chore: reorganize file structure

This commit is contained in:
2025-10-18 00:23:32 +03:00
parent 34ba481f28
commit 4d9c3fb5a0
13 changed files with 0 additions and 0 deletions

532
project-docs/dev-plan.md Normal file
View File

@@ -0,0 +1,532 @@
# Astrolabe Development Plan
**A lightweight, browser-based snippet manager for Vega-Lite visualizations**
> **For AI Context**: See `CLAUDE.md` for current status and key patterns.
> **This Document**: Complete development roadmap and technical decisions.
## Project Vision
Astrolabe is a focused tool for managing, editing, and previewing Vega-Lite visualization specifications. It emphasizes efficient snippet management with a clean, resizable three-panel interface: snippet library, Monaco editor with schema-aware autocomplete, and live preview.
## Design Principles
- **Local-first**: All data stored in browser (localStorage, future IndexedDB migration)
- **Minimal dependencies**: Vanilla JavaScript, no build tools, direct CDN imports
- **Developer-friendly**: Full JSON schema support, syntax validation, and intellisense
- **Version-aware**: Draft/published workflow for safe experimentation
- **Dataset management**: Separate storage for datasets with reference system
- **Extensible**: Clean architecture for future cloud sync and authentication
## Planned Feature Set
- Resizable three-panel layout with show/hide toggles
- Auto-saving draft system with explicit publishing
- Snippet organization with sorting and search
- Storage monitoring and size tracking
- Export/import functionality
- Dataset library and management
- (Future) Authentication and cloud synchronization
---
## Development Roadmap
### **Phase 0: Storage Architecture Design** ✅ **COMPLETE**
**Goal**: Define data structures that accommodate full feature set
- [x] Design snippet schema: `{id, name, created, modified, spec, draftSpec, comment, tags[], datasetRefs[], meta}`
- [x] Design dataset schema: `{id, name, created, modified, data, format, rowCount, columnCount, size, columns, comment, tags, meta}`
- [x] Design settings schema: `{panelWidths[], panelVisibility[], autoSaveDelay, meta}`
- [x] Plan localStorage key structure and namespacing
- [x] Define upgrade/migration path for schema changes
- [x] Document storage architecture decisions (see storage-examples.md)
**Deliverable**: Documented data model that supports all planned features
**Key Decisions Made**:
- ID generation: `Date.now() + random numbers` for uniqueness
- Auto-naming: ISO datetime format for default snippet names
- Draft/published workflow: separate `spec` and `draftSpec` fields
- Computed metadata: automatic calculation of size, rowCount, columns for datasets
- Extensibility: `meta` field in all schemas for future enhancements
- Settings persistence: panel state and UI preferences stored locally
---
### **Phase 1: Static HTML Structure** ✅ **COMPLETE**
**Goal**: Basic three-panel layout with placeholder content
**Deliverables**:
- Three-panel flexbox layout (snippet library, editor, preview)
- Header with site branding (🔭 Astrolabe) and action links
- Toggle button strip with emoji icons (📄 ✏️ 👁️)
- Retro Windows 2000 aesthetic with classic colors and square edges
- Sort and search controls in snippet panel header
- Metadata panel for snippet details
---
### **Phase 2: Resizable Panels** ✅ **COMPLETE**
**Goal**: Make panels draggable to resize
**Deliverables**:
- Drag handles between panels with hover/active states
- Mouse-based horizontal resizing with 200px minimum widths
- Panel memory system preserving preferred sizes when toggled
- Proportional redistribution when panels are hidden/shown
- localStorage persistence for panel widths, visibility, and memory
- Toggle buttons with active state indicators
---
### **Phase 3: Monaco Editor Integration** ✅ **COMPLETE**
**Goal**: Working code editor with Vega-Lite schema
**Deliverables**:
- Monaco Editor v0.47.0 loaded from CDN
- JSON language mode with Vega-Lite v5 schema validation
- Autocomplete and intellisense for Vega-Lite specifications
- Format on paste/type enabled
- Error handling for schema loading failures
---
### **Phase 4: Vega-Lite Rendering** ✅ **COMPLETE**
**Goal**: Live preview of visualizations
**Deliverables**:
- Vega, Vega-Lite, and Vega-Embed libraries loaded via CDN
- AMD loader conflict resolution between Monaco and Vega
- 1.5s debounced rendering on editor changes
- Immediate rendering on snippet selection
- Error display in preview panel for invalid specs
- SVG rendering for high-quality output
---
### **Phase 5: Data Model + LocalStorage** ✅ **COMPLETE**
**Goal**: Persist snippets and load them on page refresh
**Deliverables**:
- SnippetStorage wrapper with error handling
- Full Phase 0 schema implementation (id, name, created, modified, spec, draftSpec, comment, tags, datasetRefs, meta)
- ID generation using timestamp + random numbers
