feat: browser automated test
This commit is contained in:
Timothy
@@ -0,0 +1,241 @@
|
||||
---
|
||||
name: browser-edge-cases
|
||||
description: SOP for debugging browser automation failures on complex websites. Use when browser tools fail on specific sites like LinkedIn, Twitter/X, SPAs, or sites with Shadow DOM.
|
||||
license: MIT
|
||||
---
|
||||
|
||||
# Browser Tool Edge Cases
|
||||
|
||||
Standard Operating Procedure for debugging and fixing browser automation failures on complex websites.
|
||||
|
||||
## When to Use This Skill
|
||||
|
||||
- `browser_scroll` succeeds but page doesn't move
|
||||
- `browser_click` succeeds but no action triggered
|
||||
- `browser_type` text disappears or doesn't work
|
||||
- `browser_snapshot` hangs or returns stale content
|
||||
- `browser_navigate` loads wrong content
|
||||
|
||||
## SOP: Debugging Browser Tool Failures
|
||||
|
||||
### Phase 1: Reproduce & Isolate
|
||||
|
||||
```
|
||||
1. Create minimal test case demonstrating failure
|
||||
2. Test against simple site (example.com) to verify tool works
|
||||
3. Test against problematic site to confirm issue
|
||||
```
|
||||
|
||||
**Quick isolation test:**
|
||||
```python
|
||||
# Test 1: Does the tool work at all?
|
||||
await browser_navigate(tab_id, "https://example.com")
|
||||
result = await browser_scroll(tab_id, "down", 100)
|
||||
# Should work on simple sites
|
||||
|
||||
# Test 2: Does it fail on the problematic site?
|
||||
await browser_navigate(tab_id, "https://linkedin.com/feed")
|
||||
result = await browser_scroll(tab_id, "down", 100)
|
||||
# If this fails but example.com works → site-specific edge case
|
||||
```
|
||||
|
||||
### Phase 2: Analyze Root Cause
|
||||
|
||||
**Step 2a: Check console for errors**
|
||||
```python
|
||||
console = await browser_console(tab_id)
|
||||
# Look for: CSP violations, React errors, JavaScript exceptions
|
||||
```
|
||||
|
||||
**Step 2b: Inspect DOM structure**
|
||||
```python
|
||||
html = await browser_html(tab_id)
|
||||
snapshot = await browser_snapshot(tab_id)
|
||||
# Look for:
|
||||
# - Nested scrollable divs (overflow: scroll/auto)
|
||||
# - Shadow DOM roots
|
||||
# - iframes
|
||||
# - Custom widgets
|
||||
```
|
||||
|
||||
**Step 2c: Identify the pattern**
|
||||
|
||||
| Symptom | Likely Cause | Check |
|
||||
|---------|--------------|-------|
|
||||
| Scroll doesn't move | Nested scroll container | Look for `overflow: scroll` divs |
|
||||
| Click no effect | Element covered | Check `getBoundingClientRect` vs viewport |
|
||||
| Type clears | Autocomplete/React | Check for event listeners on input |
|
||||
| Snapshot hangs | Huge DOM | Check node count in snapshot |
|
||||
| Snapshot stale | SPA hydration | Wait after navigation |
|
||||
|
||||
### Phase 3: Implement Multi-Layer Fix
|
||||
|
||||
**Pattern: Always have fallbacks**
|
||||
|
||||
```python
|
||||
async def robust_operation(tab_id):
|
||||
# Method 1: Primary approach
|
||||
try:
|
||||
result = await primary_method(tab_id)
|
||||
if verify_success(result):
|
||||
return result
|
||||
except Exception:
|
||||
pass
|
||||
|
||||
# Method 2: CDP fallback
|
||||
try:
|
||||
result = await cdp_fallback(tab_id)
|
||||
if verify_success(result):
|
||||
return result
|
||||
except Exception:
|
||||
pass
|
||||
|
||||
# Method 3: JavaScript fallback
|
||||
return await javascript_fallback(tab_id)
|
||||
```
|
||||
|
||||
**Pattern: Always add timeouts**
|
||||
|
||||
```python
|
||||
# Bad - can hang forever
|
||||
result = await browser_snapshot(tab_id)
|
||||
|
||||
# Good - fails fast with useful error
|
||||
try:
|
||||
result = await browser_snapshot(tab_id, timeout_s=10.0)
|
||||
except asyncio.TimeoutError:
|
||||
# Handle timeout gracefully
|
||||
result = await fallback_snapshot(tab_id)
|
||||
```
|
||||
|
||||
### Phase 4: Verify Fix
|
||||
|
||||
```
|
||||
1. Run against problematic site → should work
|
||||
2. Run against simple site → should still work (regression check)
|
||||
3. Document in registry.md
|
||||
```
|
||||
|
||||
## Pattern Library
|
||||
|
||||
### P1: Nested Scrollable Containers
|
||||
|
||||
**Sites:** LinkedIn, Twitter/X, any SPA with scrollable feeds
|
||||
|
||||
**Detection:**
|
||||
```javascript
|
||||
// Find largest scrollable container
|
||||
const candidates = [];
|
||||
document.querySelectorAll('*').forEach(el => {
|
||||
const style = getComputedStyle(el);
|
||||
if (style.overflow.includes('scroll') || style.overflow.includes('auto')) {
|
||||
const rect = el.getBoundingClientRect();
|
||||
if (rect.width > 100 && rect.height > 100) {
|
||||
candidates.push({el, area: rect.width * rect.height});
|
||||
}
|
||||
}
|
||||
});
|
||||
candidates.sort((a, b) => b.area - a.area);
|
||||
return candidates[0]?.el;
|
||||
```
|
||||
|
||||
**Fix:** Dispatch scroll events at container's center, not viewport center.
|
||||
|
||||
### P2: Element Covered by Overlay
|
||||
|
||||
**Sites:** Modals, tooltips, SPAs with loading overlays
|
||||
|
||||
**Detection:**
|
||||
```javascript
|
||||
const rect = element.getBoundingClientRect();
|
||||
const centerX = rect.left + rect.width / 2;
|
||||
const centerY = rect.top + rect.height / 2;
|
||||
const topElement = document.elementFromPoint(centerX, centerY);
|
||||
return topElement === element || element.contains(topElement);
|
||||
```
|
||||
|
||||
**Fix:** Wait for overlay to disappear, or use JavaScript click.
|
||||
|
||||
### P3: React Synthetic Events
|
||||
|
||||
**Sites:** React SPAs, modern web apps
|
||||
|
||||
**Detection:** If CDP click doesn't trigger handler but manual click works.
|
||||
|
||||
**Fix:** Use JavaScript click as primary:
|
||||
```javascript
|
||||
element.click();
|
||||
```
|
||||
|
||||
### P4: Huge DOM / Accessibility Tree
|
||||
|
||||
**Sites:** LinkedIn, Facebook, Twitter (feeds with 1000s of nodes)
|
||||
|
||||
**Detection:**
|
||||
```javascript
|
||||
document.querySelectorAll('*').length > 5000
|
||||
```
|
||||
|
||||
**Fix:**
|
||||
1. Add timeout to snapshot operation
|
||||
2. Truncate tree at 2000 nodes
|
||||
3. Fall back to DOM-based snapshot if accessibility tree too large
|
||||
|
||||
### P5: SPA Hydration Delay
|
||||
|
||||
**Sites:** React, Vue, Angular SPAs after navigation
|
||||
|
||||
**Detection:**
|
||||
```javascript
|
||||
// Check if React app has hydrated
|
||||
document.querySelector('[data-reactroot]') ||
|
||||
document.querySelector('[data-reactid]')
|
||||
```
|
||||
|
||||
**Fix:** Wait for specific selector after navigation:
|
||||
```python
|
||||
await browser_navigate(tab_id, url, wait_until="load")
|
||||
await browser_wait(tab_id, selector='[data-testid="content"]', timeout_ms=5000)
|
||||
```
|
||||
|
||||
### P6: Shadow DOM
|
||||
|
||||
**Sites:** Components using Shadow DOM, Lit elements
|
||||
|
||||
**Detection:**
|
||||
```javascript
|
||||
document.querySelectorAll('*').some(el => el.shadowRoot)
|
||||
```
|
||||
|
||||
**Fix:** Pierce shadow root:
|
||||
```javascript
|
||||
function queryShadow(selector) {
|
||||
const parts = selector.split('>>>');
|
||||
let node = document;
|
||||
for (const part of parts) {
|
||||
if (node.shadowRoot) {
|
||||
node = node.shadowRoot.querySelector(part.trim());
|
||||
} else {
|
||||
node = node.querySelector(part.trim());
|
||||
}
|
||||
}
|
||||
return node;
|
||||
}
|
||||
```
|
||||
|
||||
## Quick Reference
|
||||
|
||||
| Issue | Primary Fix | Fallback |
|
||||
|-------|-------------|----------|
|
||||
| Scroll not working | Find scrollable container | Mouse wheel at container center |
|
||||
| Click no effect | JavaScript click() | CDP mouse events |
|
||||
| Type clears | Add delay_ms | Use execCommand |
|
||||
| Snapshot hangs | Add timeout_s | DOM snapshot fallback |
|
||||
| Stale content | Wait for selector | Increase wait_until timeout |
|
||||
| Shadow DOM | Pierce selector | JavaScript traversal |
|
||||
|
||||
## References
|
||||
|
||||
- [registry.md](registry.md) - Full list of known edge cases
|
||||
- [scripts/test_case.py](scripts/test_case.py) - Template for testing new cases
|
||||
- [BROWSER_USE_PATTERNS.md](../../tools/BROWSER_USE_PATTERNS.md) - Implementation patterns from browser-use
|
||||
@@ -0,0 +1,232 @@
|
||||
# Browser Edge Case Registry
|
||||
|
||||
Curated list of known browser automation edge cases with symptoms, causes, and fixes.
|
||||
|
||||
---
|
||||
|
||||
## Scroll Issues
|
||||
|
||||
### #1: LinkedIn Nested Scroll Container
|
||||
|
||||
| Attribute | Value |
|
||||
|-----------|-------|
|
||||
| **Site** | LinkedIn (linkedin.com/feed) |
|
||||
| **Symptom** | `browser_scroll()` returns `{ok: true}` but page doesn't move |
|
||||
| **Root Cause** | Content is in a nested scrollable div (`overflow: scroll`), not the main window |
|
||||
| **Detection** | `document.querySelectorAll('*')` with `overflow: scroll/auto` has large candidates |
|
||||
| **Fix** | Find largest scrollable container, dispatch mouse wheel at its center coordinates |
|
||||
| **Code** | `bridge.py:808-981` - smart scroll with container detection |
|
||||
| **Verified** | 2026-04-02 |
|
||||
|
||||
### #2: Twitter/X Lazy Loading
|
||||
|
||||
| Attribute | Value |
|
||||
|-----------|-------|
|
||||
| **Site** | Twitter/X (x.com) |
|
||||
| **Symptom** | Infinite scroll doesn't load new content |
|
||||
| **Root Cause** | Lazy loading requires content to be visible before loading more |
|
||||
| **Detection** | Scroll position at bottom but no new `[data-testid="tweet"]` elements |
|
||||
| **Fix** | Add `wait_for_selector` between scroll calls with 1s delay |
|
||||
| **Code** | Test file: `tests/test_x_page_load_repro.py` |
|
||||
| **Verified** | - |
|
||||
|
||||
### #3: Modal/Dialog Scroll Container
|
||||
|
||||
| Attribute | Value |
|
||||
|-----------|-------|
|
||||
| **Site** | Any site with modal dialogs |
|
||||
| **Symptom** | Scroll scrolls background page, not modal content |
|
||||
| **Root Cause** | Modal has its own scroll container with `overflow: scroll` |
|
||||
| **Detection** | Visible element with `position: fixed` and scrollable content |
|
||||
| **Fix** | Find visible modal container (highest z-index scrollable), scroll that |
|
||||
| **Code** | - |
|
||||
| **Verified** | - |
|
||||
|
||||
---
|
||||
|
||||
## Click Issues
|
||||
|
||||
### #4: Element Covered by Overlay
|
||||
|
||||
| Attribute | Value |
|
||||
|-----------|-------|
|
||||
| **Site** | SPAs, sites with loading overlays |
|
||||
| **Symptom** | Click succeeds but no action triggered |
|
||||
| **Root Cause** | Element is covered by transparent overlay, tooltip, or iframe |
|
||||
| **Detection** | `document.elementFromPoint(x, y) !== target` |
|
||||
| **Fix** | Wait for overlay to disappear, or use JavaScript `element.click()` |
|
||||
| **Code** | `bridge.py:394-591` - JavaScript click as primary |
|
||||
| **Verified** | - |
|
||||
|
||||
### #5: React Synthetic Events
|
||||
|
||||
| Attribute | Value |
|
||||
|-----------|-------|
|
||||
| **Site** | React applications |
|
||||
| **Symptom** | CDP click doesn't trigger React handler |
|
||||
| **Root Cause** | React uses synthetic events that don't respond to CDP events |
|
||||
| **Detection** | Site uses React (check for `__reactFiber$` or `data-reactroot`) |
|
||||
| **Fix** | Use JavaScript `element.click()` as primary method |
|
||||
| **Code** | `bridge.py:394-591` - JavaScript-first click |
|
||||
| **Verified** | - |
|
||||
|
||||
### #6: Shadow DOM Elements
|
||||
|
||||
| Attribute | Value |
|
||||
|-----------|-------|
|
||||
| **Site** | Components using Shadow DOM, Lit elements |
|
||||
| **Symptom** | `querySelector` can't find element |
|
||||
| **Root Cause** | Element is inside a shadow root, not main DOM tree |
|
||||
| **Detection** | `element.shadowRoot !== null` on parent elements |
|
||||
| **Fix** | Use piercing selector (`host >>> target`) or traverse shadow roots |
|
||||
| **Code** | See SKILL.md P6 pattern |
|
||||
| **Verified** | - |
|
||||
|
||||
---
|
||||
|
||||
## Input Issues
|
||||
|
||||
### #7: ContentEditable / Rich Text Editors
|
||||
|
||||
| Attribute | Value |
|
||||
|-----------|-------|
|
||||
| **Site** | Rich text editors (Notion, Slack web, etc.) |
|
||||
| **Symptom** | `browser_type()` doesn't insert text |
|
||||
| **Root Cause** | Element is `contenteditable`, not an `<input>` or `<textarea>` |
|
||||
| **Detection** | `element.contentEditable === 'true'` |
|
||||
| **Fix** | Focus via JavaScript, use `execCommand('insertText')` or `Input.dispatchKeyEvent` |
|
||||
| **Code** | `bridge.py:616-694` - contentEditable handling |
|
||||
| **Verified** | - |
|
||||
|
||||
### #8: Autocomplete Field Clearing
|
||||
|
||||
| Attribute | Value |
|
||||
|-----------|-------|
|
||||
| **Site** | Search fields with autocomplete, address forms |
|
||||
| **Symptom** | Typed text gets cleared immediately |
|
||||
| **Root Cause** | Field expects realistic keystroke timing for autocomplete |
|
||||
| **Detection** | Field has autocomplete listeners or dropdown appears |
|
||||
| **Fix** | Add `delay_ms=50` between keystrokes |
|
||||
| **Code** | `bridge.py:type()` - delay_ms parameter |
|
||||
| **Verified** | - |
|
||||
|
||||
### #9: Custom Date Pickers
|
||||
|
||||
| Attribute | Value |
|
||||
|-----------|-------|
|
||||
| **Site** | Forms with custom date widgets |
|
||||
| **Symptom** | Can't type date into date field |
|
||||
| **Root Cause** | Custom widget intercepts and blocks keyboard input |
|
||||
| **Detection** | Typing doesn't change field value |
|
||||
| **Fix** | Click calendar widget icon, select date from dropdown |
|
||||
| **Code** | - |
|
||||
| **Verified** | - |
|
||||
|
||||
---
|
||||
|
||||
## Snapshot Issues
|
||||
|
||||
### #10: LinkedIn Huge DOM Tree
|
||||
|
||||
| Attribute | Value |
|
||||
|-----------|-------|
|
||||
| **Site** | LinkedIn, Facebook, Twitter feeds |
|
||||
| **Symptom** | `browser_snapshot()` hangs forever |
|
||||
| **Root Cause** | 10k+ DOM nodes, accessibility tree has 50k+ nodes |
|
||||
| **Detection** | `document.querySelectorAll('*').length > 5000` |
|
||||
| **Fix** | Add timeout (10s default), truncate tree at 2000 nodes |
|
||||
| **Code** | `bridge.py:1005-1050` - timeout_s param, max_nodes limit |
|
||||
| **Verified** | 2026-04-02 |
|
||||
|
||||
### #11: SPA Hydration Delay
|
||||
|
||||
| Attribute | Value |
|
||||
|-----------|-------|
|
||||
| **Site** | React/Vue/Angular SPAs |
|
||||
| **Symptom** | Snapshot shows old content after navigation |
|
||||
| **Root Cause** | Client-side hydration hasn't completed when snapshot runs |
|
||||
| **Detection** | `document.readyState === 'complete'` but content missing |
|
||||
| **Fix** | Wait for specific selector after navigation |
|
||||
| **Code** | Test file: `tests/test_x_page_load_repro.py` |
|
||||
| **Verified** | - |
|
||||
|
||||
### #12: iframe Content Missing
|
||||
|
||||
| Attribute | Value |
|
||||
|-----------|-------|
|
||||
| **Site** | Sites with embedded content |
|
||||
| **Symptom** | Snapshot missing iframe content |
|
||||
| **Root Cause** | Accessibility tree doesn't include iframe content |
|
||||
| **Detection** | `document.querySelectorAll('iframe')` has results |
|
||||
| **Fix** | Use `DOM.getFrameOwner` + separate snapshot for each iframe |
|
||||
| **Code** | - |
|
||||
| **Verified** | - |
|
||||
|
||||
---
|
||||
|
||||
## Navigation Issues
|
||||
|
||||
### #13: SPA Navigation Events
|
||||
|
||||
| Attribute | Value |
|
||||
|-----------|-------|
|
||||
| **Site** | React Router, Vue Router SPAs |
|
||||
| **Symptom** | `wait_until="load"` fires before content ready |
|
||||
| **Root Cause** | SPA uses client-side routing, no full page load |
|
||||
| **Detection** | URL changes but `load` event already fired |
|
||||
| **Fix** | Use `wait_until="networkidle"` or `wait_for_selector` |
|
||||
| **Code** | `bridge.py:navigate()` - wait_until options |
|
||||
| **Verified** | - |
|
||||
|
||||
### #14: Cross-Origin Redirects
|
||||
|
||||
| Attribute | Value |
|
||||
|-----------|-------|
|
||||
| **Site** | OAuth flows, SSO logins |
|
||||
| **Symptom** | Navigation fails during redirect |
|
||||
| **Root Cause** | Cross-origin security prevents CDP tracking |
|
||||
| **Detection** | URL changes to different domain |
|
||||
| **Fix** | Use `wait_for_url` with pattern matching instead of exact URL |
|
||||
| **Code** | - |
|
||||
| **Verified** | - |
|
||||
|
||||
---
|
||||
|
||||
## How to Add New Edge Cases
|
||||
|
||||
1. **Reproduce** the issue with minimal test case
|
||||
2. **Document** using the template below
|
||||
3. **Implement** fix with multi-layer fallback
|
||||
4. **Verify** against both problematic and simple sites
|
||||
5. **Submit** by appending to this file
|
||||
|
||||
### Template
|
||||
|
||||
```markdown
|
||||
### #N: [Short Title]
|
||||
|
||||
| Attribute | Value |
|
||||
|-----------|-------|
|
||||
| **Site** | [URL or site type] |
|
||||
| **Symptom** | [What the user observes] |
|
||||
| **Root Cause** | [Technical explanation] |
|
||||
| **Detection** | [JavaScript to detect this case] |
|
||||
| **Fix** | [Solution approach] |
|
||||
| **Code** | [File:line reference if implemented] |
|
||||
| **Verified** | [Date or "pending"] |
|
||||
```
|
||||
|
||||
---
|
||||
|
||||
## Statistics
|
||||
|
||||
| Category | Count |
|
||||
|----------|-------|
|
||||
| Scroll Issues | 3 |
|
||||
| Click Issues | 3 |
|
||||
| Input Issues | 3 |
|
||||
| Snapshot Issues | 3 |
|
||||
| Navigation Issues | 2 |
|
||||
| **Total** | **14** |
|
||||
|
||||
Last updated: 2026-04-02
|
||||
@@ -0,0 +1,111 @@
|
||||
#!/usr/bin/env python
|
||||
"""
|
||||
Test #2: Twitter/X Lazy Loading Scroll
|
||||
|
||||
Symptom: Infinite scroll doesn't load new content
|
||||
Root Cause: Lazy loading requires content to be visible before loading more
|
||||
Fix: Add wait_for_selector between scroll calls
|
||||
"""
|
||||
|
||||
import asyncio
|
||||
import sys
|
||||
import time
|
||||
from pathlib import Path
|
||||
|
||||
sys.path.insert(0, str(Path(__file__).parent.parent.parent.parent / "tools" / "src"))
|
||||
|
||||
from gcu.browser.bridge import BeelineBridge
|
||||
|
||||
BRIDGE_PORT = 9229
|
||||
CONTEXT_NAME = "twitter-scroll-test"
|
||||
|
||||
|
||||
async def test_twitter_lazy_scroll():
|
||||
"""Test that repeated scrolls with waits load new content."""
|
||||
print("=" * 70)
|
||||
print("TEST #2: Twitter/X Lazy Loading Scroll")
|
||||
print("=" * 70)
|
||||
|
||||
bridge = BeelineBridge()
|
||||
|
||||
try:
|
||||
await bridge.start()
|
||||
|
||||
for i in range(10):
|
||||
await asyncio.sleep(1)
|
||||
if bridge.is_connected:
|
||||
print("✓ Extension connected!")
|
||||
break
|
||||
print(f"Waiting for extension... ({i+1}/10)")
|
||||
else:
|
||||
print("✗ Extension not connected")
|
||||
return
|
||||
|
||||
context = await bridge.create_context(CONTEXT_NAME)
|
||||
tab_id = context.get("tabId")
|
||||
group_id = context.get("groupId")
|
||||
print(f"✓ Created tab: {tab_id}")
|
||||
|
||||
# Navigate to Twitter/X
|
||||
print("\n--- Navigating to X.com ---")
|
||||
await bridge.navigate(tab_id, "https://x.com", wait_until="networkidle", timeout_ms=30000)
|
||||
print("✓ Page loaded")
|
||||
|
||||
# Wait for tweets to appear
|
||||
print("\n--- Waiting for tweets ---")
|
||||
await bridge.wait_for_selector(tab_id, '[data-testid="tweet"]', timeout_ms=10000)
|
||||
|
||||
# Count initial tweets
|
||||
initial_count = await bridge.evaluate(
|
||||
tab_id,
|
||||
'(function() { return document.querySelectorAll(\'[data-testid="tweet"]\').length; })()'
|
||||
)
|
||||
print(f"Initial tweet count: {initial_count.get('result', 0)}")
|
||||
|
||||
# Take screenshot of initial state
|
||||
screenshot = await bridge.screenshot(tab_id)
|
||||
print(f"Screenshot: {len(screenshot.get('data', ''))} bytes")
|
||||
|
||||
# Scroll multiple times with waits
|
||||
print("\n--- Scrolling with waits ---")
|
||||
for i in range(3):
|
||||
result = await bridge.scroll(tab_id, "down", 500)
|
||||
print(f" Scroll {i+1}: {result.get('method', 'unknown')} method")
|
||||
|
||||
# Wait for new content to load
|
||||
await asyncio.sleep(2)
|
||||
|
||||
# Count tweets after scroll
|
||||
count_result = await bridge.evaluate(
|
||||
tab_id,
|
||||
'(function() { return document.querySelectorAll(\'[data-testid="tweet"]\').length; })()'
|
||||
)
|
||||
count = count_result.get('result', 0)
|
||||
print(f" Tweet count after scroll: {count}")
|
||||
|
||||
# Final count
|
||||
final_count = await bridge.evaluate(
|
||||
tab_id,
|
||||
'(function() { return document.querySelectorAll(\'[data-testid="tweet"]\').length; })()'
|
||||
)
|
||||
final = final_count.get('result', 0)
|
||||
initial = initial_count.get('result', 0)
|
||||
|
||||
print(f"\n--- Results ---")
|
||||
print(f"Initial tweets: {initial}")
|
||||
print(f"Final tweets: {final}")
|
||||
|
||||
if final > initial:
|
||||
print(f"✓ PASS: Loaded {final - initial} new tweets")
|
||||
else:
|
||||
print("✗ FAIL: No new tweets loaded (may need login)")
|
||||
|
||||
await bridge.destroy_context(group_id)
|
||||
print("\n✓ Context destroyed")
|
||||
|
||||
finally:
|
||||
await bridge.stop()
|
||||
|
||||
|
||||
if __name__ == "__main__":
|
||||
asyncio.run(test_twitter_lazy_scroll())
|
||||
@@ -0,0 +1,97 @@
|
||||
#!/usr/bin/env python
|
||||
"""
|
||||
Test #3: Modal/Dialog Scroll Container
|
||||
|
||||
Symptom: Scroll scrolls background page, not modal content
|
||||
Root Cause: Modal has its own scroll container with overflow: scroll
|
||||
Fix: Find visible modal container (highest z-index scrollable), scroll that
|
||||
"""
|
||||
|
||||
import asyncio
|
||||
import sys
|
||||
import time
|
||||
from pathlib import Path
|
||||
|
||||
sys.path.insert(0, str(Path(__file__).parent.parent.parent.parent / "tools" / "src"))
|
||||
|
||||
from gcu.browser.bridge import BeelineBridge
|
||||
|
||||
BRIDGE_PORT = 9229
|
||||
CONTEXT_NAME = "modal-scroll-test"
|
||||
|
||||
# Test site with modal - using a demo site
|
||||
MODAL_DEMO_URL = "https://www.w3schools.com/howto/howto_css_modals.asp"
|
||||
|
||||
|
||||
async def test_modal_scroll():
|
||||
"""Test that scroll targets modal content, not background."""
|
||||
print("=" * 70)
|
||||
print("TEST #3: Modal/Dialog Scroll Container")
|
||||
print("=" * 70)
|
||||
|
||||
bridge = BeelineBridge()
|
||||
|
||||
try:
|
||||
await bridge.start()
|
||||
|
||||
for i in range(10):
|
||||
await asyncio.sleep(1)
|
||||
if bridge.is_connected:
|
||||
print("✓ Extension connected!")
