From 9b250a93937445a184dd7f92224c797ada35c6f3 Mon Sep 17 00:00:00 2001 From: Al Date: Thu, 15 Sep 2016 11:22:55 -0400 Subject: [PATCH] [openaddresses] adding zero-padding option for postcodes and using in Puerto Rico --- resources/parser/data_sets/openaddresses.yaml | 2 ++ scripts/geodata/openaddresses/formatter.py | 5 +++++ 2 files changed, 7 insertions(+) diff --git a/resources/parser/data_sets/openaddresses.yaml b/resources/parser/data_sets/openaddresses.yaml index 32d95a9b..b3e1df64 100644 --- a/resources/parser/data_sets/openaddresses.yaml +++ b/resources/parser/data_sets/openaddresses.yaml @@ -2073,6 +2073,8 @@ countries: language: es files: - filename: statewide.csv + # postcodes should be like 00901 + postcode_length: 5 ri: add: diff --git a/scripts/geodata/openaddresses/formatter.py b/scripts/geodata/openaddresses/formatter.py index 03e6bf20..6601e36a 100644 --- a/scripts/geodata/openaddresses/formatter.py +++ b/scripts/geodata/openaddresses/formatter.py @@ -227,6 +227,8 @@ class OpenAddressesFormatter(object): place_only_probability = float(self.get_property('place_only_probability', *configs)) place_and_postcode_probability = float(self.get_property('place_and_postcode_probability', *configs)) + postcode_length = int(self.get_property('postcode_length', *configs)) + drop_address_probability = place_only_probability + place_and_postcode_probability ignore_rows_missing_fields = set(self.get_property('ignore_rows_missing_fields', *configs) or []) @@ -346,6 +348,9 @@ class OpenAddressesFormatter(object): components.pop(AddressFormatter.POSTCODE) postcode = None else: + if postcode_length: + postcode = postcode.zfill(postcode_length) + components[AddressFormatter.POSTCODE] = postcode unit = components.get(AddressFormatter.UNIT, None)