Split unit designators (UNIT/APT/STE/etc) out of combined address strings into addr:unit instead of leaving them in addr:street

qgis-functions.py

@@ -21,13 +21,32 @@ def title(s):
         result = re.sub(pattern, replacement, result)
     return result
 
+# Secondary unit designators that may appear inline in a full address string,
+# e.g. "123 ROLLING ACRES RD UNIT 208" - everything from the designator on is
+# the unit, not part of the street name.
+_UNIT_DESIGNATORS = {'UNIT', 'APT', 'APARTMENT', 'STE', 'SUITE', 'BLDG', 'BUILDING', 'LOT'}
+
+def _splitunit(parts):
+    for i, part in enumerate(parts):
+        if part.upper().rstrip('.') in _UNIT_DESIGNATORS:
+            return parts[:i], (" ".join(parts[i + 1:]) or None)
+    return parts, None
+
 # @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")
+    street_parts, _ = _splitunit(parts)
     #parts = map(formatstreetname, parts)
     #return " ".join(parts)
-    return formatstreet(" ".join(parts), None, None)
+    return formatstreet(" ".join(street_parts), None, None)
+
+# @qgsfunction(args='auto', group='Custom', referenced_columns=[])
+def getunitfromaddress(value1, feature, parent):
+    parts = value1.split()
+    parts.pop(0) # Ignore the first bit (i.e. "123" in "123 N MAIN ST")
+    _, unit = _splitunit(parts)
+    return unit
 
 # @qgsfunction(args='auto', group='Custom', referenced_columns=[])
 def formatstreet(value1, feature, parent):