|
||||
break
|
||||
else:
|
||||
print("✗ Extension not connected")
|
||||
return
|
||||
|
||||
context = await bridge.create_context(CONTEXT_NAME)
|
||||
tab_id = context.get("tabId")
|
||||
group_id = context.get("groupId")
|
||||
print(f"✓ Created tab: {tab_id}")
|
||||
|
||||
# Navigate to modal demo
|
||||
print("\n--- Navigating to modal demo ---")
|
||||
await bridge.navigate(tab_id, MODAL_DEMO_URL, wait_until="load")
|
||||
print("✓ Page loaded")
|
||||
|
||||
# Take screenshot before
|
||||
screenshot_before = await bridge.screenshot(tab_id)
|
||||
print(f"Screenshot before: {len(screenshot_before.get('data', ''))} bytes")
|
||||
|
||||
# Click button to open modal
|
||||
print("\n--- Opening modal ---")
|
||||
# Find and click the "Open Modal" button
|
||||
result = await bridge.click(tab_id, '.ws-btn', timeout_ms=5000)
|
||||
print(f"Click result: {result}")
|
||||
|
||||
await asyncio.sleep(1)
|
||||
|
||||
# Take screenshot with modal open
|
||||
screenshot_modal = await bridge.screenshot(tab_id)
|
||||
print(f"Screenshot modal open: {len(screenshot_modal.get('data', ''))} bytes")
|
||||
|
||||
# Try to scroll within modal
|
||||
print("\n--- Scrolling modal content ---")
|
||||
result = await bridge.scroll(tab_id, "down", 100)
|
||||
print(f"Scroll result: {result}")
|
||||
|
||||
await asyncio.sleep(0.5)
|
||||
|
||||
# Take screenshot after scroll
|
||||
screenshot_after = await bridge.screenshot(tab_id)
|
||||
print(f"Screenshot after scroll: {len(screenshot_after.get('data', ''))} bytes")
|
||||
|
||||
# Check if modal content scrolled (not background)
|
||||
# This is a visual check - we can verify by comparing screenshots
|
||||
print("\n--- Results ---")
|
||||
print(f"Modal scroll test completed. Method used: {result.get('method', 'unknown')}")
|
||||
print("Visual verification needed: Check if modal content scrolled vs background")
|
||||
|
||||
await bridge.destroy_context(group_id)
|
||||
print("\n✓ Context destroyed")
|
||||
|
||||
finally:
|
||||
await bridge.stop()
|
||||
|
||||
|
||||
if __name__ == "__main__":
|
||||
asyncio.run(test_modal_scroll())
|
||||
@@ -0,0 +1,123 @@
|
||||
#!/usr/bin/env python
|
||||
"""
|
||||
Test #4: Element Covered by Overlay
|
||||
|
||||
Symptom: Click succeeds but no action triggered
|
||||
Root Cause: Element is covered by transparent overlay, tooltip, or iframe
|
||||
Detection: document.elementFromPoint(x, y) !== target
|
||||
Fix: Wait for overlay to disappear, or use JavaScript element.click()
|
||||
"""
|
||||
|
||||
import asyncio
|
||||
import sys
|
||||
from pathlib import Path
|
||||
|
||||
sys.path.insert(0, str(Path(__file__).parent.parent.parent.parent / "tools" / "src"))
|
||||
|
||||
from gcu.browser.bridge import BeelineBridge
|
||||
|
||||
CONTEXT_NAME = "overlay-click-test"
|
||||
|
||||
|
||||
async def test_overlay_click():
|
||||
"""Test clicking elements that are covered by overlays."""
|
||||
print("=" * 70)
|
||||
print("TEST #4: Element Covered by Overlay")
|
||||
print("=" * 70)
|
||||
|
||||
bridge = BeelineBridge()
|
||||
|
||||
try:
|
||||
await bridge.start()
|
||||
|
||||
for i in range(10):
|
||||
await asyncio.sleep(1)
|
||||
if bridge.is_connected:
|
||||
print("✓ Extension connected!")
|
||||
break
|
||||
else:
|
||||
print("✗ Extension not connected")
|
||||
return
|
||||
|
||||
context = await bridge.create_context(CONTEXT_NAME)
|
||||
tab_id = context.get("tabId")
|
||||
group_id = context.get("groupId")
|
||||
print(f"✓ Created tab: {tab_id}")
|
||||
|
||||
# Create a test page with overlay
|
||||
print("\n--- Creating test page with overlay ---")
|
||||
test_html = """
|
||||
<!DOCTYPE html>
|
||||
<html>
|
||||
<head><title>Overlay Test</title></head>
|
||||
<body>
|
||||
<button id="target-btn" onclick="alert('Clicked!')">Click Me</button>
|
||||
<div id="overlay" style="position:fixed;top:0;left:0;width:100%;height:100%;background:rgba(0,0,0,0.3);z-index:1000;"></div>
|
||||
<script>
|
||||
window.clickCount = 0;
|
||||
document.getElementById('target-btn').addEventListener('click', () => {
|
||||
window.clickCount++;
|
||||
});
|
||||
</script>
|
||||
</body>
|
||||
</html>
|
||||
"""
|
||||
|
||||
# Navigate to data URL
|
||||
import base64
|
||||
data_url = f"data:text/html;base64,{base64.b64encode(test_html.encode()).decode()}"
|
||||
await bridge.navigate(tab_id, data_url, wait_until="load")
|
||||
|
||||
# Screenshot before
|
||||
screenshot = await bridge.screenshot(tab_id)
|
||||
print(f"Screenshot: {len(screenshot.get('data', ''))} bytes")
|
||||
|
||||
# Try to click the covered button
|
||||
print("\n--- Attempting to click covered button ---")
|
||||
|
||||
# First, check if element is covered
|
||||
coverage_check = await bridge.evaluate(
|
||||
tab_id,
|
||||
"""
|
||||
(function() {
|
||||
const btn = document.getElementById('target-btn');
|
||||
const rect = btn.getBoundingClientRect();
|
||||
const centerX = rect.left + rect.width / 2;
|
||||
const centerY = rect.top + rect.height / 2;
|
||||
const topElement = document.elementFromPoint(centerX, centerY);
|
||||
return {
|
||||
isCovered: topElement !== btn && !btn.contains(topElement),
|
||||
topElement: topElement?.tagName,
|
||||
targetElement: btn.tagName
|
||||
};
|
||||
})();
|
||||
"""
|
||||
)
|
||||
print(f"Coverage check: {coverage_check.get('result', {})}")
|
||||
|
||||
# Try CDP click (may fail due to overlay)
|
||||
click_result = await bridge.click(tab_id, "#target-btn", timeout_ms=5000)
|
||||
print(f"Click result: {click_result}")
|
||||
|
||||
# Check if click registered
|
||||
count_result = await bridge.evaluate(
|
||||
tab_id,
|
||||
"(function() { return window.clickCount; })()"
|
||||
)
|
||||
count = count_result.get("result", 0)
|
||||
print(f"Click count after CDP click: {count}")
|
||||
|
||||
if count > 0:
|
||||
print("✓ PASS: JavaScript click penetrated overlay")
|
||||
else:
|
||||
print("✗ FAIL: Click did not reach button (overlay blocked it)")
|
||||
|
||||
await bridge.destroy_context(group_id)
|
||||
print("\n✓ Context destroyed")
|
||||
|
||||
finally:
|
||||
await bridge.stop()
|
||||
|
||||
|
||||
if __name__ == "__main__":
|
||||
asyncio.run(test_overlay_click())
|
||||
@@ -0,0 +1,151 @@
|
||||
#!/usr/bin/env python
|
||||
"""
|
||||
Test #6: Shadow DOM Elements
|
||||
|
||||
Symptom: querySelector can't find element
|
||||
Root Cause: Element is inside a shadow root, not main DOM tree
|
||||
Detection: element.shadowRoot !== null on parent elements
|
||||
Fix: Use piercing selector (host >>> target) or traverse shadow roots
|
||||
"""
|
||||
|
||||
import asyncio
|
||||
import sys
|
||||
import base64
|
||||
from pathlib import Path
|
||||
|
||||
sys.path.insert(0, str(Path(__file__).parent.parent.parent.parent / "tools" / "src"))
|
||||
|
||||
from gcu.browser.bridge import BeelineBridge
|
||||
|
||||
CONTEXT_NAME = "shadow-dom-test"
|
||||
|
||||
|
||||
async def test_shadow_dom():
|
||||
"""Test clicking elements inside Shadow DOM."""
|
||||
print("=" * 70)
|
||||
print("TEST #6: Shadow DOM Elements")
|
||||
print("=" * 70)
|
||||
|
||||
bridge = BeelineBridge()
|
||||
|
||||
try:
|
||||
await bridge.start()
|
||||
|
||||
for i in range(10):
|
||||
await asyncio.sleep(1)
|
||||
if bridge.is_connected:
|
||||
print("✓ Extension connected!")
|
||||
break
|
||||
else:
|
||||
print("✗ Extension not connected")
|
||||
return
|
||||
|
||||
context = await bridge.create_context(CONTEXT_NAME)
|
||||
tab_id = context.get("tabId")
|
||||
group_id = context.get("groupId")
|
||||
print(f"✓ Created tab: {tab_id}")
|
||||
|
||||
# Create test page with Shadow DOM
|
||||
print("\n--- Creating test page with Shadow DOM ---")
|
||||
test_html = """
|
||||
<!DOCTYPE html>
|
||||
<html>
|
||||
<head><title>Shadow DOM Test</title></head>
|
||||
<body>
|
||||
<div id="shadow-host"></div>
|
||||
<script>
|
||||
const host = document.getElementById('shadow-host');
|
||||
const shadow = host.attachShadow({ mode: 'open' });
|
||||
shadow.innerHTML = `
|
||||
<style>
|
||||
button { padding: 10px 20px; font-size: 16px; }
|
||||
</style>
|
||||
<button id="shadow-btn">Shadow Button</button>
|
||||
`;
|
||||
shadow.getElementById('shadow-btn').addEventListener('click', () => {
|
||||
window.shadowClickCount = (window.shadowClickCount || 0) + 1;
|
||||
console.log('Shadow button clicked:', window.shadowClickCount);
|
||||
});
|
||||
</script>
|
||||
</body>
|
||||
</html>
|
||||
"""
|
||||
|
||||
data_url = f"data:text/html;base64,{base64.b64encode(test_html.encode()).decode()}"
|
||||
await bridge.navigate(tab_id, data_url, wait_until="load")
|
||||
print("✓ Page loaded")
|
||||
|
||||
# Screenshot
|
||||
screenshot = await bridge.screenshot(tab_id)
|
||||
print(f"Screenshot: {len(screenshot.get('data', ''))} bytes")
|
||||
|
||||
# Detect Shadow DOM
|
||||
print("\n--- Detecting Shadow DOM ---")
|
||||
detection = await bridge.evaluate(
|
||||
tab_id,
|
||||
"""
|
||||
(function() {
|
||||
const hosts = [];
|
||||
document.querySelectorAll('*').forEach(el => {
|
||||
if (el.shadowRoot) {
|
||||
hosts.push({
|
||||
tag: el.tagName,
|
||||
id: el.id,
|
||||
hasButton: el.shadowRoot.querySelector('button') !== null
|
||||
});
|
||||
}
|
||||
});
|
||||
return { count: hosts.length, hosts };
|
||||
})();
|
||||
"""
|
||||
)
|
||||
print(f"Shadow DOM detection: {detection.get('result', {})}")
|
||||
|
||||
# Try to click shadow button using regular selector (should fail)
|
||||
print("\n--- Attempting click with regular selector ---")
|
||||
try:
|
||||
result = await bridge.click(tab_id, "#shadow-btn", timeout_ms=3000)
|
||||
print(f"Result: {result}")
|
||||
except Exception as e:
|
||||
print(f"Expected failure: {e}")
|
||||
|
||||
# Try to click using JavaScript that pierces shadow DOM
|
||||
print("\n--- Clicking via JavaScript shadow piercing ---")
|
||||
click_result = await bridge.evaluate(
|
||||
tab_id,
|
||||
"""
|
||||
(function() {
|
||||
const host = document.getElementById('shadow-host');
|
||||
const btn = host.shadowRoot.getElementById('shadow-btn');
|
||||
if (btn) {
|
||||
btn.click();
|
||||
return { success: true, clicked: 'shadow-btn' };
|
||||
}
|
||||
return { success: false, error: 'Button not found' };
|
||||
})();
|
||||
"""
|
||||
)
|
||||
print(f"JS click result: {click_result.get('result', {})}")
|
||||
|
||||
# Verify click was registered
|
||||
count_result = await bridge.evaluate(
|
||||
tab_id,
|
||||
"(function() { return window.shadowClickCount || 0; })()"
|
||||
)
|
||||
count = count_result.get("result", 0)
|
||||
print(f"Shadow click count: {count}")
|
||||
|
||||
if count > 0:
|
||||
print("✓ PASS: Shadow DOM element clicked successfully")
|
||||
else:
|
||||
print("✗ FAIL: Could not click Shadow DOM element")
|
||||
|
||||
await bridge.destroy_context(group_id)
|
||||
print("\n✓ Context destroyed")
|
||||
|
||||
finally:
|
||||
await bridge.stop()
|
||||
|
||||
|
||||
if __name__ == "__main__":
|
||||
asyncio.run(test_shadow_dom())
|
||||
@@ -0,0 +1,169 @@
|
||||
#!/usr/bin/env python
|
||||
"""
|
||||
Test #7: ContentEditable / Rich Text Editors
|
||||
|
||||
Symptom: browser_type() doesn't insert text
|
||||
Root Cause: Element is contenteditable, not an <input> or <textarea>
|
||||
Detection: element.contentEditable === 'true'
|
||||
Fix: Focus via JavaScript, use execCommand('insertText') or Input.dispatchKeyEvent
|
||||
"""
|
||||
|
||||
import asyncio
|
||||
import sys
|
||||
import base64
|
||||
from pathlib import Path
|
||||
|
||||
sys.path.insert(0, str(Path(__file__).parent.parent.parent.parent / "tools" / "src"))
|
||||
|
||||
from gcu.browser.bridge import BeelineBridge
|
||||
|
||||
CONTEXT_NAME = "contenteditable-test"
|
||||
|
||||
|
||||
async def test_contenteditable():
|
||||
"""Test typing into contenteditable elements."""
|
||||
print("=" * 70)
|
||||
print("TEST #7: ContentEditable / Rich Text Editors")
|
||||
print("=" * 70)
|
||||
|
||||
bridge = BeelineBridge()
|
||||
|
||||
try:
|
||||
await bridge.start()
|
||||
|
||||
for i in range(10):
|
||||
await asyncio.sleep(1)
|
||||
if bridge.is_connected:
|
||||
print("✓ Extension connected!")
|
||||
break
|
||||
else:
|
||||
print("✗ Extension not connected")
|
||||
return
|
||||
|
||||
context = await bridge.create_context(CONTEXT_NAME)
|
||||
tab_id = context.get("tabId")
|
||||
group_id = context.get("groupId")
|
||||
print(f"✓ Created tab: {tab_id}")
|
||||
|
||||
# Create test page with contenteditable
|
||||
test_html = """
|
||||
<!DOCTYPE html>
|
||||
<html>
|
||||
<head><title>ContentEditable Test</title></head>
|
||||
<body>
|
||||
<h2>ContentEditable Test</h2>
|
||||
|
||||
<h3>1. Simple contenteditable div</h3>
|
||||
<div id="editor1" contenteditable="true" style="border:1px solid #ccc;padding:10px;min-height:50px;">Start text</div>
|
||||
|
||||
<h3>2. Rich text editor (like Notion)</h3>
|
||||
<div id="editor2" contenteditable="true" style="border:1px solid #ccc;padding:10px;min-height:50px;">
|
||||
<p>Type here...</p>
|
||||
</div>
|
||||
|
||||
<h3>3. Regular input (for comparison)</h3>
|
||||
<input id="input1" type="text" placeholder="Regular input" />
|
||||
|
||||
<script>
|
||||
// Track content changes
|
||||
window.editor1Content = '';
|
||||
window.editor2Content = '';
|
||||
|
||||
document.getElementById('editor1').addEventListener('input', (e) => {
|
||||
window.editor1Content = e.target.innerText;
|
||||
});
|
||||
document.getElementById('editor2').addEventListener('input', (e) => {
|
||||
window.editor2Content = e.target.innerText;
|
||||
});
|
||||
</script>
|
||||
</body>
|
||||
</html>
|
||||
"""
|
||||
|
||||
data_url = f"data:text/html;base64,{base64.b64encode(test_html.encode()).decode()}"
|
||||
await bridge.navigate(tab_id, data_url, wait_until="load")
|
||||
print("✓ Page loaded")
|
||||
|
||||
# Screenshot
|
||||
screenshot = await bridge.screenshot(tab_id)
|
||||
print(f"Screenshot: {len(screenshot.get('data', ''))} bytes")
|
||||
|
||||
# Detect contenteditable
|
||||
print("\n--- Detecting contenteditable elements ---")
|
||||
detection = await bridge.evaluate(
|
||||
tab_id,
|
||||
"""
|
||||
(function() {
|
||||
const editables = document.querySelectorAll('[contenteditable="true"]');
|
||||
return {
|
||||
count: editables.length,
|
||||
ids: Array.from(editables).map(el => el.id)
|
||||
};
|
||||
})();
|
||||
"""
|
||||
)
|
||||
print(f"Contenteditable detection: {detection.get('result', {})}")
|
||||
|
||||
# Test 1: Type into regular input (baseline)
|
||||
print("\n--- Test 1: Regular input ---")
|
||||
await bridge.click(tab_id, "#input1")
|
||||
await bridge.type(tab_id, "#input1", "Hello input")
|
||||
input_result = await bridge.evaluate(
|
||||
tab_id,
|
||||
"(function() { return document.getElementById('input1').value; })()"
|
||||
)
|
||||
print(f"Input value: {input_result.get('result', '')}")
|
||||
|
||||
# Test 2: Type into contenteditable div
|
||||
print("\n--- Test 2: Contenteditable div ---")
|
||||
await bridge.click(tab_id, "#editor1")
|
||||
await bridge.type(tab_id, "#editor1", "Hello contenteditable", clear_first=True)
|
||||
editor_result = await bridge.evaluate(
|
||||
tab_id,
|
||||
"(function() { return document.getElementById('editor1').innerText; })()"
|
||||
)
|
||||
print(f"Editor1 innerText: {editor_result.get('result', '')}")
|
||||
|
||||
# Test 3: Use JavaScript insertText for rich editor
|
||||
print("\n--- Test 3: JavaScript insertText for rich editor ---")
|
||||
insert_result = await bridge.evaluate(
|
||||
tab_id,
|
||||
"""
|
||||
(function() {
|
||||
const editor = document.getElementById('editor2');
|
||||
editor.focus();
|
||||
document.execCommand('selectAll', false, null);
|
||||
document.execCommand('insertText', false, 'Hello from execCommand');
|
||||
return editor.innerText;
|
||||
})();
|
||||
"""
|
||||
)
|
||||
print(f"Editor2 after execCommand: {insert_result.get('result', '')}")
|
||||
|
||||
# Screenshot after
|
||||
screenshot_after = await bridge.screenshot(tab_id)
|
||||
print(f"Screenshot after: {len(screenshot_after.get('data', ''))} bytes")
|
||||
|
||||
# Results
|
||||
print("\n--- Results ---")
|
||||
input_val = input_result.get("result", "")
|
||||
editor1_val = editor_result.get("result", "")
|
||||
editor2_val = insert_result.get("result", "")
|
||||
|
||||
input_pass = "Hello input" in input_val
|
||||
editor1_pass = "Hello contenteditable" in editor1_val
|
||||
editor2_pass = "execCommand" in editor2_val
|
||||
|
||||
print(f"Input: {'✓ PASS' if input_pass else '✗ FAIL'} - {input_val}")
|
||||
print(f"Editor1: {'✓ PASS' if editor1_pass else '✗ FAIL'} - {editor1_val}")
|
||||
print(f"Editor2: {'✓ PASS' if editor2_pass else '✗ FAIL'} - {editor2_val}")
|
||||
|
||||
await bridge.destroy_context(group_id)
|
||||
print("\n✓ Context destroyed")
|
||||
|
||||
finally:
|
||||
await bridge.stop()
|
||||
|
||||
|
||||
if __name__ == "__main__":
|
||||
asyncio.run(test_contenteditable())
|
||||
@@ -0,0 +1,230 @@
|
||||
#!/usr/bin/env python
|
||||
"""
|
||||
Test #8: Autocomplete Field Clearing
|
||||
|
||||
Symptom: Typed text gets cleared immediately
|
||||
Root Cause: Field expects realistic keystroke timing for autocomplete
|
||||
Detection: Field has autocomplete listeners or dropdown appears
|
||||
Fix: Add delay_ms between keystrokes
|
||||
"""
|
||||
|
||||
import asyncio
|
||||
import sys
|
||||
import base64
|
||||
from pathlib import Path
|
||||
|
||||
sys.path.insert(0, str(Path(__file__).parent.parent.parent.parent / "tools" / "src"))
|
||||
|
||||
from gcu.browser.bridge import BeelineBridge
|
||||
|
||||
CONTEXT_NAME = "autocomplete-test"
|
||||
|
||||
|
||||
async def test_autocomplete():
|
||||
"""Test typing into fields with autocomplete behavior."""
|
||||
print("=" * 70)
|
||||
print("TEST #8: Autocomplete Field Clearing")
|
||||
print("=" * 70)
|
||||
|
||||
bridge = BeelineBridge()
|
||||
|
||||
try:
|
||||
await bridge.start()
|
||||
|
||||
for i in range(10):
|
||||
await asyncio.sleep(1)
|
||||
if bridge.is_connected:
|
||||
print("✓ Extension connected!")
|
||||
break
|
||||
else:
|
||||
print("✗ Extension not connected")
|
||||
return
|
||||
|
||||
context = await bridge.create_context(CONTEXT_NAME)
|
||||
tab_id = context.get("tabId")
|
||||
group_id = context.get("groupId")
|
||||
print(f"✓ Created tab: {tab_id}")
|
||||
|
||||
# Create test page with autocomplete behavior
|
||||
test_html = """
|
||||
<!DOCTYPE html>
|
||||
<html>
|
||||
<head><title>Autocomplete Test</title>
|
||||
<style>
|
||||
.autocomplete-items {
|
||||
position: absolute;
|
||||
border: 1px solid #d4d4d4;
|
||||
border-top: none;
|
||||
z-index: 99;
|
||||
top: 100%;
|
||||
left: 0;
|
||||
right: 0;
|
||||
max-height: 200px;
|
||||
overflow-y: auto;
|
||||
background: white;
|
||||
}
|
||||
.autocomplete-items div {
|
||||
padding: 10px;
|
||||
cursor: pointer;
|
||||
}
|
||||
.autocomplete-items div:hover {
|
||||
background-color: #e9e9e9;
|
||||
}
|
||||
.autocomplete-active {
|
||||
background-color: DodgerBlue !important;
|
||||
color: white;
|
||||
}
|
||||
.autocomplete { position: relative; display: inline-block; }
|
||||
input { width: 300px; padding: 10px; font-size: 16px; }
|
||||
</style></head>
|
||||
<body>
|
||||
<h2>Autocomplete Test</h2>
|
||||
|
||||
<div class="autocomplete">
|
||||
<input id="search" type="text" placeholder="Search countries..." autocomplete="off">
|
||||
</div>
|
||||
|
||||
<div id="log" style="margin-top:20px;font-family:monospace;"></div>
|
||||
|
||||
<script>
|
||||
const countries = ["Afghanistan","Albania","Algeria","Andorra","Angola","Argentina","Armenia","Australia","Austria","Azerbaijan","Bahamas","Bahrain","Bangladesh","Belarus","Belgium","Belize","Benin","Bhutan","Bolivia","Brazil","Canada","China","Colombia","Denmark","Egypt","France","Germany","India","Indonesia","Italy","Japan","Mexico","Netherlands","Nigeria","Norway","Pakistan","Peru","Philippines","Poland","Portugal","Russia","Spain","Sweden","Switzerland","Thailand","Turkey","Ukraine","United Kingdom","United States","Vietnam"];
|
||||
|
||||
const input = document.getElementById('search');
|
||||
const log = document.getElementById('log');
|
||||
let currentFocus = -1;
|
||||
let typingTimeout = null;
|
||||
|
||||
// Track events for testing
|
||||
window.inputEvents = [];
|
||||
window.inputValue = '';
|
||||
|
||||
function logEvent(type, value) {
|
||||
window.inputEvents.push({ type, value, time: Date.now() });
|
||||
const entry = document.createElement('div');
|
||||
entry.textContent = type + ': ' + value;
|
||||
log.insertBefore(entry, log.firstChild);
|
||||
}
|
||||
|
||||
// Simulate autocomplete that clears fast typing
|
||||
input.addEventListener('input', function(e) {
|
||||
const val = this.value;
|
||||
|
||||
// Clear previous dropdown
|
||||
closeAllLists();
|
||||
|
||||
if (!val) return;
|
||||
|
||||
// If typing too fast (autocomplete-style), clear and restart
|
||||
clearTimeout(typingTimeout);
|
||||
typingTimeout = setTimeout(() => {
|
||||
logEvent('input', val);
|
||||
window.inputValue = val;
|
||||
|
||||
// Create dropdown
|
||||
const div = document.createElement('div');
|
||||
div.setAttribute('id', this.id + 'autocomplete-list');
|
||||
div.setAttribute('class', 'autocomplete-items');
|
||||
this.parentNode.appendChild(div);
|
||||
|
||||
countries.filter(c => c.substr(0, val.length).toUpperCase() === val.toUpperCase())
|
||||
.slice(0, 5)
|
||||
.forEach(country => {
|
||||
const item = document.createElement('div');
|
||||
item.innerHTML = '<strong>' + country.substr(0, val.length) + '</strong>' + country.substr(val.length);
|
||||
item.addEventListener('click', function() {
|
||||
input.value = country;
|
||||
closeAllLists();
|
||||
logEvent('select', country);
|
||||
window.inputValue = country;
|
||||
});
|
||||
div.appendChild(item);
|
||||
});
|
||||
}, 100); // 100ms debounce
|
||||
});
|
||||
|
||||
function closeAllLists() {
|
||||
document.querySelectorAll('.autocomplete-items').forEach(el => el.remove());
|
||||
}
|
||||
|
||||
document.addEventListener('click', function() {
|
||||
closeAllLists();
|
||||
});
|
||||
</script>
|
||||
</body>
|
||||
</html>
|
||||
"""
|
||||
|
||||
data_url = f"data:text/html;base64,{base64.b64encode(test_html.encode()).decode()}"
|
||||
await bridge.navigate(tab_id, data_url, wait_until="load")
|
||||
print("✓ Page loaded")
|
||||
|
||||
# Screenshot
|
||||
screenshot = await bridge.screenshot(tab_id)
|
||||
print(f"Screenshot: {len(screenshot.get('data', ''))} bytes")
|
||||
|
||||
# Test 1: Fast typing (no delay) - may fail
|
||||
print("\n--- Test 1: Fast typing (delay_ms=0) ---")
|
||||
await bridge.click(tab_id, "#search")
|
||||
await bridge.type(tab_id, "#search", "Ger", clear_first=True, delay_ms=0)
|
||||
await asyncio.sleep(0.5)
|
||||
|
||||
fast_result = await bridge.evaluate(
|
||||
tab_id,
|
||||
"(function() { return document.getElementById('search').value; })()"
|
||||
)
|
||||
fast_value = fast_result.get("result", "")
|
||||
print(f"Value after fast typing: '{fast_value}'")
|
||||
|
||||
# Check events
|
||||
events_result = await bridge.evaluate(
|
||||
tab_id,
|
||||
"(function() { return window.inputEvents; })()"
|
||||
)
|
||||
print(f"Events logged: {events_result.get('result', [])}")
|
||||
|
||||
# Test 2: Slow typing (with delay) - should work
|
||||
print("\n--- Test 2: Slow typing (delay_ms=100) ---")
|
||||
await bridge.click(tab_id, "#search")
|
||||
await bridge.type(tab_id, "#search", "United", clear_first=True, delay_ms=100)
|
||||
await asyncio.sleep(0.5)
|
||||
|
||||
slow_result = await bridge.evaluate(
|
||||
tab_id,
|
||||
"(function() { return document.getElementById('search').value; })()"
|
||||
)
|
||||
slow_value = slow_result.get("result", "")
|
||||
print(f"Value after slow typing: '{slow_value}'")
|
||||
|
||||
# Check if dropdown appeared
|
||||
dropdown_result = await bridge.evaluate(
|
||||
tab_id,
|
||||
"(function() { return document.querySelectorAll('.autocomplete-items div').length; })()"
|
||||
)
|
||||
dropdown_count = dropdown_result.get("result", 0)
|
||||
print(f"Dropdown items: {dropdown_count}")
|
||||
|
||||
# Screenshot with dropdown
|
||||
screenshot_dropdown = await bridge.screenshot(tab_id)
|
||||
print(f"Screenshot with dropdown: {len(screenshot_dropdown.get('data', ''))} bytes")
|
||||
|
||||
# Results
|
||||
print("\n--- Results ---")
|
||||
if "United" in slow_value:
|
||||
print("✓ PASS: Slow typing with delay_ms worked")
|
||||
else:
|
||||
print("✗ FAIL: Slow typing still didn't work")
|
||||
|
||||
if dropdown_count > 0:
|
||||
print("✓ PASS: Autocomplete dropdown appeared")
|
||||
else:
|
||||
print("⚠ WARNING: No autocomplete dropdown")
|
||||
|
||||
await bridge.destroy_context(group_id)
|
||||
print("\n✓ Context destroyed")
|
||||
|
||||
finally:
|
||||
await bridge.stop()
|
||||
|
||||
|
||||
if __name__ == "__main__":
|
||||
asyncio.run(test_autocomplete())
|
||||
@@ -0,0 +1,162 @@
|
||||
#!/usr/bin/env python
|
||||
"""
|
||||
Test #10: LinkedIn Huge DOM Tree
|
||||
|
||||
Symptom: browser_snapshot() hangs forever
|
||||
Root Cause: 10k+ DOM nodes, accessibility tree has 50k+ nodes
|
||||
Detection: document.querySelectorAll('*').length > 5000
|
||||
Fix: Add timeout (10s default), truncate tree at 2000 nodes
|
||||
"""
|
||||
|
||||
import asyncio
|
||||
import sys
|
||||
import time
|
||||
import base64
|
||||
from pathlib import Path
|
||||
|
||||
sys.path.insert(0, str(Path(__file__).parent.parent.parent.parent / "tools" / "src"))
|
||||
|
||||
from gcu.browser.bridge import BeelineBridge
|
||||
|
||||
CONTEXT_NAME = "huge-dom-test"
|
||||
|
||||
|
||||
async def test_huge_dom():
|
||||
"""Test snapshot performance on huge DOM trees."""
|
||||
print("=" * 70)
|
||||
print("TEST #10: Huge DOM Tree (LinkedIn-style)")
|
||||
print("=" * 70)
|
||||
|
||||
bridge = BeelineBridge()
|
||||
|
||||
try:
|
||||
await bridge.start()
|
||||
|
||||
for i in range(10):
|
||||
await asyncio.sleep(1)
|
||||
if bridge.is_connected:
|
||||
print("✓ Extension connected!")
|
||||
break
|
||||
else:
|
||||
print("✗ Extension not connected")
|
||||
return
|
||||
|
||||
context = await bridge.create_context(CONTEXT_NAME)
|
||||
tab_id = context.get("tabId")
|
||||
group_id = context.get("groupId")
|
||||
print(f"✓ Created tab: {tab_id}")
|
||||
|
||||
# Test 1: Small DOM (baseline)
|
||||
print("\n--- Test 1: Small DOM (baseline) ---")
|
||||
small_html = """
|
||||
<!DOCTYPE html>
|
||||
<html><body>
|
||||
<h1>Small Page</h1>
|
||||
<p>A few elements</p>
|
||||
<button>Click me</button>
|
||||
</body></html>
|
||||
"""
|
||||
data_url = f"data:text/html;base64,{base64.b64encode(small_html.encode()).decode()}"
|
||||
await bridge.navigate(tab_id, data_url, wait_until="load")
|
||||
|
||||
start = time.perf_counter()
|
||||
snapshot = await bridge.snapshot(tab_id, timeout_s=5.0)
|
||||
elapsed = time.perf_counter() - start
|
||||
tree_len = len(snapshot.get("tree", ""))
|
||||
print(f"Small DOM snapshot: {elapsed:.3f}s, {tree_len} chars")
|
||||
|
||||
# Test 2: Generate huge DOM
|
||||
print("\n--- Test 2: Huge DOM (5000+ elements) ---")
|
||||
huge_html = """
|
||||
<!DOCTYPE html>
|
||||
<html><body>
|
||||
<h1>Huge DOM Test</h1>
|
||||
<div id="container"></div>
|
||||
<script>
|
||||
const container = document.getElementById('container');
|
||||
for (let i = 0; i < 5000; i++) {
|
||||
const div = document.createElement('div');
|
||||
div.className = 'item-' + i;
|
||||
div.innerHTML = '<span>Item ' + i + '</span><button>Action</button>';
|
||||
container.appendChild(div);
|
||||
}
|
||||
</script>
|
||||
</body></html>
|
||||
"""
|
||||
data_url = f"data:text/html;base64,{base64.b64encode(huge_html.encode()).decode()}"
|
||||
await bridge.navigate(tab_id, data_url, wait_until="load")
|
||||
|
||||
# Count elements
|
||||
count_result = await bridge.evaluate(
|
||||
tab_id,
|
||||
"(function() { return document.querySelectorAll('*').length; })()"
|
||||
)
|
||||
elem_count = count_result.get("result", 0)
|
||||
print(f"DOM elements: {elem_count}")
|
||||
|
||||
# Screenshot to verify page loaded
|
||||
screenshot = await bridge.screenshot(tab_id)
|
||||
print(f"Screenshot: {len(screenshot.get('data', ''))} bytes")
|
||||
|
||||
# Test snapshot with timeout
|
||||
print("\n--- Testing snapshot with 10s timeout ---")
|
||||
start = time.perf_counter()
|
||||
try:
|
||||
snapshot = await bridge.snapshot(tab_id, timeout_s=10.0)
|
||||
elapsed = time.perf_counter() - start
|
||||
tree_len = len(snapshot.get("tree", ""))
|
||||
truncated = "(truncated)" in snapshot.get("tree", "")
|
||||
print(f"✓ Huge DOM snapshot: {elapsed:.3f}s, {tree_len} chars, truncated={truncated}")
|
||||
|
||||
if elapsed < 5.0:
|
||||
print("✓ PASS: Snapshot completed quickly")
|
||||
else:
|
||||
print(f"⚠ WARNING: Snapshot took {elapsed:.1f}s")
|
||||
|
||||
if truncated:
|
||||
print("✓ PASS: Tree was truncated to prevent hang")
|
||||
else:
|
||||
print("⚠ WARNING: Tree not truncated (may need adjustment)")
|
||||
|
||||
except asyncio.TimeoutError:
|
||||
print("✗ FAIL: Snapshot timed out (this shouldn't happen)")
|
||||
|
||||
# Test 3: Real LinkedIn
|
||||
print("\n--- Test 3: Real LinkedIn Feed ---")
|
||||
await bridge.navigate(tab_id, "https://www.linkedin.com/feed", wait_until="load", timeout_ms=30000)
|
||||
await asyncio.sleep(2)
|
||||
|
||||
count_result = await bridge.evaluate(
|
||||
tab_id,
|
||||
"(function() { return document.querySelectorAll('*').length; })()"
|
||||
)
|
||||
elem_count = count_result.get("result", 0)
|
||||
print(f"LinkedIn DOM elements: {elem_count}")
|
||||
|
||||
start = time.perf_counter()
|
||||
try:
|
||||
snapshot = await bridge.snapshot(tab_id, timeout_s=15.0)
|
||||
elapsed = time.perf_counter() - start
|
||||
tree_len = len(snapshot.get("tree", ""))
|
||||
truncated = "(truncated)" in snapshot.get("tree", "")
|
||||
print(f"LinkedIn snapshot: {elapsed:.3f}s, {tree_len} chars, truncated={truncated}")
|
||||
|
||||
if elapsed < 5.0:
|
||||
print("✓ PASS: LinkedIn snapshot fast enough")
|
||||
elif elapsed < 15.0:
|
||||
print("⚠ WARNING: LinkedIn snapshot slow but within timeout")
|
||||
else:
|
||||
print("✗ FAIL: LinkedIn snapshot too slow")
|
||||
|
||||
except asyncio.TimeoutError:
|
||||
print("✗ FAIL: LinkedIn snapshot timed out")
|
||||
|
||||
await bridge.destroy_context(group_id)
|
||||
print("\n✓ Context destroyed")
|
||||
|
||||
finally:
|
||||
await bridge.stop()
|
||||
|
||||
|
||||
if __name__ == "__main__":
|
||||
asyncio.run(test_huge_dom())
|
||||
@@ -0,0 +1,184 @@
|
||||
#!/usr/bin/env python
|
||||
"""
|
||||
Test #13: SPA Navigation Events
|
||||
|
||||
Symptom: wait_until="load" fires before content ready
|
||||
Root Cause: SPA uses client-side routing, no full page load
|
||||
Detection: URL changes but load event already fired
|
||||
Fix: Use wait_until="networkidle" or wait_for_selector
|
||||
"""
|
||||
|
||||
import asyncio
|
||||
import sys
|
||||
import time
|
||||
import base64
|
||||
from pathlib import Path
|
||||
|
||||
sys.path.insert(0, str(Path(__file__).parent.parent.parent.parent / "tools" / "src"))
|
||||
|
||||
from gcu.browser.bridge import BeelineBridge
|
||||
|
||||
CONTEXT_NAME = "spa-nav-test"
|
||||
|
||||
|
||||
async def test_spa_navigation():
|
||||
"""Test navigation timing on SPA pages."""
|
||||
print("=" * 70)
|
||||
print("TEST #13: SPA Navigation Events")
|
||||
print("=" * 70)
|
||||
|
||||
bridge = BeelineBridge()
|
||||
|
||||
try:
|
||||
await bridge.start()
|
||||
|
||||
for i in range(10):
|
||||
await asyncio.sleep(1)
|
||||
if bridge.is_connected:
|
||||
print("✓ Extension connected!")
|
||||
break
|
||||
else:
|
||||
print("✗ Extension not connected")
|
||||
return
|
||||
|
||||
context = await bridge.create_context(CONTEXT_NAME)
|
||||
tab_id = context.get("tabId")
|
||||
group_id = context.get("groupId")
|
||||
print(f"✓ Created tab: {tab_id}")
|
||||
|
||||
# Create a test SPA
|
||||
spa_html = """
|
||||
<!DOCTYPE html>
|
||||
<html>
|
||||
<head>
|
||||
<title>SPA Test</title>
|
||||
<style>
|
||||
nav a { margin-right: 10px; }
|
||||
.page { padding: 20px; border: 1px solid #ccc; margin-top: 10px; }
|
||||
</style>
|
||||
</head>
|
||||
<body>
|
||||
<nav>
|
||||
<a href="#home" onclick="navigate('home')">Home</a>
|
||||
<a href="#about" onclick="navigate('about')">About</a>
|
||||
<a href="#contact" onclick="navigate('contact')">Contact</a>
|
||||
</nav>
|
||||
<div id="app" class="page">
|
||||
<h1>Loading...</h1>
|
||||
</div>
|
||||
<script>
|
||||
// Simulate SPA routing
|
||||
let currentPage = '';
|
||||
|
||||
async function navigate(page) {
|
||||
event.preventDefault();
|
||||
currentPage = page;
|
||||
|
||||
// Show loading state
|
||||
document.getElementById('app').innerHTML = '<h1>Loading...</h1>';
|
||||
|
||||
// Simulate async content loading (like real SPAs)
|
||||
await new Promise(r => setTimeout(r, 500));
|
||||
|
||||
// Render content
|
||||
const content = {
|
||||
home: '<h1>Home Page</h1><p>Welcome to the SPA!</p><button id="home-btn">Home Action</button>',
|
||||
about: '<h1>About Page</h1><p>This is a simulated SPA.</p><button id="about-btn">About Action</button>',
|
||||
contact: '<h1>Contact Page</h1><p>Contact us at test@example.com</p><button id="contact-btn">Contact Action</button>'
|
||||
};
|
||||
|
||||
document.getElementById('app').innerHTML = content[page] || '<h1>404</h1>';
|
||||
window.location.hash = page;
|
||||
}
|
||||
|
||||
// Initial load with delay (simulates SPA hydration)
|
||||
setTimeout(() => {
|
||||
navigate('home');
|
||||
}, 1000);
|
||||
|
||||
// Track for testing
|
||||
window.pageLoads = [];
|
||||
window.addEventListener('hashchange', () => {
|
||||
window.pageLoads.push(window.location.hash);
|
||||
});
|
||||
</script>
|
||||
</body>
|
||||
</html>
|
||||
"""
|
||||
|
||||
data_url = f"data:text/html;base64,{base64.b64encode(spa_html.encode()).decode()}"
|
||||
|
||||
# Test 1: wait_until="load" - may fire before content ready
|
||||
print("\n--- Test 1: wait_until='load' ---")
|
||||
start = time.perf_counter()
|
||||
await bridge.navigate(tab_id, data_url, wait_until="load")
|
||||
elapsed = time.perf_counter() - start
|
||||
print(f"Navigation completed in {elapsed:.3f}s")
|
||||
|
||||
# Check content immediately
|
||||
content = await bridge.evaluate(
|
||||
tab_id,
|
||||
"(function() { return document.getElementById('app').innerText; })()"
|
||||
)
|
||||
print(f"Content immediately after load: '{content.get('result', '')}'")
|
||||
|
||||
# Screenshot
|
||||
screenshot = await bridge.screenshot(tab_id)
|
||||
print(f"Screenshot: {len(screenshot.get('data', ''))} bytes")
|
||||
|
||||
# Wait for content
|
||||
print("\n--- Waiting for content to hydrate ---")
|
||||
await bridge.wait_for_selector(tab_id, "#home-btn", timeout_ms=5000)
|
||||
print("✓ Content loaded")
|
||||
|
||||
# Check content after wait
|
||||
content_after = await bridge.evaluate(
|
||||
tab_id,
|
||||
"(function() { return document.getElementById('app').innerText; })()"
|
||||
)
|
||||
print(f"Content after wait: '{content_after.get('result', '')}'")
|
||||
|
||||
# Test 2: SPA navigation (no full page load)
|
||||
print("\n--- Test 2: SPA client-side navigation ---")
|
||||
|
||||
# Click "About" link
|
||||
await bridge.click(tab_id, 'a[href="#about"]')
|
||||
await asyncio.sleep(1)
|
||||
|
||||
# Check if content changed
|
||||
about_content = await bridge.evaluate(
|
||||
tab_id,
|
||||
"(function() { return document.getElementById('app').innerText; })()"
|
||||
)
|
||||
print(f"Content after SPA nav: '{about_content.get('result', '')}'")
|
||||
|
||||
if "About Page" in about_content.get("result", ""):
|
||||
print("✓ PASS: SPA navigation worked")
|
||||
else:
|
||||
print("✗ FAIL: SPA navigation didn't update content")
|
||||
|
||||
# Test 3: wait_until="networkidle"
|
||||
print("\n--- Test 3: wait_until='networkidle' ---")
|
||||
await bridge.navigate(tab_id, data_url, wait_until="networkidle", timeout_ms=10000)
|
||||
|
||||
# Check content immediately
|
||||
content_networkidle = await bridge.evaluate(
|
||||
tab_id,
|
||||
"(function() { return document.getElementById('app').innerText; })()"
|
||||
)
|
||||
print(f"Content after networkidle: '{content_networkidle.get('result', '')}'")
|
||||
|
||||
if "Home Page" in content_networkidle.get("result", ""):
|
||||
print("✓ PASS: networkidle waited for content")
|
||||
else:
|
||||
print("⚠ WARNING: networkidle didn't wait long enough")
|
||||
|
||||
await bridge.destroy_context(group_id)
|
||||
print("\n✓ Context destroyed")
|
||||
|
||||
finally:
|
||||
await bridge.stop()
|
||||
|
||||
|
||||
if __name__ == "__main__":
|
||||
asyncio.run(test_spa_navigation())
|
||||
@@ -0,0 +1,327 @@
|
||||
#!/usr/bin/env python
|
||||
"""
|
||||
Browser Edge Case Test Template
|
||||
|
||||
This script provides a template for testing and debugging browser tool failures
|
||||
on specific websites. Use this to reproduce, isolate, and verify fixes.
|
||||
|
||||
Usage:
|
||||
1. Copy this file: cp test_case.py test_#[number]_[site].py
|
||||
2. Fill in the CONFIG section with your test details
|
||||
3. Run: uv run python test_#[number]_[site].py
|
||||
|
||||
Example:
|
||||
uv run python test_01_linkedin_scroll.py
|
||||
"""
|
||||
|
||||
import asyncio
|
||||
import sys
|
||||
import time
|
||||
from pathlib import Path
|
||||
|
||||
# Add tools to path
|
||||
sys.path.insert(0, str(Path(__file__).parent.parent.parent.parent / "tools" / "src"))
|
||||
|
||||
from gcu.browser.bridge import BeelineBridge
|
||||
|
||||
# ═══════════════════════════════════════════════════════════════════════════════
|
||||
# CONFIG: Fill in these values for your test case
|
||||
# ═══════════════════════════════════════════════════════════════════════════════
|
||||
|
||||
TEST_CASE = {
|
||||
"number": 1,
|
||||
"name": "LinkedIn Nested Scroll Container",
|
||||
"site": "https://www.linkedin.com/feed",
|
||||
"simple_site": "https://example.com",
|
||||
"category": "scroll", # scroll, click, input, snapshot, navigation
|
||||
"symptom": "scroll() returns success but page doesn't move",
|
||||
}
|
||||
|
||||
BRIDGE_PORT = 9229
|
||||
CONTEXT_NAME = "edge-case-test"
|
||||
|
||||
|
||||
# ═══════════════════════════════════════════════════════════════════════════════
|
||||
# TEST FUNCTIONS
|
||||
# ═══════════════════════════════════════════════════════════════════════════════
|
||||
|
||||
|
||||
async def test_simple_site(bridge: BeelineBridge, tab_id: int) -> dict:
|
||||
"""Test that the tool works on a simple site (baseline)."""
|
||||
print("\n--- Baseline Test (Simple Site) ---")
|
||||
|
||||
await bridge.navigate(tab_id, TEST_CASE["simple_site"], wait_until="load")
|
||||
await asyncio.sleep(1)
|
||||
|
||||
# Adjust this based on category
|
||||
if TEST_CASE["category"] == "scroll":
|
||||
result = await bridge.scroll(tab_id, "down", 100)
|
||||
print(f" Scroll result: {result}")
|
||||
return result
|
||||
elif TEST_CASE["category"] == "click":
|
||||
# Add click test
|
||||
pass
|
||||
elif TEST_CASE["category"] == "snapshot":
|
||||
result = await bridge.snapshot(tab_id, timeout_s=5.0)
|
||||
print(f" Snapshot length: {len(result.get('tree', ''))}")
|
||||
return result
|
||||
|
||||
return {"ok": True}
|
||||
|
||||
|
||||
async def test_problematic_site(bridge: BeelineBridge, tab_id: int) -> dict:
|
||||
"""Test the tool on the problematic site."""
|
||||
print("\n--- Problem Site Test ---")
|
||||
|
||||
await bridge.navigate(tab_id, TEST_CASE["site"], wait_until="load", timeout_ms=30000)
|
||||
await asyncio.sleep(2)
|
||||
|
||||
# Adjust this based on category
|
||||
if TEST_CASE["category"] == "scroll":
|
||||
# Get scroll positions before
|
||||
before = await bridge.evaluate(
|
||||
tab_id,
|
||||
"""
|
||||
(function() {
|
||||
const results = { window: { y: window.scrollY } };
|
||||
document.querySelectorAll('*').forEach((el, i) => {
|
||||
const style = getComputedStyle(el);
|
||||
if ((style.overflowY === 'scroll' || style.overflowY === 'auto') &&
|
||||
el.scrollHeight > el.clientHeight) {
|
||||
results['el_' + i] = {
|
||||
tag: el.tagName,
|
||||
scrollTop: el.scrollTop,
|
||||
class: el.className.substring(0, 30)
|
||||
};
|
||||
}
|
||||
});
|
||||
return results;
|
||||
})();
|
||||
"""
|
||||
)
|
||||
print(f" Before scroll: {before.get('result', {})}")
|
||||
|
||||
# Try to scroll
|
||||
result = await bridge.scroll(tab_id, "down", 500)
|
||||
print(f" Scroll result: {result}")
|
||||
|
||||
await asyncio.sleep(1)
|
||||
|
||||
# Get scroll positions after
|
||||
after = await bridge.evaluate(
|
||||
tab_id,
|
||||
"""
|
||||
(function() {
|
||||
const results = { window: { y: window.scrollY } };
|
||||
document.querySelectorAll('*').forEach((el, i) => {
|
||||
const style = getComputedStyle(el);
|
||||
if ((style.overflowY === 'scroll' || style.overflowY === 'auto') &&
|
||||
el.scrollHeight > el.clientHeight) {
|
||||
results['el_' + i] = {
|
||||
tag: el.tagName,
|
||||
scrollTop: el.scrollTop,
|
||||
class: el.className.substring(0, 30)
|
||||
};
|
||||
}
|
||||
});
|
||||
return results;
|
||||
})();
|
||||
"""
|
||||
)
|
||||
print(f" After scroll: {after.get('result', {})}")
|
||||
|
||||
# Check if anything changed
|
||||
before_data = before.get("result", {}) or {}
|
||||
after_data = after.get("result", {}) or {}
|
||||
|
||||
changed = False
|
||||
for key in after_data:
|
||||
if key in before_data:
|
||||
b_val = before_data[key].get("scrollTop", 0) if isinstance(before_data[key], dict) else 0
|
||||
a_val = after_data[key].get("scrollTop", 0) if isinstance(after_data[key], dict) else 0
|
||||
if a_val != b_val:
|
||||
print(f" ✓ CHANGE DETECTED: {key} scrolled from {b_val} to {a_val}")
|
||||
changed = True
|
||||
|
||||
if not changed:
|
||||
print(" ✗ NO CHANGE: Scroll did not affect any container")
|
||||
|
||||
return {"ok": changed, "scroll_result": result}
|
||||
|
||||
elif TEST_CASE["category"] == "snapshot":
|
||||
start = time.perf_counter()
|
||||
try:
|
||||
result = await bridge.snapshot(tab_id, timeout_s=15.0)
|
||||
elapsed = time.perf_counter() - start
|
||||
tree_len = len(result.get("tree", ""))
|
||||
print(f" Snapshot completed in {elapsed:.2f}s, {tree_len} chars")
|
||||
return {"ok": True, "elapsed": elapsed, "tree_length": tree_len}
|
||||
except asyncio.TimeoutError:
|
||||
print(" ✗ SNAPSHOT TIMED OUT")
|
||||
return {"ok": False, "error": "timeout"}
|
||||
|
||||
return {"ok": True}
|
||||
|
||||
|
||||
async def detect_root_cause(bridge: BeelineBridge, tab_id: int) -> dict:
|
||||
"""Run detection scripts to identify the root cause."""
|
||||
print("\n--- Root Cause Detection ---")
|
||||
|
||||
detections = {}
|
||||
|
||||
# Detection 1: Nested scrollable containers
|
||||
scroll_check = await bridge.evaluate(
|
||||
tab_id,
|
||||
"""
|
||||
(function() {
|
||||
const candidates = [];
|
||||
document.querySelectorAll('*').forEach(el => {
|
||||
const style = getComputedStyle(el);
|
||||
if (style.overflow.includes('scroll') || style.overflow.includes('auto')) {
|
||||
const rect = el.getBoundingClientRect();
|
||||
if (rect.width > 100 && rect.height > 100) {
|
||||
candidates.push({
|
||||
tag: el.tagName,
|
||||
area: rect.width * rect.height,
|
||||
class: el.className.substring(0, 30)
|
||||
});
|
||||
}
|
||||
}
|
||||
});
|
||||
candidates.sort((a, b) => b.area - a.area);
|
||||
return {
|
||||
count: candidates.length,
|
||||
largest: candidates[0]
|
||||
};
|
||||
})();
|
||||
"""
|
||||
)
|
||||
detections["nested_scroll"] = scroll_check.get("result", {})
|
||||
print(f" Nested scroll containers: {detections['nested_scroll']}")
|
||||
|
||||
# Detection 2: Shadow DOM
|
||||
shadow_check = await bridge.evaluate(
|
||||
tab_id,
|
||||
"""
|
||||
(function() {
|
||||
const withShadow = [];
|
||||
document.querySelectorAll('*').forEach(el => {
|
||||
if (el.shadowRoot) {
|
||||
withShadow.push(el.tagName);
|
||||
}
|
||||
});
|
||||
return { count: withShadow.length, elements: withShadow.slice(0, 5) };
|
||||
})();
|
||||
"""
|
||||
)
|
||||
detections["shadow_dom"] = shadow_check.get("result", {})
|
||||
print(f" Shadow DOM: {detections['shadow_dom']}")
|
||||
|
||||
# Detection 3: iframes
|
||||
iframe_check = await bridge.evaluate(
|
||||
tab_id,
|
||||
"""
|
||||
(function() {
|
||||
const iframes = document.querySelectorAll('iframe');
|
||||
return { count: iframes.length };
|
||||
})();
|
||||
"""
|
||||
)
|
||||
detections["iframes"] = iframe_check.get("result", {})
|
||||
print(f" iframes: {detections['iframes']}")
|
||||
|
||||
# Detection 4: DOM size
|
||||
dom_check = await bridge.evaluate(
|
||||
tab_id,
|
||||
"""
|
||||
(function() {
|
||||
return {
|
||||
elements: document.querySelectorAll('*').length,
|
||||
body_children: document.body.children.length
|
||||
};
|
||||
})();
|
||||
"""
|
||||
)
|
||||
detections["dom_size"] = dom_check.get("result", {})
|
||||
print(f" DOM size: {detections['dom_size']}")
|
||||
|
||||
# Detection 5: Framework detection
|
||||
framework_check = await bridge.evaluate(
|
||||
tab_id,
|
||||
"""
|
||||
(function() {
|
||||
return {
|
||||
react: !!document.querySelector('[data-reactroot], [data-reactid]'),
|
||||
vue: !!document.querySelector('[data-v-]'),
|
||||
angular: !!document.querySelector('[ng-app], [ng-version]')
|
||||
};
|
||||
})();
|
||||
"""
|
||||
)
|
||||
detections["frameworks"] = framework_check.get("result", {})
|
||||
print(f" Frameworks: {detections['frameworks']}")
|
||||
|
||||
return detections
|
||||
|
||||
|
||||
# ═══════════════════════════════════════════════════════════════════════════════
|
||||
# MAIN
|
||||
# ═══════════════════════════════════════════════════════════════════════════════
|
||||
|
||||
|
||||
async def main():
|
||||
print("=" * 70)
|
||||
print(f"EDGE CASE TEST #{TEST_CASE['number']}: {TEST_CASE['name']}")
|
||||
print("=" * 70)
|
||||
print(f"Site: {TEST_CASE['site']}")
|
||||
print(f"Category: {TEST_CASE['category']}")
|
||||
print(f"Symptom: {TEST_CASE['symptom']}")
|
||||
|
||||
bridge = BeelineBridge()
|
||||
|
||||
try:
|
||||
print("\n--- Starting Bridge ---")
|
||||
await bridge.start()
|
||||
|
||||
# Wait for extension connection
|
||||
for i in range(10):
|
||||
await asyncio.sleep(1)
|
||||
if bridge.is_connected:
|
||||
print("✓ Extension connected!")
|
||||
break
|
||||
print(f"Waiting for extension... ({i+1}/10)")
|
||||
else:
|
||||
print("✗ Extension not connected. Ensure Chrome with Beeline extension is running.")
