From 29f9cc9570454baf80e66f323bf66f51160f35f2 Mon Sep 17 00:00:00 2001
From: m-maryia <mmaryia@google.com>
Date: Fri, 11 Jul 2025 18:20:50 +0200
Subject: [PATCH 1/5] introduce zip prefix and suffix

---
 model/build_internal_version.sh               |  1 +
 model/countries/BR/BR-parsing-rules.yaml      | 40 +++++++++++++
 model/countries/CA/CA-formatting-rules.yaml   |  7 ++-
 model/countries/CA/CA-parsing-rules.yaml      | 44 ++++++++++++++
 model/countries/NL/NL-formatting-rules.yaml   | 10 +++-
 model/countries/NL/NL-parsing-rules.yaml      | 42 ++++++++++++-
 model/countries/PL/PL-parsing-rules.yaml      | 42 ++++++++++++-
 model/countries/US/US-formatting-rules.yaml   |  7 ++-
 model/countries/US/US-parsing-rules.yaml      | 56 +++++++++++++++++
 .../countries/global/global-descriptions.yaml |  2 +
 .../global/global-formatting-rules.yaml       |  5 ++
 model/countries/global/global-model.yaml      |  4 +-
 .../global/global-parsing-rules.yaml          | 60 +++++++++++++++++++
 model/vendor                                  |  1 +
 14 files changed, 312 insertions(+), 9 deletions(-)
 create mode 120000 model/build_internal_version.sh
 create mode 100644 model/countries/CA/CA-parsing-rules.yaml
 create mode 100644 model/countries/US/US-parsing-rules.yaml
 create mode 120000 model/vendor

diff --git a/model/build_internal_version.sh b/model/build_internal_version.sh
new file mode 120000
index 0000000..ebee58d
--- /dev/null
+++ b/model/build_internal_version.sh
@@ -0,0 +1 @@
+/Users/mmaryia/autofill-i18n-model/build_internal_version.sh
\ No newline at end of file
diff --git a/model/countries/BR/BR-parsing-rules.yaml b/model/countries/BR/BR-parsing-rules.yaml
index 9168462..744942e 100644
--- a/model/countries/BR/BR-parsing-rules.yaml
+++ b/model/countries/BR/BR-parsing-rules.yaml
@@ -19,6 +19,14 @@ regex_definitions:
 
   kFloorLiteralRe:  # Regex for literal for a floor
     regex_fragment: andar
+  
+  # Regular expression to match 5-digit zip prefix
+  kZipPrefixValueRe:
+    regex_fragment: (?:\d{5})
+
+  # Regular expression to match 3-digit zip suffix
+  kZipSuffixValueRe:
+    regex_fragment: (?:\d{3})
 
 capture_definitions:
   ParseBuildingLocation:
@@ -97,6 +105,22 @@ capture_definitions:
       parts:
         - regex_fragment: '[^,\n]+'
 
+  ParsePostalCodeExpression:
+    capture:
+      output: postal-code
+      parts:
+      - capture:
+          output: postal-code-prefix
+          parts: [ {regex_reference: kZipPrefixValueRe} ]
+      - no_capture:
+          parts:
+          - separator: {regex_reference: kZipCodeSeparatorsRe}
+          quantifier: MATCH_OPTIONAL
+      - capture:
+          output: postal-code-suffix
+          parts: [ {regex_reference: kZipSuffixValueRe} ]
+          quantifier: MATCH_OPTIONAL
+
 parsing_definitions:
   building-location:
     decomposition:
@@ -165,6 +189,10 @@ parsing_definitions:
             output: unit-name
             parts: [ {regex_reference: kUnitNameValueRe} ]
 
+  postal-code:
+    decomposition:
+      capture_reference: ParsePostalCodeExpression
+
 test_regex_definitions:
 # Tests for kBuildingValueRe
 - id: "kBuildingValueRe: plain number"
@@ -324,3 +352,15 @@ test_parsing_definitions:
     unit-name: "12"
     floor: "1"
     landmark: "foo"
+- id: "zip code with separator"
+  type: postal-code
+  input: "12345-678"
+  output:
+    postal-code-prefix: "12345"
+    postal-code-suffix: "678"
+- id: "zip code without separator"
+  type: postal-code
+  input: "12345678"
+  output:
+    postal-code-prefix: "12345"
+    postal-code-suffix: "678"
diff --git a/model/countries/CA/CA-formatting-rules.yaml b/model/countries/CA/CA-formatting-rules.yaml
index 1844c67..3ae98d6 100644
--- a/model/countries/CA/CA-formatting-rules.yaml
+++ b/model/countries/CA/CA-formatting-rules.yaml
@@ -11,6 +11,10 @@ formatting-rules:
   - skip: country  # redundant with country-name
   - skip: street-address  # redundant with street-address-alternative-1
 
+  postal-code:
+  - postal-code-prefix
+  - postal-code-suffix
+
 examples:
 - id: name
   comment: |
@@ -33,7 +37,8 @@ examples:
     address-line2: Apt. 306
     locality1: Ottawa
     admin-area1: ON
-    postal-code: M5H 2J9
+    postal-code-prefix: M5H
+    postal-code-suffix: 2J9
     country: CA
     country-name: Canada
 
