Files
offlineu/offlineu_core.py
T
rmsitzandClaude Sonnet 5 a6c49d3762 Add Recently Added, bulk rename, time remaining, library stats, backup export
Five more usability features on top of the search/thumbnails batch:

- Recently Added: dashboard card of courses by folder mtime, separate from
  Recently Viewed (watched vs. just showed up on disk).
- Bulk find/replace rename across every course/folder name at once, with a
  mandatory preview step before anything touches disk. Refactored the
  single-item rename route to share the same validate/apply logic.
- Estimated time remaining on the loaded course's stats card, computed only
  from lessons that have actually reported a duration.
- Library-wide stats overview: total courses, lessons tracked, time
  watched, daily streak - read from each course's small progress file
  rather than re-scanning course contents.
- One-click backup/export zip of settings, hidden-paths, recently-viewed,
  and every course's progress/notes.

Also fixes a real performance bug found along the way: search was routing
through list_library_directory, which computes a full recursive file count
and thumbnail lookup for every course at every level regardless of match -
turning a search into an O(every file in the library) scan. Gave search its
own lightweight directory-only walk (iter_all_courses), now shared by
Recently Added and the stats overview too, so the expensive per-course work
only runs for courses that actually match.

Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
2026-08-22 11:36:05 -04:00

2196 lines
84 KiB
Python

#!/usr/bin/env python3
"""
OfflineU - Self-hosted Course Viewer & Tracker
Enhanced version with dynamic subdirectory navigation
"""
import os
import json
import mimetypes
import re
import sys
import argparse
import io
import zipfile
from pathlib import Path
from datetime import datetime, timedelta
from dataclasses import dataclass, asdict
from typing import List, Dict, Optional, Any, Tuple, Iterator
from flask import Flask, render_template, request, jsonify, send_file, redirect, url_for
app = Flask(__name__)
app.config['SECRET_KEY'] = 'your-secret-key-change-in-production'
# Supported file types
VIDEO_EXTENSIONS = {'.mp4', '.mkv', '.avi', '.mov', '.webm', '.m4v', '.flv', '.wmv'}
AUDIO_EXTENSIONS = {'.mp3', '.wav', '.m4a', '.aac', '.ogg', '.flac'}
SUBTITLE_EXTENSIONS = {'.srt', '.vtt', '.ass', '.sub', '.sbv'}
TEXT_EXTENSIONS = {'.txt', '.md', '.html', '.htm', '.pdf', '.docx', '.doc', '.rtf'}
QUIZ_INDICATORS = {'quiz', 'exam', 'test', 'assessment', 'exercise', 'assignment', 'homework'}
IMAGE_EXTENSIONS = {'.jpg', '.jpeg', '.png', '.webp'}
THUMBNAIL_BASENAMES = ('cover', 'folder', 'thumbnail', 'thumb', 'poster')
# Base directory the "Library" browser scans for courses, so users don't have
# to type a full filesystem path. Matches the ./courses volume mount in
# docker-compose.yml by default; override with --library-path or the
# COURSES_LIBRARY_PATH env var.
LIBRARY_PATH = os.environ.get('COURSES_LIBRARY_PATH', '/app/courses')
# App-wide (non per-course) persisted data, e.g. display settings. Matches
# the ./data volume mount in docker-compose.yml.
DATA_DIR = os.environ.get('OFFLINEU_DATA_DIR', '/app/data')
SETTINGS_FILE = os.path.join(DATA_DIR, 'settings.json')
DEFAULT_SETTINGS = {
'theme': 'dark', # 'dark' | 'light'
'accent_color': '#007acc', # any #rrggbb hex
'font_family': 'system', # 'system' | 'sans' | 'serif' | 'monospace'
'font_size': 'medium', # 'small' | 'medium' | 'large' | 'xlarge'
'layout_width': 'wide', # 'normal' | 'wide' | 'full'
'density': 'comfortable', # 'comfortable' | 'compact'
'card_style': 'flat', # 'flat' | 'elevated' | 'bordered'
'corner_radius': 'rounded', # 'sharp' | 'rounded' | 'pill'
'library_path': '', # '' = use COURSES_LIBRARY_PATH/--library-path default
'video_width': '', # '' = responsive full-width; else last dragged size, in px
'video_height': '',
'playback_speed': '1', # video/audio playback rate, as a string (see SETTINGS_CHOICES)
}
# Bounds for the persisted video player size, to reject garbage values
# without capping how big/small someone can reasonably drag it.
VIDEO_SIZE_BOUNDS = {'video_width': (200, 4000), 'video_height': (120, 3000)}
# Allowed values for every setting except accent_color (validated separately
# as a hex string). Anything outside these sets is rejected/ignored on save.
SETTINGS_CHOICES = {
'theme': {
'dark', 'light', 'dracula', 'tokyo_night', 'catppuccin_mocha', 'ayu_dark',
'github_dark', 'atom_one_dark', 'houston', 'night_owl', 'nord', 'matcha',
},
'font_family': {'system', 'sans', 'serif', 'monospace', 'monaspace'},
'font_size': {'small', 'medium', 'large', 'xlarge'},
'layout_width': {'normal', 'wide', 'full'},
'density': {'comfortable', 'compact'},
'card_style': {'flat', 'elevated', 'bordered'},
'corner_radius': {'sharp', 'rounded', 'pill'},
'playback_speed': {'0.75', '1', '1.25', '1.5', '1.75', '2'},
}
# Display names for the theme dropdown - only needed where the raw key
# (e.g. 'tokyo_night') isn't already a clean label via .capitalize().
THEME_DISPLAY_NAMES = {
'dark': 'Dark', 'light': 'Light', 'dracula': 'Dracula',
'tokyo_night': 'Tokyo Night', 'catppuccin_mocha': 'Catppuccin Mocha',
'ayu_dark': 'Ayu Dark', 'github_dark': 'GitHub Dark',
'atom_one_dark': 'Atom One Dark', 'houston': 'Houston',
'night_owl': 'Night Owl', 'nord': 'Nord', 'matcha': 'Matcha',
}
# Full color palette per theme. 'accent' here is only the *default* accent
# offered when a theme is first selected - the accent_color setting is what
# actually drives --accent afterward, so it stays independently editable.
# Sourced from each theme's official palette (Dracula, Tokyo Night,
# Catppuccin Mocha, Ayu, GitHub Dark, Atom One Dark, Houston, Nord all
# verified against upstream repos/specs). Matcha is sourced from
# lucafalasco/matcha's published VS Code theme JSON (editor/sideBar/
# activityBar backgrounds, foreground, panel.border, and the
# statusBar/button accent color).
#
# Dainty used to be here instead of Nord, but it turned out to be a
# Lab-space theme *generator* (HotWordland/dainty-vscode) with no fixed
# shipped palette - its output depends on whatever base theme you feed
# it, so there was no single "official Dainty" hex set to verify our
# approximation against. Swapped for Nord, which does have one.
THEME_PALETTES = {
'dark': {
'bg-primary': '#1a1a1a', 'bg-secondary': '#2d2d2d', 'bg-tertiary': '#3d3d3d',
'bg-tertiary-hover': '#404040', 'text-primary': '#e0e0e0', 'text-muted': '#999999',
'border-color': '#555555', 'accent': '#007acc', 'accent-hover': '#005a9e',
},
'light': {
'bg-primary': '#f2f2f2', 'bg-secondary': '#ffffff', 'bg-tertiary': '#eeeeee',
'bg-tertiary-hover': '#e2e2e2', 'text-primary': '#222222', 'text-muted': '#666666',
'border-color': '#cccccc', 'accent': '#007acc', 'accent-hover': '#005a9e',
},
'dracula': {
'bg-primary': '#282a36', 'bg-secondary': '#343746', 'bg-tertiary': '#44475a',
'bg-tertiary-hover': '#4d5066', 'text-primary': '#f8f8f2', 'text-muted': '#6272a4',
'border-color': '#44475a', 'accent': '#bd93f9', 'accent-hover': '#a672f0',
},
'tokyo_night': {
'bg-primary': '#1a1b26', 'bg-secondary': '#1f2335', 'bg-tertiary': '#283457',
'bg-tertiary-hover': '#364a73', 'text-primary': '#c0caf5', 'text-muted': '#565f89',
'border-color': '#292e42', 'accent': '#7aa2f7', 'accent-hover': '#6183bb',
},
'catppuccin_mocha': {
'bg-primary': '#1e1e2e', 'bg-secondary': '#313244', 'bg-tertiary': '#45475a',
'bg-tertiary-hover': '#585b70', 'text-primary': '#cdd6f4', 'text-muted': '#a6adc8',
'border-color': '#585b70', 'accent': '#cba6f7', 'accent-hover': '#b48ee0',
},
'ayu_dark': {
'bg-primary': '#0a0e14', 'bg-secondary': '#131721', 'bg-tertiary': '#1f2430',
'bg-tertiary-hover': '#272d3b', 'text-primary': '#b3b1ad', 'text-muted': '#626d7a',
'border-color': '#1f2430', 'accent': '#e6b450', 'accent-hover': '#ffb454',
},
'github_dark': {
'bg-primary': '#0d1117', 'bg-secondary': '#161b22', 'bg-tertiary': '#21262d',
'bg-tertiary-hover': '#30363d', 'text-primary': '#e6edf3', 'text-muted': '#8b949e',
'border-color': '#30363d', 'accent': '#2f81f7', 'accent-hover': '#1f6feb',
},
'atom_one_dark': {
'bg-primary': '#282c34', 'bg-secondary': '#2c313a', 'bg-tertiary': '#3b414d',
'bg-tertiary-hover': '#454b59', 'text-primary': '#abb2bf', 'text-muted': '#5c6370',
'border-color': '#3b414d', 'accent': '#61afef', 'accent-hover': '#528bff',
},
'houston': {
'bg-primary': '#17191e', 'bg-secondary': '#2b2d33', 'bg-tertiary': '#34363d',
'bg-tertiary-hover': '#3d3f47', 'text-primary': '#eef0f9', 'text-muted': '#79808f',
'border-color': '#2b2d33', 'accent': '#acafff', 'accent-hover': '#54b9ff',
},
'night_owl': {
'bg-primary': '#011627', 'bg-secondary': '#001122', 'bg-tertiary': '#0b2942',
'bg-tertiary-hover': '#123a5c', 'text-primary': '#d6deeb', 'text-muted': '#5f7e97',
'border-color': '#102a44', 'accent': '#82aaff', 'accent-hover': '#6690e0',
},
'nord': {
'bg-primary': '#2e3440', 'bg-secondary': '#3b4252', 'bg-tertiary': '#434c5e',
'bg-tertiary-hover': '#4c566a', 'text-primary': '#d8dee9', 'text-muted': '#4c566a',
'border-color': '#434c5e', 'accent': '#88c0d0', 'accent-hover': '#5e81ac',
},
'matcha': {
'bg-primary': '#1c2427', 'bg-secondary': '#273136', 'bg-tertiary': '#323e45',
'bg-tertiary-hover': '#3c4850', 'text-primary': '#d1ded3', 'text-muted': '#7c8885',
