feat(classifier): ✨ add classifier feature on log entries

add classifier on log entries, clean test functions and uniform fixtures in windows-1252 since all logs are encoded in windows-1252

tests/test_classifier.py

@@ -4,10 +4,9 @@ from pathlib import Path
 
 import pytest
 
-from logwatcher.classifier import N2_PATTERNS
+from logwatcher.classifier import N2_PATTERNS, classify_log_entries
 from logwatcher.config import FIXTURE_PATH
-
-ORIGINAL_LOGS_DIR = FIXTURE_PATH / "original_logs"
+from logwatcher.parser import parse_log_file
 
 
 def _pattern_name_to_filename(name: str) -> str:
@@ -21,20 +20,20 @@ def _read_fixture(path: Path) -> str:
         Read a file with fallback Windows-1252.
     """
     try:
-        contenu = path.read_text(encoding="utf-8")
+        content = path.read_text(encoding="utf-8")
     except UnicodeDecodeError:
-        return path.read_text(encoding="windows-1252")
-    if "\ufffd" not in contenu:
-        return contenu
+        content = path.read_text(encoding="windows-1252")
+    if "\ufffd" not in content:
+        return content
     return path.read_text(encoding="windows-1252")  # TODO: improve coverage score
 
 @pytest.mark.parametrize("name, pattern", N2_PATTERNS.items())
-def test_pattern_matches_fixture(name, pattern):
+def test_pattern_matches_fixture(valid_log_dir, name, pattern):
     """
         Assert each N2 pattern must match at least one line in its fixture.
     """
 
-    fixture = ORIGINAL_LOGS_DIR / _pattern_name_to_filename(name)
+    fixture = valid_log_dir / _pattern_name_to_filename(name)
     if not fixture.exists():
         msg = f"No fixture named '{fixture}'"
         warnings.warn(msg)
@@ -85,4 +84,55 @@ def test_no_orphan_fixtures():
     expected = {f"CR_{name.lower()}.txt" for name in N2_PATTERNS}
     actual = {f.name for f in FIXTURE_PATH.glob("CR_*.txt")}
     orphan = actual - expected
-    assert not orphan, f"Fixtures orphelines (sans pattern) : {sorted(orphan)}"
+    assert not orphan, f"Fixtures orphelines (sans pattern) : {sorted(orphan)}"
+
+
+def test_classify_log_entries_all_relevant(valid_log_dir: Path):
+    """
+        Must return an empty irrelevant log entry list
+    """
+    log_file = valid_log_dir / "only_relevant_logs.txt"
+    log_entries = parse_log_file(log_file)
+
+    # all entries are relevant
+    relevant, irrelevant = classify_log_entries(log_entries)
+
+    assert relevant and not irrelevant
+
+
+def test_classify_log_entries_none_relevant(valid_log_dir: Path):
+    """
+        Must return an empty relevant log entry list
+    """
+    log_file = valid_log_dir / "no_relevant_error.txt"
+    log_entries = parse_log_file(log_file)
+
+    # all entries are relevant
+    relevant, irrelevant = classify_log_entries(log_entries)
+
+    assert not relevant and irrelevant
+
+
+def test_classify_log_entries_mixed(valid_log_dir: Path):
+    """
+        Test classification on relevant and irrelevant log entry list
+    """
+    log_file = valid_log_dir / "mixed_logs.txt"
+    log_entries = parse_log_file(log_file)
+
+    relevant, irrelevant = classify_log_entries(log_entries)
+
+    assert len(relevant) and len(irrelevant)
+    assert len(relevant) + len(irrelevant) == len(log_entries)
+
+
+def test_classify_log_entries_empty_log_entries(invalid_log_dir: Path):
+    """
+        Test classification on an empty log entry list
+    """
+    empty_file = invalid_log_dir / "empty_file.txt"
+    log_entries = parse_log_file(empty_file)
+
+    relevant, irrelevant = classify_log_entries(log_entries)
+
+    assert not relevant and not irrelevant