- Auto-naming with ISO datetime format (YYYY-MM-DD_HH-MM-SS)
- Dynamic snippet list rendering with relative date formatting
- Default snippet initialization on first load
- Snippet selection loading specs into editor
---
### **Phase 6: Snippet Selection & Basic CRUD** ✅ **COMPLETE**
**Goal**: Core snippet management with advanced organization
**Deliverables**:
- Snippet selection loading into editor with live preview
- Ghost card "Create New Snippet" interface at top of list
- Duplicate button creating copies with "_copy" suffix
- Delete button with confirmation dialog
- Inline name editing in metadata panel with auto-save
- Comment field with auto-save
- 1-second debounced auto-save for spec and metadata changes
- Multi-field sorting (Modified/Created/Name) with ascending/descending toggle
- Visual sort indicators with arrow icons (⬇⬆)
- Real-time search across name, comment, and spec content (300ms debounce)
- Search clear button
- Settings persistence for sort preferences
- Smart state management maintaining selection during operations
- Synchronous flag-based prevention of auto-save during programmatic updates
---
### **Phase 7: Draft/Published Workflow** ✅ **COMPLETE**
**Goal**: Safe experimentation without losing working versions
**Deliverables**:
- Draft/Published toggle buttons in editor header (merged visual design)
- Status indicator lights on snippets (🟢 green = no changes, 🟡 yellow = has draft)
- Publish button (green, copies draftSpec → spec)
- Revert button (orange, copies spec → draftSpec with confirmation)
- Context-aware button visibility (only shown in draft mode)
- Read-only published view when draft exists
- Auto-draft creation when editing published view without draft
- Auto-select first snippet on page load
- Instant status light updates after auto-save
---
### **Phase 8: Storage Monitoring** ✅ **COMPLETE**
**Goal**: Show storage usage and limits
**Deliverables**:
- Storage usage calculation using Blob API for accurate byte counting
- Progress bar display showing usage vs 5MB limit (e.g., "2.3 MB / 5 MB")
- Visual warning states: green (normal), orange (90%+ warning), red (95%+ critical)
- Storage monitor positioned at bottom of snippet panel below metadata
- Automatic updates after every save operation
- Alert dialog when localStorage quota is exceeded
- Flexbox layout ensuring monitor stays at panel bottom with scrollable snippet list
---
### **Phase 9: Export/Import** ✅ **COMPLETE**
**Goal**: Portability and backup
**Deliverables**:
- Export all snippets to JSON file with auto-generated filename (astrolabe-snippets-YYYY-MM-DD.json)
- Import snippets from JSON with automatic format detection
- Support for both Astrolabe native format and external formats
- Automatic field mapping and normalization (createdAt → created, content → spec, draft → draftSpec)
- "imported" tag automatically added to externally imported snippets
- ID conflict resolution with automatic regeneration
- Additive import (merges with existing snippets, no overwrites)
- Success/error feedback with import count
- File picker integration with header Import/Export links
- Snippet size display in list (shows KB for snippets ≥ 1 KB, right-aligned)
**Note**: Drag-and-drop import and selective export options deferred to future phases
---
### **Phase 10: Dataset Management** ✅ **COMPLETE**
**Goal**: Separate dataset storage with multiple data sources and formats
**Deliverables**:
- IndexedDB-based dataset storage (separate from snippets, no 5MB localStorage limit)
- Full CRUD operations for datasets (create, read, update, delete)
- Modal-based Dataset Manager UI (accessible via header link and 📁 toggle button)
- Support for multiple data sources:
- **Inline data**: JSON, CSV, TSV, TopoJSON stored directly
- **URL data**: Remote data sources with format specification
- **Intelligent auto-detection system**:
- Single input field for data or URL
- Automatic URL detection and content fetching
- Format detection from content (JSON, CSV, TSV, TopoJSON)
- Confidence scoring (high/medium/low)
- Visual confirmation UI with badges and preview
- Automatic metadata calculation:
- Row count, column count, column names
- Data size in bytes
- Automatic URL fetching on dataset creation
- Refresh metadata button for URL datasets (🔄)
- Dataset list with informative metadata display
- Dataset details panel with:
- Editable name and comment
- Statistics display (rows, columns, size)
- Data preview (first 5 rows or URL info)
- Copy reference button (copies `"data": {"name": "dataset-name"}`)
- Delete button with confirmation
- Auto-resolve dataset references in Vega-Lite specs during rendering
- Format-aware rendering:
- JSON: `{ values: data }`
- CSV/TSV: `{ values: data, format: { type: 'csv'/'tsv' } }`
- TopoJSON: `{ values: data, format: { type: 'topojson' } }`
- URL: `{ url: "...", format: { type: '...' } }`
- Graceful error handling for CORS and network failures
