test(analyzer): strengthen GB phone regression coverage

CHANGELOG.md

@@ -11,7 +11,7 @@ All notable changes to this project will be documented in this file.
 - Added `PhUmidRecognizer` for Philippine Unified Multi-Purpose ID (UMID/CRN) numbers in dashed and plain 12-digit formats; disabled by default (#2045) (Thanks @Surya-5555)
 
 #### Fixed
-- `PhoneRecognizer.DEFAULT_SUPPORTED_REGIONS` used `"UK"`, which is not a valid `phonenumbers` (libphonenumber) region code — region codes are ISO 3166-1 alpha-2, where the United Kingdom is `"GB"`. The `"UK"` entry was a no-op, so UK numbers in national/local format (e.g. `020 7946 0958`) were never detected by default; only international-format `+44 …` numbers matched, because they carry the country code and match under any region. Replaced `"UK"` with `"GB"`.
+- `PhoneRecognizer.DEFAULT_SUPPORTED_REGIONS` used `"UK"`, which is not a valid `phonenumbers` (libphonenumber) region code; the United Kingdom uses `"GB"`. The invalid region pass could not apply GB metadata to national-format UK numbers, so those numbers could be missed or recognized only through unrelated regional matchers. International-format `+44 …` numbers were unaffected because they carry their country code. Replaced `"UK"` with `"GB"`.
 - Language model recognizers (`BasicLangExtractRecognizer`, `AzureOpenAILangExtractRecognizer`) configured in a recognizer registry YAML now honour `config_path` (and other recognizer-specific kwargs). Previously these entries were validated by the strict `PredefinedRecognizerConfig` schema, which has no `config_path` field and does not allow extra keys, so `config_path` was silently dropped and the recognizer fell back to its bundled default model configuration. Added a `LangExtractRecognizerConfig` model (`extra="allow"`) and registered both recognizer class names in `CONFIG_MODEL_MAP`.
 - `BasicLangExtractRecognizer` now honours values under `langextract.model.provider.language_model_params` (including `timeout` and `num_ctx`). Previously these were silently dropped because `langextract.extract()` ignores its `language_model_params` argument when a pre-built `ModelConfig` is passed via `config=`, causing Ollama-backed recognizers to fall back to langextract's 120s default regardless of the configured timeout. The recognizer now merges `language_model_params` into `ModelConfig.provider_kwargs`, which is the path that reaches the provider constructor. Explicit entries under `provider.kwargs:` still take precedence. Also fixed a `TypeError` when `kwargs:` or `language_model_params:` is `null` in the YAML. (#1943, Thanks @lsternlicht)
 - Fixed `UsSsnRecognizer` over-blocking valid SSNs in the `987654320`-`987654329` range due to an 8-digit instead of full 9-digit prefix in the sample-SSN denylist check (#2074) (Thanks @AUTHENSOR)

presidio-analyzer/tests/test_phone_recognizer.py

@@ -1,6 +1,9 @@
+import phonenumbers
 import pytest
+from presidio_analyzer.predefined_recognizers.generic.phone_recognizer import (
+    PhoneRecognizer,
+)
 
-from presidio_analyzer.predefined_recognizers.generic.phone_recognizer import PhoneRecognizer
 from tests import assert_result, assert_result_with_textual_explanation
 
 
@@ -28,9 +31,6 @@ def recognizer():
         ("BR: +55 11 98456 5666", 1, ["PHONE_NUMBER"], ((4, 21), ), 0.4),
         ("My Japanese number is 090-1234-5678", 1, ["PHONE_NUMBER"],((22, 35), ), 0.4),
         ("My CN number is 13812345678", 1, ["PHONE_NUMBER"],((16, 27), ), 0.4),
-        # GB national-format number: only matched when region is the valid ISO code
-        # "GB" (not "UK"). Regression test for the DEFAULT_SUPPORTED_REGIONS fix.
-        ("My UK number is 020 7946 0958", 1, ["PHONE_NUMBER"], ((16, 29), ), 0.4),
         # fmt: on
     ],
 )
@@ -50,6 +50,38 @@ def test_when_all_phones_then_succeed(
         assert_result(res, entities[i], st_pos, fn_pos, score)
 
 
+def test_default_supported_regions_are_valid():
+    """Ensure every default phone region is supported by libphonenumber."""
+    invalid_regions = (
+        set(PhoneRecognizer.DEFAULT_SUPPORTED_REGIONS)
+        - phonenumbers.SUPPORTED_REGIONS
+    )
+    assert not invalid_regions
+
+
+def test_when_gb_national_phone_then_succeed_with_gb_explanation(
+    spacy_nlp_engine,
+):
+    """Ensure a GB national phone number is attributed to the GB region."""
+    text = "My UK number is 020 7946 0958"
+    nlp_artifacts = spacy_nlp_engine.process_text(text, "en")
+    results = PhoneRecognizer().analyze(
+        text,
+        ["PHONE_NUMBER"],
+        nlp_artifacts=nlp_artifacts,
+    )
+
+    assert len(results) == 1
+    assert_result_with_textual_explanation(
+        results[0],
+        "PHONE_NUMBER",
+        16,
+        29,
+        0.4,
+        "Recognized as GB region phone number, using PhoneRecognizer",
+    )
+
+
 @pytest.mark.parametrize(
     "text, expected_len, entities, expected_positions, score, leniency",
     [