9 Commits
11 changed files with 219 additions and 53 deletions
+6 -1
View File
@@ -647,7 +647,12 @@ except (LookupError, ValueError, AttributeError) as e:
1. **Read Before Modify**: Always read files before suggesting modifications
2. **Follow Existing Patterns**: Understand established architecture before changes
3. **Test Everything**: Run `uv run pytest` after all changes
3. **Run CI Checks After Every Code Change**: After any code modification, run the same checks as GitHub CI:
```bash
uv run pytest # all tests must pass
uv run mypy src/ --ignore-missing-imports # no type errors
```
Do NOT consider work complete until both commands succeed.
4. **Simplicity First**: Avoid over-engineering solutions
5. **Document Changes**: Update relevant documentation
+3 -3
View File
@@ -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/v0.9.7/moma-0.9.7-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/v0.9.7/moma-0.9.7-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/v0.9.7/moma-0.9.7-py3-none-any.whl
```
### Method 3: Development Installation
+1 -1
View File
@@ -1,6 +1,6 @@
[project]
name = "moma"
version = "0.9.2"
version = "0.9.7"
description = "Terminal-based media file renamer and metadata viewer"
readme = "README.md"
requires-python = ">=3.12"
+53 -15
View File
@@ -4,6 +4,7 @@ from textual.containers import Horizontal, Container, ScrollableContainer, Verti
from textual.widget import Widget
from textual.command import Provider, Hit
from rich.markup import escape
from rich.text import Text
from pathlib import Path
from functools import partial
from typing import TYPE_CHECKING, cast, Any
@@ -14,6 +15,7 @@ from .logging_config import LoggerConfig # Initialize logging singleton
from .constants import MEDIA_TYPES
from .views import OpenScreen, HelpScreen, RenameConfirmScreen, SettingsScreen, ConvertConfirmScreen, DeleteConfirmScreen
from .extractors.extractor import MediaExtractor
from .extractors.filename_extractor import FilenameExtractor
from .views import MediaPanelView, ProposedFilenameView
from .formatters.text_formatter import TextFormatter
from .formatters.catalog_formatter import CatalogFormatter
@@ -211,18 +213,56 @@ class MomaApp(App):
icons = {
'mkv': '🎥', # Video camera for MKV
'mk3d': '🕹️', # Clapper board for 3D
'mp4': '🎥', # Video camera
'mov': '🎥', # Video camera
'webm': '🎥', # Video camera
'avi': '💿', # Film frames for AVI
'wmv': '📀', # Video camera
'mp4': '🌐', # Web
'mov': '📽️', # Video camera
'webm': '🌐', # Web
'avi': '💿', # Silver compact disk
'wmv': '📀', # Gold compact disk
'm4v': '📹', # Video camera
'mpg': '📼', # Video camera
'mpeg': '📼', # Video camera
'mpg': '📼', # Video cassette
'mpeg': '📼', # Video cassette
}
return icons.get(ext, '📄') # Default to document icon
@staticmethod
def _frame_class_color(frame_class: str | None) -> str | None:
"""Return a Rich colour name for a frame class, or None if no highlight needed."""
if not frame_class:
return None
fc = frame_class.lower()
if fc in ("4320p",):
return "bright_green"
if fc in ("2160p",):
return "green"
if fc in ("1440p", "1080p", "1080i"):
return "yellow1"
if fc in ("720p",):
return "orange1"
# 576p, 480p, 480i, 360p and anything lower
return "red"
def _make_file_label(self, file_path: Path) -> Text:
"""Build a Rich Text label with resolution highlighted by quality tier."""