- Modal scrolling support for small viewports
**Dataset Schema**:
```javascript
{
id: number,
name: string,
created: ISO string,
modified: ISO string,
data: array/string/object, // Inline data or URL string
format: 'json'|'csv'|'tsv'|'topojson',
source: 'inline'|'url',
comment: string,
rowCount: number|null,
columnCount: number|null,
columns: string[],
size: number|null // bytes
}
```
**Technical Implementation**:
- IndexedDB with keyPath 'id', indexes on 'name' (unique) and 'modified'
- Async/Promise-based DatasetStorage API
- Format-specific parsing for metadata calculation
- Vega-Lite's native format parsers used for rendering
- Metadata refresh fetches live data and updates statistics
- Modal resizes with viewport (max-height: 90vh)
- **Auto-detection algorithms**:
- URL validation with protocol check (http/https)
- JSON parsing with TopoJSON identification
- CSV/TSV detection via delimiter counting and consistency checks
- Format inference from URL file extensions
- Debounced input handling for real-time feedback
---
### **Phase 11: URL State Management** ✅ **COMPLETE**
**Goal**: Shareable URLs and browser navigation support
**Deliverables**:
- Hash-based URL routing for snippets and datasets
- URL patterns:
- Snippet selection: `#snippet-123456`
- Dataset modal open: `#datasets`
- Dataset selected: `#datasets/dataset-123456`
- New dataset form: `#datasets/new`
- URLState utility in config.js with parse/update/clear methods
- Browser back/forward navigation (hashchange event listener)
- Page reload preserves selected snippet or dataset state
- Automatic state restoration after editor initialization
- Restores snippet URL when closing dataset modal
- No external dependencies (native Hash API and History API)
**Technical Implementation**:
- `URLState.parse()`: Parses hash into state object
- `URLState.update(state, replaceState)`: Updates URL with optional history.replaceState
- `URLState.clear()`: Removes hash from URL
- Integration points:
- `selectSnippet()`: Updates URL on snippet selection
- `selectDataset()`: Updates URL on dataset selection
- `openDatasetManager()`: Sets `#datasets` hash
- `closeDatasetManager()`: Restores snippet hash or clears
- `showNewDatasetForm()`: Sets `#datasets/new` hash
- `handleURLStateChange()`: Responds to hashchange events
- `initializeURLStateManagement()`: Called after Monaco editor ready
- Prevents double-prefix issues by handling prefix addition in URLState.update()
- Optional `updateURL` parameter on all functions to prevent infinite loops
**Benefits**:
- Shareable links to specific snippets or datasets
- Browser navigation works intuitively
- Page refresh preserves user context
- Better UX for multi-tab workflows
---
### **Phase 12: Advanced Dataset Features** ✅ **COMPLETE**
**Goal**: Enhanced dataset workflows
**Deliverables**:
- Recursive dataset reference extraction from Vega-Lite specs
- Extract to Dataset modal with automatic spec transformation
- Bidirectional linking between snippets and datasets
- Usage tracking with visual indicators (📁 icon, 📄 count badges)
- Dataset import/export with auto-format detection
- Table preview with type detection (🔢📅✓🔤)
- On-demand URL preview loading with session cache
- New Snippet creation from datasets
**Technical Implementation**:
- Type detection: 80% threshold for number/date/boolean/text inference
- Table rendering: Vanilla JS with sticky headers, 20-row preview
- Cell formatting: Type-specific styling and alignment
- Import: Automatic naming with timestamp collision handling
- Export: Format-specific extensions with URL content fetching
- Preview cache: In-memory storage, not persisted to database
- Modal: 95% width (max 1200px), 85% height for improved data visibility
---
### **Phase 13: Polish & UX Refinements** ✅ **COMPLETE**
**Goal**: Professional feel and usability
- [x] **Keyboard shortcuts** - Cross-platform support (Cmd/Ctrl+Shift+N, Cmd/Ctrl+K, Cmd/Ctrl+S, Escape)
- [x] **Tooltips for buttons and features** - Added to all interactive elements throughout the UI
- [x] **Better error messages and validation feedback** - Toast notification system with 4 types (error, success, warning, info)
- [x] **Enhanced Help documentation** - Comprehensive help modal with About, Features, Getting Started, Shortcuts, Storage, and Privacy sections
- [x] **User Settings System** - Configurable editor options, performance tuning, date formatting, and theme selection
- [x] **Theme System** - Light theme (Windows 2000) and experimental dark theme with CSS variables
**Deliverable**: Polished, production-ready MVP
**Completed Features**:
- Keyboard shortcut system with dual handlers (document-level + Monaco integration)
- Toast notification system (bottom-right, 4s auto-dismiss, slide animations)
- Replaced 19 `alert()` calls with contextual toasts
- Added success feedback for silent operations (publish, duplicate, delete, export, etc.)
