fix: Update version to 0.9.4 in INSTALL.md and pyproject.toml

INSTALL.md

@@ -28,7 +28,7 @@ powershell -c "irm https://astral.sh/uv/install.sh | iex"
 uv tool install https://github.com/shadoll/moma/releases/latest/download/moma-latest.tar.gz
 
 # Specific version
-uv tool install https://github.com/shadoll/moma/releases/download/v0.9.2/moma-0.9.2-py3-none-any.whl
+uv tool install https://github.com/shadoll/moma/releases/download/v0.9.4/moma-0.9.4-py3-none-any.whl
 
 # From PyPI (when published)
 uv tool install moma
@@ -40,7 +40,7 @@ uv tool install moma
 uv tool install --force https://github.com/shadoll/moma/releases/latest/download/moma-latest.tar.gz
 
 # Upgrade to a newer specific version
-uv tool install --force https://github.com/shadoll/moma/releases/download/v0.9.2/moma-0.9.2-py3-none-any.whl
+uv tool install --force https://github.com/shadoll/moma/releases/download/v0.9.4/moma-0.9.4-py3-none-any.whl
 ```
 
 #### Usage
@@ -56,7 +56,7 @@ moma /path/to/directory # Scan specific directory
 pip install https://github.com/shadoll/moma/releases/latest/download/moma-latest.tar.gz
 
 # Specific version
-pip install https://github.com/shadoll/moma/releases/download/v0.9.2/moma-0.9.2-py3-none-any.whl
+pip install https://github.com/shadoll/moma/releases/download/v0.9.4/moma-0.9.4-py3-none-any.whl
 ```
 
 ### Method 3: Development Installation

pyproject.toml

@@ -1,6 +1,6 @@
 [project]
 name = "moma"
-version = "0.9.2"
+version = "0.9.4"
 description = "Terminal-based media file renamer and metadata viewer"
 readme = "README.md"
 requires-python = ">=3.12"

src/extractors/filename_extractor.py

@@ -64,19 +64,26 @@ class FilenameExtractor:
             if dot_match:
                 year_pos = dot_match.start()
             else:
-                # Last resort: any 4-digit number
+                # Try year between mixed separators (like .1967_ or _1967.)
-                any_match = re.search(r'\b(\d{4})\b', self.file_name)
+                sep_match = re.search(r'(?<=[.\-_\s])(\d{4})(?=[.\-_\s])', self.file_name)
-                if any_match:
+                if sep_match:
-                    year = int(any_match.group(1))
+                    year_val = int(sep_match.group(1))
-                    # Basic sanity check using constants
+                    if is_valid_year(year_val):
-                    if is_valid_year(year):
+                        year_pos = sep_match.start(1)
-                        year_pos = any_match.start()  # Cut before the year for plain years
+                else:
+                    # Last resort: any 4-digit number
+                    any_match = re.search(r'\b(\d{4})\b', self.file_name)
+                    if any_match:
+                        year = int(any_match.group(1))
+                        # Basic sanity check using constants
+                        if is_valid_year(year):
+                            year_pos = any_match.start()  # Cut before the year for plain years
 
         # Find source position
         source = self.extract_source()
         if source:
             for alias in SOURCE_DICT[source]:
-                match = re.search(r'\b' + re.escape(alias) + r'\b', self.file_name, re.IGNORECASE)
+                match = re.search(r'(?<![a-zA-Z])' + re.escape(alias) + r'(?![a-zA-Z])', self.file_name, re.IGNORECASE)
                 if match:
                     source_pos = match.start()
                     break
@@ -108,26 +115,23 @@ class FilenameExtractor:
         # Remove bracketed prefixes like [01.1], [1], etc.
         title = re.sub(r'^\s*\[[^\]]+\]\s*', '', title)
         
