Add statefip (FIPS code) input support

changelog.d/add-statefip-support.added.md

@@ -0,0 +1 @@
+Add statefip (FIPS code) input support, matching TAXSIM-35 behavior.

dashboard/src/components/CsvRunner.jsx

@@ -40,7 +40,7 @@ const SAMPLE_DATASETS = [
 const fmt = (n) => Number(n).toLocaleString();
 
 const KNOWN_COLUMNS = new Set([
-  'taxsimid', 'year', 'state', 'mstat', 'page', 'sage',
+  'taxsimid', 'year', 'state', 'statefip', 'mstat', 'page', 'sage',
   'dependent_exemption', 'depx', 'pwages', 'swages',
   'psemp', 'ssemp', 'dividends', 'intrec', 'stcg', 'ltcg',
   'otherprop', 'nonprop', 'pensions', 'gssi',

policyengine_taxsim/api.py

@@ -49,6 +49,7 @@
     "taxsimid",
     "year",
     "state",
+    "statefip",
     "mstat",
     "page",
     "sage",

policyengine_taxsim/core/utils.py

@@ -284,3 +284,6 @@ def get_ordinal(n):
     50: 55,  # Wisconsin
     51: 56,  # Wyoming
 }
+
+# Reverse mapping: FIPS → SOI (for statefip input support)
+FIPS_TO_SOI_MAP = {fips: soi for soi, fips in SOI_TO_FIPS_MAP.items()}

policyengine_taxsim/runners/base_runner.py

@@ -5,8 +5,10 @@
 
 try:
     from ..core.io import write_output
+    from ..core.utils import FIPS_TO_SOI_MAP
 except ImportError:
     from policyengine_taxsim.core.io import write_output
+    from policyengine_taxsim.core.utils import FIPS_TO_SOI_MAP
 
 
 class BaseTaxRunner(ABC):
@@ -39,6 +41,17 @@ def _validate_input(self):
         if "year" not in self.input_df.columns:
             raise ValueError("Input data must contain a 'year' column")
 
+        # Convert statefip (FIPS) to state (SOI) per row.
+        # Matches TAXSIM-35 behavior: for each record, if state==0
+        # and statefip is set, convert FIPS→SOI.
+        if "statefip" in self.input_df.columns:
+            if "state" not in self.input_df.columns:
+                self.input_df["state"] = 0
+            fips_vals = self.input_df["statefip"].astype(int)
+            soi_from_fips = fips_vals.map(FIPS_TO_SOI_MAP).fillna(0).astype(int)
+            use_fips = (self.input_df["state"] == 0) & (fips_vals > 0)
+            self.input_df.loc[use_fips, "state"] = soi_from_fips[use_fips]
+
         # Auto-assign taxsimid if not present
         if "taxsimid" not in self.input_df.columns:
             # Assign sequential IDs starting from 1