From 7870d26e6951d07cda37ecdc2a7d0b6b98635248 Mon Sep 17 00:00:00 2001
From: Thanos Tsiamis <thanos.tsiamis@allwyn-lotterysolutions.com>
Date: Fri, 14 Nov 2025 20:34:49 +0200
Subject: [PATCH] feat: Add Git hooks for pre-commit and pre-push testing

---
 .githooks/pre-commit     | 18 ++++++++++++++++++
 .githooks/pre-push       | 17 +++++++++++++++++
 tests/test_algorithms.py | 17 +++++++++++++++++
 3 files changed, 52 insertions(+)
 create mode 100755 .githooks/pre-commit
 create mode 100755 .githooks/pre-push

diff --git a/.githooks/pre-commit b/.githooks/pre-commit
new file mode 100755
index 0000000..8412a6d
--- /dev/null
+++ b/.githooks/pre-commit
@@ -0,0 +1,18 @@
+#!/bin/sh
+echo "🔍 Running tests before commit..."
+
+# activate venv if needed
+if [ -f ".venv/bin/activate" ]; then
+    . .venv/bin/activate
+fi
+
+pytest tests
+STATUS=$?
+
+if [ $STATUS -ne 0 ]; then
+    echo "❌ Tests failed. Commit aborted."
+    exit 1
+fi
+
+echo "✔️ Tests passed. Proceeding with commit."
+exit 0
\ No newline at end of file
diff --git a/.githooks/pre-push b/.githooks/pre-push
new file mode 100755
index 0000000..7e3b3ab
--- /dev/null
+++ b/.githooks/pre-push
@@ -0,0 +1,17 @@
+#!/bin/sh
+echo "🚀 Running tests before push..."
+
+if [ -f ".venv/bin/activate" ]; then
+    . .venv/bin/activate
+fi
+
+pytest
+STATUS=$?
+
+if [ $STATUS -ne 0 ]; then
+    echo "❌ Tests failed. Push aborted."
+    exit 1
+fi
+
+echo "✔️ Tests passed. Proceeding with push."
+exit 0
\ No newline at end of file
diff --git a/tests/test_algorithms.py b/tests/test_algorithms.py
index 697cf4e..550e811 100644
--- a/tests/test_algorithms.py
+++ b/tests/test_algorithms.py
@@ -1,3 +1,4 @@
+import nltk
 import pytest
 
 from tests import df1, df2
@@ -23,7 +24,23 @@ def test_coma():
     # Assume the Schema and instance should provide different results
     assert matches_coma_matcher_schema != matches_coma_matcher_instances
 
+def _has_nltk_data():
+    # Check only what this test actually needs
+    needed = [
+        "tokenizers/punkt_tab/english/",  # tokenizer
+        "corpora/wordnet",                # wordnet
+        "corpora/omw-1.4",                # multilingual wordnet
+        "corpora/stopwords",              # stopwords
+    ]
+    for res in needed:
+        try:
+            nltk.data.find(res)
+        except LookupError:
+            return False
+    return True
 
+
+@pytest.mark.skipif(not _has_nltk_data(), reason="Required NLTK data not installed")
 def test_cupid():
     # Test the CUPID matcher
     cu_matcher = Cupid()