From 7ae6ee97945837c572cd8331e4c98d0ee87f0153 Mon Sep 17 00:00:00 2001
From: Atalyk Akash <atalyk@google.com>
Date: Thu, 29 Jan 2026 14:14:11 +0000
Subject: [PATCH] Add ADDRESS_HOME_ZIP_AND_CITY field to FR model

---
 model/countries/FR/FR-formatting-rules.yaml |  9 ++-
 model/countries/FR/FR-model.yaml            | 15 +++-
 model/countries/FR/FR-parsing-rules.yaml    | 78 ++++++++++++++++++++-
 3 files changed, 97 insertions(+), 5 deletions(-)

diff --git a/model/countries/FR/FR-formatting-rules.yaml b/model/countries/FR/FR-formatting-rules.yaml
index ad486ce..d0820b4 100644
--- a/model/countries/FR/FR-formatting-rules.yaml
+++ b/model/countries/FR/FR-formatting-rules.yaml
@@ -4,9 +4,7 @@ formatting-rules:
   - separator: "\n"
   - locality2
   - separator: "\n"
-  - postal-code
-  - separator: " "
-  - locality1
+  - postal-code-and-city
   - separator: "\n"
   - admin-area1
   - separator: "\n"
@@ -24,6 +22,11 @@ formatting-rules:
   - separator: " "
   - street
 
+  postal-code-and-city:
+  - postal-code
+  - separator: " "
+  - locality1
+
 examples:
 - id: name
   comment: |
diff --git a/model/countries/FR/FR-model.yaml b/model/countries/FR/FR-model.yaml
index d5359c5..746ea68 100644
--- a/model/countries/FR/FR-model.yaml
+++ b/model/countries/FR/FR-model.yaml
@@ -13,6 +13,19 @@ cut-off-tokens:
   - locality4
 
 extra-definitions:
+  address:
+    - street-address
+    - street-address-alternative-1
+    - locality2
+    - postal-code-and-city
+    - admin-area1
+    - country
+    - country-name
+
   street-address-alternative-1:
   - building-location
-  - address-overflow
\ No newline at end of file
+  - address-overflow
+
+  postal-code-and-city:
+  - postal-code
+  - locality1
\ No newline at end of file
diff --git a/model/countries/FR/FR-parsing-rules.yaml b/model/countries/FR/FR-parsing-rules.yaml
index c4b4387..b732f11 100644
--- a/model/countries/FR/FR-parsing-rules.yaml
+++ b/model/countries/FR/FR-parsing-rules.yaml
@@ -20,6 +20,9 @@ regex_definitions:
   kCommaOrNewlineSeparator:
     regex_fragment: '(?:, |\n|\r|,)+'
 
+  kZipValueRe:
+    regex_fragment: '\b\d{5}\b'
+
 capture_definitions:
   ParseBuildingLocation:
     capture:
@@ -47,6 +50,30 @@ capture_definitions:
           parts: [ {regex_fragment: '(?:[^\r\n]+)'} ]
           quantifier: MATCH_OPTIONAL
 
+  ParsePostalCodeThenCity:
+    capture:
+      output: postal-code-and-city
+      parts:
+      - capture:
+          output: postal-code
+          parts: [ {regex_reference: kZipValueRe} ]
+      - separator: {regex_reference: kCommaOrWhitespaceSeparator}
+      - capture:
+          output: locality1
+          parts: [ {regex_reference: kMultipleWordsRe} ]
+
+  ParseCityThenPostalCode:
+    capture:
+      output: postal-code-and-city
+      parts:
+      - capture:
+          output: locality1
+          parts: [ {regex_reference: kMultipleWordsRe} ]
+      - separator: {regex_reference: kCommaOrWhitespaceSeparator}
+      - capture:
+          output: postal-code
+          parts: [ {regex_reference: kZipValueRe} ]
+
 parsing_definitions:
   building-location:
     decomposition:
@@ -54,6 +81,13 @@ parsing_definitions:
   street-address-alternative-1:
     decomposition:
       capture_reference: StreetAddressDecomposition
+  postal-code-and-city:
+    decomposition_cascade:
+      alternatives:
+        - decomposition:
+            capture_reference: ParsePostalCodeThenCity
+        - decomposition:
+            capture_reference: ParseCityThenPostalCode
 
 test_parsing_definitions:
 - id: "Test 1"
@@ -155,4 +189,46 @@ test_parsing_definitions:
     building-location: "1661 Place Charles de Gaulle"
     street: "Place Charles de Gaulle"
     building: "1661"
-    address-overflow: "Floor 5, Apartment 2"
\ No newline at end of file
+    address-overflow: "Floor 5, Apartment 2"
+- id: "Test 13"
+  type: postal-code-and-city
+  input: "59491 Villeneuve-d'Ascq"
+  output:
+    postal-code-and-city: "59491 Villeneuve-d'Ascq"
+    postal-code: "59491"
+    locality1: "Villeneuve-d'Ascq"
+- id: "Test 14"
+  type: postal-code-and-city
+  input: "Paris 75002"
+  output:
+    postal-code-and-city: "75002 Paris"
+    locality1: "Paris"
+    postal-code: "75002"
+- id: "Test 15"
+  type: postal-code-and-city
+  input: "69120 Vaulx-en-Velin"
+  output:
+    postal-code-and-city: "69120 Vaulx-en-Velin"
+    postal-code: "69120"
+    locality1: "Vaulx-en-Velin"
+- id: "Test 16"
+  type: postal-code-and-city
+  input: "Aix-en-Provence 13100"
+  output:
+    postal-code-and-city: "Aix-en-Provence 13100"
+    locality1: "Aix-en-Provence"
+    postal-code: "13100"
+- id: "Test 17"
+  type: postal-code-and-city  # Matches the key in parsing_definitions
+  input: "Val de Moder 67350"
+  output:
+    postal-code-and-city: "Val de Moder 67350"
+    locality1: "Val de Moder"
+    postal-code: "67350"
+- id: "Test 18"
+  type: postal-code-and-city  # Matches the key in parsing_definitions
+  input: "42000 St. Étienne"
+  output:
+    postal-code-and-city: "42000 St. Étienne"
+    locality1: "St. Étienne"
+    postal-code: "42000"
\ No newline at end of file