|
||||
return
|
||||
|
||||
# Create browser context
|
||||
context = await bridge.create_context(CONTEXT_NAME)
|
||||
tab_id = context.get("tabId")
|
||||
group_id = context.get("groupId")
|
||||
print(f"✓ Created tab: {tab_id}")
|
||||
|
||||
# Run tests
|
||||
baseline_result = await test_simple_site(bridge, tab_id)
|
||||
problem_result = await test_problematic_site(bridge, tab_id)
|
||||
detections = await detect_root_cause(bridge, tab_id)
|
||||
|
||||
# Summary
|
||||
print("\n" + "=" * 70)
|
||||
print("SUMMARY")
|
||||
print("=" * 70)
|
||||
print(f"Baseline test: {'✓ PASS' if baseline_result.get('ok') else '✗ FAIL'}")
|
||||
print(f"Problem test: {'✓ PASS' if problem_result.get('ok') else '✗ FAIL'}")
|
||||
print(f"Root cause indicators: {list(k for k, v in detections.items() if v)}")
|
||||
|
||||
# Cleanup
|
||||
print("\n--- Cleanup ---")
|
||||
await bridge.destroy_context(group_id)
|
||||
print("✓ Context destroyed")
|
||||
|
||||
finally:
|
||||
await bridge.stop()
|
||||
print("✓ Bridge stopped")
|
||||
|
||||
|
||||
if __name__ == "__main__":
|
||||
asyncio.run(main())
|
||||
@@ -222,7 +222,7 @@ def truncate_tool_result(
|
||||
- Small results (≤ limit): full content kept + file annotation
|
||||
- Large results (> limit): preview + file reference
|
||||
- Errors: pass through unchanged
|
||||
- load_data results: truncate with pagination hint (no re-spill)
|
||||
- read_file/load_data results: truncate with pagination hint (no re-spill)
|
||||
"""
|
||||
limit = max_tool_result_chars
|
||||
|
||||
@@ -230,12 +230,12 @@ def truncate_tool_result(
|
||||
if result.is_error:
|
||||
return result
|
||||
|
||||
# load_data reads FROM spilled files — never re-spill (circular).
|
||||
# read_file/load_data reads FROM spilled files — never re-spill (circular).
|
||||
# Just truncate with a pagination hint if the result is too large.
|
||||
if tool_name == "load_data":
|
||||
if tool_name in ("load_data", "read_file"):
|
||||
if limit <= 0 or len(result.content) <= limit:
|
||||
return result # Small load_data result — pass through as-is
|
||||
# Large load_data result — truncate with smart preview
|
||||
return result # Small result — pass through as-is
|
||||
# Large result — truncate with smart preview
|
||||
PREVIEW_CAP = min(5000, max(limit - 500, limit // 2))
|
||||
|
||||
metadata_str = ""
|
||||
@@ -284,7 +284,7 @@ def truncate_tool_result(
|
||||
spill_path.mkdir(parents=True, exist_ok=True)
|
||||
filename = next_spill_filename_fn(tool_name)
|
||||
|
||||
# Pretty-print JSON content so load_data's line-based
|
||||
# Pretty-print JSON content so read_file's line-based
|
||||
# pagination works correctly.
|
||||
write_content = result.content
|
||||
parsed_json: Any = None # track for metadata extraction
|
||||
@@ -294,7 +294,10 @@ def truncate_tool_result(
|
||||
except (json.JSONDecodeError, TypeError, ValueError):
|
||||
pass # Not JSON — write as-is
|
||||
|
||||
(spill_path / filename).write_text(write_content, encoding="utf-8")
|
||||
file_path = spill_path / filename
|
||||
file_path.write_text(write_content, encoding="utf-8")
|
||||
# Use absolute path so parent agents can find files from subagents
|
||||
abs_path = str(file_path.resolve())
|
||||
|
||||
if limit > 0 and len(result.content) > limit:
|
||||
# Large result: build a small, metadata-rich preview so the
|
||||
@@ -316,14 +319,14 @@ def truncate_tool_result(
|
||||
# Assemble header with structural info + warning
|
||||
header = (
|
||||
f"[Result from {tool_name}: {len(result.content):,} chars — "
|
||||
f"too large for context, saved to '{filename}'.]\n"
|
||||
f"too large for context, saved to '{abs_path}'.]\n"
|
||||
)
|
||||
if metadata_str:
|
||||
header += f"\nData structure:\n{metadata_str}"
|
||||
header += (
|
||||
f"\n\nWARNING: The preview below is INCOMPLETE. "
|
||||
f"Do NOT draw conclusions or counts from it. "
|
||||
f"Use load_data(filename='{filename}') to read the "
|
||||
f"Use read_file(path='{abs_path}') to read the "
|
||||
f"full data before analysis."
|
||||
)
|
||||
|
||||
@@ -332,11 +335,11 @@ def truncate_tool_result(
|
||||
"Tool result spilled to file: %s (%d chars → %s)",
|
||||
tool_name,
|
||||
len(result.content),
|
||||
filename,
|
||||
abs_path,
|
||||
)
|
||||
else:
|
||||
# Small result: keep full content + annotation
|
||||
content = f"{result.content}\n\n[Saved to '{filename}']"
|
||||
# Small result: keep full content + annotation with absolute path
|
||||
content = f"{result.content}\n\n[Saved to '{abs_path}']"
|
||||
logger.info(
|
||||
"Tool result saved to file: %s (%d chars → %s)",
|
||||
tool_name,
|
||||
|
||||
@@ -151,8 +151,9 @@ class OutputAccumulator:
|
||||
if isinstance(value, (dict, list))
|
||||
else str(value)
|
||||
)
|
||||
(spill_path / filename).write_text(write_content, encoding="utf-8")
|
||||
file_size = (spill_path / filename).stat().st_size
|
||||
file_path = spill_path / filename
|
||||
file_path.write_text(write_content, encoding="utf-8")
|
||||
file_size = file_path.stat().st_size
|
||||
logger.info(
|
||||
"set_output value auto-spilled: key=%s, %d chars -> %s (%d bytes)",
|
||||
key,
|
||||
@@ -160,9 +161,11 @@ class OutputAccumulator:
|
||||
filename,
|
||||
file_size,
|
||||
)
|
||||
# Use absolute path so parent agents can find files from subagents
|
||||
abs_path = str(file_path.resolve())
|
||||
return (
|
||||
f"[Saved to '{filename}' ({file_size:,} bytes). "
|
||||
f"Use load_data(filename='{filename}') "
|
||||
f"[Saved to '{abs_path}' ({file_size:,} bytes). "
|
||||
f"Use read_file(path='{abs_path}') "
|
||||
f"to access full data.]"
|
||||
)
|
||||
|
||||
|
||||
@@ -0,0 +1,180 @@
|
||||
# Browser-Use Patterns Analysis
|
||||
|
||||
## Key Learnings from `/home/timothy/aden/browser-use`
|
||||
|
||||
### 1. Element Click Implementation
|
||||
|
||||
**browser-use approach** (`browser_use/actor/element.py`):
|
||||
|
||||
```python
|
||||
# Three fallback methods for element geometry:
|
||||
|
||||
# Method 1: DOM.getContentQuads (best for inline elements and complex layouts)
|
||||
content_quads_result = await self._client.send.DOM.getContentQuads(
|
||||
params={'backendNodeId': self._backend_node_id}, session_id=self._session_id
|
||||
)
|
||||
|
||||
# Method 2: DOM.getBoxModel (fallback)
|
||||
box_model = await self._client.send.DOM.getBoxModel(
|
||||
params={'backendNodeId': self._backend_node_id}, session_id=self._session_id
|
||||
)
|
||||
|
||||
# Method 3: JavaScript getBoundingClientRect (final fallback)
|
||||
bounds_result = await self._client.send.Runtime.callFunctionOn(
|
||||
params={
|
||||
'functionDeclaration': """
|
||||
function() {
|
||||
const rect = this.getBoundingClientRect();
|
||||
return {
|
||||
x: rect.left,
|
||||
y: rect.top,
|
||||
width: rect.width,
|
||||
height: rect.height
|
||||
};
|
||||
}
|
||||
""",
|
||||
'objectId': object_id,
|
||||
'returnByValue': True,
|
||||
},
|
||||
session_id=self._session_id,
|
||||
)
|
||||
|
||||
# Method 4: JavaScript click (if all else fails)
|
||||
await self._client.send.Runtime.callFunctionOn(
|
||||
params={
|
||||
'functionDeclaration': 'function() { this.click(); }',
|
||||
'objectId': object_id,
|
||||
},
|
||||
session_id=self._session_id,
|
||||
)
|
||||
```
|
||||
|
||||
**Key differences from our implementation:**
|
||||
- Uses `backendNodeId` instead of `nodeId` (more stable across DOM updates)
|
||||
- Tries `DOM.getContentQuads` first (better for complex layouts)
|
||||
- Multiple fallback methods with JavaScript click as final resort
|
||||
- Finds largest visible quad within viewport
|
||||
- Has timeouts for each mouse operation
|
||||
- Proper modifier key handling
|
||||
|
||||
### 2. Input/Type Text Implementation
|
||||
|
||||
**browser-use approach**:
|
||||
|
||||
```python
|
||||
# 1. Scroll element into view
|
||||
await cdp_client.send.DOM.scrollIntoViewIfNeeded(
|
||||
params={'backendNodeId': backend_node_id},
|
||||
session_id=session_id
|
||||
)
|
||||
|
||||
# 2. Get object ID
|
||||
result = await cdp_client.send.DOM.resolveNode(
|
||||
params={'backendNodeId': backend_node_id},
|
||||
session_id=session_id,
|
||||
)
|
||||
object_id = result['object']['objectId']
|
||||
|
||||
# 3. Focus via JavaScript (more reliable than CDP focus)
|
||||
await cdp_client.send.Runtime.callFunctionOn(
|
||||
params={
|
||||
'functionDeclaration': 'function() { this.focus(); }',
|
||||
'objectId': object_id,
|
||||
},
|
||||
session_id=session_id,
|
||||
)
|
||||
|
||||
# 4. Type using Input.dispatchKeyEvent
|
||||
for char in text:
|
||||
await self._client.send.Input.dispatchKeyEvent(
|
||||
params={
|
||||
'type': 'keyDown',
|
||||
'key': char,
|
||||
'text': char,
|
||||
},
|
||||
session_id=self._session_id,
|
||||
)
|
||||
```
|
||||
|
||||
### 3. Accessibility Tree (Snapshot)
|
||||
|
||||
**browser-use approach** (`browser_use/dom/service.py`):
|
||||
|
||||
- Uses `Accessibility.getFullAXTree` for accessibility data
|
||||
- Combines with DOM tree for enhanced snapshot
|
||||
- Filters by paint order (elements actually visible)
|
||||
- Handles iframes with depth limits
|
||||
- Detects hidden interactive elements and reports them
|
||||
- Uses `DOM.getFrameOwner` for iframe handling
|
||||
|
||||
### 4. CDP Domain Handling
|
||||
|
||||
**browser-use approach** (`browser_use/browser/session.py`):
|
||||
|
||||
```python
|
||||
# Session setup enables ONLY these domains:
|
||||
await self._client.send.Page.enable(session_id=self._session_id)
|
||||
await self._client.send.DOM.enable(session_id=self._session_id)
|
||||
await self._client.send.Runtime.enable(session_id=self._session_id)
|
||||
await self._client.send.Network.enable(session_id=self._session_id)
|
||||
|
||||
# Input.enable is NEVER called - it's not required!
|
||||
```
|
||||
|
||||
### 5. Element Selection
|
||||
|
||||
**browser-use approach**:
|
||||
|
||||
- Uses index-based element selection from accessibility tree
|
||||
- Maintains a map of index -> EnhancedAXNode
|
||||
- Elements have `backendNodeId` which is stable
|
||||
- Uses `DOM.pushNodesByBackendIdsToFrontend` to get fresh nodeId
|
||||
|
||||
### 6. Scroll Handling
|
||||
|
||||
**browser-use approach**:
|
||||
|
||||
```python
|
||||
# Uses multiple methods:
|
||||
# 1. DOM.scrollIntoViewIfNeeded (CDP)
|
||||
# 2. JavaScript scrollIntoView as fallback
|
||||
# 3. Mouse wheel events for smooth scrolling
|
||||
```
|
||||
|
||||
### 7. Wait Strategies
|
||||
|
||||
**browser-use approach**:
|
||||
|
||||
- `wait_for_element` uses CDP DOM queries with polling
|
||||
- Has configurable timeouts
|
||||
- Uses `DOM.getContentQuads` to verify element is visible
|
||||
- Detects page load state via `Page.loadEventFired`
|
||||
|
||||
---
|
||||
|
||||
## Improvements to Make to hive/tools
|
||||
|
||||
### 1. Bridge Updates
|
||||
|
||||
- [x] Use `backendNodeId` instead of `nodeId` where possible
|
||||
- [x] Add `DOM.getContentQuads` as primary method for element geometry
|
||||
- [x] Add JavaScript click as final fallback
|
||||
- [x] Add proper timeouts to mouse operations
|
||||
- [x] Handle modifier keys for click
|
||||
|
||||
### 2. Type Text Updates
|
||||
|
||||
- [x] Focus element via JavaScript before typing
|
||||
- [x] Use `Input.dispatchKeyEvent` for typing (more reliable than insertText)
|
||||
|
||||
### 3. Snapshot Updates
|
||||
|
||||
- [x] Use accessibility tree (CDP Accessibility domain)
|
||||
- [x] Add computed styles to detect visibility
|
||||
- [x] Report hidden interactive elements
|
||||
|
||||
### 4. Error Handling
|
||||
|
||||
- [x] Better error messages with element context
|
||||
- [x] Graceful fallbacks instead of hard failures
|
||||
- [x] Timeout handling for all CDP operations
|
||||
@@ -10,10 +10,11 @@ const HIVE_WS_URL = "ws://127.0.0.1:9229/bridge";
|
||||
|
||||
let ws = null;
|
||||
let reconnectAttempts = 0;
|
||||
const MAX_RECONNECT_DELAY = 10000; // Max 10 seconds between attempts
|
||||
|
||||
function connect() {
|
||||
// Don't try to reconnect too fast
|
||||
const delay = Math.min(reconnectAttempts * 1000, 5000);
|
||||
// Exponential backoff with cap
|
||||
const delay = Math.min(reconnectAttempts * 1000, MAX_RECONNECT_DELAY);
|
||||
|
||||
if (reconnectAttempts > 0) {
|
||||
console.log(`[Beeline] Reconnecting in ${delay}ms (attempt ${reconnectAttempts + 1})...`);
|
||||
@@ -34,18 +35,21 @@ function connect() {
|
||||
};
|
||||
|
||||
ws.onclose = (event) => {
|
||||
console.log(`[Beeline] WebSocket closed: code=${event.code}`);
|
||||
console.log(`[Beeline] WebSocket closed: code=${event.code}, reason=${event.reason}`);
|
||||
chrome.runtime.sendMessage({ _beeline: true, type: "ws_close" });
|
||||
reconnectAttempts++;
|
||||
// Reconnect after delay
|
||||
setTimeout(connect, 2000);
|
||||
};
|
||||
|
||||
ws.onerror = (error) => {
|
||||
console.error("[Beeline] WebSocket error:", error);
|
||||
ws.close();
|
||||
ws.onerror = () => {
|
||||
// Don't log the full error object - it's usually just an Event
|
||||
// The actual error will be reflected in onclose
|
||||
console.warn(`[Beeline] WebSocket connection failed (server may not be running)`);
|
||||
// Don't close here - let onclose handle cleanup
|
||||
};
|
||||
} catch (error) {
|
||||
console.error("[Beeline] Failed to create WebSocket:", error);
|
||||
console.error("[Beeline] Failed to create WebSocket:", error.message);
|
||||
reconnectAttempts++;
|
||||
setTimeout(connect, 2000);
|
||||
}
|
||||
@@ -58,9 +62,11 @@ chrome.runtime.onMessage.addListener((msg) => {
|
||||
if (ws && ws.readyState === WebSocket.OPEN) {
|
||||
ws.send(msg.data);
|
||||
} else {
|
||||
console.warn("[Beeline] Cannot send - WebSocket not connected");
|
||||
console.warn("[Beeline] Cannot send - WebSocket not connected (state: %s)",
|
||||
ws ? ws.readyState : "null");
|
||||
}
|
||||
}
|
||||
});
|
||||
|
||||
// Start connection
|
||||
connect();
|
||||
|
||||
@@ -83,11 +83,28 @@ def _find_project_root() -> str:
|
||||
|
||||
|
||||
def _resolve_path(path: str) -> str:
|
||||
"""Resolve path relative to PROJECT_ROOT. Raises ValueError if outside."""
|
||||
"""Resolve path relative to PROJECT_ROOT. Raises ValueError if outside.
|
||||
|
||||
Also allows access to ~/.hive/ directory for agent session data files.
|
||||
"""
|
||||
# Normalize slashes for cross-platform (e.g. exports/hi_agent from LLM)
|
||||
path = path.replace("/", os.sep)
|
||||
|
||||
# Expand ~ to home directory
|
||||
if path.startswith("~"):
|
||||
path = os.path.expanduser(path)
|
||||
|
||||
if os.path.isabs(path):
|
||||
resolved = os.path.abspath(path)
|
||||
|
||||
# Allow access to ~/.hive/ for agent session data
|
||||
hive_dir = os.path.expanduser("~/.hive")
|
||||
try:
|
||||
if os.path.commonpath([resolved, hive_dir]) == hive_dir:
|
||||
return resolved # Path is under ~/.hive, allow it
|
||||
except ValueError:
|
||||
pass
|
||||
|
||||
try:
|
||||
common = os.path.commonpath([resolved, PROJECT_ROOT])
|
||||
except ValueError:
|
||||
|
||||
@@ -0,0 +1,91 @@
|
||||
#!/usr/bin/env python
|
||||
"""
|
||||
Direct browser control test - uses the bridge directly.
|
||||
Run: uv run python direct_browser_test.py
|
||||
"""
|
||||
|
||||
import asyncio
|
||||
import sys
|
||||
from pathlib import Path
|
||||
|
||||
sys.path.insert(0, str(Path(__file__).parent / "src"))
|
||||
|
||||
from gcu.browser.bridge import BeelineBridge
|
||||
|
||||
|
||||
async def main():
|
||||
print("=" * 60)
|
||||
print("DIRECT BROWSER TEST")
|
||||
print("=" * 60)
|
||||
|
||||
bridge = BeelineBridge()
|
||||
await bridge.start()
|
||||
|
||||
# Wait for connection
|
||||
print("\nWaiting for extension connection...")
|
||||
for i in range(10):
|
||||
await asyncio.sleep(1)
|
||||
if bridge.is_connected:
|
||||
print("✓ Extension connected!")
|
||||
break
|
||||
print(f" Waiting... ({i+1}/10)")
|
||||
else:
|
||||
print("✗ Extension not connected")
|
||||
await bridge.stop()
|
||||
return
|
||||
|
||||
# Create a context (tab group)
|
||||
print("\n--- Creating browser context ---")
|
||||
ctx = await bridge.create_context("test-session")
|
||||
tab_id = ctx.get("tabId")
|
||||
group_id = ctx.get("groupId")
|
||||
print(f"✓ Context created: tabId={tab_id}, groupId={group_id}")
|
||||
|
||||
# Navigate
|
||||
print("\n--- Navigating to example.com ---")
|
||||
result = await bridge.navigate(tab_id, "https://example.com", wait_until="load")
|
||||
print(f"✓ Navigated: {result}")
|
||||
|
||||
await asyncio.sleep(1)
|
||||
|
||||
# Get snapshot
|
||||
print("\n--- Getting page snapshot ---")
|
||||
snapshot = await bridge.snapshot(tab_id)
|
||||
tree = snapshot.get("tree", "")
|
||||
print(f"✓ Snapshot ({len(tree)} chars):")
|
||||
print(tree[:500] + "..." if len(tree) > 500 else tree)
|
||||
|
||||
# Click the link
|
||||
print("\n--- Clicking link ---")
|
||||
result = await bridge.click(tab_id, "a", timeout_ms=5000)
|
||||
print(f"Click result: {result}")
|
||||
|
||||
if result.get("ok"):
|
||||
print("✓ Click succeeded!")
|
||||
await asyncio.sleep(2)
|
||||
# Go back
|
||||
await bridge.go_back(tab_id)
|
||||
await asyncio.sleep(1)
|
||||
|
||||
# Test type
|
||||
print("\n--- Testing type on Google ---")
|
||||
await bridge.navigate(tab_id, "https://www.google.com", wait_until="load")
|
||||
await asyncio.sleep(2)
|
||||
|
||||
result = await bridge.type_text(tab_id, "textarea[name='q']", "hello world")
|
||||
print(f"Type result: {result}")
|
||||
|
||||
if result.get("ok"):
|
||||
print("✓ Type succeeded!")
|
||||
|
||||
# Cleanup
|
||||
print("\n--- Cleaning up ---")
|
||||
await bridge.destroy_context(group_id)
|
||||
print("✓ Context destroyed")
|
||||
|
||||
await bridge.stop()
|
||||
print("✓ Bridge stopped")
|
||||
|
||||
|
||||
if __name__ == "__main__":
|
||||
asyncio.run(main())
|
||||
@@ -0,0 +1,185 @@
|
||||
#!/usr/bin/env python
|
||||
"""
|
||||
Test browser tools on LinkedIn - specifically tests scroll and snapshot fixes.
|
||||
|
||||
Run: uv run python linkedin_browser_test.py
|
||||
"""
|
||||
|
||||
import asyncio
|
||||
import sys
|
||||
from pathlib import Path
|
||||
|
||||
sys.path.insert(0, str(Path(__file__).parent / "src"))
|
||||
|
||||
from gcu.browser.bridge import BeelineBridge
|
||||
|
||||
|
||||
async def main():
|
||||
print("=" * 60)
|
||||
print("LINKEDIN BROWSER TEST")
|
||||
print("=" * 60)
|
||||
print("\nThis tests the fixes for:")
|
||||
print("1. Scroll on nested scrollable containers (LinkedIn feed)")
|
||||
print("2. Snapshot timeout on large DOM trees")
|
||||
print()
|
||||
|
||||
bridge = BeelineBridge()
|
||||
|
||||
try:
|
||||
print("Starting bridge...")
|
||||
await bridge.start()
|
||||
|
||||
for i in range(10):
|
||||
await asyncio.sleep(1)
|
||||
if bridge.is_connected:
|
||||
print("✓ Extension connected!")
|
||||
break
|
||||
print(f"Waiting for extension... ({i+1}/10)")
|
||||
else:
|
||||
print("✗ Extension not connected")
|
||||
return
|
||||
|
||||
# Create context
|
||||
context = await bridge.create_context("linkedin-test")
|
||||
tab_id = context.get("tabId")
|
||||
group_id = context.get("groupId")
|
||||
print(f"✓ Created tab: {tab_id}")
|
||||
|
||||
# Navigate to LinkedIn
|
||||
print("\n--- Navigating to LinkedIn ---")
|
||||
try:
|
||||
await bridge.navigate(tab_id, "https://www.linkedin.com", wait_until="load", timeout_ms=30000)
|
||||
print("✓ Page loaded")
|
||||
except Exception as e:
|
||||
print(f"Navigation result: {e}")
|
||||
|
||||
await asyncio.sleep(2)
|
||||
|
||||
# Test 1: Snapshot with timeout
|
||||
print("\n--- Test 1: Snapshot (with timeout protection) ---")
|
||||
try:
|
||||
import time
|
||||
start = time.perf_counter()
|
||||
snapshot = await bridge.snapshot(tab_id, timeout_s=15.0)
|
||||
elapsed = time.perf_counter() - start
|
||||
tree = snapshot.get("tree", "")
|
||||
print(f"✓ Snapshot completed in {elapsed:.2f}s")
|
||||
print(f" Tree length: {len(tree)} chars")
|
||||
if "truncated" in tree:
|
||||
print(" (Tree was truncated due to size)")
|
||||
print(f" First 300 chars:\n{tree[:300]}...")
|
||||
except asyncio.TimeoutError:
|
||||
print("✗ Snapshot timed out (this shouldn't happen with 15s timeout)")
|
||||
except Exception as e:
|
||||
print(f"✗ Snapshot error: {e}")
|
||||
|
||||
# Test 2: Scroll - should now find nested scrollable container
|
||||
print("\n--- Test 2: Scroll (finds nested scrollable container) ---")
|
||||
|
||||
# Get scroll position of ALL scrollable elements before
|
||||
get_scroll_positions = """
|
||||
(function() {
|
||||
const results = { window: { x: window.scrollX, y: window.scrollY } };
|
||||
const scrollables = document.querySelectorAll('*');
|
||||
let idx = 0;
|
||||
for (const el of scrollables) {
|
||||
const style = getComputedStyle(el);
|
||||
if ((style.overflowY === 'scroll' || style.overflowY === 'auto') &&
|
||||
el.scrollHeight > el.clientHeight) {
|
||||
const rect = el.getBoundingClientRect();
|
||||
if (rect.width > 200 && rect.height > 200) {
|
||||
results['container_' + idx] = {
|
||||
tag: el.tagName,
|
||||
class: el.className.substring(0, 30),
|
||||
scrollTop: el.scrollTop,
|
||||
scrollHeight: el.scrollHeight
|
||||
};
|
||||
idx++;
|
||||
}
|
||||
}
|
||||
}
|
||||
return results;
|
||||
})();
|
||||
"""
|
||||
|
||||
try:
|
||||
pos_before = await bridge.evaluate(tab_id, get_scroll_positions)
|
||||
before_data = pos_before.get("result", {}) if pos_before else {}
|
||||
print(f" Positions before scroll:")
|
||||
if isinstance(before_data, dict):
|
||||
for key, val in before_data.items():
|
||||
print(f" {key}: {val}")
|
||||
else:
|
||||
print(f" {before_data}")
|
||||
|
||||
result = await bridge.scroll(tab_id, "down", 500)
|
||||
print(f" Scroll result: {result}")
|
||||
|
||||
if result.get("ok"):
|
||||
method = result.get("method", "unknown")
|
||||
container = result.get("container", "unknown")
|
||||
print(f" ✓ Scroll command succeeded using {method}")
|
||||
else:
|
||||
print(f" ✗ Scroll command failed: {result.get('error')}")
|
||||
|
||||
await asyncio.sleep(1)
|
||||
|
||||
# Get scroll positions after
|
||||
pos_after = await bridge.evaluate(tab_id, get_scroll_positions)
|
||||
after_data = pos_after.get("result", {}) if pos_after else {}
|
||||
print(f" Positions after scroll:")
|
||||
if isinstance(after_data, dict):
|
||||
for key, val in after_data.items():
|
||||
print(f" {key}: {val}")
|
||||
else:
|
||||
print(f" {after_data}")
|
||||
|
||||
# Check if any position changed
|
||||
changed = False
|
||||
if isinstance(before_data, dict) and isinstance(after_data, dict):
|
||||
for key in after_data:
|
||||
if key in before_data:
|
||||
b_val = before_data[key].get("scrollTop", 0) if isinstance(before_data[key], dict) else 0
|
||||
a_val = after_data[key].get("scrollTop", 0) if isinstance(after_data[key], dict) else 0
|
||||
if a_val != b_val:
|
||||
print(f" ✓ SCROLL CONFIRMED: {key} changed from {b_val} to {a_val}")
|
||||
changed = True
|
||||
if not changed:
|
||||
print(" ✗ NO SCROLL DETECTED: No scroll positions changed")
|
||||
|
||||
except Exception as e:
|
||||
import traceback
|
||||
print(f"✗ Scroll error: {e}")
|
||||
traceback.print_exc()
|
||||
|
||||
# Test 3: Multiple scrolls
|
||||
print("\n--- Test 3: Multiple Scrolls ---")
|
||||
for i in range(3):
|
||||
try:
|
||||
result = await bridge.scroll(tab_id, "down", 200)
|
||||
print(f" Scroll {i+1}: {result.get('method', 'failed')} on {result.get('container', 'unknown')}")
|
||||
await asyncio.sleep(0.5)
|
||||
except Exception as e:
|
||||
print(f" Scroll {i+1} failed: {e}")
|
||||
|
||||
# Test 4: Snapshot after scroll
|
||||
print("\n--- Test 4: Snapshot After Scroll ---")
|
||||
try:
|
||||
snapshot = await bridge.snapshot(tab_id, timeout_s=10.0)
|
||||
tree = snapshot.get("tree", "")
|
||||
print(f"✓ Snapshot: {len(tree)} chars")
|
||||
except Exception as e:
|
||||
print(f"✗ Snapshot error: {e}")
|
||||
|
||||
# Cleanup
|
||||
print("\n=== Cleanup ===")
|
||||
await bridge.destroy_context(group_id)
|
||||
print("✓ Context destroyed")
|
||||
|
||||
finally:
|
||||
await bridge.stop()
|
||||
print("✓ Bridge stopped")
|
||||
|
||||
|
||||
if __name__ == "__main__":
|
||||
asyncio.run(main())
|
||||
@@ -0,0 +1,321 @@
|
||||
#!/usr/bin/env python
|
||||
"""
|
||||
Complex browser scenarios test - real browser interaction.