diff --git a/model/countries/CA/CA-parsing-rules.yaml b/model/countries/CA/CA-parsing-rules.yaml
new file mode 100644
index 0000000..e86e34b
--- /dev/null
+++ b/model/countries/CA/CA-parsing-rules.yaml
@@ -0,0 +1,44 @@
+regex_definitions:
+  # Regular expression to match zip prefix
+  kZipPrefixValueRe:
+    regex_fragment: (?:[ABCEGHJ-NPRSTVXY]\d[ABCEGHJ-NPRSTV-Z])
+
+  # Regular expression to match zip suffix
+  kZipSuffixValueRe:
+    regex_fragment: (?:\d[ABCEGHJ-NPRSTV-Z]\d)
+
+capture_definitions:
+  ParsePostalCodeExpression:
+    capture:
+      output: postal-code
+      parts:
+      - capture:
+          output: postal-code-prefix
+          parts: [ {regex_reference: kZipPrefixValueRe} ]
+      - no_capture:
+          parts:
+          - separator: {regex_reference: kWhitespaceSeparator}
+          quantifier: MATCH_OPTIONAL
+      - capture:
+          output: postal-code-suffix
+          parts: [ {regex_reference: kZipSuffixValueRe} ]
+          quantifier: MATCH_OPTIONAL
+
+parsing_definitions:
+  postal-code:
+    decomposition:
+      capture_reference: ParsePostalCodeExpression
+
+test_parsing_definitions:
+- id: "zip code"
+  type: postal-code
+  input: "K1A 0B1"
+  output:
+    postal-code-prefix: "K1A"
+    postal-code-suffix: "0B1"
+- id: "zip code without separator"
+  type: postal-code
+  input: "K1A0B1"
+  output:
+    postal-code-prefix: "K1A"
+    postal-code-suffix: "0B1"
diff --git a/model/countries/NL/NL-formatting-rules.yaml b/model/countries/NL/NL-formatting-rules.yaml
index 2ab81f0..f1e0fbc 100644
--- a/model/countries/NL/NL-formatting-rules.yaml
+++ b/model/countries/NL/NL-formatting-rules.yaml
@@ -24,6 +24,10 @@ formatting-rules:
   - separator: "-"
   - unit
 
+  postal-code:
+  - postal-code-prefix
+  - postal-code-suffix
+
 examples:
 - id: name
   comment: |
@@ -46,7 +50,8 @@ examples:
     unit: A
     building-and-unit: 10-A
     locality1: Amsterdam
-    postal-code: 1234 AB
+    postal-code-prefix: 1234
+    postal-code-suffix: AB
     country: NL
     country-name: Netherlands
   output:
@@ -65,7 +70,8 @@ examples:
     building: 10
     building-and-unit: 10
     locality1: Amsterdam
-    postal-code: 1234 AB
+    postal-code-prefix: 1234
+    postal-code-suffix: AB
     country: NL
     country-name: Netherlands
   output:
diff --git a/model/countries/NL/NL-parsing-rules.yaml b/model/countries/NL/NL-parsing-rules.yaml
index ae4b211..161ec94 100644
--- a/model/countries/NL/NL-parsing-rules.yaml
+++ b/model/countries/NL/NL-parsing-rules.yaml
@@ -9,6 +9,14 @@ regex_definitions:
   kHouseNumberAndUnitSeparator:
     regex_fragment: (?:[-\s/,]*)
 
+  # Regular expression to match zip prefix
+  kZipPrefixValueRe:
+    regex_fragment: (?:[1-9]\d{3})
+
+  # Regular expression to match zip suffix
+  kZipSuffixValueRe:
+    regex_fragment: (?:[A-Z]{2})
+
 capture_definitions:
   ParseBuildingLocation:
     capture:
@@ -31,7 +39,23 @@ capture_definitions:
                   - capture:
                       output: unit-name
                       parts: [ {regex_reference: kUnitValueRe} ]
-                quantifier: MATCH_OPTIONAL       
+                quantifier: MATCH_OPTIONAL
+  
+  ParsePostalCodeExpression:
+    capture:
+      output: postal-code
+      parts:
+      - capture:
+          output: postal-code-prefix
+          parts: [ {regex_reference: kZipPrefixValueRe} ]
+      - no_capture:
+          parts:
+          - separator: {regex_reference: kWhitespaceSeparator}
+          quantifier: MATCH_OPTIONAL
+      - capture:
+          output: postal-code-suffix
+          parts: [ {regex_reference: kZipSuffixValueRe} ]
+          quantifier: MATCH_OPTIONAL
 
 parsing_definitions:
   building-location:
@@ -44,6 +68,10 @@ parsing_definitions:
         output: street-address-alternative-1
         parts:
         - capture_reference: ParseBuildingLocation
+  
+  postal-code:
+    decomposition:
+      capture_reference: ParsePostalCodeExpression
 
 test_parsing_definitions:
 - id: "Test 1"
@@ -241,3 +269,15 @@ test_parsing_definitions:
     building: "146"
     unit: "A-02"
     unit-name: "A-02"
+- id: "zip code with separator"
+  type: postal-code
+  input: "1234 AB"
+  output:
+    postal-code-prefix: "1234"
+    postal-code-suffix: "AB"
+- id: "zip code without separator"
+  type: postal-code
+  input: "1234AB"
+  output:
+    postal-code-prefix: "1234"
+    postal-code-suffix: "AB"
diff --git a/model/countries/PL/PL-parsing-rules.yaml b/model/countries/PL/PL-parsing-rules.yaml
index 4df0e56..b06a13d 100644
--- a/model/countries/PL/PL-parsing-rules.yaml
+++ b/model/countries/PL/PL-parsing-rules.yaml
@@ -24,6 +24,14 @@ regex_definitions:
   # Regular expression to match separator of house/building number and unit/apartment number.
   kHouseNumberAndUnitSeparator:
     regex_fragment: (?:^|[/\s]+)
+  
+  # Regular expression to match 2-digit zip prefix
+  kZipPrefixValueRe:
+    regex_fragment: (?:\d{2})
+
+  # Regular expression to match 3-digit zip suffix
+  kZipSuffixValueRe:
+    regex_fragment: (?:\d{3})
 
 capture_definitions:
   ParseBuildingLocation:
@@ -57,6 +65,22 @@ capture_definitions:
           output: unit-name
           parts: [{regex_reference: kUnitNameValueRe}]
       quantifier: MATCH_OPTIONAL
+  
+  ParsePostalCodeExpression:
+    capture:
+      output: postal-code
+      parts:
+      - capture:
+          output: postal-code-prefix
+          parts: [ {regex_reference: kZipPrefixValueRe} ]
+      - no_capture:
+          parts:
+          - separator: {regex_reference: kZipCodeSeparatorsRe}
+          quantifier: MATCH_OPTIONAL
+      - capture:
+          output: postal-code-suffix
+          parts: [ {regex_reference: kZipSuffixValueRe} ]
+          quantifier: MATCH_OPTIONAL
 
 
 parsing_definitions:
@@ -85,6 +109,10 @@ parsing_definitions:
   unit:
     decomposition:
       capture_reference: ParseUnitWithOptionalPrefix
+  
+  postal-code:
+    decomposition:
+      capture_reference: ParsePostalCodeExpression
 
 
 test_parsing_definitions:
@@ -222,4 +250,16 @@ test_parsing_definitions:
     building: "9A"
     unit: "m.10"
     unit-type: "m."
-    unit-name: "10"
\ No newline at end of file
+    unit-name: "10"
+- id: "zip code with separator"
+  type: postal-code
+  input: "00-843"
+  output:
+    postal-code-prefix: "00"
+    postal-code-suffix: "843"
+- id: "zip code without separator"
+  type: postal-code
+  input: "00843"
+  output:
+    postal-code-prefix: "00"
+    postal-code-suffix: "843"
diff --git a/model/countries/US/US-formatting-rules.yaml b/model/countries/US/US-formatting-rules.yaml
index 6ff9c31..828d8bf 100644
--- a/model/countries/US/US-formatting-rules.yaml
+++ b/model/countries/US/US-formatting-rules.yaml
@@ -35,7 +35,8 @@ examples:
     address-line2: Apt. 10, Club of Autofillers
     locality1: New York City
     admin-area1: NY
-    postal-code: 11367
+    postal-code-prefix: 11367
+    postal-code-suffix: 4100
     country: US
     country-name: USA
 
@@ -45,5 +46,5 @@ examples:
       text: |
         1234 Main St.
         Apt. 10, Club of Autofillers
-        New York City, NY 11367
-        USA
\ No newline at end of file
+        New York City, NY 11367-4100
+        USA
diff --git a/model/countries/US/US-parsing-rules.yaml b/model/countries/US/US-parsing-rules.yaml
new file mode 100644
index 0000000..cfac58a
--- /dev/null
+++ b/model/countries/US/US-parsing-rules.yaml
@@ -0,0 +1,56 @@
+regex_definitions:
+  # Regular expression to match 5-digit zip prefix
+  kZipPrefixValueRe:
+    regex_fragment: (?:\d{5})
+
+  # Regular expression to match 4-digit zip suffix
+  kZipSuffixValueRe:
+    regex_fragment: (?:\d{4})
+
+capture_definitions:
+  ParsePostalCodeExpression:
+    capture:
+      output: postal-code
+      parts:
+      - capture:
+          output: postal-code-prefix
+          parts: [ {regex_reference: kZipPrefixValueRe} ]
+      - no_capture:
+          parts:
+          - separator: {regex_reference: kZipCodeSeparatorsRe}
+          quantifier: MATCH_OPTIONAL
+      - capture:
+          output: postal-code-suffix
+          parts: [ {regex_reference: kZipSuffixValueRe} ]
+          quantifier: MATCH_OPTIONAL
+
+parsing_definitions:
+  postal-code:
+    decomposition:
+      capture_reference: ParsePostalCodeExpression
+
+test_parsing_definitions:
+- id: "zip code with suffix"
+  type: postal-code
+  input: "90210-5555"
+  output:
+    postal-code-prefix: "90210"
+    postal-code-suffix: "5555"
+- id: "zip code without suffix"
+  type: postal-code
+  input: "90210"
+  output:
+    postal-code-prefix: "90210"
+    postal-code-suffix: ""
+- id: "zip code with extra spaces"
+  type: postal-code
+  input: "90210 - 5555"
+  output:
+    postal-code-prefix: "90210"
+    postal-code-suffix: "5555" 
+- id: "zip code without separator"
+  type: postal-code
+  input: "902105555"
+  output:
+    postal-code-prefix: "90210"
+    postal-code-suffix: "5555" 
diff --git a/model/countries/global/global-descriptions.yaml b/model/countries/global/global-descriptions.yaml
index 22690ab..84437fe 100644
--- a/model/countries/global/global-descriptions.yaml
+++ b/model/countries/global/global-descriptions.yaml
@@ -13,6 +13,8 @@ short-descriptions:
   country: 2-letter country code
   country-name: Name of a country
   postal-code: Postal code
+  postal-code-prefix: Postal code prefix
+  postal-code-suffix: Postal code suffix
   admin-area1: Biggest type of admin area if a country has multiple levels
   admin-area2: 2nd biggest type of admin area if a country has multiple levels
   admin-area3: 3rd biggest type of admin area if a country has multiple levels
diff --git a/model/countries/global/global-formatting-rules.yaml b/model/countries/global/global-formatting-rules.yaml
index 6252aaf..560cdf7 100644
--- a/model/countries/global/global-formatting-rules.yaml
+++ b/model/countries/global/global-formatting-rules.yaml
@@ -109,3 +109,8 @@ formatting-rules:
   - tel-local-prefix
   - separator: ""
   - tel-local-suffix
+
+  postal-code:
+  - postal-code-prefix
+  - separator: "-"
+  - postal-code-suffix
diff --git a/model/countries/global/global-model.yaml b/model/countries/global/global-model.yaml
index 7e2858b..fff6adb 100644
--- a/model/countries/global/global-model.yaml
+++ b/model/countries/global/global-model.yaml
@@ -56,7 +56,9 @@ concepts:
     - admin-area2
     - admin-area3
     - admin-area4
-    - postal-code
+    - postal-code:
+      - postal-code-prefix
+      - postal-code-suffix
     - country
     - country-name
   # Company related
diff --git a/model/countries/global/global-parsing-rules.yaml b/model/countries/global/global-parsing-rules.yaml
index 01ab9f4..2291d03 100644
--- a/model/countries/global/global-parsing-rules.yaml
+++ b/model/countries/global/global-parsing-rules.yaml
@@ -178,6 +178,16 @@ regex_definitions:
   kMiddleNameInitialsCharacteristicsRe:
     regex_fragment: |-
       ^(?:[A-Z]\.?(?:(?:\s|-)?[A-Z]\.?)*)$
+  
+  # Regular expression pattern to match a sequence of alphanumeric chars.
+  kAlphaNumericSequenceRe:
+    regex_fragment:
+      (?:[^\W_]+)
+
+  # Regular expression pattern to match the separator between
+  # zip code prefix and suffix.
+  kZipCodeSeparatorsRe:
+    regex_fragment: (?:[\s-]+)
 
 capture_definitions:
   # Returns an expression to parse a CJK name that includes one separator.
@@ -420,6 +430,22 @@ capture_definitions:
     no_capture:
       parts: [ {regex_fragment: '\A\s*'} ]
 
+  # Returns an expression to parse `postal-code` into `postal-code-prefix`
+  # and `postal-code-suffix`.
+  ParsePostalCodeExpression:
+    capture:
+      output: postal-code
+      parts:
+      - capture:
+          output: postal-code-prefix
+          parts: [ {regex_reference: kAlphaNumericSequenceRe} ]
+      - separator: {regex_reference: kZipCodeSeparatorsRe}
+      - capture:
+          output: postal-code-suffix
+          parts: [ {regex_reference: kAlphaNumericSequenceRe} ]
+          quantifier: MATCH_OPTIONAL
+
+
 parsing_definitions:
   name:
     # If the name is a CJK name, try to match in the following order:
@@ -462,6 +488,10 @@ parsing_definitions:
           - decomposition: {capture_reference: ParseLastCommaFirstMiddleNameExpression}
           - decomposition: {capture_reference: ParseFirstMiddleLastNameExpression}
 
+  postal-code:
+    decomposition:
+      capture_reference: ParsePostalCodeExpression
+
 test_capture_definitions:
 - id: "Chinese name, Unihan"
   capture_name: ParseSeparatedCjkNameExpression
@@ -914,3 +944,33 @@ test_parsing_definitions:
     # family-name-first: ""
     # family-name-conjunction: ""
     # family-name-second: "孫"
+- id: US zip code with suffix.
+  type: postal-code
+  input: "90210-5555"
+  output:
+    postal-code-prefix: "90210"
+    postal-code-suffix: "5555"
+- id: US zip code without suffix.
+  type: postal-code
+  input: "90210"
+  output:
+    postal-code-prefix: "90210"
+    postal-code-suffix: ""
+- id: Canadian postal code.
+  type: postal-code
+  input: "M5V 2T6"
+  output:
+    postal-code-prefix: "M5V"
+    postal-code-suffix: "2T6"
+- id: UK postal code.
+  type: postal-code
+  input: "SW1A 0AA"
+  output:
+    postal-code-prefix: "SW1A"
+    postal-code-suffix: "0AA"
+- id: Polish postal code.
+  type: postal-code
+  input: "00-950"
+  output:
+    postal-code-prefix: "00"
+    postal-code-suffix: "950"
diff --git a/model/vendor b/model/vendor
new file mode 120000
index 0000000..3f91278
--- /dev/null
+++ b/model/vendor
@@ -0,0 +1 @@
+/Users/mmaryia/autofill-i18n-model/vendor/
\ No newline at end of file

From 2868bfef7532e882a2f20669da4c82c37ea8ee12 Mon Sep 17 00:00:00 2001
From: m-maryia <mmaryia@google.com>
Date: Fri, 11 Jul 2025 18:20:50 +0200
Subject: [PATCH 2/5] introduce zip prefix and suffix

---
 model/countries/BR/BR-parsing-rules.yaml      | 24 ++++++
 model/countries/CA/CA-formatting-rules.yaml   |  7 +-
 model/countries/CA/CA-parsing-rules.yaml      | 32 +++++++
 model/countries/NL/NL-formatting-rules.yaml   | 10 ++-
 model/countries/NL/NL-parsing-rules.yaml      | 26 +++++-
 model/countries/PL/PL-parsing-rules.yaml      | 26 +++++-
 model/countries/US/US-formatting-rules.yaml   |  7 +-
 model/countries/US/US-parsing-rules.yaml      | 39 +++++++++
 .../countries/global/global-descriptions.yaml |  2 +
 .../global/global-formatting-rules.yaml       |  5 ++
 model/countries/global/global-model.yaml      |  4 +-
 .../global/global-parsing-rules.yaml          | 86 +++++++++++++++++++
 12 files changed, 259 insertions(+), 9 deletions(-)
 create mode 100644 model/countries/CA/CA-parsing-rules.yaml
 create mode 100644 model/countries/US/US-parsing-rules.yaml

diff --git a/model/countries/BR/BR-parsing-rules.yaml b/model/countries/BR/BR-parsing-rules.yaml
index 9168462..fc18dc4 100644
--- a/model/countries/BR/BR-parsing-rules.yaml
+++ b/model/countries/BR/BR-parsing-rules.yaml
@@ -19,6 +19,14 @@ regex_definitions:
 
   kFloorLiteralRe:  # Regex for literal for a floor
     regex_fragment: andar
+  
+  # Regular expression to match 5-digit zip prefix
+  kZipPrefixValueRe:
+    regex_fragment: (?:\d{5})
+
+  # Regular expression to match 3-digit zip suffix
+  kZipSuffixValueRe:
+    regex_fragment: (?:\d{3})
 
 capture_definitions:
   ParseBuildingLocation:
@@ -165,6 +173,10 @@ parsing_definitions:
             output: unit-name
             parts: [ {regex_reference: kUnitNameValueRe} ]
 
+  postal-code:
+    decomposition:
+      capture_reference: ParsePostalCodeOptionalSeparatorExpression
+
 test_regex_definitions:
 # Tests for kBuildingValueRe
 - id: "kBuildingValueRe: plain number"
@@ -324,3 +336,15 @@ test_parsing_definitions:
     unit-name: "12"
     floor: "1"
     landmark: "foo"
+- id: "zip code with separator"
+  type: postal-code
+  input: "12345-678"
+  output:
+    postal-code-prefix: "12345"
+    postal-code-suffix: "678"
+- id: "zip code without separator"
+  type: postal-code
+  input: "12345678"
+  output:
+    postal-code-prefix: "12345"
+    postal-code-suffix: "678"
diff --git a/model/countries/CA/CA-formatting-rules.yaml b/model/countries/CA/CA-formatting-rules.yaml
index 1844c67..3ae98d6 100644
--- a/model/countries/CA/CA-formatting-rules.yaml
+++ b/model/countries/CA/CA-formatting-rules.yaml
@@ -11,6 +11,10 @@ formatting-rules:
   - skip: country  # redundant with country-name
   - skip: street-address  # redundant with street-address-alternative-1
 