'border-color': '#707c4f', 'accent': '#a4b07e', 'accent-hover': '#8b966b',
},
}
# How each choice resolves to an actual CSS value. Keeping this server-side
# (rather than duplicating the mapping in JS) means the client just applies
# whatever /api/settings hands back.
FONT_FAMILY_CSS = {
'system': "'Segoe UI', Tahoma, Geneva, Verdana, sans-serif",
'sans': "Arial, Helvetica, sans-serif",
'serif': "Georgia, 'Times New Roman', serif",
'monospace': "'Courier New', Consolas, monospace",
# Monaspace is GitHub's monospace font superfamily, not a color theme -
# falls back to Consolas/monospace if the font isn't installed locally,
# same as any other web font declaration.
'monaspace': "'Monaspace Neon', 'Monaspace Argon', 'Cascadia Code', Consolas, monospace",
}
FONT_SIZE_CSS = {'small': '14px', 'medium': '16px', 'large': '18px', 'xlarge': '20px'}
LAYOUT_WIDTH_CSS = {'normal': '1000px', 'wide': '1600px', 'full': '100%'}
RADIUS_CSS = {'sharp': '2px', 'rounded': '8px', 'pill': '999px'}
HEX_COLOR_RE = re.compile(r'^#[0-9a-fA-F]{6}$')
def load_settings() -> Dict[str, Any]:
"""Load persisted display settings, filling in defaults for anything missing/invalid."""
settings = dict(DEFAULT_SETTINGS)
try:
if os.path.exists(SETTINGS_FILE):
with open(SETTINGS_FILE, 'r') as f:
saved = json.load(f)
for key, value in saved.items():
if key == 'accent_color' and isinstance(value, str) and HEX_COLOR_RE.match(value):
settings[key] = value
elif key == 'library_path' and isinstance(value, str):
# Lenient on load (directory may be transiently
# unavailable at startup) - only validated on save.
settings[key] = value
elif key in VIDEO_SIZE_BOUNDS:
if value == '' or (isinstance(value, int) and not isinstance(value, bool)):
settings[key] = value
elif key in SETTINGS_CHOICES and value in SETTINGS_CHOICES[key]:
settings[key] = value
except (json.JSONDecodeError, OSError) as e:
print(f"Could not load settings, using defaults: {e}")
return settings
def save_settings(new_settings: Dict[str, Any]) -> Dict[str, Any]:
"""Validate and persist display settings; unknown/invalid keys are ignored."""
current = load_settings()
for key, value in new_settings.items():
if key == 'accent_color' and isinstance(value, str) and HEX_COLOR_RE.match(value):
current[key] = value
elif key == 'theme' and value in SETTINGS_CHOICES['theme']:
current['theme'] = value
# Switching to a named preset also adopts its signature accent,
# unless this same request is also setting accent_color
# explicitly (in which case that wins).
if 'accent_color' not in new_settings:
current['accent_color'] = THEME_PALETTES[value]['accent']
elif key == 'library_path' and isinstance(value, str):
value = value.strip()
if value and not os.path.isdir(value):
raise ValueError(f"Directory not found: {value}")
current[key] = value
elif key in VIDEO_SIZE_BOUNDS:
if value in (None, ''):
current[key] = ''
continue
try:
num = int(value)
except (TypeError, ValueError):
raise ValueError(f"{key} must be a number")
lo, hi = VIDEO_SIZE_BOUNDS[key]
if not (lo <= num <= hi):
raise ValueError(f"{key} must be between {lo} and {hi}")
current[key] = num
elif key in SETTINGS_CHOICES and value in SETTINGS_CHOICES[key]:
current[key] = value
os.makedirs(DATA_DIR, exist_ok=True)
with open(SETTINGS_FILE, 'w') as f:
json.dump(current, f, indent=2)
return current
def settings_css_vars(settings: Dict[str, Any]) -> Dict[str, str]:
"""Resolve a settings dict into the actual CSS custom-property values."""
palette = THEME_PALETTES.get(settings['theme'], THEME_PALETTES['dark'])
return {
'--accent': settings['accent_color'],
'--accent-hover': palette['accent-hover'],
'--bg-primary': palette['bg-primary'],
'--bg-secondary': palette['bg-secondary'],
'--bg-tertiary': palette['bg-tertiary'],
'--bg-tertiary-hover': palette['bg-tertiary-hover'],
'--text-primary': palette['text-primary'],
'--text-muted': palette['text-muted'],
'--border-color': palette['border-color'],
'--font-family': FONT_FAMILY_CSS[settings['font_family']],
'--font-size-base': FONT_SIZE_CSS[settings['font_size']],
'--container-max-width': LAYOUT_WIDTH_CSS[settings['layout_width']],
'--radius': RADIUS_CSS[settings['corner_radius']],
}
@dataclass
class Lesson:
title: str
path: str
lesson_type: str # 'video', 'audio', 'text', 'quiz', 'mixed'
video_file: Optional[str] = None
audio_file: Optional[str] = None
subtitle_file: Optional[str] = None
text_files: List[str] = None
completed: bool = False
last_accessed: Optional[str] = None
progress_seconds: int = 0
duration_seconds: int = 0
order: int = 0
def __post_init__(self):
if self.text_files is None:
self.text_files = []
@dataclass
class DirectoryNode:
"""Represents a directory in the course structure"""
name: str
path: str
type: str # 'directory' or 'lesson'
children: Dict[str, 'DirectoryNode'] = None
lessons: List[Lesson] = None
completed: bool = False
last_accessed: Optional[str] = None
order: int = 0
has_content: bool = False # Whether this directory contains actual lesson content
def __post_init__(self):
if self.children is None:
self.children = {}
if self.lessons is None:
self.lessons = []
@dataclass
class Course:
name: str
path: str
root_node: DirectoryNode
progress_file: str
last_accessed_path: Optional[str] = None
completion_percentage: float = 0.0
def __post_init__(self):
if self.root_node is None:
self.root_node = DirectoryNode("", "", "directory")
class DynamicCourseParser:
"""Enhanced parser that builds a proper directory tree structure"""
@staticmethod
def scan_directory(course_path: str) -> Course:
"""Scan directory and build dynamic tree structure"""
course_path = Path(course_path)
if not course_path.exists() or not course_path.is_dir():
raise ValueError(f"Invalid course path: {course_path}")
course_name = course_path.name
print(f"Scanning course: {course_name}")
# Build the directory tree
root_node = DynamicCourseParser._build_directory_tree(course_path, course_path)
# Calculate completion statistics
stats = DynamicCourseParser._calculate_completion_stats(root_node)
progress_file = str(course_path / ".offlineu_progress.json")
return Course(
name=course_name,
path=str(course_path),
root_node=root_node,
progress_file=progress_file
)
@staticmethod
def _build_directory_tree(course_path: Path, current_path: Path, depth: int = 0) -> DirectoryNode:
"""Recursively build directory tree structure"""
if depth > 10: # Prevent infinite recursion
return DirectoryNode(current_path.name, str(current_path), "directory")
node_name = current_path.name if current_path != course_path else "Course Root"
node = DirectoryNode(
name=node_name,
path=str(current_path),
type="directory",
order=depth
)
try:
# Get all items in current directory
items = sorted(current_path.iterdir(), key=lambda x: (x.is_file(), x.name.lower()))
for item in items:
if item.name.startswith('.'):
continue
if item.is_dir():
# Recursively process subdirectory
child_node = DynamicCourseParser._build_directory_tree(course_path, item, depth + 1)
if child_node.has_content or child_node.children:
node.children[child_node.name] = child_node
node.has_content = True
elif item.is_file():
# Process file as potential lesson content
lesson = DynamicCourseParser._create_lesson_from_file(item, course_path)
if lesson:
# Add lesson directly to this node's lessons list
node.lessons.append(lesson)
node.has_content = True
except (PermissionError, OSError) as e:
print(f"Error accessing {current_path}: {e}")
return node
@staticmethod
def _create_lesson_from_file(file_path: Path, course_path: Path) -> Optional[Lesson]:
"""Create a lesson from a single file"""
ext = file_path.suffix.lower()
filename = file_path.name.lower()
# Skip non-content files
if ext in {'.log', '.tmp', '.bak', '.swp', '.DS_Store', '.Thumbs.db'}:
return None
# Determine lesson type and files
video_file = None
audio_file = None
subtitle_file = None
text_files = []
lesson_type = 'text'
# Create relative path for file serving - normalize to forward slashes
relative_path = str(file_path.relative_to(course_path)).replace('\\', '/')
if ext in VIDEO_EXTENSIONS:
video_file = relative_path
lesson_type = 'video'
elif ext in AUDIO_EXTENSIONS:
audio_file = relative_path
lesson_type = 'audio'
elif ext in SUBTITLE_EXTENSIONS:
subtitle_file = relative_path
return None # Don't create lessons for subtitle files alone
elif ext in TEXT_EXTENSIONS:
text_files.append(relative_path)
if any(indicator in filename for indicator in QUIZ_INDICATORS):
lesson_type = 'quiz'
else:
# Skip unsupported file types
return None
# Clean up lesson name for display
display_name = DynamicCourseParser._clean_lesson_name(file_path.stem)
return Lesson(
title=display_name,
path=str(file_path), # Store the actual file path, not just parent
lesson_type=lesson_type,
video_file=video_file,
audio_file=audio_file,
subtitle_file=subtitle_file,
text_files=text_files,
order=0
)
@staticmethod
def _clean_lesson_name(name: str) -> str:
"""Clean up lesson name for display"""
# Remove common patterns
name = re.sub(r'^\d+[\.\-_\s]*', '', name) # Remove leading numbers
name = re.sub(r'[-_]+', ' ', name) # Replace dashes/underscores with spaces
name = ' '.join(word.capitalize() for word in name.split() if word)
return name if name.strip() else "Untitled Lesson"
@staticmethod
def _calculate_completion_stats(node: DirectoryNode) -> Dict[str, Any]:
"""Calculate completion statistics for a directory node"""