-        # Remove order number prefixes like 01., 1., 1.1 followed by space/underscore
+        # Remove order prefix (order followed by dot or space)
-        # Only remove if the number is multi-digit or has decimal (to avoid removing single digit titles)
-        match = re.match(r'^\s*(\d+(?:\.\d+)?)\.(?=\s|_)', title)
-        if match:
-            order = match.group(1)
-            if len(order) > 1 or '.' in order:
-                title = re.sub(r'^\s*(\d+(?:\.\d+)?)\.(?=\s|_)', '', title)
-        
-        # Remove order like 1.9 where 1 is order, 9 is title
         order = self.extract_order()
         if order:
-            match = re.match(r'^' + re.escape(order) + r'\.(.+)', title)
+            match = re.match(r'^' + re.escape(order) + r'[.\s]+(.+)', title)
             if match:
                 title = match.group(1)
         
         # Clean up any remaining leading separators
         title = title.lstrip('_ \t')
         
-        # Clean up title: remove leading/trailing brackets and dots
+        # Clean up title: remove leading/trailing brackets and orphaned dots
-        title = title.strip('[](). ')
+        title = title.strip('[]. ')
+        # Only strip unmatched leading/trailing parens
+        if title.endswith(')') and title.count('(') < title.count(')'):
+            title = title.rstrip(')')
+        if title.startswith('(') and title.count('(') > title.count(')'):
+            title = title.lstrip('(')
 
         # Replace dots with spaces if they appear to be word separators
         # Only replace dots that are surrounded by letters/digits (not at edges)
@@ -151,6 +155,13 @@ class FilenameExtractor:
         if dot_match:
             return dot_match.group(1)
 
+        # Try year between mixed separators (like .1967_ or _1967.)
+        sep_match = re.search(r'(?<=[.\-_\s])(\d{4})(?=[.\-_\s])', self.file_name)
+        if sep_match:
+            year = int(sep_match.group(1))
+            if is_valid_year(year):
+                return str(year)
+
         # Last resort: any 4-digit number (but this is less reliable)
         any_match = re.search(r'\b(\d{4})\b', self.file_name)
         if any_match:
@@ -213,6 +224,14 @@ class FilenameExtractor:
             # Fallback to height-based if not in constants
             return self._get_frame_class_from_height(height)
 
+        # Check for bare resolution numbers inside brackets (e.g., [720,ukr,eng])
+        bare_match = re.search(r'[\[,](\d{3,4})(?=[,\]])', normalized_name, re.IGNORECASE)
+        if bare_match:
+            height = int(bare_match.group(1))
+            frame_class = self._get_frame_class_from_height(height)
+            if frame_class:
+                return frame_class
+        
         # If no specific resolution found, check for non-standard quality indicators
         for indicator in NON_STANDARD_QUALITY_INDICATORS:
             if re.search(r'\b' + re.escape(indicator) + r'\b', self.file_name, re.IGNORECASE):
@@ -334,8 +353,17 @@ class FilenameExtractor:
         # Remove bracketed content first
         text_without_brackets = re.sub(r'\[([^\]]+)\]', '', self.file_name)
 
-        # Split on dots, spaces, and underscores
+        # Find start of metadata section (after title) to avoid title words being
-        parts = re.split(r'[.\s_]+', text_without_brackets)
+        # misdetected as language codes (e.g. "War" from "The.War.Wagon")
+        metadata_start = 0
+        year_m = (re.search(r'\(\d{4}\)', text_without_brackets) or
+                  re.search(r'\.\d{4}\.', text_without_brackets) or
+                  re.search(r'(?<=[.\-_\s])\d{4}(?=[.\-_\s])', text_without_brackets))
+        if year_m:
+            metadata_start = year_m.start()
+
+        # Split on dots, spaces, and underscores (only in the post-title portion)
+        parts = re.split(r'[.\s_]+', text_without_brackets[metadata_start:])
 
         for part in parts:
             part = part.strip()