+  postal-code:
+  - postal-code-prefix
+  - postal-code-suffix
+
 examples:
 - id: name
   comment: |
@@ -33,7 +37,8 @@ examples:
     address-line2: Apt. 306
     locality1: Ottawa
     admin-area1: ON
-    postal-code: M5H 2J9
+    postal-code-prefix: M5H
+    postal-code-suffix: 2J9
     country: CA
     country-name: Canada
 
diff --git a/model/countries/CA/CA-parsing-rules.yaml b/model/countries/CA/CA-parsing-rules.yaml
new file mode 100644
index 0000000..81e8bb2
--- /dev/null
+++ b/model/countries/CA/CA-parsing-rules.yaml
@@ -0,0 +1,32 @@
+regex_definitions:
+  # Regular expression to match the separator between
+  # zip code prefix and suffix.
+  kZipCodeSeparatorsRe:
+    regex_reference: kWhitespaceSeparator
+
+  # Regular expression to match zip prefix
+  kZipPrefixValueRe:
+    regex_fragment: (?:[ABCEGHJ-NPRSTVXY]\d[ABCEGHJ-NPRSTV-Z])
+
+  # Regular expression to match zip suffix
+  kZipSuffixValueRe:
+    regex_fragment: (?:\d[ABCEGHJ-NPRSTV-Z]\d)
+
+parsing_definitions:
+  postal-code:
+    decomposition:
+      capture_reference: ParsePostalCodeOptionalSeparatorExpression
+
+test_parsing_definitions:
+- id: "zip code"
+  type: postal-code
+  input: "K1A 0B1"
+  output:
+    postal-code-prefix: "K1A"
+    postal-code-suffix: "0B1"
+- id: "zip code without separator"
+  type: postal-code
+  input: "K1A0B1"
+  output:
+    postal-code-prefix: "K1A"
+    postal-code-suffix: "0B1"
diff --git a/model/countries/NL/NL-formatting-rules.yaml b/model/countries/NL/NL-formatting-rules.yaml
index 2ab81f0..f1e0fbc 100644
--- a/model/countries/NL/NL-formatting-rules.yaml
+++ b/model/countries/NL/NL-formatting-rules.yaml
@@ -24,6 +24,10 @@ formatting-rules:
   - separator: "-"
   - unit
 
+  postal-code:
+  - postal-code-prefix
+  - postal-code-suffix
+
 examples:
 - id: name
   comment: |
@@ -46,7 +50,8 @@ examples:
     unit: A
     building-and-unit: 10-A
     locality1: Amsterdam
-    postal-code: 1234 AB
+    postal-code-prefix: 1234
+    postal-code-suffix: AB
     country: NL
     country-name: Netherlands
   output:
@@ -65,7 +70,8 @@ examples:
     building: 10
     building-and-unit: 10
     locality1: Amsterdam
-    postal-code: 1234 AB
+    postal-code-prefix: 1234
+    postal-code-suffix: AB
     country: NL
     country-name: Netherlands
   output:
diff --git a/model/countries/NL/NL-parsing-rules.yaml b/model/countries/NL/NL-parsing-rules.yaml
index ae4b211..badaec4 100644
--- a/model/countries/NL/NL-parsing-rules.yaml
+++ b/model/countries/NL/NL-parsing-rules.yaml
@@ -9,6 +9,14 @@ regex_definitions:
   kHouseNumberAndUnitSeparator:
     regex_fragment: (?:[-\s/,]*)
 
+  # Regular expression to match zip prefix
+  kZipPrefixValueRe:
+    regex_fragment: (?:[1-9]\d{3})
+
+  # Regular expression to match zip suffix
+  kZipSuffixValueRe:
+    regex_fragment: (?:[A-Z]{2})
+
 capture_definitions:
   ParseBuildingLocation:
     capture:
@@ -31,7 +39,7 @@ capture_definitions:
                   - capture:
                       output: unit-name
                       parts: [ {regex_reference: kUnitValueRe} ]
-                quantifier: MATCH_OPTIONAL       
+                quantifier: MATCH_OPTIONAL
 
 parsing_definitions:
   building-location:
@@ -44,6 +52,10 @@ parsing_definitions:
         output: street-address-alternative-1
         parts:
         - capture_reference: ParseBuildingLocation
+  
+  postal-code:
+    decomposition:
+      capture_reference: ParsePostalCodeOptionalSeparatorExpression
 
 test_parsing_definitions:
 - id: "Test 1"
@@ -241,3 +253,15 @@ test_parsing_definitions:
     building: "146"
     unit: "A-02"
     unit-name: "A-02"
+- id: "zip code with separator"
+  type: postal-code
+  input: "1234 AB"
+  output:
+    postal-code-prefix: "1234"
+    postal-code-suffix: "AB"
+- id: "zip code without separator"
+  type: postal-code
+  input: "1234AB"
+  output:
+    postal-code-prefix: "1234"
+    postal-code-suffix: "AB"
diff --git a/model/countries/PL/PL-parsing-rules.yaml b/model/countries/PL/PL-parsing-rules.yaml
index 4df0e56..f9a0e8a 100644
--- a/model/countries/PL/PL-parsing-rules.yaml
+++ b/model/countries/PL/PL-parsing-rules.yaml
@@ -24,6 +24,14 @@ regex_definitions:
   # Regular expression to match separator of house/building number and unit/apartment number.
   kHouseNumberAndUnitSeparator:
     regex_fragment: (?:^|[/\s]+)
+  
+  # Regular expression to match 2-digit zip prefix
+  kZipPrefixValueRe:
+    regex_fragment: (?:\d{2})
+
+  # Regular expression to match 3-digit zip suffix
+  kZipSuffixValueRe:
+    regex_fragment: (?:\d{3})
 
 capture_definitions:
   ParseBuildingLocation:
@@ -85,6 +93,10 @@ parsing_definitions:
   unit:
     decomposition:
       capture_reference: ParseUnitWithOptionalPrefix
+  
+  postal-code:
+    decomposition:
+      capture_reference: ParsePostalCodeOptionalSeparatorExpression
 
 
 test_parsing_definitions:
@@ -222,4 +234,16 @@ test_parsing_definitions:
     building: "9A"
     unit: "m.10"
     unit-type: "m."
-    unit-name: "10"
\ No newline at end of file
+    unit-name: "10"
+- id: "zip code with separator"
+  type: postal-code
+  input: "00-843"
+  output:
+    postal-code-prefix: "00"
+    postal-code-suffix: "843"
+- id: "zip code without separator"
+  type: postal-code
+  input: "00843"
+  output:
+    postal-code-prefix: "00"
+    postal-code-suffix: "843"
diff --git a/model/countries/US/US-formatting-rules.yaml b/model/countries/US/US-formatting-rules.yaml
index 6ff9c31..828d8bf 100644
--- a/model/countries/US/US-formatting-rules.yaml
+++ b/model/countries/US/US-formatting-rules.yaml
@@ -35,7 +35,8 @@ examples:
     address-line2: Apt. 10, Club of Autofillers
     locality1: New York City
     admin-area1: NY
-    postal-code: 11367
+    postal-code-prefix: 11367
+    postal-code-suffix: 4100
     country: US
     country-name: USA
 