total_lessons = 0
completed_lessons = 0
# Only lessons that have actually been played report a duration, so
# "remaining time" is only ever an estimate over what's known so far
# - there's no way to know the length of a lesson nobody's opened yet.
total_duration_seconds = 0
watched_seconds = 0
def count_lessons_recursive(n: DirectoryNode):
nonlocal total_lessons, completed_lessons, total_duration_seconds, watched_seconds
# Count lessons in this node
for lesson in n.lessons:
total_lessons += 1
if lesson.completed:
completed_lessons += 1
if lesson.duration_seconds:
total_duration_seconds += lesson.duration_seconds
watched_seconds += (
lesson.duration_seconds if lesson.completed
else min(lesson.progress_seconds, lesson.duration_seconds)
)
# Recursively count in children
for child in n.children.values():
count_lessons_recursive(child)
count_lessons_recursive(node)
completion_percentage = (completed_lessons / total_lessons * 100) if total_lessons > 0 else 0
return {
'total_lessons': total_lessons,
'completed_lessons': completed_lessons,
'completion_percentage': round(completion_percentage, 1),
'total_duration_seconds': total_duration_seconds,
'remaining_seconds': max(0, total_duration_seconds - watched_seconds)
}
def _has_direct_media(directory: Path) -> bool:
"""Check whether a directory contains media files directly (not recursively)"""
try:
for f in directory.iterdir():
if f.is_file() and f.suffix.lower() in VIDEO_EXTENSIONS | AUDIO_EXTENSIONS:
return True
except (PermissionError, OSError):
pass
return False
def find_course_thumbnail(course_path: str) -> Optional[str]:
"""
Look for a cover image directly inside a course folder (not recursive -
this only needs to catch the common 'cover.jpg next to the sections'
layout, not go hunting through every subfolder).
"""
try:
names = {f.name.lower(): f for f in Path(course_path).iterdir() if f.is_file()}
except (PermissionError, OSError):
return None
for base in THUMBNAIL_BASENAMES:
for ext in IMAGE_EXTENSIONS:
match = names.get(f'{base}{ext}')
if match:
return str(match)
return None
_SECTION_NAME_RE = re.compile(
r'^(section|module|chapter|part|unit|lesson)\b', re.IGNORECASE
)
def _looks_like_course(directory: Path) -> bool:
"""
Heuristic for 'this folder is a course, stop recursing into it':
it has media files directly, or at least two of its immediate
subfolders do (covers the common Section 1/, Section 2/... layout).
Requiring 2+ matching subfolders (rather than just 1) avoids mistaking
a publisher/grouping folder that holds a single course - e.g.
Pluralsight/Docker Deep Dive/lesson.mp4 - for the course itself.
The one exception: a directory with exactly one media-holding subfolder
whose name reads as a section label ("Section 1", "Module 2", ...)
rather than a course title. That's still a course with just one
section, not a publisher folder - Pluralsight/Docker Deep Dive doesn't
get named "Section 1", so this doesn't reopen the ambiguity above.
"""
if _has_direct_media(directory):
return True
try:
children = [
child for child in directory.iterdir()
if child.is_dir() and not child.name.startswith('.')
]
except (PermissionError, OSError):
return False
children_with_media = [child for child in children if _has_direct_media(child)]
if len(children_with_media) >= 2:
return True
if (
len(children_with_media) == 1
and _SECTION_NAME_RE.match(children_with_media[0].name.strip())
):
return True
return False
HIDDEN_PATHS_FILE = os.path.join(DATA_DIR, 'hidden_paths.json')
def get_hidden_paths() -> List[str]:
"""Load the set of course/directory paths curated out of the Library browser."""
try:
if os.path.exists(HIDDEN_PATHS_FILE):
with open(HIDDEN_PATHS_FILE, 'r') as f:
data = json.load(f)
if isinstance(data, list):
return data
except (json.JSONDecodeError, OSError) as e:
print(f"Could not load hidden paths: {e}")
return []
def set_path_hidden(path: str, hidden: bool) -> List[str]:
"""Add or remove a path from the hidden set; returns the updated list."""
paths = set(get_hidden_paths())
normalized = os.path.abspath(path)
if hidden:
paths.add(normalized)
else:
paths.discard(normalized)
result = sorted(paths)
os.makedirs(DATA_DIR, exist_ok=True)
with open(HIDDEN_PATHS_FILE, 'w') as f:
json.dump(result, f, indent=2)
return result
def _rebase_prefix(value: str, old_abs: str, new_abs: str) -> str:
"""If `value` equals or is nested under `old_abs`, rewrite that prefix to `new_abs`."""
if value == old_abs:
return new_abs
if value.startswith(old_abs + os.sep):
return new_abs + value[len(old_abs):]
return value
def rebase_library_path(old_abs: str, new_abs: str) -> None:
"""
After a directory in the library is renamed/moved on disk, rewrite any
stored absolute paths that pointed inside it (hidden_paths.json,
recent_views.json's course_path) so curation and Recently Viewed don't
silently go stale. Lesson-level progress needs no rebasing - it lives
inside the directory itself, keyed by paths relative to it, so it moves
with the rename automatically.
"""
hidden = get_hidden_paths()
rebased_hidden = [_rebase_prefix(p, old_abs, new_abs) for p in hidden]
if rebased_hidden != hidden:
os.makedirs(DATA_DIR, exist_ok=True)
with open(HIDDEN_PATHS_FILE, 'w') as f:
json.dump(sorted(set(rebased_hidden)), f, indent=2)
views = get_recent_views()
changed = False
for entry in views:
course_path = entry.get('course_path', '')
rebased = _rebase_prefix(course_path, old_abs, new_abs)
if rebased != course_path:
entry['course_path'] = rebased
changed = True
if changed:
try:
os.makedirs(DATA_DIR, exist_ok=True)
with open(RECENT_VIEWS_FILE, 'w') as f:
json.dump(views, f, indent=2)
except OSError as e:
print(f"Could not rebase recent views after rename: {e}")
def _contains_visible_course(directory: Path, hidden_set: set) -> bool:
"""
Whether `directory` leads to at least one course that isn't curated
out - directly, or via a hidden ancestor folder within this subtree.
Used by the normal (skip_hidden=True) Library browser so a folder
whose every course has been individually hidden doesn't still show up
as a drillable directory that dead-ends empty once opened. Hiding a
folder hides everything inside it, so a hidden folder short-circuits
the walk rather than counting anything beneath it as visible.
"""
if os.path.abspath(str(directory)) in hidden_set:
return False
if _looks_like_course(directory):
return True
try:
children = [
c for c in directory.iterdir()
if c.is_dir() and not c.name.startswith('.')
]
except (PermissionError, OSError):
return False
return any(_contains_visible_course(child, hidden_set) for child in children)
def list_library_directory(dir_path: str, skip_hidden: bool = True) -> Dict[str, Any]:
"""
List only the immediate children of dir_path for the lazy-loading
Library browser: subdirectories to drill into, and course folders to
load directly - without recursively scanning the whole tree. This is
what lets the UI start collapsed at the top level and expand on demand
instead of rendering hundreds of courses at once.
Directories that don't lead to any course anywhere inside them are
left out entirely, so drilling down never dead-ends on an empty folder.
