From 25fb7168e3a0b99c607af9644e152ceab443c069 Mon Sep 17 00:00:00 2001
From: Dan Paulat <dpaulat@gmail.com>
Date: Sun, 23 Oct 2022 08:19:14 -0500
Subject: [PATCH] Python script to generate counties database

---
 .gitmodules                           |   3 +
 data                                  |   1 +
 scwx-qt/tools/generate_counties_db.py | 121 ++++++++++++++++++++++++++
 3 files changed, 125 insertions(+)
 create mode 160000 data
 create mode 100644 scwx-qt/tools/generate_counties_db.py

diff --git a/.gitmodules b/.gitmodules
index ca0da9a3..6428f061 100644
--- a/.gitmodules
+++ b/.gitmodules
@@ -16,3 +16,6 @@
 [submodule "external/stb"]
 	path = external/stb
 	url = https://github.com/nothings/stb.git
+[submodule "data"]
+	path = data
+	url = ../supercell-wx-data
diff --git a/data b/data
new file mode 160000
index 00000000..19a3854e
--- /dev/null
+++ b/data
@@ -0,0 +1 @@
+Subproject commit 19a3854e7986215bf6282ccbfd173635f4ac03eb
diff --git a/scwx-qt/tools/generate_counties_db.py b/scwx-qt/tools/generate_counties_db.py
new file mode 100644
index 00000000..1cbc01bf
--- /dev/null
+++ b/scwx-qt/tools/generate_counties_db.py
@@ -0,0 +1,121 @@
+import argparse
+import geopandas as gpd
+import pathlib
+import sqlite3
+
+class DatabaseInfo:
+    def __init__(self):
+        self.sqlConnection_ = None
+        self.sqlCursor_     = None
+
+def ParseArguments():
+    parser = argparse.ArgumentParser(description='Generate counties SQLite database')
+    parser.add_argument("-c", "--county_dbf",
+                        metavar = "filename",
+                        help    = "input county database",
+                        dest    = "inputCountyDbs_",
+                        action  = "extend",
+                        nargs   = "+",
+                        default = [],
+                        type    = pathlib.Path)
+    parser.add_argument("-z", "--zone_dbf",
+                        metavar = "filename",
+                        help    = "input zone database",
+                        dest    = "inputZoneDbs_",
+                        action  = "extend",
+                        nargs   = "+",
+                        default = [],
+                        type    = pathlib.Path)
+    parser.add_argument("-o", "--output_db",
+                        metavar  = "filename",
+                        help     = "output sqlite database",
+                        dest     = "outputDb_",
+                        type     = pathlib.Path,
+                        required = True)
+    return parser.parse_args()
+
+def Prepare(dbInfo, outputDb):
+    # Truncate existing database
+    file = open(outputDb, 'w')
+    file.close()
+
+    # Establish SQLite database connection
+    dbInfo.sqlConnection_ = sqlite3.connect(outputDb)
+    
+    # Set row factory for name-based access to columns
+    dbInfo.sqlConnection_.row_factory = sqlite3.Row
+
+    dbInfo.sqlCursor_     = dbInfo.sqlConnection_.cursor()
+
+    # Create database table
+    dbInfo.sqlCursor_.execute("""CREATE TABLE counties(
+        id   TEXT NOT NULL PRIMARY KEY,
+        name TEXT)""")
+
+def ProcessCountiesDbf(dbInfo, dbfFilename):
+    # County area type
+    areaType = 'C'
+
+    # Read dataframe
+    dbfTable = gpd.read_file(filename        = dbfFilename,
+                             include_fields  = ["STATE", "FIPS", "COUNTYNAME"],
+                             ignore_geometry = True)
+    dbfTable.drop_duplicates(inplace=True)
+
+    for row in dbfTable.itertuples():
+        # Generate a FIPS ID compatible with UGC format (NWSI 10-1702)
+        fipsId = "{}{}{:03}".format(row.STATE, areaType, (int(row.FIPS) % 1000))
+
+        # Insert FIPS ID and name pair into database
+        try:
+            dbInfo.sqlCursor_.execute("INSERT INTO counties VALUES (?, ?)", (fipsId, row.COUNTYNAME))
+        except:
+            print("Skipping duplicate county:", fipsId, row.COUNTYNAME)
+
+def ProcessZoneDbf(dbInfo, dbfFilename):
+    print("Processing zone file:", dbfFilename)
+    # Zone area type
+    areaType = 'Z'
+
+    # Read dataframe
+    dbfTable = gpd.read_file(filename        = dbfFilename,
+                             include_fields  = ["ID", "STATE", "ZONE", "NAME"],
+                             ignore_geometry = True)
+    dbfTable.drop_duplicates(inplace=True)
+
+    for row in dbfTable.itertuples():
+        # Generate a FIPS ID compatible with UGC format (NWSI 10-1702)
+        if "ID" in dbfTable.columns:
+            fipsId = row.ID
+        else:
+            fipsId = "{}{}{:03}".format(row.STATE, areaType, (int(row.ZONE) % 1000))
+
+        # Insert FIPS ID and name pair into database
+        try:
+            dbInfo.sqlCursor_.execute("INSERT INTO counties VALUES (?, ?)",
+                                      (fipsId, row.NAME))
+        except:
+            # Only print warning if FIPS ID has multiple names
+            result = dbInfo.sqlCursor_.execute("SELECT name FROM counties WHERE id = :fipsId",
+                                               {"fipsId": fipsId})
+            resultRow = result.fetchone()
+            if resultRow is not None:
+                if resultRow["name"] != row.NAME:
+                    print("Skipping duplicate zone:", fipsId, row.NAME)
+
+def PostProcess(dbInfo):
+    # Commit changes and close database
+    dbInfo.sqlConnection_.commit()
+    dbInfo.sqlConnection_.close()
+
+dbInfo = DatabaseInfo()
+args   = ParseArguments()
+Prepare(dbInfo, args.outputDb_)
+
+for countyDb in args.inputCountyDbs_:
+    ProcessCountiesDbf(dbInfo, countyDb)
+
+for zoneDb in args.inputZoneDbs_:
+    ProcessZoneDbf(dbInfo, zoneDb)
+
+PostProcess(dbInfo)