From 47516de53111f49e8c2d51ec66da788f0e3ca7e6 Mon Sep 17 00:00:00 2001
From: Will Bradley <wbradley@netburner.com>
Date: Sun, 29 Jun 2025 11:55:45 -0700
Subject: [PATCH] Additional handling/improvements

---
 qgis-functions.py | 259 ++++++++++++++++++++++++++++++++++++++++++++++
 threaded.py       |  22 +++-
 2 files changed, 278 insertions(+), 3 deletions(-)
 create mode 100644 qgis-functions.py

diff --git a/qgis-functions.py b/qgis-functions.py
new file mode 100644
index 0000000..898ef51
--- /dev/null
+++ b/qgis-functions.py
@@ -0,0 +1,259 @@
+#import qgis.core
+#import qgis.gui
+import re
+
+#
+# This will keep street names like SR 574A as SR 574A however
+# will lowercase other number-digit suffixes with <2 or >4 numbers
+# or >1 suffix-letters, like 12th Street or 243rd Ave.
+#
+
+# @qgsfunction(args='auto', group='Custom', referenced_columns=[])
+def getstreetfromaddress(value1, feature, parent):
+    parts = value1.split()
+    parts.pop(0) # Ignore the first bit (i.e. "123" in "123 N MAIN ST")
+    parts = map(formatstreetname, parts)
+    return " ".join(parts)
+
+# @qgsfunction(args='auto', group='Custom', referenced_columns=[])
+def formatstreet(value1, feature, parent):
+    parts = value1.split()
+    # Handle the special case of a street name starting with "ST"
+    # which is almost always "Saint __" and not "Street __"
+    if parts[0].upper() == "ST":
+        parts[0] = "Saint"
+    if parts[0].upper() == "ROYAL" and parts[1].upper() == "ST":
+        parts[0] = "Royal"
+        parts[1] = "Saint"
+    # And "CR" as a first part (County Road) vs last part (Circle)
+    if parts[0].upper() == "CR":
+        parts[0] = "County Road"
+    if parts[0].upper() == "SR":
+        parts[0] = "State Route"
+    parts = map(formatstreetname, parts)
+    return " ".join(parts)
+
+
+# @qgsfunction(args='auto', group='Custom', referenced_columns=[])
+def formatname(value1, feature, parent):
+    parts = value1.split()
+    parts = map(formatstreetname, parts)
+    return " ".join(parts)
+
+# @qgsfunction(args='auto', group='Custom', referenced_columns=[])
+def gethighwaytype(value1, feature, parent):
+    match value1:
+        case "ALLEY":
+            return "alley"
+        case "LOCAL":
+            return "residential"
+        case "MAJOR":
+            return "trunk"
+        case "MEDIAN CUT":
+            return "primary_link"
+        case "OTHER":
+            return "unclassified"
+        case "PRIMARY":
+            return "primary"
+        case "PRIVATE":
+            return "service"
+        case "RAMP":
+            return "trunk_link"
+        case "SECONDARY":
+            return "secondary"
+        case "TURN LANE":
+            return "primary_link"
+        case "VEHICULAR TRAIL":
+            return "track"
+
+# Internal function
+def formatstreetname(name):
+    nameUp = name.upper()
+    # Acronyms
+    if nameUp == "SR":
+        return "SR" # State Route
+    if nameUp == "NFS":
+        return "NFS" # National Forest Service?
+    if nameUp == "US":
+        return "US"
+    # Directions
+    if nameUp == "N":
+        return "North"
+    if nameUp == "NE":
+        return "Northeast"
+    if nameUp == "E":
+        return "East"
+    if nameUp == "SE":
+        return "Southeast"
+    if nameUp == "S":
+        return "South"
+    if nameUp == "SW":
+        return "Southwest"
+    if nameUp == "W":
+        return "West"
+    if nameUp == "NW":
+        return "Northwest"
+    # Names
+    if nameUp == "MACLEAY":
+        return "MacLeay"
+    if nameUp == "MCCLAINE":
+        return "McClaine"
+    if nameUp == "MCAHREN":
+        return "McAhren"
+    if nameUp == "MCCAMMON":
+        return "McCammon"
+    if nameUp == "MCCLELLAN":
+        return "McClellan"
+    if nameUp == "MCCOY":
+        return "McCoy"
+    if nameUp == "MCDONALD":
+        return "McDonald"
+    if nameUp == "MCGEE":
+        return "McGee"
+    if nameUp == "MCGILCHRIST":
+        return "McGilchrist"
+    if nameUp == "MCINTOSH":
+        return "McIntosh"
+    if nameUp == "MCKAY":
+        return "McKay"
+    if nameUp == "MCKEE":
+        return "McKee"
+    if nameUp == "MCKENZIE":
+        return "McKenzie"
+    if nameUp == "MCKILLOP":
+        return "McKillop"
+    if nameUp == "MCKINLEY":
+        return "McKinley"
+    if nameUp == "MCKNIGHT":
+        return "McKnight"
+    if nameUp == "MCLAUGHLIN":
+        return "McLaughlin"
+    if nameUp == "MCLEOD":
+        return "McLeod"
+    if nameUp == "MCMASTER":
+        return "McMaster"
+    if nameUp == "MCNARY":
+        return "McNary"
+    if nameUp == "MCNAUGHT":
+        return "McNaught"
+    if nameUp == "O'BRIEN":
+        return "O'Brien"
+    if nameUp == "O'CONNOR":
+        return "O'Connor"
+    if nameUp == "O'NEIL":
+        return "O'Neil"
+    if nameUp == "O'TOOLE":
+        return "O'Toole"
+    # Suffixes
+    if nameUp == "ALY":
+        return "Alley"
+    if nameUp == "AV":
+        return "Avenue"
+    if nameUp == "AVE":
+        return "Avenue"
+    if nameUp == "BAY":
+        return "Bay"
+    if nameUp == "BLF":
+        return "Bluff"
+    if nameUp == "BLVD":
+        return "Boulevard"
+    if nameUp == "BV":
+        return "Boulevard"
+    if nameUp == "BND":
+        return "Bend"
+    if nameUp == "CIR":
+        return "Circle"
+    if nameUp == "CR":
+        return "Circle"
+    if nameUp == "CRK":
+        return "Creek"
+    if nameUp == "CRST":
+        return "Crest"
+    if nameUp == "CT":
+        return "Court"
+    if nameUp == "CURV":
+        return "Curve"
+    if nameUp == "CV":
+        return "Curve"
+    if nameUp == "DR":
+        return "Drive"
+    if nameUp == "FLDS":
+        return "Fields"
+    if nameUp == "GLN":
+        return "Glenn"
+    if nameUp == "GRV":
+        return "Grove"
+    if nameUp == "HL":
+        return "Hill"
+    if nameUp == "HOLW":
+        return "Hollow"
+    if nameUp == "HTS":
+        return "Heights"
+    if nameUp == "HW":
+        return "Highway"
+    if nameUp == "HWY":
+        return "Highway"
+    if nameUp == "HY":
+        return "Highway"
+    if nameUp == "LN":
+        return "Lane"
+    if nameUp == "LNDG":
+        return "Landing"
+    if nameUp == "LOOP":
+        return "Loop"
+    if nameUp == "LP":
+        return "Loop"
+    if nameUp == "MNR":
+        return "Manor"
+    if nameUp == "MT":
+        return "Mount"
+    if nameUp == "MTN":
+        return "Mountain"
+    if nameUp == "PARK":
+        return "Park"
+    if nameUp == "PASS":
+        return "Pass"
+    if nameUp == "PATH":
+        return "Path"
+    if nameUp == "PKWY":
+        return "Parkway"
+    if nameUp == "PL":
+        return "Place"
+    if nameUp == "PLZ":
+        return "Plaza"
+    if nameUp == "PS":
+        return "Pass"
+    if nameUp == "PT":
+        return "Point"
+    if nameUp == "RD":
+        return "Road"
+    if nameUp == "RDG":
+        return "Ridge"
+    if nameUp == "RUN":
+        return "Run"
+    if nameUp == "SHRS":
+        return "Shores"
+    if nameUp == "SQ":
+        return "Square"
+    if nameUp == "ST":
+        return "Street"
+    if nameUp == "TER":
+        return "Terrace"
+    if nameUp == "TR":
+        return "Trail"
+    if nameUp == "TRL":
+        return "Trail"
+    if nameUp == "VW":
+        return "View"
+    if nameUp == "WALK":
+        return "Walk"
+    if nameUp == "WAY":
+        return "Way"
+    if nameUp == "WY":
+        return "Way"
+    if nameUp == "XING":
+        return "Crossing"
+    if re.match('^[0-9]{2,4}[A-Za-z]$', name) != None:
+        return name
+
+    return name #.capitalize()
\ No newline at end of file
diff --git a/threaded.py b/threaded.py
index bb05b24..ff70cb4 100644
--- a/threaded.py
+++ b/threaded.py
@@ -8,6 +8,10 @@ Compares two GeoJSON files containing road data and identifies:
 