Items the user has curated out (see get_hidden_paths) are excluded
when skip_hidden=True (normal browsing). When skip_hidden=False, they
are included but flagged with 'hidden': True instead - used by the
Settings-page curation UI, which needs to show hidden items so they
can be un-hidden.
"""
directory = Path(dir_path)
items: List[Dict[str, Any]] = []
hidden_set = set(get_hidden_paths())
if not directory.exists() or not directory.is_dir():
return {'items': items, 'errors': [f"Directory not found: {dir_path}"]}
try:
entries = sorted(
(p for p in directory.iterdir() if p.is_dir() and not p.name.startswith('.')),
key=lambda p: p.name.lower()
)
except (PermissionError, OSError) as e:
return {'items': items, 'errors': [f"Could not read {dir_path}: {e}"]}
for entry in entries:
entry_path_str = str(entry)
is_hidden = os.path.abspath(entry_path_str) in hidden_set
if skip_hidden and is_hidden:
continue
if _looks_like_course(entry):
media_count = len([
f for f in entry.rglob('*')
if f.is_file() and f.suffix.lower() in VIDEO_EXTENSIONS | AUDIO_EXTENSIONS
])
items.append({
'type': 'course',
'name': entry.name,
'path': entry_path_str,
'media_files': media_count,
'hidden': is_hidden,
'has_thumbnail': find_course_thumbnail(entry_path_str) is not None
})
else:
if skip_hidden:
# Only count courses that aren't themselves curated out -
# otherwise a folder whose courses are all hidden
# individually would still show up, empty, once opened.
has_course_inside = _contains_visible_course(entry, hidden_set)
else:
has_course_inside = any(
f.is_file() and f.suffix.lower() in VIDEO_EXTENSIONS | AUDIO_EXTENSIONS
for f in entry.rglob('*')
)
if has_course_inside:
items.append({
'type': 'directory',
'name': entry.name,
'path': entry_path_str,
'hidden': is_hidden
})
return {'items': items, 'errors': []}
def get_library_root() -> str:
"""
The effective root the Library browser scans: the persisted
'library_path' setting if it's set and exists, otherwise the
COURSES_LIBRARY_PATH/--library-path default.
"""
settings = load_settings()
override = (settings.get('library_path') or '').strip()
if override and os.path.isdir(override):
return override
return LIBRARY_PATH
def iter_all_courses(dir_path: str) -> Iterator[Path]:
"""
Recursively yield every course directory under dir_path, respecting
hidden paths exactly like normal browsing - without touching file
contents/counts. This is the cheap directory-only walk (iterdir +
_looks_like_course) that search, "Recently Added," and library stats
all need; the expensive per-course work (media_count via rglob,
thumbnail lookup, progress-file reads) is left to each caller to do
only for the courses it actually ends up using.
"""
hidden_set = set(get_hidden_paths())
def walk(directory: Path) -> Iterator[Path]:
try:
entries = sorted(
(p for p in directory.iterdir() if p.is_dir() and not p.name.startswith('.')),
key=lambda p: p.name.lower()
)
except (PermissionError, OSError):
return
for entry in entries:
if os.path.abspath(str(entry)) in hidden_set:
continue
if _looks_like_course(entry):
yield entry
else:
yield from walk(entry)
yield from walk(Path(dir_path))
def _course_summary(course_dir: Path) -> Dict[str, Any]:
"""Build the {type, name, path, media_files, hidden, has_thumbnail} shape
shared by the Library browser, search results, and Recently Added."""
media_count = len([
f for f in course_dir.rglob('*')
if f.is_file() and f.suffix.lower() in VIDEO_EXTENSIONS | AUDIO_EXTENSIONS
])
return {
'type': 'course',
'name': course_dir.name,
'path': str(course_dir),
'media_files': media_count,
'hidden': False,
'has_thumbnail': find_course_thumbnail(str(course_dir)) is not None
}
def search_library_courses(dir_path: str, query: str) -> List[Dict[str, Any]]:
"""
Search the library for courses whose name contains `query`
(case-insensitive). The expensive per-course lookups (media_count,
thumbnail) only run for courses whose name actually matches - see
iter_all_courses.
"""
query_lower = query.lower()
return [
_course_summary(course)
for course in iter_all_courses(dir_path)
if query_lower in course.name.lower()
]
def _humanize_days_ago(dt: datetime) -> str:
"""'Today' / 'Yesterday' / 'N days ago' / a plain date once it's old enough to matter less."""
days = (datetime.now().date() - dt.date()).days
if days <= 0:
return 'Today'
if days == 1:
return 'Yesterday'
if days < 14:
return f'{days} days ago'
return dt.strftime('%b %d')
def get_recently_added_courses(limit: int = 5) -> List[Dict[str, Any]]:
"""
Courses whose folder was most recently created/modified on disk, for
the dashboard's "Recently Added" card - separate from Recently Viewed
(which tracks what you've *watched*, not what showed up in the library).
"""
library_root = get_library_root()
dated = []
for course_dir in iter_all_courses(library_root):
try:
mtime = course_dir.stat().st_mtime
except OSError:
continue
dated.append((mtime, course_dir))
dated.sort(key=lambda pair: pair[0], reverse=True)
results = []
for mtime, course_dir in dated[:limit]:
item = _course_summary(course_dir)
item['added_display'] = _humanize_days_ago(datetime.fromtimestamp(mtime))
results.append(item)
return results
def get_library_stats() -> Dict[str, Any]:
"""
Library-wide overview for the dashboard: total courses, lessons with
any progress record, completed lessons, total time watched, and the
current daily streak. Reads each course's small progress JSON file
directly rather than re-scanning course directory contents, so this
stays cheap regardless of how many files are inside each course.
"""
total_courses = 0
lessons_tracked = 0
completed_lessons = 0
watched_seconds = 0
active_dates = set()
for course_dir in iter_all_courses(get_library_root()):
total_courses += 1
try:
with open(course_dir / '.offlineu_progress.json', 'r') as f:
progress = json.load(f)
except (FileNotFoundError, json.JSONDecodeError, OSError):
continue
for key, entry in progress.items():
if key == 'last_accessed_path' or not isinstance(entry, dict):
continue
lessons_tracked += 1
if entry.get('completed'):
completed_lessons += 1
watched_seconds += entry.get('duration_seconds') or 0
else:
watched_seconds += entry.get('progress_seconds') or 0
last_accessed = entry.get('last_accessed')
if last_accessed:
try:
active_dates.add(datetime.fromisoformat(last_accessed).date())
except ValueError:
pass
streak_days = 0
day = datetime.now().date()
while day in active_dates:
streak_days += 1
day -= timedelta(days=1)
return {
'total_courses': total_courses,
'lessons_tracked': lessons_tracked,
'completed_lessons': completed_lessons,
'watched_display': format_duration(watched_seconds),
'streak_days': streak_days
}
RECENT_VIEWS_FILE = os.path.join(DATA_DIR, 'recent_views.json')
MAX_RECENT_VIEWS = 20
def record_recent_view(course_name: str, course_path: str, lesson_path: str, lesson_title: str) -> None:
"""
Record a lesson view for the cross-course 'Recently Viewed' list on the
dashboard, so jumping back to where you left off doesn't require
re-browsing the library - even for a different course than whichever
one happens to be loaded right now.
"""
entries = get_recent_views()
# Drop any existing entry for this exact lesson so it moves to the
# front instead of appearing twice.
entries = [
e for e in entries
if not (e.get('course_path') == course_path and e.get('lesson_path') == lesson_path)
]
entries.insert(0, {
'course_name': course_name,
'course_path': course_path,
'lesson_path': lesson_path,
'lesson_title': lesson_title,
'viewed_at': datetime.now().isoformat()
})
entries = entries[:MAX_RECENT_VIEWS]
try:
os.makedirs(DATA_DIR, exist_ok=True)
with open(RECENT_VIEWS_FILE, 'w') as f:
json.dump(entries, f, indent=2)
except OSError as e:
print(f"Could not save recent views: {e}")
def get_recent_views() -> List[Dict[str, Any]]:
"""Load the persisted 'Recently Viewed' list, newest first."""
try:
if os.path.exists(RECENT_VIEWS_FILE):
with open(RECENT_VIEWS_FILE, 'r') as f:
return json.load(f)
except (json.JSONDecodeError, OSError) as e:
print(f"Could not load recent views: {e}")
return []
def _read_lesson_progress_entry(course_path: str, lesson_path: str) -> Dict[str, Any]:
"""
Read a single lesson's progress entry directly from its course's own
progress file, without needing that course to be the currently loaded
one. Used to show watch progress for 'Recently Viewed' entries that may
belong to a different course than whatever's active right now.
