release: v1.5.0 — Desktop GUI complete

README.md

@@ -1,6 +1,6 @@
 # NeuraScreen
 
-![Version](https://img.shields.io/badge/version-1.4.0-blue)
+![Version](https://img.shields.io/badge/version-1.5.0-blue)
 ![PyPI](https://img.shields.io/badge/pip_install-neurascreen-3775A9?logo=pypi&logoColor=white)
 ![Python](https://img.shields.io/badge/python-3.12+-3776AB?logo=python&logoColor=white)
 ![Playwright](https://img.shields.io/badge/playwright-1.52-2EAD33?logo=playwright&logoColor=white)
@@ -56,7 +56,13 @@ Watch NeuraScreen generate a full demo video from a JSON scenario — preview, T
 
 NeuraScreen includes an optional desktop application (PySide6) for editing scenarios, previewing TTS audio and managing configuration visually.
 
-![NeuraScreen GUI](docs/images/gui-screenshot.png)
+| Welcome | Editor | Configuration |
+|---------|--------|---------------|
+| ![Welcome](docs/images/capture_1.png) | ![Editor](docs/images/capture_2.png) | ![Config](docs/images/capture_5.png) |
+
+| Output Browser | Macro Recorder | JSON + Split View |
+|----------------|----------------|-------------------|
+| ![Output](docs/images/capture_12.png) | ![Macro](docs/images/capture_16.png) | ![Split](docs/images/capture_4.png) |
 
 Install with GUI support: `pip install neurascreen[gui]` then launch with `neurascreen gui`.
 
@@ -405,10 +411,15 @@ neurascreen gui
 - **Configuration manager** — visual .env editor with 7 tabs (Application, Browser, Screen Capture, TTS, Selectors, Directories), validation, import/export
 - **TTS & audio preview** — per-provider voice config (`~/.neurascreen/voices.json`), per-step audio preview, pronunciation helper, narration statistics
 - **Output browser** — browse generated videos with integrated video player (QMediaPlayer), SRT subtitles viewer, YouTube chapters viewer
-- **Theme engine** — dark teal (default) and light themes, switchable via Ctrl+T. Create custom themes as JSON files in `~/.neurascreen/themes/`
+- **Macro recorder** — record browser interactions from the GUI with live event feed, cleanup options, and direct import into the editor (Ctrl+R)
+- **Selector validator** — verify scenario selectors against the real DOM using Playwright headless, with found/not found/multiple status and suggestions (Ctrl+Shift+V)
+- **Scenario statistics** — steps count, actions breakdown, narration metrics, estimated duration, unique URLs and selectors
+- **Scenario diff** — compare two scenario files side by side with added/removed/modified/unchanged status
+- **Autosave & recovery** — periodic autosave every 60s, recovery prompt on startup
+- **Theme engine** — dark teal (default) and light themes with full Fusion style support, switchable via Ctrl+T. Create custom themes as JSON files in `~/.neurascreen/themes/`
 - **Step templates** — insert common patterns (navigation, drag & configure, form fill) from the context menu
 - **Undo/redo** — full undo history for all editing operations
-- **Keyboard shortcuts** — Ctrl+N/O/S, F5-F8 for commands, Ctrl+Shift+O for output browser, Ctrl+Shift+T for TTS panel
+- **Keyboard shortcuts** — 20+ shortcuts including Ctrl+N/O/S, F5-F8, Ctrl+R (record), Ctrl+Shift+V (validate selectors)
 
 The GUI is optional — the CLI remains the primary interface and works without PySide6.
 
@@ -465,7 +476,9 @@ neurascreen/
 │       ├── execution/    # Command execution panel
 │       ├── config/       # Configuration manager (.env editor)
 │       ├── tts/          # TTS panel, audio preview, voices, pronunciation
-│       └── output/       # Output browser, video player, SRT/chapters viewers
+│       ├── output/       # Output browser, video player, SRT/chapters viewers
+│       ├── macro/        # Macro recorder dialog, event feed, cleanup
+│       └── advanced/     # Selector validator, statistics, diff, autosave
 ├── tests/                # Unit tests (pytest)
 ├── examples/             # Example scenarios
 ├── docs/                 # Documentation

ROADMAP.md

@@ -38,20 +38,23 @@
 - [x] Macro recorder: record browser interactions → JSON scenario (#9)
 - [x] Docker container for headless generation (#13)
 
-## v1.5 — Desktop GUI (in progress)
+## v1.5 — Desktop GUI (done)
 
 Optional PySide6 desktop interface: `pip install neurascreen[gui]`
 
 - [x] GUI foundation: main window, menu bar, toolbar, sidebar, status bar (#25)
-- [x] Theme engine: JSON palettes → dynamic QSS, dark teal + light themes (#25)
+- [x] Theme engine: JSON palettes → dynamic QSS, dark teal + light themes, Fusion style (#25)
 - [x] Scenario editor: visual step list, adaptive detail panel, JSON source view (#26)
 - [x] Execution panel: run commands from GUI with real-time console logs (#27)
 - [x] Configuration manager: visual .env editor with 7 tabs (#29)
 - [x] TTS & audio preview: per-provider voice config, per-step preview, pronunciation helper (#28)
 - [x] Output browser: video list, integrated player, SRT/chapters/YouTube viewers (#30)
 - [x] App icon and cross-platform process name
-- [ ] Macro recorder integration: record browser interactions from GUI (#31)
-- [ ] Advanced: selector validator, scenario stats, diff viewer, autosave (#32)
+- [x] Macro recorder integration: record from GUI with live event feed, cleanup, editor import (#31)
+- [x] Selector validator: verify selectors against real DOM with Playwright headless (#32)
+- [x] Scenario statistics: steps, actions, narration metrics, duration, URLs (#32)
+- [x] Scenario diff: compare two scenarios side by side (#32)
+- [x] Autosave & recovery: periodic save, recovery on startup (#32)
 
 ## v1.6 — CLI Enhancements (planned)
 

docs/gui.md

@@ -18,6 +18,8 @@ neurascreen gui
 
 ## Overview
 
+![Welcome screen](images/capture_1.png)
+
 The GUI has six main areas:
 
 | Area | Description |
@@ -31,6 +33,8 @@ The GUI has six main areas:
 
 ## Editing scenarios
 
+![Scenario editor — Detail view](images/capture_2.png)
+
 ### Create or open
 
 - `Ctrl+N` — new empty scenario
@@ -70,10 +74,14 @@ Common fields (always visible): Wait after, Narration, Screenshot after step.
 
 ### JSON view
 
+![JSON view with syntax highlighting](images/capture_3.png)
+
 Switch to the **JSON** tab to edit the raw JSON directly. Changes sync bidirectionally with the visual editor. Syntax highlighting colors keys, strings, numbers and booleans.
 
 The **Split** tab shows both views side by side.
 
+![Split view — Detail + JSON](images/capture_4.png)
+
 ### Templates
 
 Right-click in the step list → **Insert template** to add common patterns:
@@ -106,6 +114,14 @@ When running from the GUI without the Headless checkbox, the browser always open
 
 Open via **Tools > Configuration** or **Ctrl+,**.
 
+| Application | Browser | Screen Capture | TTS |
+|-------------|---------|----------------|-----|
+| ![App](images/capture_5.png) | ![Browser](images/capture_6.png) | ![Capture](images/capture_7.png) | ![TTS](images/capture_8.png) |
+
+| Login Selectors | Canvas Selectors | Directories |
+|-----------------|------------------|-------------|
+| ![Login](images/capture_9.png) | ![Canvas](images/capture_10.png) | ![Dirs](images/capture_11.png) |
+
 The configuration dialog provides a visual editor for all `.env` variables, organized in 7 tabs:
 
 | Tab | Settings |
@@ -144,6 +160,10 @@ The panel shows: steps narrated / total, word count, estimated reading time (~13
 
 Open via **View > Output Browser** or **Ctrl+Shift+O**.
 
