Reverting census parser output csv header

---

(.venv) MP-LF4T2CC2CX:main/pyhmda tignort$ diff CensusFlatFile2024-parsed.txt CensusFlatFile2024-parsed.txt.BEFORE
1c1
< Collection Year|MSA/MD|State|County|Census Tract|FFIEC Median Family Income|Population|Minority Population %|Number of Owner Occupied Units|Number of 1 to 4 Family Units|Tract MFI|Tract to MSA Income %|Median Age|Small County|MSA/MD Name
---
> CollectionYear|CBSACode|FIPSStateCode|FIPSCountyCode|CensusTract|FFIECMedianFamilyIncome|Population|MinorityPopulationPct|NumOwnerOccupiedUnits|Num1To4FamilyUnits|TractMFI|TractToMSAIncomePct|MedianAge|SmallCounty|MSAOrMDTitle
(.venv) MP-LF4T2CC2CX:main/pyhmda tignort$
(.venv) MP-LF4T2CC2CX:main/pyhmda tignort$ head -1 CensusFlatFile2024-parsed.txt > reverted-header.txt
(.venv) MP-LF4T2CC2CX:main/pyhmda tignort$ cat reverted-header.txt
Collection Year|MSA/MD|State|County|Census Tract|FFIEC Median Family Income|Population|Minority Population %|Number of Owner Occupied Units|Number of 1 to 4 Family Units|Tract MFI|Tract to MSA Income %|Median Age|Small County|MSA/MD Name
(.venv) MP-LF4T2CC2CX:main/pyhmda tignort$ cat expected.txt
Collection Year|MSA/MD|State|County|Census Tract|FFIEC Median Family Income|Population|Minority Population %|Number of Owner Occupied Units|Number of 1 to 4 Family Units|Tract MFI|Tract to MSA Income %|Median Age|Small County|MSA/MD Name
(.venv) MP-LF4T2CC2CX:main/pyhmda tignort$ diff reverted-header.txt expected.txt
(.venv) MP-LF4T2CC2CX:main/pyhmda tignort$

common/src/main/pyhmda/parse_census_file.py

@@ -24,15 +24,14 @@ def conv_scf(val: str) -> str:
     if val not in ["T", "S", "I"]: raise ValueError(f"invalid scf: \"{val}\"")
     return val    
 
-
 census_file_columns = {
-    0: ["CollectionYear", conv_num], 1: ["CBSACode", conv_dgstr], 2: ["FIPSStateCode", conv_dgstr],
-    3: ["FIPSCountyCode", conv_dgstr], 4: ["CensusTract", conv_dgstr],
-    13: ["FFIECMedianFamilyIncome", conv_num], 22: ["Population", conv_optnum],
-    28: ["MinorityPopulationPct", conv_optpct], 879: ["NumOwnerOccupiedUnits", conv_optnum],
-    899: ["Num1To4FamilyUnits", conv_optnum], 585: ["TractMFI", conv_optnum],
-    12: ["TractToMSAIncomePct", conv_optpct], 1057: ["MedianAge", conv_optnum],
-    6: ["SmallCounty", conv_scf]
+    0: ["Collection Year", conv_num], 1: ["MSA/MD", conv_dgstr], 2: ["State", conv_dgstr],
+    3: ["County", conv_dgstr], 4: ["Census Tract", conv_dgstr],
+    13: ["FFIEC Median Family Income", conv_num], 22: ["Population", conv_optnum],
+    28: ["Minority Population %", conv_optpct], 879: ["Number of Owner Occupied Units", conv_optnum],
+    899: ["Number of 1 to 4 Family Units", conv_optnum], 585: ["Tract MFI", conv_optnum],
+    12: ["Tract to MSA Income %", conv_optpct], 1057: ["Median Age", conv_optnum],
+    6: ["Small County", conv_scf]
 }
 cfkeys = census_file_columns.keys()
 cfcolnames = {k: v[0] for k, v in census_file_columns.items()}
@@ -50,7 +49,7 @@ def conv_scf(val: str) -> str:
 
 delineation_file_columns = {
     3: ["CBSATitle", conv_str], 5: ["MDTitle", conv_optstr],
-    9: ["FIPSStateCode", conv_dgstr], 10: ["FIPSCountyCode", conv_dgstr]
+    9: ["State", conv_dgstr], 10: ["County", conv_dgstr]
 }
 dfkeys = delineation_file_columns.keys()
 dfcolnames = {k: v[0] for k, v in delineation_file_columns.items()}
@@ -59,17 +58,17 @@ def conv_scf(val: str) -> str:
                               converters=dfconverters).rename(dfcolnames, axis=1)
 logging.info(f"Parsed {prepared_file}")
 
-parsed_delin_df["MSAOrMDTitle"] = parsed_delin_df.apply(lambda row:
+parsed_delin_df["MSA/MD Name"] = parsed_delin_df.apply(lambda row:
     row.MDTitle if pd.notna(row.MDTitle) else row.CBSATitle, axis=1)
 parsed_delin_df.drop(columns=["CBSATitle", "MDTitle"], inplace=True)
-logging.info("Calculated MSAOrMDTitles")
+logging.info("Calculated MSA/MD Names")
 
 output_file = args.output_file if args.output_file \
     else f"{os.path.splitext(args.censusfile)[0]}-parsed.txt"
 output_df = parsed_census_df.merge(parsed_delin_df,
-    how="left", on=["FIPSStateCode", "FIPSCountyCode"])
-output_df["MSAOrMDTitle"] = output_df.apply(lambda row:
-    "" if row.CBSACode == "99999" else row.MSAOrMDTitle, axis=1)
+    how="left", on=["State", "County"])
+output_df["MSA/MD Name"] = output_df.apply(lambda row:
+    "" if row["MSA/MD"] == "99999" else row["MSA/MD Name"], axis=1)
 output_df.to_csv(output_file, sep='|', index=False)
 logging.info(f"Wrote output file {output_file}")
 os.remove(prepared_file)

common/src/main/pyhmda/utils.py

@@ -87,7 +87,7 @@ def apply_authorized_modifications(modmap: dict, df: pd.DataFrame) -> pd.DataFra
 # Census Flat File Modifications
 
 def replace_MedianAge_2002_values(df: pd.DataFrame) -> pd.DataFrame:
-    df.loc[df["MedianAge"] == 2002, "MedianAge"] = 6
+    df.loc[df["Median Age"] == 2002, "Median Age"] = 6
     return df