featurelayer.gdf.rename(
         columns={"property_count": "n_properties_owned"}, inplace=True
     )
-
     primary_featurelayer.gdf["negligent_dev"] = (
         primary_featurelayer.gdf["n_properties_owned"] > 5
-    )
+    ) & (primary_featurelayer.gdf["city_owner_agency"].isna())
 
     return primary_featurelayer