"""
if not course_path or not lesson_path:
return {}
progress_file = os.path.join(course_path, '.offlineu_progress.json')
try:
with open(progress_file, 'r') as f:
progress = json.load(f)
return progress.get(lesson_path, {})
except (FileNotFoundError, json.JSONDecodeError, OSError):
return {}
def get_recent_views_for_display() -> List[Dict[str, Any]]:
"""Recent views with a human-friendly timestamp and watch progress added."""
entries = get_recent_views()
for entry in entries:
try:
dt = datetime.fromisoformat(entry['viewed_at'])
entry['viewed_display'] = dt.strftime('%b %d, %I:%M %p').replace(' 0', ' ')
except (KeyError, ValueError):
entry['viewed_display'] = ''
lesson_progress = _read_lesson_progress_entry(entry.get('course_path', ''), entry.get('lesson_path', ''))
completed = lesson_progress.get('completed', False)
progress_seconds = lesson_progress.get('progress_seconds', 0)
duration_seconds = lesson_progress.get('duration_seconds', 0)
entry['completed'] = completed
if completed:
entry['percent_watched'] = 100
elif duration_seconds:
entry['percent_watched'] = max(0, min(100, round(100 * progress_seconds / duration_seconds)))
else:
entry['percent_watched'] = 0
entry['has_thumbnail'] = find_course_thumbnail(entry.get('course_path', '')) is not None
return entries
class ProgressTracker:
"""Handles progress tracking and persistence"""
@staticmethod
def load_progress(course: Course) -> Dict[str, Any]:
"""Load progress from JSON file"""
try:
with open(course.progress_file, 'r') as f:
return json.load(f)
except (FileNotFoundError, json.JSONDecodeError):
return {}
@staticmethod
def save_progress(course: Course, progress_data: Dict[str, Any]):
"""Save progress to JSON file"""
try:
with open(course.progress_file, 'w') as f:
json.dump(progress_data, f, indent=2)
except Exception as e:
print(f"Error saving progress: {e}")
@staticmethod
def update_lesson_progress(course: Course, lesson_path: str, completed: bool = False,
progress_seconds: int = 0, duration_seconds: Optional[int] = None):
"""Update progress for specific lesson by path"""
progress = ProgressTracker.load_progress(course)
existing = progress.get(lesson_path, {})
entry = {
'completed': completed,
'progress_seconds': progress_seconds,
'last_accessed': datetime.now().isoformat()
}
# Preserve a previously-known duration if this particular save
# didn't report one, rather than clobbering it back to unknown.
if duration_seconds:
entry['duration_seconds'] = duration_seconds
elif existing.get('duration_seconds'):
entry['duration_seconds'] = existing['duration_seconds']
# Same for a note - this call has no opinion on it, so don't let a
# routine playback-progress save wipe one out.
if existing.get('note'):
entry['note'] = existing['note']
progress[lesson_path] = entry
# Update last accessed path
progress['last_accessed_path'] = lesson_path
ProgressTracker.save_progress(course, progress)
@staticmethod
def update_lesson_note(course: Course, lesson_path: str, note: str):
"""Save (or clear) a lesson's note, without touching its playback progress."""
progress = ProgressTracker.load_progress(course)
entry = progress.setdefault(lesson_path, {})
if note:
entry['note'] = note
else:
entry.pop('note', None)
ProgressTracker.save_progress(course, progress)
@staticmethod
def mark_all_completed(course: Course):
"""Mark every lesson in the course as completed, in one save."""
progress = ProgressTracker.load_progress(course)
def mark_node(node: DirectoryNode):
for lesson in node.lessons:
lesson_path = os.path.relpath(lesson.path, course.path).replace('\\', '/')
if lesson_path.startswith('/'):
lesson_path = lesson_path[1:]
entry = progress.setdefault(lesson_path, {})
entry['completed'] = True
entry['last_accessed'] = datetime.now().isoformat()
entry.setdefault('progress_seconds', entry.get('duration_seconds', 0))
for child in node.children.values():
mark_node(child)
mark_node(course.root_node)
ProgressTracker.save_progress(course, progress)
@staticmethod
def apply_progress_to_tree(course: Course):
"""Apply saved progress to the course tree"""
progress = ProgressTracker.load_progress(course)
def apply_to_node(node: DirectoryNode):
# Apply progress to lessons in this node
for lesson in node.lessons:
lesson_path = os.path.relpath(lesson.path, course.path)
lesson_path = lesson_path.replace('\\', '/')
if lesson_path.startswith('/'):
lesson_path = lesson_path[1:]
# Check both the base path and path with title
lesson_path_with_title = f"{lesson_path}/{lesson.title.replace(' ', '_')}"
if lesson_path in progress:
lesson.completed = progress[lesson_path].get('completed', False)
lesson.last_accessed = progress[lesson_path].get('last_accessed')
lesson.progress_seconds = progress[lesson_path].get('progress_seconds', 0)
lesson.duration_seconds = progress[lesson_path].get('duration_seconds', 0)
elif lesson_path_with_title in progress:
lesson.completed = progress[lesson_path_with_title].get('completed', False)
lesson.last_accessed = progress[lesson_path_with_title].get('last_accessed')
lesson.progress_seconds = progress[lesson_path_with_title].get('progress_seconds', 0)
lesson.duration_seconds = progress[lesson_path_with_title].get('duration_seconds', 0)
# Recursively apply to children
for child in node.children.values():
apply_to_node(child)
apply_to_node(course.root_node)
course.last_accessed_path = progress.get('last_accessed_path')
@staticmethod
def get_completion_stats(course: Course) -> Dict[str, Any]:
"""Calculate completion statistics"""
return DynamicCourseParser._calculate_completion_stats(course.root_node)
# Global course storage
current_course = None
def _split_continue_watching(all_views: List[Dict[str, Any]]) -> Tuple[List[Dict[str, Any]], List[Dict[str, Any]]]:
"""
Split the recent-views list (already carrying percent_watched/completed,
see get_recent_views_for_display) into 'Continue Watching' - genuinely
in-progress lessons - and the remaining 'Recently Viewed' entries, each
capped at 5 for the dashboard. Reuses the same underlying history rather
than a separate library-wide progress index (see plan notes).
"""
continue_watching = [v for v in all_views if 0 < v['percent_watched'] < 100][:5]
continue_ids = {(v.get('course_path'), v.get('lesson_path')) for v in continue_watching}
recent_views = [
v for v in all_views
if (v.get('course_path'), v.get('lesson_path')) not in continue_ids
][:5]
return continue_watching, recent_views
def format_duration(seconds: int) -> str:
"""Render a duration in seconds as e.g. '3h 24m' or '45m' (under an hour)."""
total_minutes = max(0, int(seconds)) // 60
hours, minutes = divmod(total_minutes, 60)
return f"{hours}h {minutes}m" if hours else f"{minutes}m"
@app.route('/')
def index():
"""Main dashboard"""
global current_course
continue_watching, recent_views = _split_continue_watching(get_recent_views_for_display())
if current_course is None:
# Show dashboard with course selection option
return render_template('course_dashboard.html',
course=None,
stats={'total_lessons': 0, 'completed_lessons': 0, 'completion_percentage': 0},
continue_watching=continue_watching,
recent_views=recent_views,
recently_added=get_recently_added_courses(),
library_stats=get_library_stats())
# Apply progress data to tree
ProgressTracker.apply_progress_to_tree(current_course)
stats = ProgressTracker.get_completion_stats(current_course)
if stats.get('total_duration_seconds'):
stats['remaining_display'] = format_duration(stats['remaining_seconds'])
return render_template('course_dashboard.html',
course=current_course,
stats=stats,
continue_watching=continue_watching,
recent_views=recent_views)
@app.route('/browse')
def browse_directories():
"""Browse directories for course selection"""
path = request.args.get('path', '')
try:
# If no path specified, start with available drives on Windows
if not path:
import platform
if platform.system() == 'Windows':
# Get available drives on Windows
import string
drives = []
for letter in string.ascii_uppercase:
drive = f"{letter}:\\"
if os.path.exists(drive):
drives.append({
'name': f"Drive {letter}:",
'path': drive,
'media_files': 0,
'is_course_candidate': False
})
print(f"Returning {len(drives)} drives")
return jsonify({
'current_path': 'Select a Drive',
'parent_path': None,
'directories': drives
})
else:
# On other systems, start from home directory
path = str(Path.home())
current_path = Path(path)
if not current_path.exists() or not current_path.is_dir():
# Fallback to home directory if path doesn't exist
current_path = Path.home()
print(f"Browsing directory: {current_path}")
# Get directories and basic info
directories = []
try:
for item in sorted(current_path.iterdir()):
if item.is_dir() and not item.name.startswith('.'):
try:
# Check if this looks like a course directory
media_count = len([f for f in item.rglob('*')
if f.suffix.lower() in VIDEO_EXTENSIONS | AUDIO_EXTENSIONS])
directories.append({
'name': item.name,
'path': str(item),
'media_files': media_count,
'is_course_candidate': media_count > 0
})
except (PermissionError, OSError):
directories.append({
'name': item.name + " (Access Denied)",
'path': str(item),
'media_files': 0,
'is_course_candidate': False
})
except (PermissionError, OSError) as e:
print(f"Access denied to {current_path}: {str(e)}")
return jsonify({'error': f'Access denied to {current_path}: {str(e)}'}), 403
# Determine parent path
parent = None
if current_path.parent != current_path:
try:
parent = str(current_path.parent)
except (PermissionError, OSError):
pass
print(f"Found {len(directories)} directories")
return jsonify({
'current_path': str(current_path),
'parent_path': parent,
'directories': directories
})
except Exception as e:
print(f"Error in browse_directories: {str(e)}")
return jsonify({'error': str(e)}), 500
@app.route('/library')
def browse_library():
"""
Lazily list one directory level of the courses library at a time, so
the UI can start collapsed at the top level and drill down on click
instead of scanning/rendering the whole tree upfront. Defaults to the
effective library root (persisted setting, or the
COURSES_LIBRARY_PATH/--library-path default); the 'path' query param
lets the browser descend, but is restricted to stay within that root.
