Added char validation

Author: TimoLassmann

python-kalign/__init__.py

@@ -154,6 +154,14 @@ def align(
                 f"Sequence at index {i} contains invalid control characters"
             )
 
+        # Check for digits and other problematic characters that cause platform-specific segfaults
+        invalid_chars = set(char for char in cleaned_seq if char.isdigit())
+        if invalid_chars:
+            raise ValueError(
+                f"Sequence at index {i} contains invalid characters: {sorted(invalid_chars)}. "
+                f"Sequences should only contain valid biological sequence characters."
+            )
+
     # Warn about very short sequences (like C CLI warnings)
     very_short_sequences = [
         i for i, seq in enumerate(sequences) if len(seq.strip()) < 3

tests/python/test_edge_cases.py

@@ -55,7 +55,15 @@ def test_mixed_case_sequences(self, mixed_case):
 
     def test_large_sequence_count(self):
         """Test with many sequences."""
-        many_seqs = [f"ATCG{i:02d}ATCG" for i in range(20)]
+        # Use valid DNA characters only - create variation with different bases
+        base_patterns = ["ATCG", "TACG", "GATC", "CGTA", "TGCA", "ACGT", "GCTA", "CTAG"]
+        many_seqs = []
+        for i in range(20):
+            pattern = base_patterns[i % len(base_patterns)]
+            # Add length variation with valid DNA bases
+            suffix = "A" * (i % 4)  # Add 0-3 A's for length variation
+            many_seqs.append(f"{pattern}{suffix}TG")
+
         aligned = kalign.align(many_seqs, seq_type="dna")
 
         assert len(aligned) == 20