|
||||
|
||||
Tests complex selectors and interactions similar to:
|
||||
- LinkedIn profile scrolling and data extraction
|
||||
- Twitter/X infinite timeline
|
||||
- YouTube video controls
|
||||
|
||||
Prerequisites:
|
||||
1. Chrome with Beeline extension installed
|
||||
2. Logged into LinkedIn/Twitter/YouTube (for some tests)
|
||||
3. Run: uv run python manual_browser_complex_test.py
|
||||
"""
|
||||
|
||||
import asyncio
|
||||
import sys
|
||||
from pathlib import Path
|
||||
|
||||
sys.path.insert(0, str(Path(__file__).parent / "src"))
|
||||
|
||||
from gcu.browser.bridge import BeelineBridge
|
||||
|
||||
|
||||
async def wait_for_bridge(bridge: BeelineBridge, timeout: int = 5) -> bool:
|
||||
"""Wait for extension connection."""
|
||||
await bridge.start()
|
||||
for i in range(timeout):
|
||||
await asyncio.sleep(1)
|
||||
if bridge.is_connected:
|
||||
return True
|
||||
return False
|
||||
|
||||
|
||||
async def test_linkedin_profile_scroll(bridge: BeelineBridge, tab_id: int) -> dict:
|
||||
"""Test LinkedIn-style infinite scroll and data extraction."""
|
||||
print("\n=== LinkedIn: Profile Scroll Test ===")
|
||||
|
||||
try:
|
||||
# Navigate to a LinkedIn page (public profile, no login required)
|
||||
await bridge.navigate(tab_id, "https://www.linkedin.com/in/williamhgates/", wait_until="networkidle")
|
||||
await asyncio.sleep(3)
|
||||
|
||||
results = {"steps": []}
|
||||
|
||||
# Scroll down to load more content
|
||||
for i in range(3):
|
||||
result = await bridge.scroll(tab_id, "down", 400)
|
||||
results["steps"].append(f"scroll_{i}: {result.get('ok')}")
|
||||
await asyncio.sleep(1)
|
||||
|
||||
# Extract profile data using complex selectors
|
||||
profile_script = """
|
||||
const name = document.querySelector('h1.text-heading-xlarge')?.innerText ||
|
||||
document.querySelector('h1')?.innerText || 'Not found';
|
||||
const headline = document.querySelector('.text-body-medium')?.innerText || 'Not found';
|
||||
return { name, headline };
|
||||
"""
|
||||
result = await bridge.evaluate(tab_id, profile_script)
|
||||
profile_data = result.get("result", {}).get("value", {})
|
||||
results["profile"] = profile_data
|
||||
print(f" Profile: {profile_data}")
|
||||
|
||||
# Check if we found real content
|
||||
if profile_data.get("name") and profile_data.get("name") != "Not found":
|
||||
results["ok"] = True
|
||||
print(" ✓ Successfully extracted LinkedIn profile data")
|
||||
else:
|
||||
results["ok"] = False
|
||||
print(" ✗ Could not extract profile data (may need login)")
|
||||
|
||||
return results
|
||||
|
||||
except Exception as e:
|
||||
print(f" ✗ Error: {e}")
|
||||
return {"ok": False, "error": str(e)}
|
||||
|
||||
|
||||
async def test_twitter_timeline(bridge: BeelineBridge, tab_id: int) -> dict:
|
||||
"""Test Twitter/X timeline interaction."""
|
||||
print("\n=== Twitter/X: Timeline Test ===")
|
||||
|
||||
try:
|
||||
# Navigate to Twitter explore (doesn't require login)
|
||||
await bridge.navigate(tab_id, "https://twitter.com/explore", wait_until="networkidle")
|
||||
await asyncio.sleep(3)
|
||||
|
||||
results = {"steps": []}
|
||||
|
||||
# Try to find and interact with content
|
||||
extraction_script = """
|
||||
// Twitter has complex selectors
|
||||
const tweets = document.querySelectorAll('article[data-testid="tweet"]');
|
||||
const titles = document.querySelectorAll('h2');
|
||||
return {
|
||||
tweetCount: tweets.length,
|
||||
titleCount: titles.length,
|
||||
pageTitle: document.title
|
||||
};
|
||||
"""
|
||||
result = await bridge.evaluate(tab_id, extraction_script)
|
||||
data = result.get("result", {}).get("value", {})
|
||||
results["data"] = data
|
||||
print(f" Page data: {data}")
|
||||
|
||||
# Scroll to load more
|
||||
await bridge.scroll(tab_id, "down", 500)
|
||||
await asyncio.sleep(2)
|
||||
results["steps"].append("scrolled")
|
||||
|
||||
results["ok"] = data.get("pageTitle", "").lower().find("x") >= 0 or data.get("tweetCount", 0) >= 0
|
||||
print(f" {'✓' if results['ok'] else '✗'} Twitter page loaded")
|
||||
return results
|
||||
|
||||
except Exception as e:
|
||||
print(f" ✗ Error: {e}")
|
||||
return {"ok": False, "error": str(e)}
|
||||
|
||||
|
||||
async def test_youtube_controls(bridge: BeelineBridge, tab_id: int) -> dict:
|
||||
"""Test YouTube video player interaction."""
|
||||
print("\n=== YouTube: Video Controls Test ===")
|
||||
|
||||
try:
|
||||
# Navigate to a YouTube video
|
||||
await bridge.navigate(tab_id, "https://www.youtube.com/watch?v=dQw4w9WgXcQ", wait_until="networkidle")
|
||||
await asyncio.sleep(3)
|
||||
|
||||
results = {"steps": []}
|
||||
|
||||
# Get player state
|
||||
player_script = """
|
||||
const video = document.querySelector('video');
|
||||
if (video) {
|
||||
return {
|
||||
hasVideo: true,
|
||||
paused: video.paused,
|
||||
currentTime: Math.round(video.currentTime),
|
||||
duration: Math.round(video.duration),
|
||||
muted: video.muted
|
||||
};
|
||||
}
|
||||
return { hasVideo: false };
|
||||
"""
|
||||
result = await bridge.evaluate(tab_id, player_script)
|
||||
state = result.get("result", {}).get("value", {})
|
||||
results["initialState"] = state
|
||||
print(f" Initial state: {state}")
|
||||
|
||||
if state.get("hasVideo"):
|
||||
# Try clicking play/pause button
|
||||
click_result = await bridge.click(tab_id, "button.ytp-play-button", timeout_ms=5000)
|
||||
results["steps"].append(f"click_play: {click_result.get('ok')}")
|
||||
await asyncio.sleep(1)
|
||||
|
||||
# Check state after click
|
||||
result = await bridge.evaluate(tab_id, player_script)
|
||||
new_state = result.get("result", {}).get("value", {})
|
||||
results["afterClickState"] = new_state
|
||||
print(f" After click: {new_state}")
|
||||
|
||||
results["ok"] = True
|
||||
print(" ✓ YouTube video controls working")
|
||||
else:
|
||||
results["ok"] = False
|
||||
print(" ✗ Video element not found")
|
||||
|
||||
return results
|
||||
|
||||
except Exception as e:
|
||||
print(f" ✗ Error: {e}")
|
||||
return {"ok": False, "error": str(e)}
|
||||
|
||||
|
||||
async def test_form_interaction(bridge: BeelineBridge, tab_id: int) -> dict:
|
||||
"""Test complex form filling with various input types."""
|
||||
print("\n=== Form: Complex Input Test ===")
|
||||
|
||||
try:
|
||||
# Navigate to a form testing page
|
||||
await bridge.navigate(tab_id, "https://httpbin.org/forms/post", wait_until="load")
|
||||
await asyncio.sleep(2)
|
||||
|
||||
results = {"steps": []}
|
||||
|
||||
# Fill text input
|
||||
result = await bridge.type_text(tab_id, "input[name='custname']", "Test Customer")
|
||||
results["steps"].append(f"type_name: {result.get('ok')}")
|
||||
|
||||
# Fill textarea
|
||||
result = await bridge.type_text(tab_id, "textarea[name='comments']", "This is a test comment with multiple lines.\nLine 2.\nLine 3.")
|
||||
results["steps"].append(f"type_comments: {result.get('ok')}")
|
||||
|
||||
# Click radio button
|
||||
result = await bridge.click(tab_id, "input[value='medium']")
|
||||
results["steps"].append(f"click_radio: {result.get('ok')}")
|
||||
|
||||
# Click checkbox
|
||||
result = await bridge.click(tab_id, "input[name='topping'][value='cheese']")
|
||||
results["steps"].append(f"click_checkbox: {result.get('ok')}")
|
||||
|
||||
# Verify form state
|
||||
verify_script = """
|
||||
return {
|
||||
name: document.querySelector("input[name='custname']")?.value,
|
||||
comments: document.querySelector("textarea[name='comments']")?.value,
|
||||
medium: document.querySelector("input[value='medium']")?.checked,
|
||||
cheese: document.querySelector("input[name='topping'][value='cheese']")?.checked
|
||||
};
|
||||
"""
|
||||
result = await bridge.evaluate(tab_id, verify_script)
|
||||
form_state = result.get("result", {}).get("value", {})
|
||||
results["formState"] = form_state
|
||||
print(f" Form state: {form_state}")
|
||||
|
||||
# Check all fields are filled correctly
|
||||
results["ok"] = (
|
||||
form_state.get("name") == "Test Customer" and
|
||||
form_state.get("medium") is True and
|
||||
form_state.get("cheese") is True
|
||||
)
|
||||
|
||||
print(f" {'✓' if results['ok'] else '✗'} Form interaction")
|
||||
return results
|
||||
|
||||
except Exception as e:
|
||||
print(f" ✗ Error: {e}")
|
||||
return {"ok": False, "error": str(e)}
|
||||
|
||||
|
||||
async def test_drag_drop(bridge: BeelineBridge, tab_id: int) -> dict:
|
||||
"""Test drag and drop functionality."""
|
||||
print("\n=== Drag & Drop Test ===")
|
||||
|
||||
try:
|
||||
# Navigate to a drag-drop demo page
|
||||
await bridge.navigate(tab_id, "https://www.w3schools.com/html/html5_draganddrop.asp", wait_until="load")
|
||||
await asyncio.sleep(2)
|
||||
|
||||
results = {"steps": []}
|
||||
|
||||
# Scroll to demo
|
||||
await bridge.scroll(tab_id, "down", 600)
|
||||
await asyncio.sleep(1)
|
||||
|
||||
# Try drag operation - this page has draggable elements
|
||||
# Note: HTML5 drag-drop via CDP is limited, this tests mouse events
|
||||
result = await bridge.evaluate(tab_id, """
|
||||
// Check if drag elements exist
|
||||
const drag1 = document.getElementById('drag1');
|
||||
const div2 = document.getElementById('div2');
|
||||
return {
|
||||
hasDragElement: !!drag1,
|
||||
hasDropZone: !!div2
|
||||
};
|
||||
""")
|
||||
elements = result.get("result", {}).get("value", {})
|
||||
results["elements"] = elements
|
||||
print(f" Elements found: {elements}")
|
||||
|
||||
results["ok"] = elements.get("hasDragElement") and elements.get("hasDropZone")
|
||||
print(f" {'✓' if results['ok'] else '✗'} Drag elements found")
|
||||
return results
|
||||
|
||||
except Exception as e:
|
||||
print(f" ✗ Error: {e}")
|
||||
return {"ok": False, "error": str(e)}
|
||||
|
||||
|
||||
async def main():
|
||||
print("=" * 60)
|
||||
print("COMPLEX BROWSER SCENARIOS TEST")
|
||||
print("=" * 60)
|
||||
print("\nThis tests complex interactions on real websites.")
|
||||
print("Some tests may fail if not logged in to the respective sites.\n")
|
||||
|
||||
bridge = BeelineBridge()
|
||||
|
||||
try:
|
||||
if not await wait_for_bridge(bridge):
|
||||
print("❌ Extension not connected. Ensure Chrome extension is running.")
|
||||
return
|
||||
|
||||
# Create context
|
||||
context = await bridge.create_context("complex-test")
|
||||
tab_id = context.get("tabId")
|
||||
group_id = context.get("groupId")
|
||||
print(f"✓ Created context: tabId={tab_id}")
|
||||
|
||||
# Run tests
|
||||
results = []
|
||||
results.append(("LinkedIn Profile", await test_linkedin_profile_scroll(bridge, tab_id)))
|
||||
results.append(("Twitter Timeline", await test_twitter_timeline(bridge, tab_id)))
|
||||
results.append(("YouTube Controls", await test_youtube_controls(bridge, tab_id)))
|
||||
results.append(("Form Interaction", await test_form_interaction(bridge, tab_id)))
|
||||
results.append(("Drag & Drop", await test_drag_drop(bridge, tab_id)))
|
||||
|
||||
# Cleanup
|
||||
print("\n=== Cleanup ===")
|
||||
await bridge.destroy_context(group_id)
|
||||
print("✓ Context destroyed")
|
||||
|
||||
# Summary
|
||||
print("\n" + "=" * 60)
|
||||
print("RESULTS")
|
||||
print("=" * 60)
|
||||
passed = sum(1 for _, r in results if r.get("ok"))
|
||||
for name, result in results:
|
||||
status = "✓" if result.get("ok") else "✗"
|
||||
print(f" {status} {name}")
|
||||
if not result.get("ok") and result.get("error"):
|
||||
print(f" Error: {result['error']}")
|
||||
print(f"\nTotal: {passed}/{len(results)} passed")
|
||||
|
||||
finally:
|
||||
await bridge.stop()
|
||||
print("\nBridge stopped.")
|
||||
|
||||
|
||||
if __name__ == "__main__":
|
||||
asyncio.run(main())
|
||||
@@ -0,0 +1,280 @@
|
||||
#!/usr/bin/env python
|
||||
"""
|
||||
Manual browser tools test - connects to real Chrome extension.
|
||||
|
||||
Prerequisites:
|
||||
1. Chrome with Beeline extension installed and enabled
|
||||
2. Run: uv run python manual_browser_test.py
|
||||
|
||||
This will test:
|
||||
- Bridge connection
|
||||
- Tab group creation
|
||||
- Navigation
|
||||
- Click, type, scroll interactions
|
||||
- Snapshot/screenshot
|
||||
- Complex JS execution (LinkedIn-style selectors)
|
||||
"""
|
||||
|
||||
import asyncio
|
||||
import json
|
||||
import sys
|
||||
from pathlib import Path
|
||||
|
||||
# Add src to path
|
||||
sys.path.insert(0, str(Path(__file__).parent / "src"))
|
||||
|
||||
from gcu.browser.bridge import BeelineBridge
|
||||
|
||||
|
||||
async def test_connection(bridge: BeelineBridge) -> bool:
|
||||
"""Test 1: Extension connection."""
|
||||
print("\n=== Test 1: Extension Connection ===")
|
||||
print("Starting bridge on port 9229...")
|
||||
await bridge.start()
|
||||
|
||||
for i in range(5):
|
||||
await asyncio.sleep(1)
|
||||
if bridge.is_connected:
|
||||
print("✓ Extension connected!")
|
||||
return True
|
||||
print(f" Waiting... ({i+1}/5)")
|
||||
|
||||
print("✗ Extension not connected. Ensure Chrome extension is installed.")
|
||||
return False
|
||||
|
||||
|
||||
async def test_context_creation(bridge: BeelineBridge) -> dict | None:
|
||||
"""Test 2: Create tab group/context."""
|
||||
print("\n=== Test 2: Create Tab Group ===")
|
||||
try:
|
||||
result = await bridge.create_context("manual-test-agent")
|
||||
print(f"✓ Created context: groupId={result.get('groupId')}, tabId={result.get('tabId')}")
|
||||
return result
|
||||
except Exception as e:
|
||||
print(f"✗ Failed: {e}")
|
||||
return None
|
||||
|
||||
|
||||
async def test_navigation(bridge: BeelineBridge, tab_id: int) -> bool:
|
||||
"""Test 3: Navigate to example.com."""
|
||||
print("\n=== Test 3: Navigation ===")
|
||||
try:
|
||||
result = await bridge.navigate(tab_id, "https://example.com", wait_until="load")
|
||||
print(f"✓ Navigated to: {result.get('url')}")
|
||||
await asyncio.sleep(1)
|
||||
return True
|
||||
except Exception as e:
|
||||
print(f"✗ Failed: {e}")
|
||||
return False
|
||||
|
||||
|
||||
async def test_snapshot(bridge: BeelineBridge, tab_id: int) -> bool:
|
||||
"""Test 4: Get accessibility snapshot."""
|
||||
print("\n=== Test 4: Accessibility Snapshot ===")
|
||||
try:
|
||||
result = await bridge.snapshot(tab_id)
|
||||
tree = result.get("tree", "")
|
||||
lines = tree.split("\n")[:10]
|
||||
print(f"✓ Got snapshot ({len(tree)} chars)")
|
||||
print(" First 10 lines:")
|
||||
for line in lines:
|
||||
print(f" {line}")
|
||||
return True
|
||||
except Exception as e:
|
||||
print(f"✗ Failed: {e}")
|
||||
return False
|
||||
|
||||
|
||||
async def test_click(bridge: BeelineBridge, tab_id: int) -> bool:
|
||||
"""Test 5: Click a link."""
|
||||
print("\n=== Test 5: Click Element ===")
|
||||
try:
|
||||
# example.com has a link to "More information..."
|
||||
result = await bridge.click(tab_id, "a", timeout_ms=5000)
|
||||
if result.get("ok"):
|
||||
print(f"✓ Clicked link at ({result.get('x')}, {result.get('y')})")
|
||||
await asyncio.sleep(2)
|
||||
# Go back
|
||||
await bridge.go_back(tab_id)
|
||||
await asyncio.sleep(1)
|
||||
return True
|
||||
else:
|
||||
print(f"✗ Click failed: {result.get('error')}")
|
||||
return False
|
||||
except Exception as e:
|
||||
print(f"✗ Failed: {e}")
|
||||
return False
|
||||
|
||||
|
||||
async def test_type_and_clear(bridge: BeelineBridge, tab_id: int) -> bool:
|
||||
"""Test 6: Type into an input field."""
|
||||
print("\n=== Test 6: Type Text ===")
|
||||
try:
|
||||
# Navigate to a page with an input
|
||||
await bridge.navigate(tab_id, "https://www.google.com", wait_until="load")
|
||||
await asyncio.sleep(2)
|
||||
|
||||
# Type in search box
|
||||
result = await bridge.type_text(tab_id, "textarea[name='q']", "hello world")
|
||||
if result.get("ok"):
|
||||
print("✓ Typed 'hello world' into search box")
|
||||
await asyncio.sleep(0.5)
|
||||
|
||||
# Clear and type something else
|
||||
await bridge.press_key(tab_id, "Control+a")
|
||||
await asyncio.sleep(0.2)
|
||||
await bridge.type_text(tab_id, "textarea[name='q']", "new search")
|
||||
print("✓ Replaced with 'new search'")
|
||||
return True
|
||||
else:
|
||||
print(f"✗ Type failed: {result.get('error')}")
|
||||
return False
|
||||
except Exception as e:
|
||||
print(f"✗ Failed: {e}")
|
||||
return False
|
||||
|
||||
|
||||
async def test_scroll(bridge: BeelineBridge, tab_id: int) -> bool:
|
||||
"""Test 7: Scroll page."""
|
||||
print("\n=== Test 7: Scroll ===")
|
||||
try:
|
||||
# Scroll down
|
||||
result = await bridge.scroll(tab_id, "down", 500)
|
||||
if result.get("ok"):
|
||||
print("✓ Scrolled down 500px")
|
||||
await asyncio.sleep(0.5)
|
||||
|
||||
# Scroll up
|
||||
await bridge.scroll(tab_id, "up", 250)
|
||||
print("✓ Scrolled up 250px")
|
||||
return True
|
||||
else:
|
||||
print(f"✗ Scroll failed: {result.get('error')}")
|
||||
return False
|
||||
except Exception as e:
|
||||
print(f"✗ Failed: {e}")
|
||||
return False
|
||||
|
||||
|
||||
async def test_evaluate_js(bridge: BeelineBridge, tab_id: int) -> bool:
|
||||
"""Test 8: Execute JavaScript."""
|
||||
print("\n=== Test 8: JavaScript Execution ===")
|
||||
try:
|
||||
# Simple return
|
||||
result = await bridge.evaluate(tab_id, "return document.title;")
|
||||
print(f"✓ Page title: {result.get('result', {}).get('value')}")
|
||||
|
||||
# Complex selector (LinkedIn-style)
|
||||
complex_script = """
|
||||
const links = document.querySelectorAll('a');
|
||||
return {
|
||||
total: links.length,
|
||||
first: links[0]?.href || null
|
||||
};
|
||||
"""
|
||||
result = await bridge.evaluate(tab_id, complex_script)
|
||||
value = result.get("result", {}).get("value", {})
|
||||
print(f"✓ Found {value.get('total')} links, first: {value.get('first', 'N/A')[:50]}...")
|
||||
return True
|
||||
except Exception as e:
|
||||
print(f"✗ Failed: {e}")
|
||||
return False
|
||||
|
||||
|
||||
async def test_screenshot(bridge: BeelineBridge, tab_id: int) -> bool:
|
||||
"""Test 9: Take screenshot."""
|
||||
print("\n=== Test 9: Screenshot ===")
|
||||
try:
|
||||
result = await bridge.screenshot(tab_id, full_page=False)
|
||||
if result.get("ok"):
|
||||
data = result.get("data", "")
|
||||
print(f"✓ Screenshot captured ({len(data)} chars base64)")
|
||||
return True
|
||||
else:
|
||||
print(f"✗ Screenshot failed: {result.get('error')}")
|
||||
return False
|
||||
except Exception as e:
|
||||
print(f"✗ Failed: {e}")
|
||||
return False
|
||||
|
||||
|
||||
async def test_tab_management(bridge: BeelineBridge, group_id: int, tab_id: int) -> bool:
|
||||
"""Test 10: Create and close tabs."""
|
||||
print("\n=== Test 10: Tab Management ===")
|
||||
try:
|
||||
# Create new tab
|
||||
new_tab = await bridge.create_tab(group_id, "https://httpbin.org")
|
||||
new_tab_id = new_tab.get("tabId")
|
||||
print(f"✓ Created new tab: {new_tab_id}")
|
||||
await asyncio.sleep(2)
|
||||
|
||||
# List tabs
|
||||
tabs = await bridge.list_tabs(group_id)
|
||||
print(f"✓ Group has {len(tabs.get('tabs', []))} tabs")
|
||||
|
||||
# Close the new tab
|
||||
await bridge.close_tab(new_tab_id)
|
||||
print(f"✓ Closed tab {new_tab_id}")
|
||||
return True
|
||||
except Exception as e:
|
||||
print(f"✗ Failed: {e}")
|
||||
return False
|
||||
|
||||
|
||||
async def main():
|
||||
print("=" * 60)
|
||||
print("MANUAL BROWSER TOOLS TEST")
|
||||
print("=" * 60)
|
||||
|
||||
bridge = BeelineBridge()
|
||||
|
||||
try:
|
||||
# Test 1: Connection
|
||||
if not await test_connection(bridge):
|
||||
print("\n❌ Cannot proceed without extension connection")
|
||||
return
|
||||
|
||||
# Test 2: Context creation
|
||||
context = await test_context_creation(bridge)
|
||||
if not context:
|
||||
print("\n❌ Cannot proceed without context")
|
||||
return
|
||||
|
||||
tab_id = context.get("tabId")
|
||||
group_id = context.get("groupId")
|
||||
|
||||
results = []
|
||||
|
||||
# Run all tests
|
||||
results.append(("Navigation", await test_navigation(bridge, tab_id)))
|
||||
results.append(("Snapshot", await test_snapshot(bridge, tab_id)))
|
||||
results.append(("Click", await test_click(bridge, tab_id)))
|
||||
results.append(("Type", await test_type_and_clear(bridge, tab_id)))
|
||||
results.append(("Scroll", await test_scroll(bridge, tab_id)))
|
||||
results.append(("Evaluate JS", await test_evaluate_js(bridge, tab_id)))
|
||||
results.append(("Screenshot", await test_screenshot(bridge, tab_id)))
|
||||
results.append(("Tab Management", await test_tab_management(bridge, group_id, tab_id)))
|
||||
|
||||
# Cleanup
|
||||
print("\n=== Cleanup ===")
|
||||
await bridge.destroy_context(group_id)
|
||||
print("✓ Destroyed context")
|
||||
|
||||
# Summary
|
||||
print("\n" + "=" * 60)
|
||||
print("RESULTS SUMMARY")
|
||||
print("=" * 60)
|
||||
passed = sum(1 for _, r in results if r)
|
||||
total = len(results)
|
||||
for name, result in results:
|
||||
status = "✓ PASS" if result else "✗ FAIL"
|
||||
print(f" {status}: {name}")
|
||||
print(f"\nTotal: {passed}/{total} passed")
|
||||
|
||||
finally:
|
||||
await bridge.stop()
|
||||
print("\nBridge stopped.")
|
||||
|
||||
|
||||
if __name__ == "__main__":
|
||||
asyncio.run(main())
|
||||
@@ -0,0 +1,148 @@
|
||||
#!/usr/bin/env python
|
||||
"""
|
||||
Debug browser click - specifically tests the Input.enable domain issue.
|
||||
|
||||
Run: uv run python manual_click_debug.py
|
||||
"""
|
||||
|
||||
import asyncio
|
||||
import sys
|
||||
from pathlib import Path
|
||||
|
||||
sys.path.insert(0, str(Path(__file__).parent / "src"))
|
||||
|
||||
from gcu.browser.bridge import BeelineBridge
|
||||
|
||||
|
||||
async def main():
|
||||
print("=" * 60)
|
||||
print("BROWSER CLICK DEBUG")
|
||||
print("=" * 60)
|
||||
print("\nThis tests the click functionality and CDP domain handling.\n")
|
||||
|
||||
bridge = BeelineBridge()
|
||||
|
||||
try:
|
||||
print("Starting bridge...")
|
||||
await bridge.start()
|
||||
|
||||
for i in range(5):
|
||||
await asyncio.sleep(1)
|
||||
if bridge.is_connected:
|
||||
print("✓ Extension connected!")
|
||||
break
|
||||
print(f"Waiting for extension... ({i+1}/5)")
|
||||
else:
|
||||
print("✗ Extension not connected")
|
||||
return
|
||||
|
||||
# Create context
|
||||
context = await bridge.create_context("click-debug")
|
||||
tab_id = context.get("tabId")
|
||||
group_id = context.get("groupId")
|
||||
print(f"✓ Created tab: {tab_id}")
|
||||
|
||||
# Navigate to a simple page
|
||||
print("\nNavigating to example.com...")
|
||||
await bridge.navigate(tab_id, "https://example.com", wait_until="load")
|
||||
await asyncio.sleep(1)
|
||||
print("✓ Page loaded")
|
||||
|
||||
# Test 1: Get snapshot first
|
||||
print("\n--- Test 1: Snapshot ---")
|
||||
try:
|
||||
snapshot = await bridge.snapshot(tab_id)
|
||||
print(f"✓ Snapshot: {snapshot.get('tree', '')[:200]}...")