 Only reports differences that are significant (above minimum length threshold).
 Optimized for performance with parallel processing and spatial indexing.
+
+TODO:
+- put properties properly on removed roads, so they're visible in JOSM
+- handle polygons properly (on previous geojson step?) for circular roads
 """
 
 import json
@@ -26,9 +30,19 @@ import numpy as np
 from concurrent.futures import ProcessPoolExecutor, as_completed
 import gc
 
+import importlib  
+qgisfunctions = importlib.import_module("qgis-functions")
+
 # Suppress warnings for cleaner output
 warnings.filterwarnings('ignore')
 
+import re
+def titlecase(s):
+    return re.sub(
+        r"[A-Za-z]+('[A-Za-z]+)?",
+        lambda word: word.group(0).capitalize(),
+        s)
+
 class RoadComparator:
     def __init__(self, tolerance_feet: float = 50.0, min_gap_length_feet: float = 100.0, 
                  n_jobs: int = None, chunk_size: int = 1000):
@@ -307,11 +321,13 @@ class RoadComparator:
 
                         for key, value in original_properties.items():
                             if key == 'NAME':
-                                properties['name'] = str(value).title() if value is not None else None
+                                properties['name'] = titlecase(qgisfunctions.formatstreet(value,None,None)) if value is not None else None
                             elif key == 'SpeedLimit':
                                 properties['maxspeed'] = f"{value} mph" if value is not None else None
                             elif key == 'RoadClass':
-                                if value.startswith('PRIMARY'):
+                                if value is None:
+                                    properties['highway'] = 'residential'
+                                elif value.startswith('PRIMARY'):
                                     properties['highway'] = 'trunk'
                                 elif value.startswith('MAJOR'):
                                     properties['highway'] = 'primary'
@@ -328,6 +344,7 @@ class RoadComparator:
                         })
                         
             except Exception as e:
+                print(e)
                 continue  # Skip problematic geometries
         
         return added_roads
@@ -420,7 +437,6 @@ class RoadComparator:
                     if field in segment and pd.notna(segment[field]):
                         road_name = str(segment[field])
                         break
-                
                 if road_name not in removed_by_road:
                     removed_by_road[road_name] = []
                 removed_by_road[road_name].append(length_feet)