- Comprehensive tooltips with keyboard shortcut hints
- Data persistence warning in Help section
- Help modal with 6 sections covering full app functionality (About, Features, Getting Started, Keyboard Shortcuts, Storage, Privacy & Data)
- User settings with localStorage persistence and validation
- Light theme (Windows 2000 aesthetic) and experimental dark theme
- Automatic editor theme synchronization with UI theme
- Custom date formatting with smart/relative times, locale, ISO, and token-based formats
- Size sorting for snippets (blob-based calculation)
- Accessibility improvements with proper ARIA labels and color contrast
---
### **Phase 14: Advanced Snippet Features**
**Goal**: Power user functionality
- [ ] Basic tagging system with tag filtering
- [ ] Snippet templates/starter library
- [ ] Bulk operations (delete multiple, export selected)
- [ ] Size calculation and display for snippets
**Deliverable**: Enhanced snippet discovery and organization
**Note**: Search/filter and sorting already implemented in Phase 6
---
### **Phase 15: Authentication & Backend** _(Future)_
**Goal**: Multi-device sync and sharing
- [ ] Design minimal backend API
- [ ] User authentication (email/password or OAuth)
- [ ] Sync localStorage ↔ cloud storage
- [ ] Conflict resolution strategy
- [ ] Share snippets via URL
- [ ] Public snippet gallery (optional)
**Deliverable**: Cloud-backed, multi-device Astrolabe
---
## Technical Stack
- **Frontend**: Vanilla JavaScript (ES6+), HTML5, CSS3
- **Editor**: Monaco Editor v0.47.0 (via CDN)
- **Visualization**: Vega-Embed v6 (includes Vega v5 & Vega-Lite v5)
- **Storage**: LocalStorage → IndexedDB (when needed)
- **Architecture**: Modular script organization with simple file separation
- **Backend** _(future)_: TBD (minimal REST API)
## Project Structure
```
/
├── index.html # Main HTML structure and markup
├── CLAUDE.md # Project instructions for Claude Code
├── src/
│ ├── styles.css # Retro Windows 2000 aesthetic styling
│ └── js/ # Modular JavaScript organization
│ ├── config.js # Global variables, settings, & sample data
│ ├── snippet-manager.js # Snippet storage, CRUD operations & localStorage wrapper
│ ├── dataset-manager.js # Dataset storage, CRUD operations & IndexedDB wrapper
│ ├── panel-manager.js # Panel resize, toggle & memory system
│ ├── editor.js # Monaco Editor initialization, dataset resolution & Vega-Lite rendering
│ └── app.js # Application initialization & event handlers
└── docs/
├── dev-plan.md # This development roadmap
└── storage-examples.md # Data model specifications
```
## Development Principles
- **Iterative**: Each phase produces working, testable functionality
- **Lean**: No frameworks, no build step, minimal dependencies
- **Data-first**: Storage schema designed upfront for extensibility
- **User-focused**: Auto-save, clear state, forgiving UX
- **Maintainable**: Clean code organization with logical separation of concerns
- **Simple**: Favor code removal over addition; avoid over-engineering
## Code Organization
**Web Application** (`/web` directory):
- **index.html**: Main HTML structure with semantic markup
- **src/app.js**: Application initialization and event handler registration
- **src/config.js**: Global variables, settings management, and sample data
- **src/snippet-manager.js**: Storage wrapper and all snippet CRUD operations
- **src/panel-manager.js**: Layout resizing, toggling, and persistence
- **src/editor.js**: Monaco and Vega library loading and rendering logic
- **src/dataset-manager.js**: Dataset CRUD and IndexedDB operations
- **src/user-settings.js**: Settings management with localStorage persistence and validation
- **src/styles.css**: Retro Windows 2000 aesthetic with CSS variables for theming
**Documentation** (`/docs` directory - not served on GitHub Pages):
- **dev-plan.md**: Complete development roadmap and technical decisions
- **features-list.md**: Feature matrix and implementation status
- **storage-examples.md**: Data structure examples and schema documentation
- **CLAUDE.md**: Context and status summary for AI assistance
---
## Current Status
**Completed**: Phases 0-13 (All core features, dataset management, advanced features, and UX refinements)
**In Progress**: GitHub Pages deployment preparation and project restructuring for public distribution
**Next**: Phase 14 - Advanced Snippet Features (tagging system, templates, bulk operations) or public launch
**See**: `CLAUDE.md` for concise current state summary
---
## Implemented Features
### Core Capabilities (Phases 0-11)
- Three-panel resizable layout with memory and persistence
- Monaco Editor v0.47.0 with Vega-Lite v5 schema validation
- Live Vega-Lite rendering with debounced updates and error display
- localStorage-based snippet management with full CRUD