@@ -45,5 +46,5 @@ examples:
       text: |
         1234 Main St.
         Apt. 10, Club of Autofillers
-        New York City, NY 11367
-        USA
\ No newline at end of file
+        New York City, NY 11367-4100
+        USA
diff --git a/model/countries/US/US-parsing-rules.yaml b/model/countries/US/US-parsing-rules.yaml
new file mode 100644
index 0000000..317e1d1
--- /dev/null
+++ b/model/countries/US/US-parsing-rules.yaml
@@ -0,0 +1,39 @@
+regex_definitions:
+  # Regular expression to match 5-digit zip prefix
+  kZipPrefixValueRe:
+    regex_fragment: (?:\d{5})
+
+  # Regular expression to match 4-digit zip suffix
+  kZipSuffixValueRe:
+    regex_fragment: (?:\d{4})
+
+parsing_definitions:
+  postal-code:
+    decomposition:
+      capture_reference: ParsePostalCodeOptionalSeparatorExpression
+
+test_parsing_definitions:
+- id: "zip code with suffix"
+  type: postal-code
+  input: "90210-5555"
+  output:
+    postal-code-prefix: "90210"
+    postal-code-suffix: "5555"
+- id: "zip code without suffix"
+  type: postal-code
+  input: "90210"
+  output:
+    postal-code-prefix: "90210"
+    postal-code-suffix: ""
+- id: "zip code with extra spaces"
+  type: postal-code
+  input: "90210 - 5555"
+  output:
+    postal-code-prefix: "90210"
+    postal-code-suffix: "5555" 
+- id: "zip code without separator"
+  type: postal-code
+  input: "902105555"
+  output:
+    postal-code-prefix: "90210"
+    postal-code-suffix: "5555" 
diff --git a/model/countries/global/global-descriptions.yaml b/model/countries/global/global-descriptions.yaml
index 22690ab..84437fe 100644
--- a/model/countries/global/global-descriptions.yaml
+++ b/model/countries/global/global-descriptions.yaml
@@ -13,6 +13,8 @@ short-descriptions:
   country: 2-letter country code
   country-name: Name of a country
   postal-code: Postal code
+  postal-code-prefix: Postal code prefix
+  postal-code-suffix: Postal code suffix
   admin-area1: Biggest type of admin area if a country has multiple levels
   admin-area2: 2nd biggest type of admin area if a country has multiple levels
   admin-area3: 3rd biggest type of admin area if a country has multiple levels
diff --git a/model/countries/global/global-formatting-rules.yaml b/model/countries/global/global-formatting-rules.yaml
index 6252aaf..560cdf7 100644
--- a/model/countries/global/global-formatting-rules.yaml
+++ b/model/countries/global/global-formatting-rules.yaml
@@ -109,3 +109,8 @@ formatting-rules:
   - tel-local-prefix
   - separator: ""
   - tel-local-suffix
+
+  postal-code:
+  - postal-code-prefix
+  - separator: "-"
+  - postal-code-suffix
diff --git a/model/countries/global/global-model.yaml b/model/countries/global/global-model.yaml
index 7e2858b..fff6adb 100644
--- a/model/countries/global/global-model.yaml
+++ b/model/countries/global/global-model.yaml
@@ -56,7 +56,9 @@ concepts:
     - admin-area2
     - admin-area3
     - admin-area4
-    - postal-code
+    - postal-code:
+      - postal-code-prefix
+      - postal-code-suffix
     - country
     - country-name
   # Company related
diff --git a/model/countries/global/global-parsing-rules.yaml b/model/countries/global/global-parsing-rules.yaml
index 01ab9f4..1121071 100644
--- a/model/countries/global/global-parsing-rules.yaml
+++ b/model/countries/global/global-parsing-rules.yaml
@@ -178,6 +178,24 @@ regex_definitions:
   kMiddleNameInitialsCharacteristicsRe:
     regex_fragment: |-
       ^(?:[A-Z]\.?(?:(?:\s|-)?[A-Z]\.?)*)$
+  
+  # Regular expression pattern to match a sequence of alphanumeric chars.
+  kAlphaNumericSequenceRe:
+    regex_fragment:
+      (?:[^\W_]+)
+
+  # Regular expression pattern to match the separator between
+  # zip code prefix and suffix.
+  kZipCodeSeparatorsRe:
+    regex_fragment: (?:[\s-]+)
+
+  # Regular expression to match zip prefix
+  kZipPrefixValueRe:
+    regex_reference: kAlphaNumericSequenceRe
+
+  # Regular expression to match zip suffix
+  kZipSuffixValueRe:
+    regex_reference: kAlphaNumericSequenceRe 
 
 capture_definitions:
   # Returns an expression to parse a CJK name that includes one separator.
@@ -420,6 +438,40 @@ capture_definitions:
     no_capture:
       parts: [ {regex_fragment: '\A\s*'} ]
 
+  # Returns an expression to parse `postal-code` into `postal-code-prefix`
+  # and `postal-code-suffix`, separator is mandatory.
+  ParsePostalCodeMandatorySeparatorExpression:
+    capture:
+      output: postal-code
+      parts:
+      - capture:
+          output: postal-code-prefix
+          parts: [ {regex_reference: kZipPrefixValueRe} ]
+      - separator: {regex_reference: kZipCodeSeparatorsRe}
+      - capture:
+          output: postal-code-suffix
+          parts: [ {regex_reference: kZipSuffixValueRe} ]
+          quantifier: MATCH_OPTIONAL
+
+  # Returns an expression to parse `postal-code` into `postal-code-prefix`
+  # and `postal-code-suffix`, separator is optional.
+  ParsePostalCodeOptionalSeparatorExpression:
+    capture:
+      output: postal-code
+      parts:
+      - capture:
+          output: postal-code-prefix
+          parts: [ {regex_reference: kZipPrefixValueRe} ]
+      - no_capture:
+          parts:
+          - separator: {regex_reference: kZipCodeSeparatorsRe}
+          quantifier: MATCH_OPTIONAL
+      - capture:
+          output: postal-code-suffix
+          parts: [ {regex_reference: kZipSuffixValueRe} ]
+          quantifier: MATCH_OPTIONAL
+
+
 parsing_definitions:
   name:
     # If the name is a CJK name, try to match in the following order:
@@ -462,6 +514,10 @@ parsing_definitions:
           - decomposition: {capture_reference: ParseLastCommaFirstMiddleNameExpression}
           - decomposition: {capture_reference: ParseFirstMiddleLastNameExpression}
 