"""
library_root = os.path.abspath(get_library_root())
requested_path = request.args.get('path', library_root)
target_path = os.path.abspath(requested_path)
if not (target_path == library_root or target_path.startswith(library_root + os.sep)):
return jsonify({'error': 'Path outside library root', 'items': [], 'errors': ['Access denied']}), 403
result = list_library_directory(target_path)
return jsonify({
'library_path': library_root,
'current_path': target_path,
'items': result['items'],
'errors': result['errors']
})
@app.route('/library/manage')
def browse_library_manage():
"""
Same as /library, but includes items the user has hidden (flagged
'hidden': true rather than filtered out), so the Settings-page
curation UI can browse the whole tree and toggle visibility.
"""
library_root = os.path.abspath(get_library_root())
requested_path = request.args.get('path', library_root)
target_path = os.path.abspath(requested_path)
if not (target_path == library_root or target_path.startswith(library_root + os.sep)):
return jsonify({'error': 'Path outside library root', 'items': [], 'errors': ['Access denied']}), 403
result = list_library_directory(target_path, skip_hidden=False)
return jsonify({
'library_path': library_root,
'current_path': target_path,
'items': result['items'],
'errors': result['errors']
})
@app.route('/library/thumbnail')
def library_thumbnail():
"""Serve a course's cover image (see find_course_thumbnail), if it has one."""
library_root = os.path.abspath(get_library_root())
course_path = request.args.get('path', '')
target_path = os.path.abspath(course_path)
if not (target_path == library_root or target_path.startswith(library_root + os.sep)):
return '', 403
thumbnail = find_course_thumbnail(target_path)
if not thumbnail:
return '', 404
return send_file(thumbnail)
@app.route('/library/search')
def library_search():
"""Recursively search course names in the library (respects hidden paths)."""
query = request.args.get('q', '').strip()
library_root = os.path.abspath(get_library_root())
if not query:
return jsonify({'library_path': library_root, 'results': []})
results = search_library_courses(library_root, query)
return jsonify({'library_path': library_root, 'results': results})
@app.route('/api/hidden-paths', methods=['GET'])
def get_hidden_paths_api():
"""List currently-hidden course/directory paths, with display names."""
paths = get_hidden_paths()
return jsonify({
'hidden_paths': [
{'path': p, 'name': os.path.basename(p.rstrip(os.sep)) or p}
for p in paths
]
})
@app.route('/api/hidden-paths', methods=['POST'])
def set_hidden_path_api():
"""Hide or un-hide a course/directory from the Library browser."""
data = request.json or {}
path = data.get('path', '')
hidden = bool(data.get('hidden', True))
if not path:
return jsonify({'error': 'path is required'}), 400
library_root = os.path.abspath(get_library_root())
target = os.path.abspath(path)
if not (target == library_root or target.startswith(library_root + os.sep)):
return jsonify({'error': 'Path outside library root'}), 403
updated = set_path_hidden(target, hidden)
return jsonify({'success': True, 'hidden_paths': updated})
def validate_new_name(new_name: str) -> Optional[str]:
"""Validate a proposed directory name; returns an error message, or None if valid."""
if not new_name:
return 'New name cannot be empty'
if '/' in new_name or '\\' in new_name or '\x00' in new_name or new_name in ('.', '..'):
return 'New name cannot contain path separators'
if new_name.startswith('.'):
return 'New name cannot start with a dot'
return None
def perform_rename(old_path: str, new_name: str) -> Dict[str, Any]:
"""
Validate and apply a single directory rename within the library root:
rebases any hidden-path/recent-view references that pointed inside it,
and resets the active course if it (or an ancestor) was the thing
renamed. Shared by the single-item and bulk rename routes, always
returning a dict with a 'status' key for the caller to respond with.
"""
global current_course
error = validate_new_name(new_name)
if error:
return {'success': False, 'error': error, 'status': 400}
library_root = os.path.abspath(get_library_root())
old_abs = os.path.abspath(old_path)
if not (old_abs == library_root or old_abs.startswith(library_root + os.sep)):
return {'success': False, 'error': 'Path outside library root', 'status': 403}
if old_abs == library_root:
return {'success': False, 'error': 'Cannot rename the library root itself', 'status': 400}
if not os.path.isdir(old_abs):
return {'success': False, 'error': 'Directory not found', 'status': 404}
new_abs = os.path.join(os.path.dirname(old_abs), new_name)
if os.path.exists(new_abs):
return {'success': False, 'error': f'"{new_name}" already exists here', 'status': 409}
try:
os.rename(old_abs, new_abs)
except OSError as e:
return {'success': False, 'error': f'Rename failed: {e}', 'status': 500}
rebase_library_path(old_abs, new_abs)
active_course_reset = False
if current_course is not None:
course_abs = os.path.abspath(current_course.path)
if course_abs == old_abs or course_abs.startswith(old_abs + os.sep):
current_course = None
active_course_reset = True
return {
'success': True,
'new_path': new_abs,
'active_course_reset': active_course_reset,
'status': 200
}
@app.route('/api/rename-path', methods=['POST'])
def rename_path_api():
"""Rename a course/folder directory in the library, in place on disk."""
data = request.json or {}
path = data.get('path', '')
new_name = (data.get('new_name') or '').strip()
if not path:
return jsonify({'error': 'path is required'}), 400
result = perform_rename(path, new_name)
status = result.pop('status')
return jsonify(result), status
def _iter_all_directories(directory: Path, hidden_set: set) -> Iterator[Path]:
"""
Yield every directory in the library tree - both course folders and the
category/group folders above them - skipping hidden ones and not
descending into a course's own internal sections (Section 1, etc. -
those aren't independently manageable library items anywhere else in
the app either, so bulk rename shouldn't touch them).
"""
try:
entries = sorted(
(p for p in directory.iterdir() if p.is_dir() and not p.name.startswith('.')),
key=lambda p: p.name.lower()
)
except (PermissionError, OSError):
return
for entry in entries:
if os.path.abspath(str(entry)) in hidden_set:
continue
yield entry
if not _looks_like_course(entry):
yield from _iter_all_directories(entry, hidden_set)
def find_bulk_rename_matches(library_root: str, pattern: str, replacement: str) -> List[Dict[str, str]]:
"""Every directory (course or folder) in the library whose name contains `pattern`."""
hidden_set = set(get_hidden_paths())
return [
{
'path': str(directory),
'old_name': directory.name,
'new_name': directory.name.replace(pattern, replacement)
}
for directory in _iter_all_directories(Path(library_root), hidden_set)
if pattern in directory.name
]
@app.route('/api/bulk-rename/preview')
def bulk_rename_preview_api():
"""Preview a find/replace rename across the whole library, without touching disk."""
pattern = request.args.get('pattern', '')
replacement = request.args.get('replacement', '')
if not pattern:
return jsonify({'error': 'pattern is required'}), 400
library_root = os.path.abspath(get_library_root())
matches = find_bulk_rename_matches(library_root, pattern, replacement)
return jsonify({'matches': matches})
@app.route('/api/bulk-rename/apply', methods=['POST'])
def bulk_rename_apply_api():
"""
Apply a bulk rename. Takes the *exact* {path, new_name} list the client
got back from /api/bulk-rename/preview, rather than re-deriving matches
from the pattern again - what gets renamed is provably what the user
saw and approved, and can't drift if the library changed in between.
Continues past individual failures (e.g. a collision) instead of
aborting the whole batch, reporting a per-item result.
"""
data = request.json or {}
items = data.get('items') or []
if not isinstance(items, list) or not items:
return jsonify({'error': 'items is required'}), 400
results = []
for item in items:
path = item.get('path', '')
new_name = item.get('new_name', '')
outcome = perform_rename(path, new_name)
outcome.pop('status', None)
outcome['path'] = path
outcome['old_name'] = item.get('old_name', '')
outcome['new_name'] = new_name
results.append(outcome)
return jsonify({'results': results})
@app.route('/settings')
def settings_page():
"""Render the display-settings page."""
current = load_settings()
named_presets = sorted(
(t for t in SETTINGS_CHOICES['theme'] if t not in ('dark', 'light')),
key=lambda t: THEME_DISPLAY_NAMES[t]
)
return render_template(
'settings.html',
settings=current,
default_library_path=LIBRARY_PATH,
theme_choices=['dark', 'light'] + named_presets,
theme_display_names=THEME_DISPLAY_NAMES,
font_family_choices=['system', 'sans', 'serif', 'monospace', 'monaspace'],
font_size_choices=['small', 'medium', 'large', 'xlarge'],
layout_width_choices=['normal', 'wide', 'full'],
density_choices=['comfortable', 'compact'],
card_style_choices=['flat', 'elevated', 'bordered'],
corner_radius_choices=['sharp', 'rounded', 'pill'],
)
@app.route('/api/settings', methods=['GET'])
def get_settings_api():
"""Return current settings plus their resolved CSS values, for the shared theme script."""
current = load_settings()
return jsonify({
'settings': current,
'css_vars': settings_css_vars(current)
})
@app.route('/api/settings', methods=['POST'])
def save_settings_api():
"""Persist updated display settings."""
try:
new_settings = request.get_json(force=True) or {}
saved = save_settings(new_settings)
return jsonify({
'success': True,
'settings': saved,
'css_vars': settings_css_vars(saved)
})
except Exception as e:
return jsonify({'success': False, 'error': str(e)}), 400
@app.route('/api/settings/reset', methods=['POST'])
def reset_settings_api():
"""Reset settings back to defaults."""
os.makedirs(DATA_DIR, exist_ok=True)
with open(SETTINGS_FILE, 'w') as f:
json.dump(DEFAULT_SETTINGS, f, indent=2)
return jsonify({
'success': True,
'settings': DEFAULT_SETTINGS,
'css_vars': settings_css_vars(DEFAULT_SETTINGS)
})
@app.route('/api/backup')
def download_backup():
"""
Bundle everything that isn't recoverable from the course files
themselves - settings, hidden-path curation, recently-viewed history,
and every course's progress/notes file - into a single downloadable
zip. Cheap insurance before a NAS migration or a docker volume mistake.
