Indexer

Task Genius uses the modern Dataflow Architecture to efficiently index and manage tasks from multiple sources. This event-driven system provides real-time updates, smart caching, and exceptional performance.

Quick Configuration

Open Index Settings

If the link doesn't work: Obsidian → Settings → Community Plugins → Task Genius

How it Works

The Dataflow architecture orchestrates multiple components for optimal performance:

Core Components

1. Orchestrator: Central coordinator managing all data sources and events
2. Data Sources:
   • ObsidianSource: Monitors file changes in your vault
   • IcsSource: Integrates external calendar events
   • FileSource: Recognizes entire files as tasks
3. Repository: Maintains the task index and handles queries
4. Storage: Multi-tier caching system for instant startup
5. Workers: Background processing for large operations

Processing Flow

1. File Monitoring: ObsidianSource watches for changes (create, modify, delete)
2. Smart Caching: Repository checks cache validity using content hashing
3. Parallel Processing:
   • WorkerOrchestrator distributes parsing to Web Workers
   • Multiple files processed simultaneously
   • Automatic fallback to main thread if needed
4. Event Broadcasting: Changes trigger task-genius:task-cache-updated events
5. View Synchronization: All views update automatically via event subscription

Performance Features

• Snapshot Loading: ~100ms startup for 1000+ tasks
• Incremental Updates: Only changed files reprocessed
• Content Hashing: Detect actual changes vs. metadata updates
• Loop Prevention: Sequence-based event tracking prevents infinite updates

Data Sources

The system integrates multiple data sources seamlessly:

• File Tasks: Regular tasks from Markdown and Canvas files
• ICS Events: External calendar events displayed as tasks
• File-level Tasks: Entire files treated as tasks based on metadata

Troubleshooting

If the index seems outdated:

• Run Task Genius: Force reindex all tasks to rebuild completely
• Use Task Genius: Refresh task index for a quick refresh
• Check console for any error messages

Managing the Indexer

Disabling

Disable indexing in Settings → Task Genius → Index → "Enable Indexing"

Performance Tuning

• Worker Count: Adjust based on vault size (Settings → Index)
• Cache Duration: Configure persistence time
• Batch Size: Control processing chunks

Configuration

Customize indexing behavior to match your workflow:

Metadata Format

Choose how to parse task metadata:

• Tasks: Emoji format (🛫 2025-01-01, ⏫, 🔁 every week)
• Dataview: Inline fields ([due:: 2025-01-01], [priority:: high])

The system automatically falls back to the other format when needed.

Daily Notes Integration

Automatically assign dates to tasks in daily notes:

• Enable: Extract dates from daily note file paths
• Path: Daily notes folder (e.g., dailies, journal/daily)
• Format: Date pattern in filenames (yyyy-MM-dd, dd-MM-yyyy)
• Assign As: Target field for extracted date (due, start, scheduled)

Heading Filters

Control which tasks are indexed based on their headings:

• Ignore Under: Exclude tasks under specific headings (## Archive, ## Done)
• Focus On: Only index tasks under specified headings (## Project, ## Inbox)

Enter headings separated by commas. Focus takes priority over ignore.