+  postal-code:
+    decomposition:
+      capture_reference: ParsePostalCodeMandatorySeparatorExpression
+
 test_capture_definitions:
 - id: "Chinese name, Unihan"
   capture_name: ParseSeparatedCjkNameExpression
@@ -914,3 +970,33 @@ test_parsing_definitions:
     # family-name-first: ""
     # family-name-conjunction: ""
     # family-name-second: "孫"
+- id: US zip code with suffix.
+  type: postal-code
+  input: "90210-5555"
+  output:
+    postal-code-prefix: "90210"
+    postal-code-suffix: "5555"
+- id: US zip code without suffix.
+  type: postal-code
+  input: "90210"
+  output:
+    postal-code-prefix: "90210"
+    postal-code-suffix: ""
+- id: Canadian postal code.
+  type: postal-code
+  input: "M5V 2T6"
+  output:
+    postal-code-prefix: "M5V"
+    postal-code-suffix: "2T6"
+- id: UK postal code.
+  type: postal-code
+  input: "SW1A 0AA"
+  output:
+    postal-code-prefix: "SW1A"
+    postal-code-suffix: "0AA"
+- id: Polish postal code.
+  type: postal-code
+  input: "00-950"
+  output:
+    postal-code-prefix: "00"
+    postal-code-suffix: "950"

From 714afb7d005112a2fd8f4241ebc101f9c6d703cd Mon Sep 17 00:00:00 2001
From: m-maryia <mmaryia@google.com>
Date: Mon, 14 Jul 2025 16:24:09 +0200
Subject: [PATCH 3/5] remove symlinks and redundant parsing definition

---
 model/build_internal_version.sh          |  1 -
 model/countries/PL/PL-parsing-rules.yaml | 16 ----------------
 model/vendor                             |  1 -
 3 files changed, 18 deletions(-)
 delete mode 120000 model/build_internal_version.sh
 delete mode 120000 model/vendor

diff --git a/model/build_internal_version.sh b/model/build_internal_version.sh
deleted file mode 120000
index ebee58d..0000000
--- a/model/build_internal_version.sh
+++ /dev/null
@@ -1 +0,0 @@
-/Users/mmaryia/autofill-i18n-model/build_internal_version.sh
\ No newline at end of file
diff --git a/model/countries/PL/PL-parsing-rules.yaml b/model/countries/PL/PL-parsing-rules.yaml
index d1dbab8..f9a0e8a 100644
--- a/model/countries/PL/PL-parsing-rules.yaml
+++ b/model/countries/PL/PL-parsing-rules.yaml
@@ -65,22 +65,6 @@ capture_definitions:
           output: unit-name
           parts: [{regex_reference: kUnitNameValueRe}]
       quantifier: MATCH_OPTIONAL
-  
-  ParsePostalCodeExpression:
-    capture:
-      output: postal-code
-      parts:
-      - capture:
-          output: postal-code-prefix
-          parts: [ {regex_reference: kZipPrefixValueRe} ]
-      - no_capture:
-          parts:
-          - separator: {regex_reference: kZipCodeSeparatorsRe}
-          quantifier: MATCH_OPTIONAL
-      - capture:
-          output: postal-code-suffix
-          parts: [ {regex_reference: kZipSuffixValueRe} ]
-          quantifier: MATCH_OPTIONAL
 
 
 parsing_definitions:
diff --git a/model/vendor b/model/vendor
deleted file mode 120000
index 3f91278..0000000
--- a/model/vendor
+++ /dev/null
@@ -1 +0,0 @@
-/Users/mmaryia/autofill-i18n-model/vendor/
\ No newline at end of file

From d4c7a2ddd84e9c75b13ddb67e63aad23cd6248e2 Mon Sep 17 00:00:00 2001
From: m-maryia <mmaryia@google.com>
Date: Mon, 14 Jul 2025 16:48:51 +0200
Subject: [PATCH 4/5] fix test ids

---
 model/countries/BR/BR-parsing-rules.yaml         |  4 ++--
 model/countries/CA/CA-parsing-rules.yaml         |  4 ++--
 model/countries/NL/NL-parsing-rules.yaml         |  4 ++--
 model/countries/PL/PL-parsing-rules.yaml         |  4 ++--
 model/countries/US/US-parsing-rules.yaml         |  8 ++++----
 model/countries/global/global-parsing-rules.yaml | 10 +++++-----
 6 files changed, 17 insertions(+), 17 deletions(-)

diff --git a/model/countries/BR/BR-parsing-rules.yaml b/model/countries/BR/BR-parsing-rules.yaml
index fc18dc4..14aeb1a 100644
--- a/model/countries/BR/BR-parsing-rules.yaml
+++ b/model/countries/BR/BR-parsing-rules.yaml
@@ -336,13 +336,13 @@ test_parsing_definitions:
     unit-name: "12"
     floor: "1"
     landmark: "foo"
-- id: "zip code with separator"
+- id: "Zip code with separator"
   type: postal-code
   input: "12345-678"
   output:
     postal-code-prefix: "12345"
     postal-code-suffix: "678"
-- id: "zip code without separator"
+- id: "Zip code without separator"
   type: postal-code
   input: "12345678"
   output:
diff --git a/model/countries/CA/CA-parsing-rules.yaml b/model/countries/CA/CA-parsing-rules.yaml
index 81e8bb2..4c7110c 100644
--- a/model/countries/CA/CA-parsing-rules.yaml
+++ b/model/countries/CA/CA-parsing-rules.yaml
@@ -18,13 +18,13 @@ parsing_definitions:
       capture_reference: ParsePostalCodeOptionalSeparatorExpression
 