"""
buffer = io.BytesIO()
with zipfile.ZipFile(buffer, 'w', zipfile.ZIP_DEFLATED) as zf:
for name, path in (
('settings.json', SETTINGS_FILE),
('hidden_paths.json', HIDDEN_PATHS_FILE),
('recent_views.json', RECENT_VIEWS_FILE),
):
if os.path.exists(path):
zf.write(path, name)
library_root = get_library_root()
for course_dir in iter_all_courses(library_root):
progress_file = course_dir / '.offlineu_progress.json'
if progress_file.exists():
relative = os.path.relpath(str(course_dir), library_root)
zf.write(progress_file, os.path.join('progress', relative, '.offlineu_progress.json'))
buffer.seek(0)
filename = f"offlineu-backup-{datetime.now().strftime('%Y-%m-%d')}.zip"
return send_file(buffer, mimetype='application/zip', as_attachment=True, download_name=filename)
@app.route('/load_course', methods=['POST'])
def load_course():
"""Load course from selected directory"""
global current_course
data = request.json
course_path = data.get('course_path')
if not course_path or not os.path.exists(course_path):
return jsonify({'error': 'Invalid course path'}), 400
try:
current_course = DynamicCourseParser.scan_directory(course_path)
return jsonify({'success': True, 'course_name': current_course.name})
except Exception as e:
return jsonify({'error': str(e)}), 500
@app.route('/recent/open')
def open_recent():
"""
Jump straight to a "Recently Viewed" entry from the dashboard: load its
course if it isn't already the active one, then go to that lesson.
"""
global current_course
course_path = request.args.get('course_path', '')
lesson_path = request.args.get('lesson_path', '')
if not course_path or not lesson_path or not os.path.exists(course_path):
return redirect(url_for('index'))
try:
if not current_course or current_course.path != course_path:
current_course = DynamicCourseParser.scan_directory(course_path)
except Exception as e:
print(f"Could not load course for recent view: {e}")
return redirect(url_for('index'))
return redirect(url_for('view_lesson', lesson_path=lesson_path))
@app.route('/help')
def help_page():
"""Static help page: how to use the app, supported file types."""
return render_template('help.html')
@app.route('/lesson/<path:lesson_path>')
def view_lesson(lesson_path: str):
"""View specific lesson by path"""
global current_course
if not current_course:
return redirect(url_for('index'))
# Find the lesson in the tree
lesson = find_lesson_in_tree(current_course.root_node, lesson_path)
if not lesson:
return redirect(url_for('index'))
# Get all lessons for navigation
all_lessons = get_all_lessons(current_course.root_node)
current_index = -1
# Find current lesson index
for i, (path, lesson_obj) in enumerate(all_lessons):
if lesson_obj == lesson:
current_index = i
break
# Get next and previous lessons
prev_lesson = None
next_lesson = None
if current_index > 0:
prev_lesson = all_lessons[current_index - 1][0]
if current_index < len(all_lessons) - 1:
next_lesson = all_lessons[current_index + 1][0]
# Update last accessed
ProgressTracker.update_lesson_progress(current_course, lesson_path)
# Record for the cross-course "Recently Viewed" list on the dashboard
record_recent_view(current_course.name, current_course.path, lesson_path, lesson.title)
# Read the note directly from the progress file rather than the Lesson
# object - apply_progress_to_tree (which populates Lesson fields) isn't
# called on this code path, only on the dashboard's tree render.
note = ProgressTracker.load_progress(current_course).get(lesson_path, {}).get('note', '')
return render_template('lesson_view.html',
course=current_course,
lesson=lesson,
lesson_path=lesson_path,
lesson_note=note,
prev_lesson=prev_lesson,
next_lesson=next_lesson)
def get_lesson_url(lesson: Lesson, course_path: str) -> str:
"""Generate the URL for a lesson"""
# Create relative path from course root
lesson_file_path = os.path.relpath(lesson.path, course_path)
lesson_file_path = lesson_file_path.replace('\\', '/')
if lesson_file_path.startswith('/'):
lesson_file_path = lesson_file_path[1:]
# Append lesson title for uniqueness
lesson_url = f"{lesson_file_path}/{lesson.title.replace(' ', '_')}"
return lesson_url
def find_lesson_in_tree(node: DirectoryNode, target_path: str) -> Optional[Lesson]:
"""Find a lesson in the tree by path"""
# Check lessons in current node
for lesson in node.lessons:
lesson_url = get_lesson_url(lesson, current_course.path)
# Check multiple possible path formats
lesson_file_path = os.path.relpath(lesson.path, current_course.path)
lesson_file_path = lesson_file_path.replace('\\', '/')
if lesson_file_path.startswith('/'):
lesson_file_path = lesson_file_path[1:]
# Also check with lesson title appended
lesson_path_with_title = f"{lesson_file_path}/{lesson.title.replace(' ', '_')}"
if (lesson_url == target_path or
lesson_file_path == target_path or
lesson_path_with_title == target_path):
return lesson
# Recursively search children
for child in node.children.values():
result = find_lesson_in_tree(child, target_path)
if result:
return result
return None
def get_all_lessons(node: DirectoryNode) -> List[Tuple[str, Lesson]]:
"""Get all lessons from the tree with their paths"""
lessons = []
def collect_lessons(n: DirectoryNode, current_path: str = ""):
# Add lessons from this node
for lesson in n.lessons:
lesson_url = get_lesson_url(lesson, current_course.path)
lessons.append((lesson_url, lesson))
# Recursively collect from children
for child in n.children.values():
collect_lessons(child, current_path)
collect_lessons(node)
return lessons
@app.route('/api/progress', methods=['POST'])
def update_progress():
"""API endpoint to update lesson progress"""
global current_course
if not current_course:
return jsonify({'error': 'No course loaded'}), 400
data = request.json
lesson_path = data.get('lesson_path')
completed = data.get('completed', False)
progress_seconds = data.get('progress_seconds', 0)
duration_seconds = data.get('duration_seconds') or None
try:
ProgressTracker.update_lesson_progress(
current_course, lesson_path, completed, progress_seconds, duration_seconds
)
return jsonify({'success': True})
except Exception as e:
return jsonify({'error': str(e)}), 500
@app.route('/api/lesson-note', methods=['POST'])
def update_lesson_note_api():
"""API endpoint to save (or clear) a lesson's note"""
global current_course
if not current_course:
return jsonify({'error': 'No course loaded'}), 400
data = request.json or {}
lesson_path = data.get('lesson_path')
note = (data.get('note') or '').strip()
if not lesson_path:
return jsonify({'error': 'lesson_path is required'}), 400
ProgressTracker.update_lesson_note(current_course, lesson_path, note)
return jsonify({'success': True})
@app.route('/api/course/mark-watched', methods=['POST'])
def mark_course_watched_api():
"""Mark every lesson in the currently loaded course as completed."""
global current_course
if not current_course:
return jsonify({'error': 'No course loaded'}), 400
ProgressTracker.mark_all_completed(current_course)
return jsonify({'success': True})
@app.route('/files/<path:filepath>')
def serve_file(filepath):
"""Serve course files"""
global current_course
if not current_course:
return "No course loaded", 404
# Security: ensure file is within course directory
try:
# URL decode the filepath and normalize it
from urllib.parse import unquote
decoded_filepath = unquote(filepath)
# Construct the full path relative to the course directory
full_path = os.path.join(current_course.path, decoded_filepath)
full_path = os.path.abspath(full_path)
course_path = os.path.abspath(current_course.path)
print(f"File request: {filepath}")
print(f"Decoded filepath: {decoded_filepath}")
print(f"Full path: {full_path}")
print(f"Course path: {course_path}")
# Security check: ensure file is within course directory
if not full_path.startswith(course_path):
print(f"Access denied: {full_path} not in {course_path}")
return "Access denied", 403
if not os.path.exists(full_path):
print(f"File not found: {full_path}")
return "File not found", 404
print(f"Serving file: {full_path}")
# Determine MIME type. Don't rely solely on mimetypes.guess_type() -
# its results can vary by OS/container base image depending on what
# system mime databases are present. Force the types that matter for
# inline preview (PDF above all) so this can't silently regress.
ext = os.path.splitext(full_path)[1].lower()
KNOWN_MIME_TYPES = {
'.pdf': 'application/pdf',
'.html': 'text/html',
'.htm': 'text/html',
'.txt': 'text/plain',
'.md': 'text/plain',
}
if ext in KNOWN_MIME_TYPES:
mime_type = KNOWN_MIME_TYPES[ext]
else:
mime_type, _ = mimetypes.guess_type(full_path)
if mime_type is None:
mime_type = 'application/octet-stream'
