Pipelines Persistence

Pipeline persistence in TangleML Studio App works through a dual-layer system that balances local editing flexibility with centralized storage for collaboration. This guide explains how your pipelines are saved, shared, and managed across different stages of their lifecycle.

tip

Think of the Editor as your personal drafting space and Pipeline Runs as your published, shareable versions.

The Two-Layer Persistence Model

Local Storage (Editor Mode)

The Pipeline Editor automatically saves all your changes to your browser's local storage. This provides:

• Instant saving - Every change is preserved immediately
• Draft-like workflow - Work on pipelines without committing to a final version
• Privacy - Your work-in-progress remains local to your machine
• Offline capability - Continue editing without backend connectivity

warning

Editor changes are stored locally in your browser. You cannot share pipelines directly via URL links while they're in the Editor. To share your work, you must either export the pipeline or submit it as a Pipeline Run.

Backend Storage (Pipeline Runs)

When you submit a pipeline for execution, it becomes a Pipeline Run stored on the backend:

• Permanent storage - Preserved on the server infrastructure
• Shareable - Each run gets a unique identifier for sharing
• Version history - Track different iterations of your pipeline
• Team collaboration - Other team members can access and review runs

info

All Pipeline Runs are available to anyone with the access to your instance of the TangleML Studio App.

Key Features Comparison

Feature	Editor (Local)	Pipeline Run (Backend)
Storage Location	Browser local storage	Server database
Shareable	❌ No	✅ Yes
Requires Network	❌ No	✅ Yes
Version Control	Limited (current version only)	Full history
Persistence	Until browser data cleared	Permanent
Collaboration	Via export/import	Direct sharing

Auto-Save Functionality

The Editor continuously saves your work without manual intervention:

• Changes are saved immediately as you make them
• No need to manually save your progress
• Visual indicators confirm when changes are saved
• Recovery from browser crashes or accidental closures

tip

The auto-save feature ensures you never lose work due to unexpected interruptions. Your pipeline state is preserved even if you accidentally close the browser tab.

Undo/Redo System

The Editor maintains a history of your recent actions:

• Up to 50 actions stored in memory
• Session-based - Available during your current editing session
• Quick recovery - Instantly revert or reapply changes
• Keyboard shortcuts - Use standard undo/redo shortcuts for efficiency

warning

Undo/Redo history is session-only and will be lost when:

• The page is refreshed
• The browser tab is closed
• You navigate away from the Editor

Managing Multiple Pipeline Versions

Save As Feature

Create copies of your pipeline while preserving the original:

1. Click on the pipeline name or menu
2. Select "Save Pipeline As"
3. Enter a new name for the copy
4. The new version becomes your active pipeline

tip

Use "Save As" to experiment with different pipeline configurations without affecting your original design.

Export and Import

Share pipelines before creating runs through the export/import workflow:

Exporting Pipelines

1. Open your pipeline in the Editor
2. Click "Export Pipeline" from the pipeline actions menu
3. Save the YAML file to your local machine

Importing Pipelines

1. Click "Import Pipeline" in the Editor
2. Either:
   • Select a YAML file from your computer
   • Paste YAML content directly

Accessing Your Pipelines

My Pipelines Tab

All your locally saved pipelines are accessible from the Home screen:

1. Navigate to the Home screen
2. Click on the "My Pipelines" tab
3. View all pipelines stored in your browser
4. Click any pipeline to open it in the Editor

tip

The "My Pipelines" tab shows only pipelines saved in your current browser. If you use multiple devices or browsers, each will have its own set of locally saved pipelines.

Summary

warning

Browser Data Clearing: Pipelines stored in the Editor will be lost if you:

• Clear browser data/cache
• Use browser incognito/private mode
• Uninstall or reset your browser

Always export important pipelines or submit them as runs for permanent storage.

The dual-layer persistence model in TangleML Studio App provides flexibility for individual work while enabling team collaboration:

• Editor Mode - Your personal workspace with instant local saves
• Pipeline Runs - Committed, shareable versions stored on the backend
• Export/Import - Bridge the gap when you need to share drafts
• Auto-save & Undo/Redo - Productivity features for efficient editing

Remember: treat the Editor as your drafting space and Pipeline Runs as your published work. This mental model will help you effectively manage your pipeline development workflow.