- Multi-field sorting (Modified/Created/Name) with direction toggle
- Real-time search across snippet name, comment, and spec content
- Auto-save system (1s debounce) for specs and metadata
- Ghost card interface for snippet creation
- Draft/Published workflow with version control
- Status indicator lights (green/yellow) showing draft state
- Context-aware Publish/Revert buttons with color coding
- Storage monitoring with visual progress bar and warning states
- Export/Import functionality with format auto-detection
- Snippet size display (right-aligned, shown for ≥ 1 KB)
- **Dataset Management (Phase 10)**:
- IndexedDB storage for datasets (unlimited size)
- Multi-format support: JSON, CSV, TSV, TopoJSON
- Multi-source support: Inline data and URL references
- Modal-based Dataset Manager with full CRUD
- Intelligent auto-detection (URL/format/confidence)
- Visual confirmation UI with badges and preview
- Automatic metadata calculation and display
- URL metadata fetching and refresh
- Dataset reference resolution in Vega-Lite specs
- **URL State Management (Phase 11)**:
- Hash-based routing for snippets and datasets
- Browser back/forward navigation support
- Page reload preserves state
- Shareable URLs for specific snippets/datasets
- Restores snippet URL when closing dataset modal
- **Advanced Dataset Features (Phase 12)**:
- Bidirectional linking with usage tracking and navigation
- Extract to Dataset with automatic spec transformation
- Import/Export with format detection and URL content fetching
- Table preview with type detection and formatting (🔢📅✓🔤)
- On-demand URL preview loading with session cache
- New Snippet creation from datasets
- **Polish & UX Refinements (Phase 13)**:
- Cross-platform keyboard shortcuts (Cmd/Ctrl+Shift+N, Cmd/Ctrl+K, Cmd/Ctrl+S, Escape)
- Dual keyboard handlers (document-level + Monaco integration)
- Toast notification system (error, success, warning, info)
- Contextual feedback for all major operations
- Comprehensive tooltips with keyboard hints
- Enhanced Help modal with 6 sections (About, Features, Getting Started, Shortcuts, Storage, Privacy & Data)
- Data persistence warning in Help section
- User Settings System with localStorage persistence
- Configurable editor options (font size, tab size, minimap, word wrap, line numbers)
- Performance tuning (render debounce delay)
- Date formatting options (smart/relative, locale, ISO, custom tokens)
- Theme system (Light and experimental Dark themes)
- Automatic editor theme synchronization
- Size sorting for snippets
- Retro Windows 2000 aesthetic throughout
- Component-based CSS architecture with base classes
### Technical Implementation
- **State Management**: Synchronous `isUpdatingEditor` flag prevents unwanted auto-saves
- **View Modes**: `currentViewMode` tracks draft vs published state
- **Read-only Logic**: Monaco editor locked in published view when draft exists
- **Auto-draft Creation**: Editing published without draft auto-switches to draft mode
- **Debouncing**: 1.5s render, 1s auto-save, 300ms search
- **AMD Resolution**: Temporary `window.define` disabling for Vega library loading
- **Panel Memory**: localStorage persistence for sizes and visibility across sessions
- **Data Model**: Phase 0 schema with `spec` (published) and `draftSpec` (working) fields
- **Storage Calculation**: Blob API for accurate byte counting of snippet data
- **Flexbox Layout**: Scrollable snippet list with fixed metadata and storage monitor at bottom
- **Import/Export**: Format detection, field normalization, ID conflict resolution, additive merging
- **Size Display**: Per-snippet size calculation with conditional rendering (≥ 1 KB threshold)
- **Dataset Storage**: IndexedDB with async/Promise-based API, unique name constraint
- **Dataset Resolution**: Async spec transformation before rendering, format-aware data injection
- **URL Metadata**: Fetch on creation with graceful CORS error handling
- **Modal UI**: Flexbox with overflow:auto, max-height responsive to viewport (80vh fixed height)
- **Auto-detection**: URL validation, JSON/CSV/TSV parsing, confidence scoring, real-time feedback
- **URL State Management**: Native Hash API with hashchange listener, initialized after editor ready
- **CSS Architecture**: Component-based with base classes (.btn, .input, .preview-box) and modifiers
- **Keyboard Shortcuts**: Dual system with document listeners + Monaco commands for full coverage
- **Toast Notifications**: Bottom-right positioned, 4s auto-dismiss, slide animations, 4 types (error/success/warning/info)
- **User Feedback**: Replaced 19 alert() calls with contextual toasts, kept confirm() for destructive actions

View File

@@ -0,0 +1,344 @@