src/test/datasets/filenames/filename_patterns.json

@@ -2,6 +2,24 @@
   "description": "Comprehensive test dataset for filename metadata extraction",
   "version": "2.0",
   "test_cases": [
+    {
+      "filename": "The.War.Wagon.1967_BDRip Ukr_Eng[Hurtom].mkv",
+      "expected": {
+        "order": null,
+        "title": "The War Wagon",
+        "year": "1967",
+        "source": "BDRip",
+        "frame_class": null,
+        "hdr": null,
+        "movie_db": null,
+        "special_info": null,
+        "audio_langs": "ukr,eng",
+        "extension": "mkv"
+      },
+      "testname": "edge-multi-lang-001",
+      "category": "edge_cases",
+      "description": "Multiple languages without brackets"
+    },
     {
       "filename": "Le Jaguar.(1996).[1080i,3ukr,fra].mkv",
       "expected": {
@@ -799,7 +817,7 @@
       "filename": "Movie.Title (2020) BDRip [1080p,ukr,eng].mkv",
       "expected": {
         "order": null,
-        "title": "Movie.Title",
+        "title": "Movie Title",
         "year": "2020",
         "source": "BDRip",
         "frame_class": "1080p",
@@ -810,7 +828,7 @@
         "extension": "mkv"
       },
       "category": "edge_cases",
-      "description": "Title with dots"
+      "description": "Title with dot separator (dot replaced with space by extractor)"
     },
     {
       "testname": "edge-no-brackets-001",

src/test/test_filename_extractor.py

@@ -15,6 +15,22 @@ def load_test_filenames():
     return []
 
 
+def load_test_cases():
+    """Load full test cases (testname, filename, expected) from dataset"""
+    dataset_file = Path(__file__).parent / "datasets" / "filenames" / "filename_patterns.json"
+    if dataset_file.exists():
+        with open(dataset_file, 'r', encoding='utf-8') as f:
+            data = json.load(f)
+            return [
+                pytest.param(
+                    case['filename'],
+                    case['expected'],
+                    id=case.get('testname', case['filename'])
+                )
+                for case in data['test_cases']
+            ]
+    return []
+
 @pytest.mark.parametrize("filename", load_test_filenames())
 def test_extract_title(filename):
     """Test title extraction from filename"""
@@ -129,3 +145,49 @@ def test_extract_audio_tracks(filename):
     for track in audio_tracks:
         assert isinstance(track, dict)
         assert 'language' in track
+
+
+# ---------------------------------------------------------------------------
+# Dataset-based value-checking tests (check against filename_patterns.json)
+# ---------------------------------------------------------------------------
+
+@pytest.mark.parametrize("filename,expected", load_test_cases())
+def test_expected_title(filename, expected):
+    """Test that extracted title matches the expected value from dataset."""
+    extractor = FilenameExtractor(Path(filename), use_cache=False)
+    assert extractor.extract_title() == expected['title']
+
+
+@pytest.mark.parametrize("filename,expected", load_test_cases())
+def test_expected_year(filename, expected):
+    """Test that extracted year matches the expected value from dataset."""
+    extractor = FilenameExtractor(Path(filename), use_cache=False)
+    assert extractor.extract_year() == expected['year']
+
+
+@pytest.mark.parametrize("filename,expected", load_test_cases())
+def test_expected_source(filename, expected):
+    """Test that extracted source matches the expected value from dataset."""
+    extractor = FilenameExtractor(Path(filename), use_cache=False)
+    assert extractor.extract_source() == expected['source']
+
+
+@pytest.mark.parametrize("filename,expected", load_test_cases())
+def test_expected_frame_class(filename, expected):
+    """Test that extracted frame_class matches the expected value from dataset."""
+    extractor = FilenameExtractor(Path(filename), use_cache=False)
+    assert extractor.extract_frame_class() == expected['frame_class']
+
+
+@pytest.mark.parametrize("filename,expected", load_test_cases())
+def test_expected_audio_langs(filename, expected):
+    """Test that extracted audio_langs matches the expected value from dataset."""
+    extractor = FilenameExtractor(Path(filename), use_cache=False)
+    assert extractor.extract_audio_langs() == expected['audio_langs']
+
+
+@pytest.mark.parametrize("filename,expected", load_test_cases())
+def test_expected_movie_db(filename, expected):
+    """Test that extracted movie_db matches the expected value from dataset."""
+    extractor = FilenameExtractor(Path(filename), use_cache=False)
+    assert extractor.extract_movie_db() == expected['movie_db']

uv.lock

@@ -208,7 +208,7 @@ wheels = [
 
 [[package]]
 name = "moma"
-version = "0.9.2"
+version = "0.9.3"
 source = { editable = "." }
 dependencies = [
     { name = "langcodes" },