# as_attachment=False (the default) sends Content-Disposition: inline
# so the browser renders PDFs/HTML in the iframe instead of prompting
# a download. Being explicit here so this can't drift.
return send_file(full_path, mimetype=mime_type, as_attachment=False)
except Exception as e:
print(f"Error serving file: {str(e)}")
return f"Error serving file: {str(e)}", 500
@app.route('/health')
def healthcheck():
"""Healthcheck endpoint for Docker"""
return jsonify({"status": "healthy"}), 200
@app.route('/reset_course')
def reset_course():
"""Reset current course selection"""
global current_course
current_course = None
return redirect(url_for('index'))
def create_templates():
"""Create basic template files if they don't exist"""
templates_dir = Path('templates')
templates_dir.mkdir(exist_ok=True)
# Basic select course template
select_template = '''<!DOCTYPE html>
<html>
<head>
<title>OfflineU - Select Course</title>
<style>
body { font-family: Arial, sans-serif; margin: 40px; }
.container { max-width: 800px; margin: 0 auto; }
.directory { padding: 10px; border: 1px solid #ddd; margin: 5px 0; cursor: pointer; }
.directory:hover { background-color: #f0f0f0; }
.course-candidate { background-color: #e8f5e8; }
</style>
</head>
<body>
<div class="container">
<h1>OfflineU - Course Selection</h1>
<div id="browser"></div>
<script>
// Basic directory browser implementation
function loadDirectories(path = '') {
fetch(`/browse?path=${encodeURIComponent(path)}`)
.then(r => r.json())
.then(data => {
const browser = document.getElementById('browser');
browser.innerHTML = `
<h3>Current: ${data.current_path}</h3>
${data.parent_path ? `<div class="directory" onclick="loadDirectories('${data.parent_path}')">📁 .. (Parent)</div>` : ''}
${data.directories.map(dir => `
<div class="directory ${dir.is_course_candidate ? 'course-candidate' : ''}"
onclick="${dir.is_course_candidate ? `loadCourse('${dir.path}')` : `loadDirectories('${dir.path}')`}">
📁 ${dir.name} ${dir.media_files > 0 ? `(${dir.media_files} media files)` : ''}
</div>
`).join('')}
`;
});
}
function loadCourse(path) {
fetch('/load_course', {
method: 'POST',
headers: {'Content-Type': 'application/json'},
body: JSON.stringify({course_path: path})
})
.then(r => r.json())
.then(data => {
if (data.success) {
location.reload();
} else {
alert('Error: ' + data.error);
}
});
}
loadDirectories();
</script>
</div>
</body>
</html>'''
# Basic course dashboard template
dashboard_template = '''<!DOCTYPE html>
<html>
<head>
<title>OfflineU - {{ course.name }}</title>
<style>
body { font-family: Arial, sans-serif; margin: 20px; }
.container { max-width: 1200px; margin: 0 auto; }
.module { margin: 20px 0; border: 1px solid #ddd; padding: 15px; }
.lesson { padding: 8px; margin: 5px 0; border-left: 4px solid #ddd; }
.lesson.completed { border-left-color: #4CAF50; background-color: #f8fff8; }
.lesson a { text-decoration: none; color: #333; }
.lesson:hover { background-color: #f0f0f0; }
.progress { background-color: #f0f0f0; height: 20px; border-radius: 10px; overflow: hidden; }
.progress-bar { background-color: #4CAF50; height: 100%; transition: width 0.3s; }
</style>
</head>
<body>
<div class="container">
<h1>{{ course.name }}</h1>
<div class="progress">
<div class="progress-bar" style="width: {{ stats.completion_percentage }}%"></div>
</div>
<p>Progress: {{ stats.completed_lessons }}/{{ stats.total_lessons }} lessons ({{ stats.completion_percentage }}%)</p>
{% for module_idx, module in course.modules|enumerate %}
<div class="module">
<h2>{{ module.title }}</h2>
{% for lesson_idx, lesson in module.lessons|enumerate %}
<div class="lesson {% if lesson.completed %}completed{% endif %}">
<a href="/lesson/{{ module_idx }}/{{ lesson_idx }}">
{{ lesson.title }}
<small>({{ lesson.lesson_type }})</small>
{% if lesson.completed %}✓{% endif %}
</a>
</div>
{% endfor %}
</div>
{% endfor %}
<p><a href="/reset_course">Select Different Course</a></p>
</div>
</body>
</html>'''
# Basic lesson view template
lesson_template = '''<!DOCTYPE html>
<html>
<head>
<title>{{ lesson.title }} - {{ course.name }}</title>
<style>
body { font-family: Arial, sans-serif; margin: 20px; }
.container { max-width: 1000px; margin: 0 auto; }
video, audio { width: 100%; max-width: 800px; }
.content { margin: 20px 0; }
.navigation { margin: 20px 0; }
button { padding: 10px 20px; margin: 5px; cursor: pointer; }
.file-link { display: block; margin: 5px 0; padding: 5px; background: #f0f0f0; text-decoration: none; color: #333; }
.file-link:hover { background: #e0e0e0; }
</style>
</head>
<body>
<div class="container">
<h1>{{ lesson.title }}</h1>
<div class="navigation">
<a href="/">← Back to Course</a>
<button onclick="markCompleted()">Mark as Completed</button>
</div>
<div class="content">
{% if lesson.video_file %}
<h3>Video</h3>
<video controls preload="metadata" id="video-player">
<source src="/files/{{ lesson.video_file }}" type="video/mp4">
{% if lesson.subtitle_file %}
<track kind="subtitles" src="/files/{{ lesson.subtitle_file }}" srclang="en" label="English">
{% endif %}
Your browser does not support the video tag.
</video>
{% endif %}
{% if lesson.audio_file %}
<h3>Audio</h3>
<audio controls preload="metadata" id="audio-player">
<source src="/files/{{ lesson.audio_file }}" type="audio/mp3">
Your browser does not support the audio tag.
</audio>
{% endif %}
{% if lesson.text_files %}
<h3>Additional Resources</h3>
{% for text_file in lesson.text_files %}
<a href="/files/{{ text_file }}" class="file-link" target="_blank">
📄 {{ text_file.split('/')[-1] }}
</a>
{% endfor %}
{% endif %}
</div>
<script>
function markCompleted() {
fetch('/api/progress', {
method: 'POST',
headers: {'Content-Type': 'application/json'},
body: JSON.stringify({
lesson_path: '{{ lesson_path }}',
completed: true,
progress_seconds: getProgressSeconds()
})
})
.then(r => r.json())
.then(data => {
if (data.success) {
alert('Lesson marked as completed!');
window.location.href = '/';
}
});
}
function getProgressSeconds() {
const video = document.getElementById('video-player');
const audio = document.getElementById('audio-player');
if (video && !video.paused) return Math.floor(video.currentTime);
if (audio && !audio.paused) return Math.floor(audio.currentTime);
return 0;
}
// Auto-save progress periodically
setInterval(() => {
const seconds = getProgressSeconds();
if (seconds > 0) {
fetch('/api/progress', {
method: 'POST',
headers: {'Content-Type': 'application/json'},
body: JSON.stringify({
lesson_path: '{{ lesson_path }}',
completed: false,
progress_seconds: seconds
})
});
}
}, 30000); // Save every 30 seconds
</script>
</div>
</body>
</html>'''
# Write templates to files
template_files = {
'select_course.html': select_template,
'course_dashboard.html': dashboard_template,
'lesson_view.html': lesson_template
}
for filename, content in template_files.items():
template_path = templates_dir / filename
if not template_path.exists():
with open(template_path, 'w', encoding='utf-8') as f:
f.write(content)
print(f"Created template: {template_path}")
if __name__ == '__main__':
parser = argparse.ArgumentParser(description='OfflineU Course Viewer & Tracker')
parser.add_argument('--host', default='0.0.0.0', help='Host to bind to')
parser.add_argument('--port', type=int, default=5000, help='Port to bind to')
parser.add_argument('--debug', action='store_true', help='Enable debug mode')
parser.add_argument('--create-templates', action='store_true', help='Create basic templates')
parser.add_argument('--library-path', default=None,
help='Base directory to scan for courses in the Library browser '
'(default: $COURSES_LIBRARY_PATH or /app/courses)')
parser.add_argument('course_path', nargs='?', help='Path to course directory')
args = parser.parse_args()
if args.library_path:
LIBRARY_PATH = args.library_path
# Create templates if requested
if args.create_templates:
create_templates()
print("Templates created successfully!")
if not args.course_path:
sys.exit(0)
# Auto-load course if provided
if args.course_path or os.environ.get('AUTO_LOAD_COURSE'):
course_path = args.course_path or os.environ.get('AUTO_LOAD_COURSE')
if not os.path.exists(course_path):
print(f"Error: Course path does not exist: {course_path}", file=sys.stderr)
sys.exit(1)
try:
current_course = DynamicCourseParser.scan_directory(course_path)
print(f"Auto-loaded course: {current_course.name}")
print(f"Built dynamic directory tree with {len(current_course.root_node.children)} top-level items")
except Exception as e:
print(f"Error loading course: {e}", file=sys.stderr)
if args.debug:
import traceback
traceback.print_exc()
sys.exit(1)
# Create templates directory if it doesn't exist
if not Path('templates').exists():
print("Templates directory not found. Creating basic templates...")
create_templates()
print(f"Starting OfflineU on http://{args.host}:{args.port}")
print("Use --create-templates to regenerate template files")
try:
app.run(debug=args.debug, host=args.host, port=args.port)
except KeyboardInterrupt:
print("\nShutting down OfflineU...")
except Exception as e:
print(f"Error starting server: {e}")
if args.debug:
import traceback
traceback.print_exc()
sys.exit(1)