+| Video Player | SRT Subtitles | Chapters |
+|-------------|---------------|----------|
+| ![Player](images/capture_12.png) | ![SRT](images/capture_13.png) | ![Chapters](images/capture_14.png) |
+
 Browse generated videos in the `output/` directory:
 
 - **File table** — name, size, date, SRT/chapters indicators, sortable, searchable
@@ -156,6 +176,78 @@ Actions: double-click to play, right-click for context menu (open, copy path, de
 
 The file list auto-refreshes when new videos are generated.
 
+## Macro Recorder
+
+Open via **Tools > Record Macro** or **Ctrl+R**.
+
+![Macro Recorder](images/capture_16.png)
+
+Record browser interactions and convert them to a scenario:
+
+1. Enter the **Start URL** and an optional **Title**
+2. Click **Start Recording** — a Chromium browser opens
+3. Interact normally — clicks, navigations, scrolls and key presses are captured
+4. The **live event feed** shows captured events in real time (color-coded by type)
+5. Close the browser or click **Stop Recording**
+6. Review the results and apply **cleanup options**:
+   - **Dedup clicks** — remove rapid double-clicks on the same element
+   - **Merge navigations** — keep only the last of consecutive navigation events
+   - **Cap waits** — limit pauses to 5 seconds, remove pauses under 500ms
+7. Click **Open in Editor** to load the scenario directly, or **Save as...** to export as JSON
+
+## Selector Validator
+
+Open via **Tools > Validate Selectors** or **Ctrl+Shift+V** (requires an open scenario).
+
+![Selector Validator](images/capture_18.png)
+
+Verifies that all CSS selectors and click texts in your scenario exist in the real DOM:
+
+1. Enter the **Base URL** of your application
+2. Click **Validate** — Playwright launches headless and checks each selector
+3. Results appear in real time with status:
+   - **found** — selector matches exactly one element
+   - **not found** — selector not found (with suggestions if a similar selector exists)
+   - **multiple** — selector matches more than one element
+   - **skipped** — page could not be loaded
+4. **Double-click** a result to jump to that step in the editor
+
+## Scenario Statistics
+
+Open via **Tools > Statistics** (requires an open scenario).
+
+![Scenario Statistics](images/capture_17.png)
+
+Shows metrics for the current scenario:
+
+- Total steps and breakdown by action type
+- Narrated vs silent steps
+- Word count and estimated reading time (~130 words/min)
+- Estimated total duration (reading + waits)
+- Unique URLs visited
+- Unique CSS selectors used
+
+## Scenario Diff
+
+Open via **Tools > Compare Scenarios**.
+
+![Compare Scenarios](images/capture_19.png)
+
+Compare two JSON scenario files side by side:
+
+1. Select **File A** and **File B**
+2. Click **Compare**
+3. A table shows each step with status: unchanged, modified, added, or removed
+4. Modified steps list the changed fields (e.g., "selector", "narration")
+
+## Autosave & Recovery
+
+NeuraScreen automatically saves your work every 60 seconds to `~/.neurascreen/autosave/`.
+
+- If you quit without saving, a **recovery dialog** appears on next launch
+- Choose **Recover** to reload the autosaved scenario, or **Discard** to start fresh
+- Autosave is cleared when you save manually
+
 ## Themes
 
 Two built-in themes:
@@ -211,6 +303,7 @@ The theme appears in the **View > Theme** menu on next launch.
 | F7 | Run |
 | F8 | Full (with TTS) |
 | Ctrl+R | Record macro |
+| Ctrl+Shift+V | Validate selectors |
 | Ctrl+, | Configuration |
 | Ctrl+Shift+O | Output browser |
 | Ctrl+Shift+T | TTS panel |

neurascreen/__init__.py

@@ -1,3 +1,3 @@
 """NeuraScreen — Automated screen walkthrough video generator."""
 
-__version__ = "1.4.0"
+__version__ = "1.5.0"

pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "neurascreen"
-version = "1.4.0"
+version = "1.5.0"
 description = "Automated screen walkthrough video generator using Playwright, JSON scenarios and Text-to-Speech"
 readme = "README.md"
 license = "MIT"