# Astrolabe - Current Features List
> **Purpose**: Comprehensive inventory of all implemented features for code review and optimization
> **Created**: 2025-10-15
> **Updated**: 2025-10-16
> **Status**: Phases 0-12 Complete, Phase 13 In Progress (Keyboard Shortcuts, Notifications, Tooltips, Help Documentation)
---
## 🔭 **ASTROLABE - Current Features Inventory**
### **1. Layout & UI Structure**
- Three-panel resizable layout (Snippet Library | Editor | Preview)
- Manual drag-to-resize with 200px minimum widths
- Panel show/hide toggle buttons (📄 ✏️ 👁️ 📁)
- Panel memory system (remembers sizes when toggled)
- Proportional width redistribution when panels are hidden/shown
- Layout persistence to localStorage across sessions
- Retro Windows 2000 aesthetic styling
**Files**: `panel-manager.js`, `styles.css`, `index.html`
---
### **2. Monaco Code Editor**
- Monaco Editor v0.47.0 integration via CDN
- JSON syntax highlighting and validation
- Vega-Lite v5 schema autocomplete and intellisense
- Format-on-paste and format-on-type
- Read-only mode (when viewing published with draft)
- Automatic layout resizing
**Files**: `app.js` (lines 31-82), `index.html` (line 11)
---
### **3. Vega-Lite Rendering**
- Live preview panel with Vega-Embed v6
- SVG rendering for high quality
- 1.5-second debounced rendering on editor changes
- Immediate rendering on snippet selection
- Error display with helpful messages
- Dataset reference resolution (inline & URL)
- Recursive resolution for layered/concat/faceted specs
**Files**: `editor.js`, `app.js` (rendering calls)
---
### **4. Snippet Management (CRUD)**
- Create new snippet (ghost card interface)
- Duplicate snippet (with "_copy" suffix)
- Delete snippet (with confirmation dialog)
- Inline name editing with auto-save
- Comment field with auto-save
- Auto-generated ISO datetime names (YYYY-MM-DD_HH-MM-SS)
- Unique ID generation (timestamp + random)
- Created/Modified timestamps with formatted display
**Files**: `snippet-manager.js` (lines 7-36, 621-701), `app.js` (event handlers)
---
### **5. Snippet Organization**
- Multi-field sorting (Modified/Created/Name)
- Ascending/descending toggle with arrow indicators (⬇⬆)
- Real-time search across name, comment, and spec content
- 300ms debounced search
- Search clear button with enabled/disabled state
- Visual sort indicators on active button
- Sort preferences persisted to localStorage
- Snippet size display (shows KB for ≥1KB snippets, right-aligned)
**Files**: `snippet-manager.js` (lines 96-405), `app.js` (initialization)
---
### **6. Draft/Published Workflow**
- Dual-spec storage (spec = published, draftSpec = working)
- View mode toggle (Draft/Published buttons)
- Status indicator lights (🟢 green = clean, 🟡 yellow = has draft)
- Publish button (copies draftSpec → spec)
- Revert button (copies spec → draftSpec with confirmation)
- Context-aware button visibility (only in draft mode)
- Auto-draft creation when editing published view
- Read-only published view when draft exists
- 1-second debounced auto-save for draft changes
**Files**: `snippet-manager.js` (lines 494-810), `config.js` (currentViewMode)
---
### **7. Storage Management**
- localStorage for snippets (5MB limit awareness)
- Storage usage monitor with progress bar
- Visual warning states (green/orange/red at 90%/95%)
- Accurate byte counting using Blob API
- Storage monitor positioned below metadata panel
- Real-time updates after save operations
- Quota exceeded alerts
**Files**: `snippet-manager.js` (lines 3-5, 812-853), `index.html` (storage monitor UI)
---
### **8. Import/Export**
- Export all snippets to JSON with auto-generated filename
- Import from JSON with format auto-detection
- Support for Astrolabe native format
- Support for external formats with field mapping
- Automatic "imported" tag for external snippets
- ID conflict resolution with regeneration
- Additive import (no overwrites)
- Success/error feedback with count
**Files**: `snippet-manager.js` (lines 855-976), `app.js` (lines 92-112)
---
### **9. Dataset Management**
- IndexedDB storage (unlimited size, separate from snippets)
- Multi-format support: JSON, CSV, TSV, TopoJSON
- Multi-source support: Inline data & URL references
- Full CRUD operations for datasets
- Modal-based Dataset Manager UI
- Dataset list with metadata display
- Dataset details panel with editable name/comment
- Statistics display (rows, columns, size)
- Data preview (first 5 rows or URL info)
- Copy reference button (generates `"data": {"name": "..."}`)
- Delete dataset with confirmation
- Refresh metadata button for URL datasets (🔄)
- **Auto-detection system**:
- Single input field for data or URL
- Automatic URL detection and content fetching
- Format detection (JSON, CSV, TSV, TopoJSON)
- Confidence scoring (high/medium/low)