|
||||
except Exception as e:
|
||||
print(f"✗ Snapshot failed: {e}")
|
||||
|
||||
# Test 2: Click the "More information" link (example.com has <p> with link inside)
|
||||
print("\n--- Test 2: Click Link ---")
|
||||
try:
|
||||
# First, let's see what elements are on the page
|
||||
check_script = """
|
||||
const links = document.querySelectorAll('a');
|
||||
return Array.from(links).map(a => ({
|
||||
href: a.href,
|
||||
text: a.innerText.substring(0, 50)
|
||||
}));
|
||||
"""
|
||||
result = await bridge.evaluate(tab_id, check_script)
|
||||
# The evaluate method returns {"ok": True, "result": value}
|
||||
links = result.get("result", [])
|
||||
print(f" Found {len(links) if isinstance(links, list) else 0} links: {links}")
|
||||
|
||||
if links and isinstance(links, list) and len(links) > 0:
|
||||
# example.com structure: <p><a href="...">More information...</a></p>
|
||||
result = await bridge.click(tab_id, "a", timeout_ms=5000)
|
||||
print(f" Click result: {result}")
|
||||
if result.get("ok"):
|
||||
print(" ✓ Click succeeded!")
|
||||
await asyncio.sleep(2)
|
||||
# Go back
|
||||
await bridge.go_back(tab_id)
|
||||
await asyncio.sleep(1)
|
||||
else:
|
||||
print(f" ✗ Click failed: {result.get('error')}")
|
||||
else:
|
||||
print(" No links found to click")
|
||||
except Exception as e:
|
||||
print(f" ✗ Click exception: {e}")
|
||||
|
||||
# Test 3: Click at coordinates
|
||||
print("\n--- Test 3: Click Coordinates ---")
|
||||
try:
|
||||
result = await bridge.click_coordinate(tab_id, 100, 100)
|
||||
print(f"Click coordinate result: {result}")
|
||||
except Exception as e:
|
||||
print(f"✗ Click coordinate exception: {e}")
|
||||
|
||||
# Test 4: Type text (requires input)
|
||||
print("\n--- Test 4: Type Text ---")
|
||||
try:
|
||||
# Navigate to Google
|
||||
await bridge.navigate(tab_id, "https://www.google.com", wait_until="load")
|
||||
await asyncio.sleep(2)
|
||||
|
||||
result = await bridge.type_text(tab_id, "textarea[name='q']", "test query")
|
||||
print(f"Type result: {result}")
|
||||
if result.get("ok"):
|
||||
print("✓ Type succeeded!")
|
||||
else:
|
||||
print(f"✗ Type failed: {result.get('error')}")
|
||||
except Exception as e:
|
||||
print(f"✗ Type exception: {e}")
|
||||
|
||||
# Test 5: Hover (on a visible button)
|
||||
print("\n--- Test 5: Hover ---")
|
||||
try:
|
||||
# Stay on Google, hover over the "Google Search" button
|
||||
result = await bridge.hover(tab_id, "input[value='Google Search']", timeout_ms=5000)
|
||||
print(f"Hover result: {result}")
|
||||
if result.get("ok"):
|
||||
print("✓ Hover succeeded!")
|
||||
else:
|
||||
# Try hovering over the search input instead
|
||||
print(f" First hover failed: {result.get('error')}")
|
||||
result = await bridge.hover(tab_id, "textarea[name='q']", timeout_ms=3000)
|
||||
print(f" Hover input result: {result}")
|
||||
if result.get("ok"):
|
||||
print("✓ Hover on input succeeded!")
|
||||
else:
|
||||
print(f"✗ Hover failed: {result.get('error')}")
|
||||
except Exception as e:
|
||||
import traceback
|
||||
print(f"✗ Hover exception: {e}")
|
||||
traceback.print_exc()
|
||||
|
||||
# Cleanup
|
||||
print("\n=== Cleanup ===")
|
||||
await bridge.destroy_context(group_id)
|
||||
print("✓ Done")
|
||||
|
||||
finally:
|
||||
await bridge.stop()
|
||||
|
||||
|
||||
if __name__ == "__main__":
|
||||
asyncio.run(main())
|
||||
@@ -26,6 +26,7 @@ import subprocess
|
||||
import sys
|
||||
import tempfile
|
||||
from collections.abc import Callable
|
||||
from contextvars import ContextVar
|
||||
from pathlib import Path
|
||||
|
||||
from fastmcp import FastMCP
|
||||
@@ -106,6 +107,87 @@ BINARY_EXTENSIONS = frozenset(
|
||||
}
|
||||
)
|
||||
|
||||
# ── Context-aware sandboxing ─────────────────────────────────────────────────
|
||||
|
||||
# Context variable for additional allowed paths (beyond base_root)
|
||||
_allowed_paths_ctx: ContextVar[list[str]] = ContextVar("allowed_paths", default=[])
|
||||
|
||||
|
||||
def set_allowed_paths(paths: list[str]) -> None:
|
||||
"""Set additional allowed paths for file operations in this context.
|
||||
|
||||
Use this to grant access to paths beyond the base root (e.g., ~/.hive/
|
||||
for cross-agent file access).
|
||||
"""
|
||||
_allowed_paths_ctx.set(list(paths))
|
||||
|
||||
|
||||
def get_allowed_paths() -> list[str]:
|
||||
"""Get current allowed paths including ~/.hive/."""
|
||||
paths = list(_allowed_paths_ctx.get())
|
||||
hive_dir = os.path.expanduser("~/.hive")
|
||||
if hive_dir not in paths:
|
||||
paths.append(hive_dir)
|
||||
return paths
|
||||
|
||||
|
||||
def create_sandboxed_resolver(
|
||||
base_root: str,
|
||||
allowed_paths: list[str] | None = None,
|
||||
) -> Callable[[str], str]:
|
||||
"""Create a path resolver that enforces sandbox boundaries.
|
||||
|
||||
Args:
|
||||
base_root: The primary allowed directory (e.g., PROJECT_ROOT or data_dir).
|
||||
allowed_paths: Additional allowed paths. If None, uses get_allowed_paths()
|
||||
which includes ~/.hive/ by default.
|
||||
|
||||
Returns:
|
||||
A path resolver function that raises ValueError for paths outside allowed scopes.
|
||||
|
||||
The resolver:
|
||||
- Resolves relative paths against base_root
|
||||
- Allows absolute paths under base_root or any allowed_path
|
||||
- Blocks access outside allowed scopes with a helpful error message
|
||||
"""
|
||||
hive_dir = os.path.expanduser("~/.hive")
|
||||
|
||||
def resolve(path: str) -> str:
|
||||
# Normalize slashes for cross-platform
|
||||
path = path.replace("/", os.sep)
|
||||
|
||||
# Expand ~ to home directory
|
||||
if path.startswith("~"):
|
||||
path = os.path.expanduser(path)
|
||||
|
||||
# Resolve to absolute path
|
||||
if os.path.isabs(path):
|
||||
resolved = os.path.abspath(path)
|
||||
else:
|
||||
resolved = os.path.abspath(os.path.join(base_root, path))
|
||||
|
||||
# Build allowed paths list
|
||||
extra_paths = allowed_paths if allowed_paths is not None else get_allowed_paths()
|
||||
all_allowed = [base_root] + extra_paths
|
||||
|
||||
# Check against all allowed paths
|
||||
for allowed_path in all_allowed:
|
||||
try:
|
||||
if os.path.commonpath([resolved, allowed_path]) == allowed_path:
|
||||
return resolved
|
||||
except ValueError:
|
||||
continue
|
||||
|
||||
# Block and remind
|
||||
allowed_str = ", ".join(f"'{p}'" for p in all_allowed)
|
||||
raise ValueError(
|
||||
f"Access denied: '{path}' is not accessible. "
|
||||
f"Allowed paths: {allowed_str}"
|
||||
)
|
||||
|
||||
return resolve
|
||||
|
||||
|
||||
# ── Private helpers ───────────────────────────────────────────────────────
|
||||
|
||||
|
||||
|
||||
@@ -1,377 +1,295 @@
|
||||
"""
|
||||
Data Tools - Load, save, and list data files for agent pipelines.
|
||||
Data Tools - Read and write files with data_dir sandboxing.
|
||||
|
||||
These tools let agents store large intermediate results in files and
|
||||
retrieve them with pagination, keeping the LLM conversation context small.
|
||||
Used in conjunction with the spillover system: when a tool result is too
|
||||
large, the framework writes it to a file and the agent can load it back
|
||||
with load_data().
|
||||
These tools let agents read and write files within their session's data directory
|
||||
and access files in ~/.hive/ for cross-agent file sharing.
|
||||
|
||||
Uses context injection for data_dir - the parameter is auto-injected by the
|
||||
framework and doesn't need to be provided by the LLM.
|
||||
"""
|
||||
|
||||
from __future__ import annotations
|
||||
|
||||
import os
|
||||
from pathlib import Path
|
||||
|
||||
from mcp.server.fastmcp import FastMCP
|
||||
|
||||
from aden_tools.credentials.browser import open_browser
|
||||
# ~/.hive/ is always allowed for cross-agent file access
|
||||
HIVE_DIR = os.path.expanduser("~/.hive")
|
||||
|
||||
|
||||
def _resolve_path(path: str, data_dir: str | None) -> str:
|
||||
"""Resolve and validate a path against the allowed directories.
|
||||
|
||||
Args:
|
||||
path: The path to resolve (can be relative or absolute)
|
||||
data_dir: The session's data directory from context
|
||||
|
||||
Returns:
|
||||
The resolved absolute path
|
||||
|
||||
Raises:
|
||||
ValueError: If path is outside allowed directories
|
||||
"""
|
||||
if not data_dir:
|
||||
raise ValueError("data_dir is not configured")
|
||||
|
||||
# Normalize path
|
||||
path = path.replace("/", os.sep)
|
||||
|
||||
# Expand ~ to home directory
|
||||
if path.startswith("~"):
|
||||
path = os.path.expanduser(path)
|
||||
|
||||
# Resolve to absolute path
|
||||
if os.path.isabs(path):
|
||||
resolved = os.path.abspath(path)
|
||||
else:
|
||||
resolved = os.path.abspath(os.path.join(data_dir, path))
|
||||
|
||||
# Check against allowed paths
|
||||
allowed_paths = [data_dir, HIVE_DIR]
|
||||
for allowed in allowed_paths:
|
||||
try:
|
||||
if os.path.commonpath([resolved, allowed]) == allowed:
|
||||
return resolved
|
||||
except ValueError:
|
||||
continue
|
||||
|
||||
# Block and remind
|
||||
allowed_str = ", ".join(f"'{p}'" for p in allowed_paths)
|
||||
raise ValueError(
|
||||
f"Access denied: '{path}' is not accessible. "
|
||||
f"Allowed paths: {allowed_str}"
|
||||
)
|
||||
|
||||
|
||||
def register_tools(mcp: FastMCP) -> None:
|
||||
"""Register data management tools with the MCP server."""
|
||||
"""Register file management tools with the MCP server."""
|
||||
|
||||
@mcp.tool()
|
||||
def save_data(filename: str, data: str, data_dir: str) -> dict:
|
||||
"""
|
||||
Purpose
|
||||
Save data to a file for later retrieval by this or downstream nodes.
|
||||
def read_file(
|
||||
path: str,
|
||||
offset: int = 1,
|
||||
limit: int = 0,
|
||||
data_dir: str = "",
|
||||
) -> str:
|
||||
"""Read file contents with line numbers.
|
||||
|
||||
When to use
|
||||
Store large results (search results, profiles, analysis) instead
|
||||
of passing them inline through set_output.
|
||||
Returns a brief summary with the filename to reference later.
|
||||
|
||||
Rules & Constraints
|
||||
filename must be a simple name like 'results.json' — no paths or '..'
|
||||
data_dir must be the absolute path to the data directory
|
||||
Files are read from the session's data directory or ~/.hive/.
|
||||
Large files are automatically truncated at 2000 lines or 50KB.
|
||||
Use offset and limit to paginate through large files.
|
||||
|
||||
Args:
|
||||
filename: Simple filename like 'github_users.json'. No paths or '..'.
|
||||
data: The string data to write (typically JSON).
|
||||
data_dir: Absolute path to the data directory.
|
||||
|
||||
Returns:
|
||||
Dict with success status and file metadata, or error dict
|
||||
path: File path to read. Can be relative to data_dir or absolute.
|
||||
offset: Starting line number, 1-indexed (default: 1).
|
||||
limit: Max lines to return, 0 = up to 2000 (default: 0).
|
||||
data_dir: Auto-injected - the session's data directory.
|
||||
"""
|
||||
if not filename or ".." in filename or "/" in filename or "\\" in filename:
|
||||
return {"error": "Invalid filename. Use simple names like 'users.json'"}
|
||||
if not data_dir:
|
||||
return {"error": "data_dir is required"}
|
||||
try:
|
||||
resolved = _resolve_path(path, data_dir)
|
||||
except ValueError as e:
|
||||
return f"Error: {e}"
|
||||
|
||||
if os.path.isdir(resolved):
|
||||
entries = []
|
||||
for entry in sorted(os.listdir(resolved)):
|
||||
full = os.path.join(resolved, entry)
|
||||
suffix = "/" if os.path.isdir(full) else ""
|
||||
entries.append(f" {entry}{suffix}")
|
||||
total = len(entries)
|
||||
return f"Directory: {path} ({total} entries)\n" + "\n".join(entries[:200])
|
||||
|
||||
if not os.path.isfile(resolved):
|
||||
return f"Error: File not found: {path}"
|
||||
|
||||
# Check for binary files
|
||||
try:
|
||||
with open(resolved, "rb") as f:
|
||||
chunk = f.read(4096)
|
||||
if b"\x00" in chunk:
|
||||
size = os.path.getsize(resolved)
|
||||
return f"Binary file: {path} ({size:,} bytes). Cannot display binary content."
|
||||
except OSError:
|
||||
pass
|
||||
|
||||
try:
|
||||
dir_path = Path(data_dir)
|
||||
dir_path.mkdir(parents=True, exist_ok=True)
|
||||
path = dir_path / filename
|
||||
path.write_text(data, encoding="utf-8")
|
||||
lines = data.count("\n") + 1
|
||||
return {
|
||||
"success": True,
|
||||
"filename": filename,
|
||||
"size_bytes": len(data.encode("utf-8")),
|
||||
"lines": lines,
|
||||
"preview": data[:200] + ("..." if len(data) > 200 else ""),
|
||||
}
|
||||
except Exception as e:
|
||||
return {"error": f"Failed to save data: {str(e)}"}
|
||||
with open(resolved, encoding="utf-8", errors="replace") as f:
|
||||
content = f.read()
|
||||
|
||||
@mcp.tool()
|
||||
def load_data(
|
||||
filename: str,
|
||||
data_dir: str,
|
||||
offset_bytes: int = 0,
|
||||
limit_bytes: int = 10000,
|
||||
) -> dict:
|
||||
"""
|
||||
Purpose
|
||||
Load data from a previously saved file with byte-based pagination.
|
||||
Efficient for files of any size (1 byte to 1 TB).
|
||||
Automatically detects safe UTF-8 boundaries to prevent character splitting.
|
||||
all_lines = content.splitlines()
|
||||
total_lines = len(all_lines)
|
||||
start_idx = max(0, offset - 1)
|
||||
effective_limit = limit if limit > 0 else 2000
|
||||
end_idx = min(start_idx + effective_limit, total_lines)
|
||||
|
||||
When to use
|
||||
Retrieve large tool results that were spilled to disk.
|
||||
Read data saved by save_data or by the spillover system.
|
||||
Page through large files without loading everything into context.
|
||||
max_bytes = 50 * 1024
|
||||
output_lines = []
|
||||
byte_count = 0
|
||||
|
||||
Rules & Constraints
|
||||
filename must match a file in data_dir
|
||||
Uses byte offsets for O(1) seeking (works with huge files)
|
||||
Automatically trims to valid UTF-8 character boundaries
|
||||
Returns exactly limit_bytes or less (rounded to safe boundary)
|
||||
for i in range(start_idx, end_idx):
|
||||
line = all_lines[i]
|
||||
if len(line) > 2000:
|
||||
line = line[:2000] + "..."
|
||||
formatted = f"{i + 1:>6}\t{line}"
|
||||
line_bytes = len(formatted.encode("utf-8")) + 1
|
||||
if byte_count + line_bytes > max_bytes:
|
||||
break
|
||||
output_lines.append(formatted)
|
||||
byte_count += line_bytes
|
||||
|
||||
Args:
|
||||
filename: The filename to load (as shown in spillover messages or save_data results).
|
||||
data_dir: Absolute path to the data directory.
|
||||
offset_bytes: Byte offset to start reading from. Default 0.
|
||||
limit_bytes: Max number of bytes to return. Default 10000 (10KB).
|
||||
result = "\n".join(output_lines)
|
||||
lines_shown = len(output_lines)
|
||||
actual_end = start_idx + lines_shown
|
||||
|
||||
Returns:
|
||||
Dict with content, pagination info, and metadata
|
||||
|
||||
Examples:
|
||||
load_data('emails.jsonl', '/data') # first 10KB
|
||||
load_data('emails.jsonl', '/data', offset_bytes=10000) # next 10KB
|
||||
load_data('large.txt', '/data', limit_bytes=50000) # first 50KB
|
||||
"""
|
||||
if not filename or ".." in filename or "/" in filename or "\\" in filename:
|
||||
return {"error": "Invalid filename"}
|
||||
if not data_dir:
|
||||
return {"error": "data_dir is required"}
|
||||
|
||||
try:
|
||||
offset_bytes = int(offset_bytes)
|
||||
limit_bytes = int(limit_bytes)
|
||||
path = Path(data_dir) / filename
|
||||
if not path.exists():
|
||||
return {"error": f"File not found: {filename}"}
|
||||
|
||||
file_size = path.stat().st_size
|
||||
|
||||
# Handle edge case: offset beyond file size
|
||||
if offset_bytes >= file_size:
|
||||
return {
|
||||
"success": True,
|
||||
"filename": filename,
|
||||
"content": "",
|
||||
"offset_bytes": offset_bytes,
|
||||
"bytes_read": 0,
|
||||
"next_offset_bytes": file_size,
|
||||
"file_size_bytes": file_size,
|
||||
"has_more": False,
|
||||
}
|
||||
|
||||
with open(path, "rb") as f:
|
||||
# O(1) seek to byte offset
|
||||
f.seek(offset_bytes)
|
||||
|
||||
# Read exactly limit_bytes
|
||||
raw_bytes = f.read(limit_bytes)
|
||||
|
||||
# Trim to valid UTF-8 boundary
|
||||
# Scan backwards max 4 bytes to find valid UTF-8 start
|
||||
chunk = raw_bytes
|
||||
text = None
|
||||
for i in range(min(4, len(raw_bytes)) + 1):
|
||||
try:
|
||||
slice_end = len(raw_bytes) - i if i > 0 else len(raw_bytes)
|
||||
text = raw_bytes[:slice_end].decode("utf-8")
|
||||
chunk = raw_bytes[:slice_end]
|
||||
break
|
||||
except UnicodeDecodeError:
|
||||
continue
|
||||
|
||||
# If we couldn't decode at all, return error
|
||||
if text is None:
|
||||
return {"error": "Could not decode file as UTF-8"}
|
||||
|
||||
# UTF-8 boundary is already handled above
|
||||
next_offset = offset_bytes + len(chunk)
|
||||
|
||||
return {
|
||||
"success": True,
|
||||
"filename": filename,
|
||||
"content": text,
|
||||
"offset_bytes": offset_bytes,
|
||||
"bytes_read": len(chunk),
|
||||
"next_offset_bytes": next_offset,
|
||||
"file_size_bytes": file_size,
|
||||
"has_more": next_offset < file_size,
|
||||
}
|
||||
except Exception as e:
|
||||
return {"error": f"Failed to load data: {str(e)}"}
|
||||
|
||||
@mcp.tool()
|
||||
def serve_file_to_user(
|
||||
filename: str, data_dir: str, label: str = "", open_in_browser: bool = False
|
||||
) -> dict:
|
||||
"""
|
||||
Purpose
|
||||
Resolve a sandboxed file path to a fully qualified file URI
|
||||
that the user can click to open in their system viewer.
|
||||
|
||||
When to use
|
||||
After saving a file (HTML report, CSV export, etc.) with save_data,
|
||||
call this to give the user a clickable link to open it.
|
||||
The TUI will render the file:// URI as a clickable link.
|
||||
Set open_in_browser=True to also auto-open the file in the
|
||||
user's default browser.
|
||||
|
||||
Rules & Constraints
|
||||
filename must be a simple name — no paths or '..'
|
||||
The file must already exist in data_dir
|
||||
Returns a file:// URI the agent should include in its response
|
||||
|
||||
Args:
|
||||
filename: The filename to serve (must exist in data_dir).
|
||||
data_dir: Absolute path to the data directory.
|
||||
label: Optional display label (defaults to filename).
|
||||
open_in_browser: If True, auto-open the file in the default browser.
|
||||
|
||||
Returns:
|
||||
Dict with file_uri, file_path, label, and optionally browser_opened
|
||||
"""
|
||||
if not filename or ".." in filename or "/" in filename or "\\" in filename:
|
||||
return {"error": "Invalid filename. Use simple names like 'report.html'"}
|
||||
if not data_dir:
|
||||
return {"error": "data_dir is required"}
|
||||
|
||||
try:
|
||||
path = Path(data_dir) / filename
|
||||
if not path.exists():
|
||||
return {"error": f"File not found: {filename}"}
|
||||
|
||||
full_path = str(path.resolve())
|
||||
file_uri = f"file://{full_path}"
|
||||
result = {
|
||||
"success": True,
|
||||
"file_uri": file_uri,
|
||||
"file_path": full_path,
|
||||
"label": label or filename,
|
||||
}
|
||||
|
||||
if open_in_browser:
|
||||
opened, msg = open_browser(file_uri)
|
||||
result["browser_opened"] = opened
|
||||
result["browser_message"] = msg
|
||||
if actual_end < total_lines:
|
||||
result += f"\n\n(Showing lines {start_idx + 1}-{actual_end} of {total_lines}. Use offset={actual_end + 1} to continue reading.)"
|
||||
|
||||
return result
|
||||
except Exception as e:
|
||||
return {"error": f"Failed to serve file: {str(e)}"}
|
||||
return f"Error reading file: {e}"
|
||||
|
||||
@mcp.tool()
|
||||
def list_data_files(data_dir: str) -> dict:
|
||||
"""
|
||||
Purpose
|
||||
List all data files in the data directory.
|
||||
def write_file(
|
||||
path: str,
|
||||
content: str,
|
||||
data_dir: str = "",
|
||||
) -> str:
|
||||
"""Create or overwrite a file with the given content.
|
||||
|
||||
When to use
|
||||
Discover what intermediate results or spillover files are available.
|
||||
Check what data was saved by previous nodes in the pipeline.
|
||||
Automatically creates parent directories. Files are written to
|
||||
the session's data directory or ~/.hive/.
|
||||
|
||||
Args:
|
||||
data_dir: Absolute path to the data directory.
|
||||
|
||||
Returns:
|
||||
Dict with list of files and their sizes
|
||||
path: File path to write. Can be relative to data_dir or absolute.
|
||||
content: Complete file content to write.
|
||||
data_dir: Auto-injected - the session's data directory.
|
||||
"""
|
||||
if not data_dir:
|
||||
return {"error": "data_dir is required"}
|
||||
try:
|
||||
resolved = _resolve_path(path, data_dir)
|
||||
except ValueError as e:
|
||||
return f"Error: {e}"
|
||||
|
||||
try:
|
||||
dir_path = Path(data_dir)
|
||||
if not dir_path.exists():
|
||||
return {"files": []}
|
||||
resolved_path = Path(resolved)
|
||||
resolved_path.parent.mkdir(parents=True, exist_ok=True)
|
||||
|
||||
files = []
|
||||
for f in sorted(dir_path.iterdir()):
|
||||
if f.is_file():
|
||||
files.append(
|
||||
{
|
||||
"filename": f.name,
|
||||
"size_bytes": f.stat().st_size,
|
||||
}
|
||||
)
|
||||
return {"files": files}
|
||||
existed = resolved_path.is_file()
|
||||
content_str = content if content is not None else ""
|
||||
with open(resolved_path, "w", encoding="utf-8") as f:
|
||||
f.write(content_str)
|
||||
f.flush()
|
||||
os.fsync(f.fileno())
|
||||
|
||||
line_count = content_str.count("\n") + (
|
||||
1 if content_str and not content_str.endswith("\n") else 0
|
||||
)
|
||||
action = "Updated" if existed else "Created"
|
||||
return f"{action} {path} ({len(content_str):,} bytes, {line_count} lines)"
|
||||
except Exception as e:
|
||||
return {"error": f"Failed to list data files: {str(e)}"}
|
||||
return f"Error writing file: {e}"
|
||||
|
||||
@mcp.tool()
|
||||
def append_data(filename: str, data: str, data_dir: str) -> dict:
|
||||
"""
|
||||
Purpose
|
||||
Append data to the end of an existing file, or create it if it
|
||||
doesn't exist yet.
|
||||
def list_files(
|
||||
path: str = ".",
|
||||
recursive: bool = False,
|
||||
data_dir: str = "",
|
||||
) -> str:
|
||||
"""List directory contents with type indicators.
|
||||
|
||||
When to use
|
||||
Build large files incrementally instead of writing everything in
|
||||
one save_data call. For example, write an HTML skeleton first,
|
||||
then append each section separately to stay within token limits.
|
||||
|
||||
Rules & Constraints
|
||||
filename must be a simple name like 'report.html' — no paths or '..'
|
||||
Directories have a / suffix. Hidden files and common build directories
|
||||
are skipped.
|
||||
|
||||
Args:
|
||||
filename: Simple filename to append to. No paths or '..'.
|
||||
data: The string data to append.
|
||||
data_dir: Absolute path to the data directory.
|
||||
|
||||
Returns:
|
||||
Dict with success status, new total size, and bytes appended
|
||||
path: Directory path (default: data_dir).
|
||||
recursive: List recursively (default: false).
|
||||
data_dir: Auto-injected - the session's data directory.
|
||||
"""
|
||||
if not filename or ".." in filename or "/" in filename or "\\" in filename:
|
||||
return {"error": "Invalid filename. Use simple names like 'report.html'"}
|
||||
if not data_dir:
|
||||
return {"error": "data_dir is required"}
|
||||
try:
|
||||
resolved = _resolve_path(path, data_dir)
|
||||
except ValueError as e:
|
||||
return f"Error: {e}"
|
||||
|
||||
if not os.path.isdir(resolved):
|
||||
return f"Error: Directory not found: {path}"
|
||||
|
||||
try:
|
||||
dir_path = Path(data_dir)
|
||||
dir_path.mkdir(parents=True, exist_ok=True)
|
||||
path = dir_path / filename
|
||||
with open(path, "a", encoding="utf-8") as f:
|
||||
f.write(data)
|
||||
appended_bytes = len(data.encode("utf-8"))
|
||||
total_bytes = path.stat().st_size
|
||||
return {
|
||||
"success": True,
|
||||
"filename": filename,
|
||||
"size_bytes": total_bytes,
|
||||
"appended_bytes": appended_bytes,
|
||||
}
|
||||
skip = {".git", "__pycache__", "node_modules", ".venv", ".tox"}
|
||||
entries: list[str] = []
|
||||
|
||||
if recursive:
|
||||
for root, dirs, files in os.walk(resolved):
|
||||
dirs[:] = sorted(d for d in dirs if d not in skip and not d.startswith("."))
|
||||
rel_root = os.path.relpath(root, resolved)
|
||||
if rel_root == ".":
|
||||
rel_root = ""
|
||||
for f in sorted(files):
|
||||
if f.startswith("."):
|
||||
continue
|
||||
entries.append(os.path.join(rel_root, f) if rel_root else f)
|
||||
if len(entries) >= 500:
|
||||
entries.append("... (truncated at 500 entries)")
|
||||
return "\n".join(entries)
|
||||
else:
|
||||
for entry in sorted(os.listdir(resolved)):
|
||||
if entry.startswith(".") or entry in skip:
|
||||
continue
|
||||
full = os.path.join(resolved, entry)
|
||||
suffix = "/" if os.path.isdir(full) else ""
|
||||
entries.append(f"{entry}{suffix}")
|
||||
|
||||
return "\n".join(entries) if entries else "(empty directory)"
|
||||
except Exception as e:
|
||||
return {"error": f"Failed to append data: {str(e)}"}
|
||||
return f"Error listing directory: {e}"
|
||||
|
||||
@mcp.tool()
|
||||
def edit_data(filename: str, old_text: str, new_text: str, data_dir: str) -> dict:
|
||||
"""
|
||||
Purpose
|
||||
Find and replace a specific text segment in an existing file.
