X-Git-Url: https://wannabe.guru.org/gitweb/?a=blobdiff_plain;f=geocode.py;h=e9e5c35c5fbec6a272518351c461ec5a4fed4243;hb=e46158e49121b8a955bb07b73f5bcf9928b79c90;hp=39593609f31bd4f3f00fa1556099d986844f3bc5;hpb=62b4b777a68a99c8e446294f7e7e35ebd9668989;p=python_utils.git

diff --git a/geocode.py b/geocode.py
index 3959360..e9e5c35 100644
--- a/geocode.py
+++ b/geocode.py
@@ -3,10 +3,11 @@
 # © Copyright 2022, Scott Gasch
 
 """Wrapper around US Census address geocoder API described here:
-https://www2.census.gov/geo/pdfs/maps-data/data/Census_Geocoder_User_Guide.pdf
-https://geocoding.geo.census.gov/geocoder/Geocoding_Services_API.pdf
 
-Also try:
+* https://www2.census.gov/geo/pdfs/maps-data/data/Census_Geocoder_User_Guide.pdf
+* https://geocoding.geo.census.gov/geocoder/Geocoding_Services_API.pdf
+
+Also try::
 
     $ curl --form addressFile=@localfile.csv \
            --form benchmark=2020 \
@@ -27,9 +28,24 @@ logger = logging.getLogger(__name__)
 
 
 def geocode_address(address: str) -> Optional[Dict[str, Any]]:
-    """Send a single address to the US Census geocoding API.  The response
-    is a parsed JSON chunk of data with N addressMatches in the result
-    section and the details of each match within it.  Returns None on error.
+    """Send a single address to the US Census geocoding API in order to
+    lookup relevant data about it (including, if possible, its
+    lat/long).  The response is a parsed JSON chunk of data with N
+    addressMatches in the result section and the details of each match
+    within it.
+
+    Args:
+        address: the full address to lookup in the form: "STREET
+        ADDRESS, CITY, STATE, ZIPCODE".  These components may be
+        omitted and the service will make educated guesses but
+        the commas delimiting each component must be included.
+
+    Returns:
+        A parsed json dict with a bunch of information about the
+            address contained within it.  Each 'addressMatch'
+            in the JSON describes the details of a possible match.
+            Returns None if there was an error or the address is
+            not known.
 
     >>> json = geocode_address('4600 Silver Hill Rd,, 20233')
     >>> json['result']['addressMatches'][0]['matchedAddress']
@@ -37,7 +53,6 @@ def geocode_address(address: str) -> Optional[Dict[str, Any]]:
 
     >>> json['result']['addressMatches'][0]['coordinates']
     {'x': -76.9274328556918, 'y': 38.845989080537514}
-
     """
     url = 'https://geocoding.geo.census.gov/geocoder/geographies/onelineaddress'
     url += f'?address={address}'
@@ -58,15 +73,25 @@ def geocode_address(address: str) -> Optional[Dict[str, Any]]:
     return r.json()
 
 
-def batch_geocode_addresses(addresses: List[str]):
-    """Send up to addresses for batch geocoding.  Each line of the input
-    list should be a single address of the form: STREET ADDRESS, CITY,
-    STATE, ZIP.  Components may be omitted but the commas may not be.
-    Result is an array of the same size as the input array with one
-    answer record per line.  Returns None on error.
+def batch_geocode_addresses(addresses: List[str]) -> Optional[List[str]]:
+    """Send a list of addresses for batch geocoding to a web service
+    operated by the US Census Bureau.
+
+    Args:
+        addresses: a list of addresses to geocode.  Each line of the
+            input list should be a single address in the form: "STREET
+            ADDRESS, CITY, STATE, ZIPCODE".  Individual address components
+            may be omitted and the service will make educated guesses but
+            the commas delimiters between address components may not be
+            omitted.
+
+    Returns:
+        An array of the same size as the input array with one
+        answer record per line.  Returns None on error.
 
-    This code will deal with requests >10k addresses by chunking them
-    internally because the census website disallows requests > 10k lines.
+    Note: this code will deal with requests >10k addresses by chunking
+    them internally because the census website disallows requests >
+    10k lines.
 
     >>> batch_geocode_addresses(
     ...     [