 test_parsing_definitions:
-- id: "zip code"
+- id: "Zip code with separator"
   type: postal-code
   input: "K1A 0B1"
   output:
     postal-code-prefix: "K1A"
     postal-code-suffix: "0B1"
-- id: "zip code without separator"
+- id: "Zip code without separator"
   type: postal-code
   input: "K1A0B1"
   output:
diff --git a/model/countries/NL/NL-parsing-rules.yaml b/model/countries/NL/NL-parsing-rules.yaml
index badaec4..b7a689e 100644
--- a/model/countries/NL/NL-parsing-rules.yaml
+++ b/model/countries/NL/NL-parsing-rules.yaml
@@ -253,13 +253,13 @@ test_parsing_definitions:
     building: "146"
     unit: "A-02"
     unit-name: "A-02"
-- id: "zip code with separator"
+- id: "Test 19: zip code with separator"
   type: postal-code
   input: "1234 AB"
   output:
     postal-code-prefix: "1234"
     postal-code-suffix: "AB"
-- id: "zip code without separator"
+- id: "Test 20: zip code without separator"
   type: postal-code
   input: "1234AB"
   output:
diff --git a/model/countries/PL/PL-parsing-rules.yaml b/model/countries/PL/PL-parsing-rules.yaml
index f9a0e8a..7ede4eb 100644
--- a/model/countries/PL/PL-parsing-rules.yaml
+++ b/model/countries/PL/PL-parsing-rules.yaml
@@ -235,13 +235,13 @@ test_parsing_definitions:
     unit: "m.10"
     unit-type: "m."
     unit-name: "10"
-- id: "zip code with separator"
+- id: "Test 15: zip code with separator"
   type: postal-code
   input: "00-843"
   output:
     postal-code-prefix: "00"
     postal-code-suffix: "843"
-- id: "zip code without separator"
+- id: "Test 16: zip code without separator"
   type: postal-code
   input: "00843"
   output:
diff --git a/model/countries/US/US-parsing-rules.yaml b/model/countries/US/US-parsing-rules.yaml
index 317e1d1..cbd6893 100644
--- a/model/countries/US/US-parsing-rules.yaml
+++ b/model/countries/US/US-parsing-rules.yaml
@@ -13,25 +13,25 @@ parsing_definitions:
       capture_reference: ParsePostalCodeOptionalSeparatorExpression
 
 test_parsing_definitions:
-- id: "zip code with suffix"
+- id: "Zip code with suffix"
   type: postal-code
   input: "90210-5555"
   output:
     postal-code-prefix: "90210"
     postal-code-suffix: "5555"
-- id: "zip code without suffix"
+- id: "Zip code without suffix"
   type: postal-code
   input: "90210"
   output:
     postal-code-prefix: "90210"
     postal-code-suffix: ""
-- id: "zip code with extra spaces"
+- id: "Zip code with extra spaces"
   type: postal-code
   input: "90210 - 5555"
   output:
     postal-code-prefix: "90210"
     postal-code-suffix: "5555" 
-- id: "zip code without separator"
+- id: "Zip code without separator"
   type: postal-code
   input: "902105555"
   output:
diff --git a/model/countries/global/global-parsing-rules.yaml b/model/countries/global/global-parsing-rules.yaml
index 1121071..d28508d 100644
--- a/model/countries/global/global-parsing-rules.yaml
+++ b/model/countries/global/global-parsing-rules.yaml
@@ -970,31 +970,31 @@ test_parsing_definitions:
     # family-name-first: ""
     # family-name-conjunction: ""
     # family-name-second: "孫"
-- id: US zip code with suffix.
+- id: US zip code with suffix
   type: postal-code
   input: "90210-5555"
   output:
     postal-code-prefix: "90210"
     postal-code-suffix: "5555"
-- id: US zip code without suffix.
+- id: US zip code without suffix
   type: postal-code
   input: "90210"
   output:
     postal-code-prefix: "90210"
     postal-code-suffix: ""
-- id: Canadian postal code.
+- id: Canadian postal code
   type: postal-code
   input: "M5V 2T6"
   output:
     postal-code-prefix: "M5V"
     postal-code-suffix: "2T6"
-- id: UK postal code.
+- id: UK postal code
   type: postal-code
   input: "SW1A 0AA"
   output:
     postal-code-prefix: "SW1A"
     postal-code-suffix: "0AA"
-- id: Polish postal code.
+- id: Polish postal code
   type: postal-code
   input: "00-950"
   output:

From fdf34327f3ae80633f64785a31560571298bd154 Mon Sep 17 00:00:00 2001
From: m-maryia <mmaryia@google.com>
Date: Mon, 21 Jul 2025 18:05:29 +0200
Subject: [PATCH 5/5] move kAlphaNumericSequenceRe to the top of file

---
 model/countries/global/global-parsing-rules.yaml | 10 +++++-----
 1 file changed, 5 insertions(+), 5 deletions(-)

diff --git a/model/countries/global/global-parsing-rules.yaml b/model/countries/global/global-parsing-rules.yaml
index d28508d..34fd886 100644
--- a/model/countries/global/global-parsing-rules.yaml
+++ b/model/countries/global/global-parsing-rules.yaml
@@ -7,6 +7,11 @@ regex_definitions:
   kCommaOrWhitespaceSeparator:
     regex_fragment: (?:^|[,\s]+)
 
+  # Regular expression pattern to match a sequence of alphanumeric chars.
+  kAlphaNumericSequenceRe:
+    regex_fragment:
+      (?:[^\W_]+)
+
   # Regular expressions pattern of common two-character CJK last names.
   # Korean names are written in Hangul.
   # Chinese names are written in their traditional and simplified version.
@@ -178,11 +183,6 @@ regex_definitions:
   kMiddleNameInitialsCharacteristicsRe:
     regex_fragment: |-
       ^(?:[A-Z]\.?(?:(?:\s|-)?[A-Z]\.?)*)$
-  
-  # Regular expression pattern to match a sequence of alphanumeric chars.
-  kAlphaNumericSequenceRe:
-    regex_fragment:
-      (?:[^\W_]+)
 
   # Regular expression pattern to match the separator between
   # zip code prefix and suffix.