From e86e5059d06ed212e882a285823fecc3fdf04c5d Mon Sep 17 00:00:00 2001
From: Sebastian Wolfschmidt <sebastian.wolfschmidt@ext.aleph-alpha.com>
Date: Wed, 17 Apr 2024 14:54:42 +0200
Subject: [PATCH 1/2] add test for case where language cannot be detected

---
 .pre-commit-config.yaml            |  2 +-
 tests/core/test_detect_language.py | 13 +++++++++++++
 2 files changed, 14 insertions(+), 1 deletion(-)

diff --git a/.pre-commit-config.yaml b/.pre-commit-config.yaml
index 3af4d8720..fe59b8084 100644
--- a/.pre-commit-config.yaml
+++ b/.pre-commit-config.yaml
@@ -51,7 +51,7 @@ repos:
         args:
           [
             "-L",
-            "newyorker,te,responde,ist,als,oder,technik,sie,rouge,unter,juli,fiel,couldn,mke, vor",
+            "newyorker,te,responde,ist,als,oder,technik,sie,rouge,unter,juli,fiel,couldn,mke, vor,fille,ans",
           ]
         exclude: '^(poetry\.lock|trace-viewer/.*|tests/connectors/retrievers/test_document_index_retriever\.py|src/intelligence_layer/use_cases/qa/multiple_chunk_qa.py|src/intelligence_layer/use_cases/summarize/.*|tests/connectors/retrievers/test_document_index_retriever\.py|src/intelligence_layer/use_cases/classify/keyword_extract.py|tests/use_cases/summarize/test_single_chunk_few_shot_summarize.py|tests/use_cases/summarize/very_long_text.txt)$'
   - repo: https://github.com/akaihola/darglint2
diff --git a/tests/core/test_detect_language.py b/tests/core/test_detect_language.py
index cfe9b1980..2d16d73cf 100644
--- a/tests/core/test_detect_language.py
+++ b/tests/core/test_detect_language.py
@@ -17,3 +17,16 @@ def test_detect_language_returns_correct_language() -> None:
     output = task.run(input, tracer)
 
     assert output.best_fit == Language("en")
+
+
+def test_detect_language_returns_non_if_no_language_can_be_detected() -> None:
+    text = "Je m’appelle Jessica. Je suis une fille, je suis française et j’ai treize ans."  # codespell:ignore
+    task = DetectLanguage()
+    input = DetectLanguageInput(
+        text=text,
+        possible_languages=[Language(lang) for lang in ["en", "de"]],
+    )
+    tracer = NoOpTracer()
+    output = task.run(input, tracer)
+
+    assert output.best_fit is None

From f6ea3134211ceb7be2ca6701da0b295d7c507cd7 Mon Sep 17 00:00:00 2001
From: Sebastian Wolfschmidt <sebastian.wolfschmidt@ext.aleph-alpha.com>
Date: Wed, 17 Apr 2024 15:07:32 +0200
Subject: [PATCH 2/2] add parameterized for different languages

---
 tests/core/test_detect_language.py | 36 ++++++++++++++++++++++++++----
 1 file changed, 32 insertions(+), 4 deletions(-)

diff --git a/tests/core/test_detect_language.py b/tests/core/test_detect_language.py
index 2d16d73cf..7c46b94d2 100644
--- a/tests/core/test_detect_language.py
+++ b/tests/core/test_detect_language.py
@@ -1,3 +1,5 @@
+import pytest
+
 from intelligence_layer.core import (
     DetectLanguage,
     DetectLanguageInput,
@@ -6,17 +8,43 @@
 )
 
 
-def test_detect_language_returns_correct_language() -> None:
-    text = "Hello, my name is Niklas. I am working with Pit on this language detection piece."
+@pytest.mark.parametrize(
+    "text_input,expected_language",
+    [
+        (
+            "Hello, my name is Niklas. I am working with Pit on this language detection piece.",
+            Language("en"),
+        ),
+        (
+            "Hola, mi nombre es Niklas. Estoy trabajando con Pit en esta pieza de detección de idioma.",
+            Language("es"),
+        ),
+        (
+            "Ciao, mi chiamo Niklas. Sto lavorando con Pit su questo pezzo di rilevamento della lingua.",
+            Language("it"),
+        ),
+        (
+            "Hallo, mein Name ist Niklas. Ich arbeite mit Pit an diesem Stück zur Spracherkennung.",
+            Language("de"),
+        ),
+        (
+            "Bonjour, je m'appelle Niklas. Je travaille avec Pit sur cette pièce de détection de langue.",
+            Language("fr"),
+        ),
+    ],
+)
+def test_detect_language_returns_correct_language(
+    text_input: str, expected_language: Language
+) -> None:
     task = DetectLanguage()
     input = DetectLanguageInput(
-        text=text,
+        text=text_input,
         possible_languages=[Language(lang) for lang in ["en", "de", "fr", "it", "es"]],
     )
     tracer = NoOpTracer()
     output = task.run(input, tracer)
 
-    assert output.best_fit == Language("en")
+    assert output.best_fit == expected_language
 
 
 def test_detect_language_returns_non_if_no_language_can_be_detected() -> None: