Add Flow Catalog system with registration, runs, and favorites#285
Merged
Edwardvaneechoud merged 18 commits intomainfrom Feb 2, 2026
Merged
Add Flow Catalog system with registration, runs, and favorites#285Edwardvaneechoud merged 18 commits intomainfrom
Edwardvaneechoud merged 18 commits intomainfrom
Conversation
…d Unity Catalog-style namespaces Introduces a comprehensive flow management system: Backend: - New DB models: CatalogNamespace (catalog/schema hierarchy), FlowRegistration, FlowRun (with YAML version snapshots), FlowFavorite, FlowFollow - New Pydantic schemas in catalog_schema.py for all CRUD operations - New /catalog router with endpoints for namespaces, flow registration, run history, favorites, follows, and stats dashboard - Run tracking hooked into existing flow/run endpoint - every execution now persists a FlowRun record with the flow version snapshot and node-level results Frontend: - CatalogView with 4 tabs: Catalog tree, Favorites, Following, Run History - Tree navigation with catalog -> schema -> flow hierarchy - Flow detail panel showing metadata, run history table, and version links - Run detail panel with node-level results and flow snapshot viewer - Stats overview panel with counts and recent activity - Full Pinia store, typed API client, and TypeScript interfaces - Sidebar navigation entry and i18n support https://claude.ai/code/session_01THQ6XyXSppra3my7wt9WQE
✅ Deploy Preview for flowfile-wasm canceled.
|
The backend now checks os.path.exists() for each registered flow's file path and returns a file_exists boolean. When a flow file has been deleted from storage: - Tree view: flow row is dimmed with an amber warning icon - List view (favorites/following): shows "File missing" instead of run count - Detail panel: prominent amber banner explains the file is gone and that run history is still accessible https://claude.ai/code/session_01THQ6XyXSppra3my7wt9WQE
|
Codecov Report❌ Patch coverage is 📢 Thoughts on this report? Let us know! |
…and refresh fixes
- Seed default 'General' > 'user_flows' namespace on DB init
- Replace text input with FileBrowser component in flow registration modal
- Auto-assign flows to default namespace when no explicit namespace given
- Add 'Open this version' button in run detail to load snapshot into designer
- Add GET /catalog/default-namespace-id and POST /catalog/runs/{run_id}/open endpoints
- Fix stats/tree not refreshing after namespace create, flow register, and favorite/follow toggle
https://claude.ai/code/session_01THQ6XyXSppra3my7wt9WQE
The stats endpoint was counting all CatalogNamespace rows (both catalogs and schemas), causing double-counting. Now filters by level=0 to count only catalogs. https://claude.ai/code/session_01THQ6XyXSppra3my7wt9WQE
flow_file_handler is exported from flowfile_core, not flowfile_core.flowfile. https://claude.ai/code/session_01THQ6XyXSppra3my7wt9WQE
When a flow is created (POST /editor/create_flow/) or imported (GET /import_flow/), it is now automatically registered in the General > user_flows namespace if that namespace exists. Duplicate paths are skipped. Also moved inline imports to module level. https://claude.ai/code/session_01THQ6XyXSppra3my7wt9WQE
- FlowDetailPanel: add "Open in Designer" button that imports the flow and navigates to the designer view (hidden when file is missing) - RunDetailPanel: make the flow name a clickable link that navigates back to the flow detail in the catalog tab - CatalogView: wire up openFlowInDesigner and navigateToFlow handlers - Add test_catalog.py with tests for namespace CRUD, flow registration, favorites, follows, runs, stats counting, and default namespace seeding https://claude.ai/code/session_01THQ6XyXSppra3my7wt9WQE
Recent runs in the overview now navigate to the run detail view, and favorite flows navigate to the flow detail in the catalog tab. https://claude.ai/code/session_01THQ6XyXSppra3my7wt9WQE
When selecting a flow via stats panel, run detail, or other navigation, the tree node containing that flow now auto-expands so the selection is visible in the sidebar. https://claude.ai/code/session_01THQ6XyXSppra3my7wt9WQE
…r tests - Fix "Follow flows to get notified" text (no notification system exists) - Remove unused flowId variable in openRunSnapshot - Add 6 tests for _auto_register_flow covering: happy path, duplicate path dedup, None user_id, None flow_path, missing default namespace, and filename stem fallback https://claude.ai/code/session_01THQ6XyXSppra3my7wt9WQE
Console logs at key points to trace selection flow: - navigateToFlow: logs the registrationId - selectFlow: logs flowId and whether it resolved in allFlows - setActiveTab: logs tab switch - CatalogTreeNode watch: logs whether the node contains the selected flow All prefixed with [Catalog] for easy filtering. https://claude.ai/code/session_01THQ6XyXSppra3my7wt9WQE
- CatalogView: log on mount and after initialization with allFlows/tree counts - CatalogTreeNode: log on flow item click with flow id and name This ensures [Catalog] logs appear as soon as the catalog page loads, confirming the logging code is active. https://claude.ai/code/session_01THQ6XyXSppra3my7wt9WQE
MenuAccordion's activeIndex was set once on mount via ref(route.name), so it never updated when navigating programmatically with router.push(). Changed to computed(() => route.name) so the sidebar reactively tracks the current route. https://claude.ai/code/session_01THQ6XyXSppra3my7wt9WQE
Both openFlowInDesigner and openRunSnapshot were ignoring the flow ID returned by the import API and navigating to the designer without setting the active flow. The designer then loaded whichever flow was previously active instead of the newly imported one. Now both functions call flowStore.setFlowId(flowId) before navigating so the designer loads the correct flow. https://claude.ai/code/session_01THQ6XyXSppra3my7wt9WQE
- Remove all console.log [Catalog] debug statements from catalog-store, CatalogView, and CatalogTreeNode - Move inline imports (init_db, User, _auto_register_flow) to top of test_catalog.py https://claude.ai/code/session_01THQ6XyXSppra3my7wt9WQE
db.query(Model).get(id) is deprecated in SQLAlchemy 2.0. Replaced all occurrences in catalog.py and test_catalog.py with db.get(Model, id). https://claude.ai/code/session_01THQ6XyXSppra3my7wt9WQE
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
3 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
Implements a complete Flow Catalog system enabling users to register, organize, and track flow executions. Introduces a Unity Catalog-style hierarchical namespace structure (catalog → schema) for organizing flows, persistent run history with versioned snapshots, and social features (favorites/follows).
Key Changes
Backend
Database Models (
models.py):CatalogNamespace: Two-level hierarchy (catalog/schema) with unique constraint on name+parentFlowRegistration: Persistent flow metadata linked to namespacesFlowRun: Complete execution history with YAML snapshots and node resultsFlowFavorite&FlowFollow: User engagement tracking with unique constraintsAPI Routes (
routes/catalog.py):Run Tracking (
routes/routes.py):_run_and_track()wrapper persists flow executions to databaseFrontend
catalog.api.ts): Type-safe HTTP client for all catalog endpointscatalog-store.ts): Centralized state management for tree, flows, runs, favorites, statscatalog.types.ts): Complete type definitions matching backend schemas/catalogroute with lazy-loaded CatalogView componentNotable Implementation Details
_enrich_flow()helper attaches computed fields (is_favorite, is_following, run_count, last_run_at) to flow objects