import re
icon = self._get_file_icon(file_path)
name = file_path.name
frame_class = FilenameExtractor(file_path, use_cache=False).extract_frame_class()
color = self._frame_class_color(frame_class)
label = Text(f"{icon} ")
if color and frame_class:
# Highlight the exact resolution token (e.g. "1080p") in the filename
m = re.search(re.escape(frame_class), name, re.IGNORECASE)
if m:
label.append(name[:m.start()])
label.append(name[m.start():m.end()], style=color)
label.append(name[m.end():])
return label
label.append(name)
return label
def build_tree(self, path: Path, node):
try:
for item in sorted(path.iterdir()):
@@ -231,16 +271,14 @@ class MomaApp(App):
if item.name.startswith(".") or item.name == "lost+found":
continue
# Add folder icon before directory name
label = f"󰉋 {escape(item.name)}"
subnode = node.add(label, data=item)
dir_label: str = f"󰉋 {escape(item.name)}"
subnode = node.add(dir_label, data=item)
self.build_tree(item, subnode)
elif item.is_file() and item.suffix.lower() in {
f".{ext}" for ext in MEDIA_TYPES
}:
# Add file type icon before filename
icon = self._get_file_icon(item)
label = f"{icon} {escape(item.name)}"
node.add(label, data=item)
# Add file type icon before filename with resolution colour
node.add(self._make_file_label(item), data=item)
except PermissionError:
pass
except PermissionError:
@@ -645,7 +683,7 @@ By Category:"""
logging.info(f"Found node for {old_path}, updating to {new_path.name}")
# Update label with icon
icon = self._get_file_icon(new_path)
node.label = f"{icon} {escape(new_path.name)}"
node.label = self._make_file_label(new_path) # type: ignore[assignment]
node.data = new_path
logging.info(f"After update: node.data = {node.data}, node.label = {node.label}")
# Ensure cursor stays on the renamed file
@@ -708,7 +746,7 @@ By Category:"""
# Get icon for the file
icon = self._get_file_icon(file_path)
label = f"{icon} {escape(file_path.name)}"
label = self._make_file_label(file_path)
# Add the new file node in alphabetically sorted position
new_node = None
+49 -22
View File
@@ -64,19 +64,26 @@ class FilenameExtractor:
if dot_match:
year_pos = dot_match.start()
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
# 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_val = int(sep_match.group(1))
if is_valid_year(year_val):
year_pos = sep_match.start(1)
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
# 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
# Remove order prefix (order followed by dot or space)
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
title = title.strip('[](). ')
# Clean up title: remove leading/trailing brackets and orphaned dots
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 +224,13 @@ 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:
bare_fc = self._get_frame_class_from_height(int(bare_match.group(1)))
if bare_fc:
return bare_fc
# 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 +352,17 @@ class FilenameExtractor:
# Remove bracketed content first
text_without_brackets = re.sub(r'\[([^\]]+)\]', '', self.file_name)
# Split on dots, spaces, and underscores
parts = re.split(r'[.\s_]+', text_without_brackets)
# Find start of metadata section (after title) to avoid title words being
# 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()
+13 -4
View File
@@ -195,7 +195,7 @@ class MediaInfoExtractor:
resolution = self.extract_resolution()
if not resolution:
return None
height, width = resolution
width, height = resolution
logger.debug(
f"[{self.file_path.name}] Frame class detection - Resolution: {width}x{height}"
@@ -249,11 +249,14 @@ class MediaInfoExtractor:
effective_height = height
# First, try to match width to typical widths
# Use a larger tolerance (10 pixels) to handle cinema/ultrawide aspect ratios
# Use proportional tolerance (2% of typical width, min 10px) to handle
# cinema/ultrawide aspect ratios where encoded width may differ slightly
# (e.g. 3820×1592 scope 4K → 2160p, not a non-standard 1592p)
width_matches = []
for frame_class, info in FRAME_CLASSES.items():
for tw in info["typical_widths"]:
if abs(width - tw) <= 10 and frame_class.endswith(scan_type):
width_tolerance = max(10, int(tw * 0.02))
if abs(width - tw) <= width_tolerance and frame_class.endswith(scan_type):
diff = abs(height - info["nominal_height"])
width_matches.append((frame_class, diff))
@@ -329,7 +332,10 @@ class MediaInfoExtractor:
return None
langs = []
for a in tracks:
lang_code = getattr(a, "language", "und") or "und"
lang_code = getattr(a, "language", None)
# Skip tracks with no language tag or 'und' (undetermined)
if not lang_code or lang_code.lower() in ("und", "undefined"):
continue
try:
# Try to get the 3-letter code
lang_obj = langcodes.Language.get(lang_code.lower())
@@ -340,6 +346,9 @@ class MediaInfoExtractor:
logger.debug(f"Invalid language code '{lang_code}': {e}")
langs.append(lang_code.lower()[:3])
if not langs:
return None # No meaningful language info — let Filename extractor try
lang_counts = Counter(langs)
audio_langs = [
f"{count}{lang}" if count > 1 else lang
@@ -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",
@@ -154,5 +154,12 @@
"interlaced": false,
"expected_frame_class": "1080p",
"testname": "test-mistakenly-high-height-2"
},
{
"testname": "test-2160p-scope-240",
"resolution": [3820, 1592],
"interlaced": false,
"expected_frame_class": "2160p",
"description": "4K cinema scope 2.40:1 - width slightly under 3840, height non-standard 1592"
}
]
+62
View File
@@ -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']
+1 -1
View File
@@ -39,7 +39,7 @@ def test_frame_class_detection(test_case):
extractor.video_tracks = [mock_track]
extractor._get_tracks.return_value = [mock_track] # satisfies @requires_tracks_type decorator
extractor._get_track.return_value = mock_track
extractor.extract_resolution.return_value = (height, width)
extractor.extract_resolution.return_value = (width, height)
extractor.extract_interlaced.return_value = interlaced
# Test the method
Generated
+1 -1
View File
@@ -208,7 +208,7 @@ wheels = [
[[package]]
name = "moma"
version = "0.9.2"
version = "0.9.7"
source = { editable = "." }
dependencies = [
{ name = "langcodes" },