- Visual confirmation with badges and preview
- URL fetching with CORS error handling
- Unique dataset name constraint (IndexedDB index)
- Empty state message for no datasets
**Files**: `dataset-manager.js`, `editor.js` (resolution), `app.js` (event handlers), `index.html` (modal UI)
---
### **10. Settings & Persistence**
- Settings storage in localStorage (separate from snippets)
- Default settings (sortBy: modified, sortOrder: desc)
- Settings API (load, save, get, set)
- Panel layout persistence
- Panel memory persistence
- Sort preferences persistence
- Auto-restore on page load
**Files**: `config.js` (lines 19-65), `panel-manager.js` (lines 76-131)
---
### **11. Auto-Save System**
- 1-second debounced auto-save for editor changes
- 1-second debounced auto-save for name/comment changes
- Synchronous flag (`isUpdatingEditor`) to prevent unwanted saves
- Auto-save only in draft mode
- Instant status light updates after save
- Storage monitor updates after save
**Files**: `snippet-manager.js` (lines 494-619), `app.js` (lines 73-75)
---
### **12. URL State Management**
- Hash-based URL routing for snippets and datasets
- Snippet selection persists in URL (`#snippet-123456`)
- Dataset modal state persists in URL:
- Modal open: `#datasets`
- Dataset selected: `#datasets/dataset-123456`
- New dataset form: `#datasets/new`
- Browser back/forward navigation support
- Page reload preserves state (selected snippet/dataset)
- URL sharing for specific snippets or datasets
- Automatic state restoration on page load
- Restores snippet URL when closing dataset modal
- No external libraries (native Hash API)
**Files**: `config.js` (URLState utility), `snippet-manager.js` (snippet URLs), `dataset-manager.js` (dataset URLs), `app.js` (hashchange listener)
---
### **13. User Experience Enhancements**
- Auto-select first snippet on page load
- Relative date formatting (Today/Yesterday/X days ago)
- Full datetime display in metadata panel
- Empty state placeholders
- "No snippets match your search" for search results
- Confirmation dialogs for destructive actions
- Flexbox layout with scrollable snippet list
- Fixed metadata and storage monitor at bottom
**Files**: `snippet-manager.js` (formatting functions), `app.js` (auto-select), `styles.css` (layout)
---
### **13. Advanced Dataset Features (Phase 12)**
- **Dataset Dependencies & Linking**:
- Recursive dataset reference extraction from Vega-Lite specs
- Bidirectional snippet ↔ dataset linking with clickable links
- Dataset usage tracking and count badges (📄 count)
- Visual indicators (📁) for snippets using datasets
- Linked datasets section in snippet metadata
- Linked snippets section in dataset details
- Navigation between snippets and datasets via links
- **Extract to Dataset**:
- Inline data detection in draft specs
- Extract modal with name generation and preview
- Automatic spec transformation (inline → reference)
- Auto-update snippet with dataset reference
- Conditional Extract button (shows when inline data detected)
- **Import/Export Datasets**:
- Import from file (.json, .csv, .tsv, .txt)
- Auto-format detection from content and filename
- Automatic naming from filename with duplicate handling
- Timestamp suffix for duplicate names (e.g., "data_123456")
- Export to format-specific files (.json, .csv, .tsv, .topojson)
- URL dataset export fetches and downloads live content
- **Table Preview with Type Detection**:
- Raw/Table toggle for tabular data (JSON arrays, CSV, TSV)
- Vanilla JS table rendering (first 20 rows)
- Sticky headers with scrollable content
- Column type detection (80% threshold): number, date, boolean, text
- Type icons in headers: 🔢 📅 ✓ 🔤
- Type-specific formatting: italic numbers (right-aligned), italic dates, bold booleans
- Color coding: blue numbers, green dates, orange booleans, gray nulls
- Monospace font (Consolas/Monaco/Courier New)
- **URL Preview Loading**:
- On-demand fetch with "Load Preview" button
- In-memory cache for fetched data (session-only)
- Full table view and type detection for fetched data
- Error handling with retry option
- Data not saved to database (preview only)
- **New Snippet from Dataset**:
- Button in dataset details to create snippet
- Auto-generated minimal Vega-Lite spec with dataset reference
- Pre-populated comment and datasetRefs
- **UI Enhancements**:
- Larger modal (95% width, max 1200px, 85% height)
- Actions moved to top of dataset details
- Dataset list with usage badges
**Files**: `dataset-manager.js` (lines 19-1165), `snippet-manager.js` (dataset tracking), `app.js` (event handlers), `index.html` (UI), `styles.css` (table styles)
---
## 📊 **Feature Statistics**
- **Core Feature Groups**: 14
- **Total Individual Capabilities**: ~100+
- **Storage Systems**: 2 (localStorage for snippets, IndexedDB for datasets)