|
||||
Works like a surgical diff — only the matched portion changes.
|
||||
def search_files(
|
||||
pattern: str,
|
||||
path: str = ".",
|
||||
data_dir: str = "",
|
||||
) -> str:
|
||||
"""Search file contents using regex.
|
||||
|
||||
When to use
|
||||
Update a section of a previously saved file without rewriting
|
||||
the entire content. For example, replace a placeholder in an
|
||||
HTML report or fix a specific paragraph.
|
||||
|
||||
Rules & Constraints
|
||||
old_text must appear exactly once in the file. If it appears
|
||||
zero times or more than once, the edit is rejected with an
|
||||
error message.
|
||||
Results sorted by file with line numbers. Searches within
|
||||
the session's data directory or ~/.hive/.
|
||||
|
||||
Args:
|
||||
filename: The file to edit. Must exist in data_dir.
|
||||
old_text: The exact text to find (must match exactly once).
|
||||
new_text: The replacement text.
|
||||
data_dir: Absolute path to the data directory.
|
||||
|
||||
Returns:
|
||||
Dict with success status and updated file size
|
||||
pattern: Regex pattern to search for.
|
||||
path: Directory path to search (default: data_dir).
|
||||
data_dir: Auto-injected - the session's data directory.
|
||||
"""
|
||||
if not filename or ".." in filename or "/" in filename or "\\" in filename:
|
||||
return {"error": "Invalid filename. Use simple names like 'report.html'"}
|
||||
if not data_dir:
|
||||
return {"error": "data_dir is required"}
|
||||
import re
|
||||
|
||||
try:
|
||||
path = Path(data_dir) / filename
|
||||
if not path.exists():
|
||||
return {"error": f"File not found: {filename}"}
|
||||
resolved = _resolve_path(path, data_dir)
|
||||
except ValueError as e:
|
||||
return f"Error: {e}"
|
||||
|
||||
content = path.read_text(encoding="utf-8")
|
||||
count = content.count(old_text)
|
||||
if not os.path.isdir(resolved):
|
||||
return f"Error: Directory not found: {path}"
|
||||
|
||||
if count == 0:
|
||||
return {
|
||||
"error": (
|
||||
"old_text not found in the file. "
|
||||
"Make sure you're matching the exact text, "
|
||||
"including whitespace and newlines."
|
||||
)
|
||||
}
|
||||
if count > 1:
|
||||
return {
|
||||
"error": (
|
||||
f"old_text found {count} times — it must be unique. "
|
||||
"Include more surrounding context to match exactly once."
|
||||
)
|
||||
}
|
||||
try:
|
||||
compiled = re.compile(pattern)
|
||||
matches: list[str] = []
|
||||
skip_dirs = {".git", "__pycache__", "node_modules", ".venv"}
|
||||
|
||||
updated = content.replace(old_text, new_text, 1)
|
||||
path.write_text(updated, encoding="utf-8")
|
||||
for root, dirs, files in os.walk(resolved):
|
||||
dirs[:] = [d for d in dirs if d not in skip_dirs]
|
||||
for fname in files:
|
||||
fpath = os.path.join(root, fname)
|
||||
display_path = os.path.relpath(fpath, resolved)
|
||||
try:
|
||||
with open(fpath, encoding="utf-8", errors="ignore") as f:
|
||||
for i, line in enumerate(f, 1):
|
||||
stripped = line.rstrip()
|
||||
if compiled.search(stripped):
|
||||
matches.append(f"{display_path}:{i}:{stripped[:2000]}")
|
||||
if len(matches) >= 100:
|
||||
return "\n".join(matches) + "\n... (truncated)"
|
||||
except (OSError, UnicodeDecodeError):
|
||||
continue
|
||||
|
||||
return {
|
||||
"success": True,
|
||||
"filename": filename,
|
||||
"size_bytes": len(updated.encode("utf-8")),
|
||||
"replacements": 1,
|
||||
}
|
||||
except Exception as e:
|
||||
return {"error": f"Failed to edit data: {str(e)}"}
|
||||
return "\n".join(matches) if matches else "No matches found."
|
||||
except re.error as e:
|
||||
return f"Error: Invalid regex: {e}"
|
||||
|
||||
+614
-116
@@ -78,8 +78,20 @@ class BeelineBridge:
|
||||
return
|
||||
|
||||
try:
|
||||
self._server = await websockets.serve(self._handle_connection, "127.0.0.1", port)
|
||||
logger.info("Beeline bridge listening on ws://127.0.0.1:%d/bridge", port)
|
||||
# Suppress noisy websockets logging for invalid upgrade attempts
|
||||
# by providing a null logger
|
||||
import logging
|
||||
null_logger = logging.getLogger("websockets.null")
|
||||
null_logger.setLevel(logging.CRITICAL)
|
||||
null_logger.addHandler(logging.NullHandler())
|
||||
|
||||
self._server = await websockets.serve(
|
||||
self._handle_connection,
|
||||
"127.0.0.1",
|
||||
port,
|
||||
logger=null_logger,
|
||||
)
|
||||
logger.info("Beeline bridge listening on ws://127.0.0.1:%d", port)
|
||||
except OSError as e:
|
||||
logger.warning("Beeline bridge could not start on port %d: %s", port, e)
|
||||
|
||||
@@ -170,6 +182,21 @@ class BeelineBridge:
|
||||
log_cdp_command(tab_id, method, params, error=str(e), duration_ms=duration_ms)
|
||||
raise
|
||||
|
||||
async def _try_enable_domain(self, tab_id: int, domain: str) -> None:
|
||||
"""Try to enable a CDP domain, ignoring errors if not available.
|
||||
|
||||
Some domains (like Input) may not be available on certain page types
|
||||
(e.g., chrome:// URLs, extension pages, or restricted sites).
|
||||
"""
|
||||
try:
|
||||
await self._cdp(tab_id, f"{domain}.enable")
|
||||
except RuntimeError as e:
|
||||
# Log but don't fail - domain may not be available on all pages
|
||||
if "wasn't found" in str(e) or "not found" in str(e).lower():
|
||||
logger.debug("CDP domain %s.enable not available for tab %s", domain, tab_id)
|
||||
else:
|
||||
raise
|
||||
|
||||
# ── Context (Tab Group) Management ─────────────────────────────────────────
|
||||
|
||||
async def create_context(self, agent_id: str) -> dict:
|
||||
@@ -374,12 +401,15 @@ class BeelineBridge:
|
||||
) -> dict:
|
||||
"""Click an element by selector.
|
||||
|
||||
Uses DOM.getDocument + DOM.querySelector to find the element,
|
||||
then DOM.getBoxModel to get coordinates, then Input.dispatchMouseEvent.
|
||||
Uses multiple fallback methods for robustness:
|
||||
1. CDP mouse events with JavaScript bounds
|
||||
2. JavaScript click() as fallback
|
||||
|
||||
Inspired by browser-use's robust click implementation.
|
||||
"""
|
||||
await self.cdp_attach(tab_id)
|
||||
await self._cdp(tab_id, "DOM.enable")
|
||||
await self._cdp(tab_id, "Input.enable")
|
||||
await self._try_enable_domain(tab_id, "DOM")
|
||||
await self._try_enable_domain(tab_id, "Input")
|
||||
|
||||
# Get document and find element
|
||||
doc = await self._cdp(tab_id, "DOM.getDocument")
|
||||
@@ -400,56 +430,172 @@ class BeelineBridge:
|
||||
if not node_id:
|
||||
return {"ok": False, "error": f"Element not found: {selector}"}
|
||||
|
||||
# Get box model for coordinates
|
||||
box = await self._cdp(tab_id, "DOM.getBoxModel", {"nodeId": node_id})
|
||||
content = box.get("content", [])
|
||||
if len(content) < 4:
|
||||
# Scroll into view FIRST to ensure element is rendered
|
||||
try:
|
||||
await self._cdp(
|
||||
tab_id,
|
||||
"DOM.scrollIntoViewIfNeeded",
|
||||
{"nodeId": node_id},
|
||||
)
|
||||
await asyncio.sleep(0.05) # Wait for scroll to complete
|
||||
except Exception:
|
||||
pass # Best effort - continue even if scroll fails
|
||||
|
||||
# Get viewport dimensions for bounds checking
|
||||
viewport_script = """
|
||||
(function() {
|
||||
return {
|
||||
width: window.innerWidth,
|
||||
height: window.innerHeight
|
||||
};
|
||||
})();
|
||||
"""
|
||||
viewport_result = await self.evaluate(tab_id, viewport_script)
|
||||
viewport = viewport_result.get("result", {}).get("value", {})
|
||||
viewport_width = viewport.get("width", 1920)
|
||||
viewport_height = viewport.get("height", 1080)
|
||||
|
||||
# Method 1: Use JavaScript to get element bounds and click
|
||||
# This is more reliable than CDP for complex layouts
|
||||
click_script = f"""
|
||||
(function() {{
|
||||
const el = document.querySelector({json.dumps(selector)});
|
||||
if (!el) return {{ error: 'Element not found' }};
|
||||
|
||||
// Check if element is visible
|
||||
const rect = el.getBoundingClientRect();
|
||||
if (rect.width === 0 || rect.height === 0) {{
|
||||
return {{ error: 'Element has zero dimensions' }};
|
||||
}}
|
||||
|
||||
// Check if element is within viewport
|
||||
if (rect.bottom < 0 || rect.top > {viewport_height} ||
|
||||
rect.right < 0 || rect.left > {viewport_width}) {{
|
||||
return {{ error: 'Element not in viewport' }};
|
||||
}}
|
||||
|
||||
// Get center for metadata
|
||||
const x = rect.x + rect.width / 2;
|
||||
const y = rect.y + rect.height / 2;
|
||||
|
||||
// Perform the click
|
||||
el.click();
|
||||
|
||||
return {{ x: x, y: y, width: rect.width, height: rect.height }};
|
||||
}})();
|
||||
"""
|
||||
|
||||
try:
|
||||
result = await self.evaluate(tab_id, click_script)
|
||||
value = result.get("result", {}).get("value")
|
||||
|
||||
if isinstance(value, dict) and "error" not in value:
|
||||
# JavaScript click succeeded
|
||||
return {
|
||||
"ok": True,
|
||||
"action": "click",
|
||||
"selector": selector,
|
||||
"x": value.get("x", 0),
|
||||
"y": value.get("y", 0),
|
||||
"method": "javascript"
|
||||
}
|
||||
|
||||
# If JavaScript click failed, try CDP approach
|
||||
if isinstance(value, dict) and value.get("error"):
|
||||
logger.debug("JS click failed: %s, trying CDP", value["error"])
|
||||
except Exception as e:
|
||||
logger.debug("JS click exception: %s, trying CDP", e)
|
||||
|
||||
# Method 2: CDP mouse events (fallback)
|
||||
# Get element bounds via JavaScript (more reliable than CDP getBoxModel)
|
||||
bounds_script = f"""
|
||||
(function() {{
|
||||
const el = document.querySelector({json.dumps(selector)});
|
||||
if (!el) return null;
|
||||
const rect = el.getBoundingClientRect();
|
||||
return {{
|
||||
x: rect.x + rect.width / 2,
|
||||
y: rect.y + rect.height / 2,
|
||||
width: rect.width,
|
||||
height: rect.height
|
||||
}};
|
||||
}})();
|
||||
"""
|
||||
bounds_result = await self.evaluate(tab_id, bounds_script)
|
||||
bounds_value = bounds_result.get("result", {}).get("value")
|
||||
|
||||
if not bounds_value:
|
||||
return {"ok": False, "error": f"Could not get element bounds: {selector}"}
|
||||
|
||||
# Calculate center of element (content quad is [x1,y1, x2,y2, x3,y3, x4,y4])
|
||||
x = (content[0] + content[2] + content[4] + content[6]) / 4
|
||||
y = (content[1] + content[3] + content[5] + content[7]) / 4
|
||||
x = bounds_value.get("x", 0)
|
||||
y = bounds_value.get("y", 0)
|
||||
|
||||
# Scroll into view first
|
||||
await self._cdp(
|
||||
tab_id,
|
||||
"DOM.scrollIntoViewIfNeeded",
|
||||
{"nodeId": node_id},
|
||||
)
|
||||
# Clamp coordinates to viewport bounds
|
||||
x = max(0, min(viewport_width - 1, x))
|
||||
y = max(0, min(viewport_height - 1, y))
|
||||
|
||||
# Dispatch mouse events
|
||||
# Dispatch mouse events with proper timing
|
||||
button_map = {"left": "left", "right": "right", "middle": "middle"}
|
||||
cdp_button = button_map.get(button, "left")
|
||||
|
||||
await self._cdp(
|
||||
tab_id,
|
||||
"Input.dispatchMouseEvent",
|
||||
{
|
||||
"type": "mousePressed",
|
||||
"x": x,
|
||||
"y": y,
|
||||
"button": cdp_button,
|
||||
"clickCount": click_count,
|
||||
},
|
||||
)
|
||||
await self._cdp(
|
||||
tab_id,
|
||||
"Input.dispatchMouseEvent",
|
||||
{
|
||||
"type": "mouseReleased",
|
||||
"x": x,
|
||||
"y": y,
|
||||
"button": cdp_button,
|
||||
"clickCount": click_count,
|
||||
},
|
||||
)
|
||||
try:
|
||||
# Move mouse to element first
|
||||
await self._cdp(
|
||||
tab_id,
|
||||
"Input.dispatchMouseEvent",
|
||||
{"type": "mouseMoved", "x": x, "y": y},
|
||||
)
|
||||
await asyncio.sleep(0.05)
|
||||
|
||||
return {"ok": True, "action": "click", "selector": selector, "x": x, "y": y}
|
||||
# Mouse down
|
||||
try:
|
||||
await asyncio.wait_for(
|
||||
self._cdp(
|
||||
tab_id,
|
||||
"Input.dispatchMouseEvent",
|
||||
{
|
||||
"type": "mousePressed",
|
||||
"x": x,
|
||||
"y": y,
|
||||
"button": cdp_button,
|
||||
"clickCount": click_count,
|
||||
},
|
||||
),
|
||||
timeout=1.0,
|
||||
)
|
||||
except asyncio.TimeoutError:
|
||||
pass # Continue even if timeout
|
||||
|
||||
await asyncio.sleep(0.08)
|
||||
|
||||
# Mouse up
|
||||
try:
|
||||
await asyncio.wait_for(
|
||||
self._cdp(
|
||||
tab_id,
|
||||
"Input.dispatchMouseEvent",
|
||||
{
|
||||
"type": "mouseReleased",
|
||||
"x": x,
|
||||
"y": y,
|
||||
"button": cdp_button,
|
||||
"clickCount": click_count,
|
||||
},
|
||||
),
|
||||
timeout=3.0,
|
||||
)
|
||||
except asyncio.TimeoutError:
|
||||
pass # Continue even if timeout
|
||||
|
||||
return {"ok": True, "action": "click", "selector": selector, "x": x, "y": y, "method": "cdp"}
|
||||
|
||||
except Exception as e:
|
||||
return {"ok": False, "error": f"Click failed: {e}"}
|
||||
|
||||
async def click_coordinate(self, tab_id: int, x: float, y: float, button: str = "left") -> dict:
|
||||
"""Click at specific coordinates."""
|
||||
await self.cdp_attach(tab_id)
|
||||
await self._cdp(tab_id, "Input.enable")
|
||||
await self._try_enable_domain(tab_id, "Input")
|
||||
|
||||
button_map = {"left": "left", "right": "right", "middle": "middle"}
|
||||
cdp_button = button_map.get(button, "left")
|
||||
@@ -476,44 +622,59 @@ class BeelineBridge:
|
||||
delay_ms: int = 0,
|
||||
timeout_ms: int = 30000,
|
||||
) -> dict:
|
||||
"""Type text into an element."""
|
||||
"""Type text into an element.
|
||||
|
||||
Uses JavaScript focus for reliability, then CDP key events.
|
||||
"""
|
||||
await self.cdp_attach(tab_id)
|
||||
await self._cdp(tab_id, "DOM.enable")
|
||||
await self._cdp(tab_id, "Input.enable")
|
||||
await self._try_enable_domain(tab_id, "DOM")
|
||||
await self._try_enable_domain(tab_id, "Input")
|
||||
await self._try_enable_domain(tab_id, "Runtime")
|
||||
|
||||
# Get document and find element
|
||||
doc = await self._cdp(tab_id, "DOM.getDocument")
|
||||
root_id = doc.get("root", {}).get("nodeId")
|
||||
# First, scroll into view and focus via JavaScript (more reliable than CDP)
|
||||
focus_script = f"""
|
||||
(function() {{
|
||||
const el = document.querySelector({json.dumps(selector)});
|
||||
if (!el) return false;
|
||||
|
||||
deadline = asyncio.get_event_loop().time() + timeout_ms / 1000
|
||||
node_id = None
|
||||
while asyncio.get_event_loop().time() < deadline:
|
||||
result = await self._cdp(
|
||||
tab_id, "DOM.querySelector", {"nodeId": root_id, "selector": selector}
|
||||
)
|
||||
node_id = result.get("nodeId")
|
||||
if node_id:
|
||||
break
|
||||
await asyncio.sleep(0.1)
|
||||
// Scroll into view
|
||||
el.scrollIntoView({{ block: 'center' }});
|
||||
|
||||
if not node_id:
|
||||
return {"ok": False, "error": f"Element not found: {selector}"}
|
||||
// Focus the element
|
||||
el.focus();
|
||||
|
||||
# Focus the element
|
||||
await self._cdp(tab_id, "DOM.focus", {"nodeId": node_id})
|
||||
// Clear if requested
|
||||
if ({str(clear_first).lower()}) {{
|
||||
if (el.value !== undefined) {{
|
||||
el.value = '';
|
||||
}} else if (el.isContentEditable) {{
|
||||
el.textContent = '';
|
||||
}}
|
||||
}}
|
||||
|
||||
# Clear if requested
|
||||
if clear_first:
|
||||
await self._cdp(
|
||||
tab_id,
|
||||
"Runtime.evaluate",
|
||||
{
|
||||
"expression": f"document.querySelector({json.dumps(selector)}).value = ''",
|
||||
"returnByValue": True,
|
||||
},
|
||||
)
|
||||
return true;
|
||||
}})();
|
||||
"""
|
||||
|
||||
# Type each character
|
||||
focus_result = await self.evaluate(tab_id, focus_script)
|
||||
success = focus_result.get("result", {}).get("value", False)
|
||||
|
||||
if not success:
|
||||
# Element not found - wait and retry
|
||||
deadline = asyncio.get_event_loop().time() + timeout_ms / 1000
|
||||
while asyncio.get_event_loop().time() < deadline:
|
||||
result = await self.evaluate(tab_id, focus_script)
|
||||
if result.get("result", {}).get("value", False):
|
||||
success = True
|
||||
break
|
||||
await asyncio.sleep(0.1)
|
||||
|
||||
if not success:
|
||||
return {"ok": False, "error": f"Element not found: {selector}"}
|
||||
|
||||
await asyncio.sleep(0.05) # Wait for focus to take effect
|
||||
|
||||
# Type each character using CDP key events
|
||||
for char in text:
|
||||
# Dispatch key down
|
||||
await self._cdp(
|
||||
@@ -540,7 +701,7 @@ class BeelineBridge:
|
||||
selector: Optional selector to focus first
|
||||
"""
|
||||
await self.cdp_attach(tab_id)
|
||||
await self._cdp(tab_id, "Input.enable")
|
||||
await self._try_enable_domain(tab_id, "Input")
|
||||
|
||||
if selector:
|
||||
doc = await self._cdp(tab_id, "DOM.getDocument")
|
||||
@@ -585,43 +746,73 @@ class BeelineBridge:
|
||||
return {"ok": True, "action": "press", "key": key}
|
||||
|
||||
async def hover(self, tab_id: int, selector: str, timeout_ms: int = 30000) -> dict:
|
||||
"""Hover over an element."""
|
||||
"""Hover over an element.
|
||||
|
||||
Uses JavaScript for bounds (more reliable than CDP getBoxModel).
|
||||
"""
|
||||
await self.cdp_attach(tab_id)
|
||||
await self._cdp(tab_id, "DOM.enable")
|
||||
await self._cdp(tab_id, "Input.enable")
|
||||
await self._try_enable_domain(tab_id, "DOM")
|
||||
await self._try_enable_domain(tab_id, "Input")
|
||||
await self._try_enable_domain(tab_id, "Runtime")
|
||||
|
||||
doc = await self._cdp(tab_id, "DOM.getDocument")
|
||||
root_id = doc.get("root", {}).get("nodeId")
|
||||
# Use JavaScript to scroll into view and get bounds
|
||||
hover_script = f"""
|
||||
(function() {{
|
||||
const el = document.querySelector({json.dumps(selector)});
|
||||
if (!el) return null;
|
||||
|
||||
// Scroll into view
|
||||
el.scrollIntoView({{ block: 'center' }});
|
||||
|
||||
const rect = el.getBoundingClientRect();
|
||||
return {{
|
||||
x: rect.x + rect.width / 2,
|
||||
y: rect.y + rect.height / 2,
|
||||
width: rect.width,
|
||||
height: rect.height
|
||||
}};
|
||||
}})();
|
||||
"""
|
||||
|
||||
# Wait for element and get bounds
|
||||
deadline = asyncio.get_event_loop().time() + timeout_ms / 1000
|
||||
node_id = None
|
||||
bounds_value = None
|
||||
|
||||
while asyncio.get_event_loop().time() < deadline:
|
||||
result = await self._cdp(
|
||||
tab_id, "DOM.querySelector", {"nodeId": root_id, "selector": selector}
|
||||
)
|
||||
node_id = result.get("nodeId")
|
||||
if node_id:
|
||||
result = await self.evaluate(tab_id, hover_script)
|
||||
bounds_value = result.get("result", {}).get("value")
|
||||
if bounds_value:
|
||||
break
|
||||
await asyncio.sleep(0.1)
|
||||
|
||||
if not node_id:
|
||||
if not bounds_value:
|
||||
return {"ok": False, "error": f"Element not found: {selector}"}
|
||||
|
||||
box = await self._cdp(tab_id, "DOM.getBoxModel", {"nodeId": node_id})
|
||||
content = box.get("content", [])
|
||||
x = (content[0] + content[2] + content[4] + content[6]) / 4
|
||||
y = (content[1] + content[3] + content[5] + content[7]) / 4
|
||||
x = bounds_value.get("x", 0)
|
||||
y = bounds_value.get("y", 0)
|
||||
|
||||
if x == 0 and y == 0:
|
||||
return {"ok": False, "error": f"Element has zero dimensions: {selector}"}
|
||||
|
||||
await asyncio.sleep(0.05) # Wait for scroll
|
||||
|
||||
# Dispatch mouse moved event
|
||||
await self._cdp(
|
||||
tab_id,
|
||||
"Input.dispatchMouseEvent",
|
||||
{"type": "mouseMoved", "x": x, "y": y},
|
||||
)
|
||||
|
||||
return {"ok": True, "action": "hover", "selector": selector}
|
||||
return {"ok": True, "action": "hover", "selector": selector, "x": x, "y": y}
|
||||
|
||||
async def scroll(self, tab_id: int, direction: str = "down", amount: int = 500) -> dict:
|
||||
"""Scroll the page."""
|
||||
"""Scroll the page.
|
||||
|
||||
Uses multiple methods for robustness:
|
||||
1. Find and scroll the largest scrollable container (handles SPAs like LinkedIn)
|
||||
2. Fallback to window scroll
|
||||
3. Fallback to mouse wheel events via CDP
|
||||
"""
|
||||
await self.cdp_attach(tab_id)
|
||||
|
||||
delta_x = 0
|
||||
@@ -635,16 +826,159 @@ class BeelineBridge:
|
||||
elif direction == "left":
|
||||
delta_x = -amount
|
||||
|
||||
await self._cdp(
|
||||
tab_id,
|
||||
"Runtime.evaluate",
|
||||
{
|
||||
"expression": f"window.scrollBy({delta_x}, {delta_y})",
|
||||
"returnByValue": True,
|
||||
},
|
||||
)
|
||||
# Method 1: Find and scroll the largest scrollable container
|
||||
# This handles SPAs like LinkedIn where content is in a nested scrollable div
|
||||
smart_scroll_script = f"""
|
||||
(function() {{
|
||||
// Find the largest scrollable container
|
||||
function findScrollableContainer() {{
|
||||
const candidates = [];
|
||||
|
||||
return {"ok": True, "action": "scroll", "direction": direction, "amount": amount}
|
||||
// Check all elements with overflow scroll/auto
|
||||
const allElements = document.querySelectorAll('*');
|
||||
for (const el of allElements) {{
|
||||
const style = getComputedStyle(el);
|
||||
const overflow = style.overflow + style.overflowY;
|
||||
|
||||
if (overflow.includes('scroll') || overflow.includes('auto')) {{
|
||||
const rect = el.getBoundingClientRect();
|
||||
// Must be visible and reasonably large
|
||||
if (rect.width > 100 && rect.height > 100 &&
|
||||
el.scrollHeight > el.clientHeight + 100) {{
|
||||
candidates.push({{
|
||||
el: el,
|
||||
area: rect.width * rect.height,
|
||||
scrollable: el.scrollHeight - el.clientHeight
|
||||
}});
|
||||
}}
|
||||
}}
|
||||
}}
|
||||
|
||||
// Sort by area (largest first) and return best candidate
|
||||
candidates.sort((a, b) => b.area - a.area);
|
||||
return candidates.length > 0 ? candidates[0].el : null;
|
||||
}}
|
||||
|
||||
const container = findScrollableContainer();
|
||||
|
||||
if (container) {{
|
||||
container.scrollBy({{
|
||||
top: {delta_y},
|
||||
left: {delta_x},
|
||||
behavior: 'smooth'
|
||||
}});
|
||||
return {{
|
||||
method: 'container-smooth',
|
||||
success: true,
|
||||
containerTag: container.tagName,
|
||||
containerClass: container.className.substring(0, 50)
|
||||
}};
|
||||
}}
|
||||
|
||||
// Fallback to window scroll
|
||||
if ('scrollBehavior' in document.documentElement.style) {{
|
||||
window.scrollBy({{
|
||||
top: {delta_y},
|
||||
left: {delta_x},
|
||||
behavior: 'smooth'
|
||||
}});
|
||||
return {{ method: 'window-smooth', success: true }};
|
||||
}}
|
||||
|
||||
window.scrollBy({delta_x}, {delta_y});
|
||||
return {{ method: 'window-instant', success: true }};
|
||||
}})();
|
||||
"""
|
||||
|
||||
try:
|
||||
result = await self.evaluate(tab_id, smart_scroll_script)
|
||||
value = result.get("result", {})
|
||||
if value and value.get("success"):
|
||||
return {
|
||||
"ok": True,
|
||||
"action": "scroll",
|
||||
"direction": direction,
|
||||
"amount": amount,
|
||||
"method": value.get("method", "js"),
|
||||
"container": value.get("containerTag", "window")
|
||||
}
|
||||
except Exception as e:
|
||||
logger.debug("Smart scroll script failed: %s", e)
|
||||
|
||||
# Method 2: Find scrollable container and use mouse wheel at its center
|
||||
try:
|
||||
# Find the largest scrollable container and its position
|
||||
find_container_script = """
|
||||
(function() {
|
||||
const candidates = [];
|
||||
const allElements = document.querySelectorAll('*');
|
||||
for (const el of allElements) {
|
||||
const style = getComputedStyle(el);
|
||||
const overflow = style.overflow + style.overflowY;
|
||||
if (overflow.includes('scroll') || overflow.includes('auto')) {
|
||||
const rect = el.getBoundingClientRect();
|
||||
if (rect.width > 100 && rect.height > 100 &&
|
||||
el.scrollHeight > el.clientHeight + 100) {
|
||||
candidates.push({
|
||||
x: Math.round(rect.left + rect.width / 2),
|
||||
y: Math.round(rect.top + rect.height / 2),
|
||||
area: rect.width * rect.height,
|
||||
tag: el.tagName
|
||||
});
|
||||
}
|
||||
}
|
||||
}
|
||||
candidates.sort((a, b) => b.area - a.area);
|
||||
return candidates.length > 0 ? candidates[0] : null;
|
||||
})();
|
||||
"""
|
||||
container_result = await self._cdp(
|
||||
tab_id,
|
||||
"Runtime.evaluate",
|
||||
{"expression": find_container_script, "returnByValue": True},
|
||||
)
|
||||
container_info = container_result.get("result", {}).get("value", {})
|
||||
|
||||
if container_info and isinstance(container_info, dict):
|
||||
x = container_info.get("x", 400)
|
||||
y = container_info.get("y", 300)
|
||||
else:
|
||||
# Fallback to viewport center
|
||||
viewport_result = await self._cdp(
|
||||
tab_id,
|
||||
"Runtime.evaluate",
|
||||
{
|
||||
"expression": "({w: window.innerWidth, h: window.innerHeight})",
|
||||
"returnByValue": True,
|
||||
},
|
||||
)
|
||||
vp = viewport_result.get("result", {}).get("value", {})
|
||||
x = vp.get("w", 800) // 2
|
||||
y = vp.get("h", 600) // 2
|
||||
|
||||
# Dispatch mouse wheel event at container center
|
||||
await self._cdp(
|
||||
tab_id,
|
||||
"Input.dispatchMouseEvent",
|
||||
{
|
||||
"type": "mouseWheel",
|
||||
"x": x,
|
||||
"y": y,
|
||||
"deltaX": -delta_x,
|
||||
"deltaY": -delta_y,
|
||||
},
|
||||
)
|
||||
return {
|
||||
"ok": True,
|
||||
"action": "scroll",
|
||||
"direction": direction,
|
||||
"amount": amount,
|
||||
"method": "mouseWheel",
|
||||
"target": f"({x}, {y})"
|
||||
}
|
||||
except Exception as e:
|
||||
logger.warning("Scroll failed: %s", e)
|
||||
return {"ok": False, "error": str(e)}
|
||||
|
||||
async def select_option(self, tab_id: int, selector: str, values: list[str]) -> dict:
|
||||
"""Select options in a select element."""