- **UI Panels**: 3 main + 1 modal
- **Auto-save Points**: 3 (draft spec, name, comment)
- **Data Formats**: 4 (JSON, CSV, TSV, TopoJSON)
- **Data Sources**: 2 (inline, URL)
- **Type Detection**: 4 types (number, date, boolean, text)
- **Import/Export**: Snippets + Datasets
---
## 🗂️ **Code Organization**
```
src/
├── js/
│ ├── config.js # Global variables, settings, sample data
│ ├── snippet-manager.js # Snippet CRUD, storage, search, sort, extract (1,100+ lines)
│ ├── dataset-manager.js # Dataset CRUD, IndexedDB, preview, types (1,200+ lines)
│ ├── panel-manager.js # Layout resizing, toggling, persistence (200 lines)
│ ├── editor.js # Monaco setup, Vega rendering, dataset resolution (150 lines)
│ └── app.js # Event handlers, initialization (250+ lines)
└── styles.css # Retro Windows 2000 aesthetic (280+ lines)
```
**Total JS Lines**: ~2,900+ lines (excluding comments and blank lines)
---
## 🎯 **Use This Document For**
1. **Feature Review**: Select a feature to audit code relevance
2. **Code Cleanup**: Identify potential leftovers or redundancies
3. **Optimization**: Find opportunities to reduce codebase size
4. **Documentation**: Reference for what's actually implemented
5. **Planning**: Determine what to polish vs what to add
---
### **13. Polish & UX Refinements (Phase 13)** ⏳ _In Progress_
**Keyboard Shortcuts**:
- Cmd/Ctrl+Shift+N: Create new snippet
- Cmd/Ctrl+K: Toggle dataset manager
- Cmd/Ctrl+S: Publish draft
- Escape: Close any open modal
- Dual handler system (document-level + Monaco integration)
- Cross-platform support (Cmd on Mac, Ctrl on Windows/Linux)
**Toast Notification System**:
- Four types: error, success, warning, info
- Bottom-right positioning, 4-second auto-dismiss
- Slide-in/slide-out animations
- Manual close button
- Replaced 19 `alert()` calls with contextual toasts
- Kept `confirm()` dialogs for destructive actions
**Tooltips**:
- Added to all interactive elements (buttons, links, controls)
- Keyboard shortcut hints included where applicable
- Consistent tooltip text across the application
**Enhanced Help Documentation**:
- Comprehensive Help modal with 6 sections
- About Astrolabe (project vision and approach)
- Key Features (scannable highlights)
- Getting Started (4-step workflow guide)
- Keyboard Shortcuts (reference table)
- Storage & Limits (with data persistence warning)
- Tips & Tricks (power user features)
**Files**: `app.js` (keyboard handlers), `config.js` (Toast utility), `styles.css` (toast/help styles), `index.html` (help modal)
---
## 📝 **Next Steps**
### Phase 13 In Progress! Keyboard shortcuts, notifications, tooltips, and help documentation complete.
**Remaining Phase 13 Tasks**:
- [ ] Loading states for rendering
- [ ] Empty states (no snippets, no datasets)
- [ ] Improved visual design polish
- [ ] Cross-browser testing
**Future Phases**:
- **Phase 14**: Advanced Snippet Features (tagging system, templates, bulk operations)
- **Phase 15**: Authentication & Backend (cloud sync, sharing)

View File

@@ -0,0 +1,92 @@
Snippet example:
```json
{
id: 1760200058.123123, // Date.now() + random numbers
name: "2025-10-13_14-23-45", // auto-populated
created: "2025-10-13T14:23:45.123Z",
modified: "2025-10-13T14:25:30.456Z",
spec: {
"$schema": "https://vega.github.io/schema/vega-lite/v5.json",
"data": {"values": [...]},
"mark": "bar",
"encoding": {
"x": {"field": "category", "type": "nominal"},
"y": {"field": "value", "type": "quantitative"}
}
},
draftSpec: {
"$schema": "https://vega.github.io/schema/vega-lite/v5.json",
"data": {"values": [...]},
"mark": "bar",
"encoding": {
"x": {"field": "category", "type": "nominal"},
"y": {"field": "value", "type": "quantitative"},
"color": {"field": "category", "type": "nominal"} // draft addition
}
},
comment: "SELECT category, SUM(value) FROM sales GROUP BY category",
tags: [],
datasetRefs: [],
meta: {}
}
```
Dataset example:
```json
{
// Core identity
id: 1760200058.123123, // Date.now() + random numbers
name: "sales_2024_q1", // editable
// Timestamps
created: 1760200058, // timestamp
modified: 1760304958 //timestamp,
// Data (stored as-is, format-agnostic)
data: [
{"category": "A", "value": 100},
{"category": "B", "value": 200}
],
format: "json", // "json" | "csv" | "tsv" (for future import/export)
// Computed on save
rowCount: 2,
columnCount: 2,
size: 78, // bytes: JSON.stringify(data).length
columns: ["category", "value"], // inferred from first row
// Metadata
comment: "",
tags: [],
// Extensibility
meta: {}
}
```
Settings example:
```json
{
// UI State
panelWidths: [25, 25, 50], // % for [list, editor, preview]
panelVisibility: [true, true, true],
// App behavior (empty for MVP, but structure exists)
autoSaveDelay: 1000,
// Extensibility
meta: {}
}
```