|
||||
@@ -675,6 +1009,8 @@ class BeelineBridge:
|
||||
async def evaluate(self, tab_id: int, script: str) -> dict:
|
||||
"""Execute JavaScript in the page."""
|
||||
await self.cdp_attach(tab_id)
|
||||
await self._try_enable_domain(tab_id, "Runtime")
|
||||
|
||||
# Wrap in IIFE to allow return statements at top level
|
||||
wrapped_script = f"(function() {{ {script} }})()"
|
||||
result = await self._cdp(
|
||||
@@ -683,31 +1019,176 @@ class BeelineBridge:
|
||||
{"expression": wrapped_script, "returnByValue": True, "awaitPromise": True},
|
||||
)
|
||||
|
||||
if result is None:
|
||||
return {"ok": False, "error": "CDP returned no result"}
|
||||
|
||||
if "exceptionDetails" in result:
|
||||
return {
|
||||
"ok": False,
|
||||
"error": result["exceptionDetails"].get("text", "Script error"),
|
||||
}
|
||||
|
||||
# The CDP response structure is {result: {type: ..., value: ...}}
|
||||
# But our bridge returns just the inner result object
|
||||
inner_result = result.get("result", {})
|
||||
value = inner_result.get("value") if isinstance(inner_result, dict) else None
|
||||
|
||||
return {
|
||||
"ok": True,
|
||||
"action": "evaluate",
|
||||
"result": result.get("result", {}).get("value"),
|
||||
"result": value,
|
||||
}
|
||||
|
||||
async def snapshot(self, tab_id: int) -> dict:
|
||||
async def snapshot(self, tab_id: int, timeout_s: float = 10.0) -> dict:
|
||||
"""Get an accessibility snapshot of the page.
|
||||
|
||||
Uses CDP Accessibility.getFullAXTree and formats it as a readable tree.
|
||||
Uses a hybrid approach:
|
||||
1. CDP Accessibility.getFullAXTree for semantic structure
|
||||
2. DOM queries for visibility and computed styles
|
||||
3. Falls back to DOM tree if accessibility returns mostly ignored
|
||||
|
||||
Args:
|
||||
tab_id: The tab ID to snapshot
|
||||
timeout_s: Maximum time to spend building snapshot (default 10s)
|
||||
"""
|
||||
await self.cdp_attach(tab_id)
|
||||
await self._cdp(tab_id, "Accessibility.enable")
|
||||
async with asyncio.timeout(timeout_s):
|
||||
await self.cdp_attach(tab_id)
|
||||
await self._try_enable_domain(tab_id, "Accessibility")
|
||||
await self._try_enable_domain(tab_id, "DOM")
|
||||
await self._try_enable_domain(tab_id, "Runtime")
|
||||
|
||||
result = await self._cdp(tab_id, "Accessibility.getFullAXTree")
|
||||
nodes = result.get("nodes", [])
|
||||
# Try accessibility tree first
|
||||
result = await self._cdp(tab_id, "Accessibility.getFullAXTree")
|
||||
nodes = result.get("nodes", [])
|
||||
|
||||
# Format the tree
|
||||
snapshot = self._format_ax_tree(nodes)
|
||||
# Count non-ignored nodes
|
||||
visible_count = sum(1 for n in nodes if not n.get("ignored", False))
|
||||
|
||||
# If tree is too large or mostly ignored, use DOM-based snapshot
|
||||
if len(nodes) > 5000:
|
||||
logger.debug(
|
||||
"Accessibility tree too large (%d nodes), using DOM snapshot",
|
||||
len(nodes),
|
||||
)
|
||||
return await self._dom_snapshot(tab_id)
|
||||
|
||||
if visible_count < 10 and len(nodes) > 50:
|
||||
logger.debug(
|
||||
"Accessibility tree has only %d/%d visible nodes, falling back to DOM snapshot",
|
||||
visible_count,
|
||||
len(nodes),
|
||||
)
|
||||
return await self._dom_snapshot(tab_id)
|
||||
|
||||
# Format the accessibility tree (with node limit)
|
||||
snapshot = self._format_ax_tree(nodes, max_nodes=2000)
|
||||
|
||||
# Get URL
|
||||
url_result = await self._cdp(
|
||||
tab_id,
|
||||
"Runtime.evaluate",
|
||||
{"expression": "window.location.href", "returnByValue": True},
|
||||
)
|
||||
url = url_result.get("result", {}).get("value", "")
|
||||
|
||||
return {
|
||||
"ok": True,
|
||||
"tabId": tab_id,
|
||||
"url": url,
|
||||
"tree": snapshot,
|
||||
}
|
||||
|
||||
async def _dom_snapshot(self, tab_id: int) -> dict:
|
||||
"""Fallback: build snapshot from DOM tree with visibility info."""
|
||||
# Get all interactive elements using DOM queries
|
||||
script = """
|
||||
(function() {
|
||||
const interactiveSelectors = [
|
||||
'a', 'button', 'input', 'textarea', 'select', 'option',
|
||||
'[onclick]', '[role="button"]', '[role="link"]',
|
||||
'[contenteditable="true"]', 'summary', 'details',
|
||||
'a[href]', 'button[type]', 'input[type]',
|
||||
'label', 'form', 'nav', 'nav a', 'nav button',
|
||||
'[aria-label]', '[aria-labelledby]', '[tabindex]',
|
||||
'h1', 'h2', 'h3', 'h4', 'h5', 'h6'
|
||||
].join(', ');
|
||||
|
||||
const elements = document.querySelectorAll(interactiveSelectors);
|
||||
const results = [];
|
||||
|
||||
for (const el of elements) {
|
||||
const rect = el.getBoundingClientRect();
|
||||
const styles = window.getComputedStyle(el);
|
||||
|
||||
// Skip invisible elements
|
||||
if (rect.width === 0 || rect.height === 1 ||
|
||||
styles.display === 'none' ||
|
||||
styles.visibility === 'hidden' ||
|
||||
styles.opacity === '0') {
|
||||
continue;
|
||||
}
|
||||
|
||||
// Skip elements outside viewport
|
||||
if (rect.bottom < 0 || rect.top > window.innerHeight ||
|
||||
rect.right < 0 || rect.left > window.innerWidth) {
|
||||
continue;
|
||||
}
|
||||
|
||||
const tag = el.tagName.toLowerCase();
|
||||
const text = (el.innerText || el.value || el.placeholder || el.getAttribute('aria-label') || '').substring(0, 80);
|
||||
const type = el.type || tag;
|
||||
const role = el.getAttribute('role') || tag;
|
||||
const name = el.name || el.id || '';
|
||||
const href = el.href || '';
|
||||
const className = el.className || '';
|
||||
|
||||
results.push({
|
||||
tag,
|
||||
type,
|
||||
role,
|
||||
text: text.trim(),
|
||||
name,
|
||||
href,
|
||||
className: className.split(' ').slice(0, 3).join(' '),
|
||||
rect: {
|
||||
x: Math.round(rect.x),
|
||||
y: Math.round(rect.y),
|
||||
width: Math.round(rect.width),
|
||||
height: Math.round(rect.height)
|
||||
}
|
||||
});
|
||||
}
|
||||
|
||||
return results;
|
||||
})();
|
||||
"""
|
||||
|
||||
result = await self.evaluate(tab_id, script)
|
||||
elements = result.get("result", [])
|
||||
|
||||
if not elements:
|
||||
return {
|
||||
"ok": True,
|
||||
"tabId": tab_id,
|
||||
"tree": "(no visible interactive elements found)",
|
||||
}
|
||||
|
||||
# Format as tree
|
||||
lines = []
|
||||
for i in range(0, min(100, len(elements))):
|
||||
el = elements[i]
|
||||
ref = f"e{i}"
|
||||
tag = el.get("tag", "unknown")
|
||||
text = el.get("text", "")
|
||||
role = el.get("role", tag)
|
||||
|
||||
desc = f"{role}"
|
||||
if text:
|
||||
desc += f' "{text[:40]}"'
|
||||
if el.get("href"):
|
||||
desc += f' [href]'
|
||||
desc += f" [ref={ref}]"
|
||||
lines.append(f" - {desc}")
|
||||
|
||||
# Get URL
|
||||
url_result = await self._cdp(
|
||||
@@ -715,17 +1196,22 @@ class BeelineBridge:
|
||||
"Runtime.evaluate",
|
||||
{"expression": "window.location.href", "returnByValue": True},
|
||||
)
|
||||
url = url_result.get("result", {}).get("result", {}).get("value", "")
|
||||
url = url_result.get("result", {}).get("value", "")
|
||||
|
||||
return {
|
||||
"ok": True,
|
||||
"tabId": tab_id,
|
||||
"url": url,
|
||||
"snapshot": snapshot,
|
||||
"tree": "\n".join(lines),
|
||||
}
|
||||
|
||||
def _format_ax_tree(self, nodes: list[dict]) -> str:
|
||||
"""Format a CDP Accessibility.getFullAXTree result."""
|
||||
def _format_ax_tree(self, nodes: list[dict], max_nodes: int = 2000) -> str:
|
||||
"""Format a CDP Accessibility.getFullAXTree result.
|
||||
|
||||
Args:
|
||||
nodes: List of accessibility tree nodes
|
||||
max_nodes: Maximum number of nodes to process (prevents hangs on huge trees)
|
||||
"""
|
||||
if not nodes:
|
||||
return "(empty tree)"
|
||||
|
||||
@@ -737,9 +1223,14 @@ class BeelineBridge:
|
||||
|
||||
lines: list[str] = []
|
||||
ref_counter = [0] # Use list to allow mutation in nested function
|
||||
node_counter = [0] # Track total nodes processed
|
||||
ref_map: dict[str, str] = {}
|
||||
|
||||
def _walk(node_id: str, depth: int) -> None:
|
||||
# Stop if we've processed enough nodes
|
||||
if node_counter[0] >= max_nodes:
|
||||
return
|
||||
|
||||
node = by_id.get(node_id)
|
||||
if not node:
|
||||
return
|
||||
@@ -760,6 +1251,8 @@ class BeelineBridge:
|
||||
_walk(cid, depth)
|
||||
return
|
||||
|
||||
node_counter[0] += 1
|
||||
|
||||
name_info = node.get("name", {})
|
||||
name = name_info.get("value", "") if isinstance(name_info, dict) else str(name_info)
|
||||
|
||||
@@ -807,6 +1300,11 @@ class BeelineBridge:
|
||||
_walk(cid, depth + 1)
|
||||
|
||||
_walk(nodes[0]["nodeId"], 0)
|
||||
|
||||
# Add truncation notice if we hit the limit
|
||||
if node_counter[0] >= max_nodes:
|
||||
lines.append("... (tree truncated, too many nodes)")
|
||||
|
||||
return "\n".join(lines) if lines else "(empty tree)"
|
||||
|
||||
async def get_text(self, tab_id: int, selector: str, timeout_ms: int = 30000) -> dict:
|
||||
|
||||
@@ -0,0 +1,878 @@
|
||||
"""Comprehensive tests for browser tools with FastMCP fixtures.
|
||||
|
||||
Tests cover:
|
||||
- Multiple subagents with multiple tab groups
|
||||
- Complex script execution for LinkedIn, Twitter, YouTube
|
||||
- Tab lifecycle management
|
||||
- Navigation and interactions
|
||||
- Error handling and edge cases
|
||||
"""
|
||||
|
||||
from __future__ import annotations
|
||||
|
||||
import asyncio
|
||||
import json
|
||||
from collections.abc import Callable
|
||||
from typing import Any
|
||||
from unittest.mock import AsyncMock, MagicMock, patch
|
||||
|
||||
import pytest
|
||||
from fastmcp import FastMCP
|
||||
|
||||
from gcu.browser.bridge import BeelineBridge
|
||||
from gcu.browser.tools.advanced import register_advanced_tools
|
||||
from gcu.browser.tools.inspection import register_inspection_tools
|
||||
from gcu.browser.tools.interactions import register_interaction_tools
|
||||
from gcu.browser.tools.lifecycle import register_lifecycle_tools
|
||||
from gcu.browser.tools.navigation import register_navigation_tools
|
||||
from gcu.browser.tools.tabs import register_tab_tools
|
||||
|
||||
|
||||
# ─────────────────────────────────────────────────────────────────────────────
|
||||
# Fixtures
|
||||
# ─────────────────────────────────────────────────────────────────────────────
|
||||
|
||||
|
||||
@pytest.fixture
|
||||
def mcp() -> FastMCP:
|
||||
"""Create a fresh FastMCP instance for testing."""
|
||||
return FastMCP("test-browser-comprehensive")
|
||||
|
||||
|
||||
@pytest.fixture
|
||||
def mock_bridge() -> MagicMock:
|
||||
"""Create a mock BeelineBridge with common methods pre-configured."""
|
||||
bridge = MagicMock(spec=BeelineBridge)
|
||||
bridge.is_connected = True
|
||||
bridge._cdp_attached = set()
|
||||
|
||||
# Context management
|
||||
bridge.create_context = AsyncMock(return_value={"groupId": 1, "tabId": 100})
|
||||
bridge.destroy_context = AsyncMock(return_value={"ok": True})
|
||||
|
||||
# Tab management
|
||||
bridge.create_tab = AsyncMock(return_value={"tabId": 101})
|
||||
bridge.close_tab = AsyncMock(return_value={"ok": True})
|
||||
bridge.list_tabs = AsyncMock(return_value={"tabs": []})
|
||||
bridge.activate_tab = AsyncMock(return_value={"ok": True})
|
||||
|
||||
# Navigation
|
||||
bridge.navigate = AsyncMock(return_value={"ok": True, "url": "https://example.com"})
|
||||
bridge.go_back = AsyncMock(return_value={"ok": True})
|
||||
bridge.go_forward = AsyncMock(return_value={"ok": True})
|
||||
bridge.reload = AsyncMock(return_value={"ok": True})
|
||||
|
||||
# Interactions
|
||||
bridge.click = AsyncMock(return_value={"ok": True})
|
||||
bridge.click_coordinate = AsyncMock(return_value={"ok": True})
|
||||
bridge.type_text = AsyncMock(return_value={"ok": True})
|
||||
bridge.press_key = AsyncMock(return_value={"ok": True})
|
||||
bridge.hover = AsyncMock(return_value={"ok": True})
|
||||
bridge.scroll = AsyncMock(return_value={"ok": True})
|
||||
bridge.select_option = AsyncMock(return_value={"ok": True, "selected": ["option1"]})
|
||||
bridge.drag = AsyncMock(return_value={"ok": True})
|
||||
|
||||
# Inspection
|
||||
bridge.evaluate = AsyncMock(return_value={"result": {"value": True}})
|
||||
bridge.snapshot = AsyncMock(return_value={"tree": "mock_accessibility_tree"})
|
||||
bridge.screenshot = AsyncMock(return_value={"data": "base64imagedata"})
|
||||
bridge.get_text = AsyncMock(return_value={"text": "sample text"})
|
||||
bridge.get_attribute = AsyncMock(return_value={"value": "attribute_value"})
|
||||
|
||||
# Advanced
|
||||
bridge.wait_for_selector = AsyncMock(return_value={"ok": True})
|
||||
bridge.wait_for_text = AsyncMock(return_value={"ok": True})
|
||||
bridge.resize = AsyncMock(return_value={"ok": True})
|
||||
bridge.upload_file = AsyncMock(return_value={"ok": True})
|
||||
bridge.handle_dialog = AsyncMock(return_value={"ok": True})
|
||||
bridge.cdp_attach = AsyncMock(return_value={"ok": True})
|
||||
bridge.cdp_detach = AsyncMock(return_value={"ok": True})
|
||||
|
||||
return bridge
|
||||
|
||||
|
||||
# ─────────────────────────────────────────────────────────────────────────────
|
||||
# Test Classes
|
||||
# ─────────────────────────────────────────────────────────────────────────────
|
||||
|
||||
|
||||
class TestMultipleSubagentsTabGroups:
|
||||
"""Tests for multiple subagents creating and managing multiple tab groups."""
|
||||
|
||||
@pytest.mark.asyncio
|
||||
async def test_multiple_agents_create_separate_tab_groups(
|
||||
self, mcp: FastMCP, mock_bridge: MagicMock
|
||||
):
|
||||
"""Multiple subagents should each create their own tab group."""
|
||||
call_count = 0
|
||||
|
||||
async def mock_create_context(agent_id: str) -> dict:
|
||||
nonlocal call_count
|
||||
call_count += 1
|
||||
return {"groupId": call_count, "tabId": 100 + call_count}
|
||||
|
||||
mock_bridge.create_context = AsyncMock(side_effect=mock_create_context)
|
||||
|
||||
# Register tools first
|
||||
register_lifecycle_tools(mcp)
|
||||
browser_start = mcp._tool_manager._tools["browser_start"].fn
|
||||
|
||||
# Now patch for execution
|
||||
with patch("gcu.browser.tools.lifecycle.get_bridge", return_value=mock_bridge):
|
||||
# Simulate 3 different subagents starting browsers
|
||||
results = await asyncio.gather(
|
||||
browser_start(profile="agent_1"),
|
||||
browser_start(profile="agent_2"),
|
||||
browser_start(profile="agent_3"),
|
||||
)
|
||||
|
||||
# Each should have created a separate context
|
||||
assert mock_bridge.create_context.call_count == 3
|
||||
assert all(r.get("ok") for r in results)
|
||||
|
||||
@pytest.mark.asyncio
|
||||
async def test_concurrent_tab_operations_different_groups(
|
||||
self, mcp: FastMCP, mock_bridge: MagicMock
|
||||
):
|
||||
"""Tab operations in different groups should not interfere."""
|
||||
group1_tabs = [
|
||||
{"id": 101, "url": "https://site1.com", "title": "Site 1"},
|
||||
{"id": 102, "url": "https://site2.com", "title": "Site 2"},
|
||||
]
|
||||
group2_tabs = [
|
||||
{"id": 201, "url": "https://site3.com", "title": "Site 3"},
|
||||
{"id": 202, "url": "https://site4.com", "title": "Site 4"},
|
||||
]
|
||||
|
||||
def mock_list_tabs(group_id: int) -> dict:
|
||||
if group_id == 1:
|
||||
return {"tabs": group1_tabs}
|
||||
elif group_id == 2:
|
||||
return {"tabs": group2_tabs}
|
||||
return {"tabs": []}
|
||||
|
||||
mock_bridge.list_tabs = AsyncMock(side_effect=mock_list_tabs)
|
||||
|
||||
register_tab_tools(mcp)
|
||||
browser_tabs = mcp._tool_manager._tools["browser_tabs"].fn
|
||||
|
||||
with patch("gcu.browser.tools.tabs.get_bridge", return_value=mock_bridge):
|
||||
with patch(
|
||||
"gcu.browser.tools.tabs._get_context",
|
||||
side_effect=lambda p: {
|
||||
"groupId": 1 if p == "agent_1" else 2,
|
||||
"activeTabId": 101 if p == "agent_1" else 201,
|
||||
},
|
||||
):
|
||||
# Concurrent tab listing from different agents
|
||||
results = await asyncio.gather(
|
||||
browser_tabs(profile="agent_1"),
|
||||
browser_tabs(profile="agent_2"),
|
||||
)
|
||||
|
||||
# Each should see only their own tabs
|
||||
assert len(results[0].get("tabs", [])) == 2
|
||||
assert len(results[1].get("tabs", [])) == 2
|
||||
assert results[0]["tabs"][0]["id"] == 101
|
||||
assert results[1]["tabs"][0]["id"] == 201
|
||||
|
||||
@pytest.mark.asyncio
|
||||
async def test_tab_group_isolation(self, mcp: FastMCP, mock_bridge: MagicMock):
|
||||
"""Closing a tab in one group should not affect other groups."""
|
||||
closed_tabs = []
|
||||
|
||||
async def mock_close_tab(tab_id: int) -> dict:
|
||||
closed_tabs.append(tab_id)
|
||||
return {"ok": True}
|
||||
|
||||
mock_bridge.close_tab = AsyncMock(side_effect=mock_close_tab)
|
||||
|
||||
register_tab_tools(mcp)
|
||||
browser_close = mcp._tool_manager._tools["browser_close"].fn
|
||||
|
||||
with patch("gcu.browser.tools.tabs.get_bridge", return_value=mock_bridge):
|
||||
with patch(
|
||||
"gcu.browser.tools.tabs._get_context",
|
||||
return_value={"groupId": 1, "activeTabId": 101},
|
||||
):
|
||||
result = await browser_close(tab_id=101, profile="agent_1")
|
||||
|
||||
assert result.get("ok") is True
|
||||
assert 101 in closed_tabs
|
||||
|
||||
|
||||
class TestComplexScriptExecution:
|
||||
"""Tests for complex JavaScript execution patterns on real-world sites."""
|
||||
|
||||
@pytest.mark.asyncio
|
||||
async def test_linkedin_scroll_infinite_feed(self, mcp: FastMCP, mock_bridge: MagicMock):
|
||||
"""Test LinkedIn-style infinite feed scrolling with lazy loading."""
|
||||
scroll_calls = []
|
||||
|
||||
async def mock_scroll(tab_id: int, direction: str, amount: int = 500) -> dict:
|
||||
scroll_calls.append((tab_id, direction, amount))
|
||||
return {"ok": True}
|
||||
|
||||
mock_bridge.scroll = AsyncMock(side_effect=mock_scroll)
|
||||
|
||||
register_interaction_tools(mcp)
|
||||
browser_scroll = mcp._tool_manager._tools["browser_scroll"].fn
|
||||
|
||||
with patch("gcu.browser.tools.interactions.get_bridge", return_value=mock_bridge):
|
||||
with patch(
|
||||
"gcu.browser.tools.interactions._get_context",
|
||||
return_value={"groupId": 1, "activeTabId": 100},
|
||||
):
|
||||
# Simulate infinite scroll - multiple scroll operations
|
||||
for _ in range(3):
|
||||
await browser_scroll(direction="down", amount=500)
|
||||
|
||||
assert len(scroll_calls) == 3
|
||||
|
||||
@pytest.mark.asyncio
|
||||
async def test_linkedin_profile_data_extraction(self, mcp: FastMCP, mock_bridge: MagicMock):
|
||||
"""Test extracting LinkedIn profile data using complex selectors."""
|
||||
profile_data = {
|
||||
"name": "John Doe",
|
||||
"title": "Software Engineer at Tech Corp",
|
||||
}
|
||||
|
||||
mock_bridge.evaluate = AsyncMock(return_value={"result": {"value": profile_data}})
|
||||
|
||||
register_advanced_tools(mcp)
|
||||
browser_evaluate = mcp._tool_manager._tools["browser_evaluate"].fn
|
||||
|
||||
with patch("gcu.browser.tools.advanced.get_bridge", return_value=mock_bridge):
|
||||
with patch(
|
||||
"gcu.browser.tools.advanced._get_context",
|
||||
return_value={"groupId": 1, "activeTabId": 100},
|
||||
):
|
||||
# Extract profile data via JavaScript
|
||||
extraction_script = """
|
||||
const name = document.querySelector('.text-heading-xlarge')?.innerText;
|
||||
const title = document.querySelector('.text-body-medium')?.innerText;
|
||||
return { name, title };
|
||||
"""
|
||||
result = await browser_evaluate(script=extraction_script)
|
||||
|
||||
# browser_evaluate returns the raw result from bridge.evaluate
|
||||
assert "result" in result
|
||||
assert result["result"]["value"] == profile_data
|
||||
|
||||
@pytest.mark.asyncio
|
||||
async def test_twitter_x_infinite_timeline(self, mcp: FastMCP, mock_bridge: MagicMock):
|
||||
"""Test Twitter/X infinite timeline scrolling with tweet loading."""
|
||||
tweets_loaded = ["tweet_0", "tweet_1", "tweet_2", "tweet_3", "tweet_4"]
|
||||
|
||||
mock_bridge.evaluate = AsyncMock(return_value={"result": {"value": tweets_loaded}})
|
||||
mock_bridge.scroll = AsyncMock(return_value={"ok": True})
|
||||
|
||||
register_interaction_tools(mcp)
|
||||
register_advanced_tools(mcp)
|
||||
|
||||
browser_scroll = mcp._tool_manager._tools["browser_scroll"].fn
|
||||
browser_evaluate = mcp._tool_manager._tools["browser_evaluate"].fn
|
||||
|
||||
with patch("gcu.browser.tools.interactions.get_bridge", return_value=mock_bridge):
|
||||
with patch(
|
||||
"gcu.browser.tools.interactions._get_context",
|
||||
return_value={"groupId": 1, "activeTabId": 100},
|
||||
):
|
||||
# Simulate Twitter timeline scroll
|
||||
await browser_scroll(direction="down", amount=800)
|
||||
|
||||
with patch("gcu.browser.tools.advanced.get_bridge", return_value=mock_bridge):
|
||||
with patch(
|
||||
"gcu.browser.tools.advanced._get_context",
|
||||
return_value={"groupId": 1, "activeTabId": 100},
|
||||
):
|
||||
extract_script = """
|
||||
return Array.from(document.querySelectorAll('article[data-testid="tweet"]'))
|
||||
.slice(0, 5)
|
||||
.map(t => t.innerText);
|
||||
"""
|
||||
result = await browser_evaluate(script=extract_script)
|
||||
|
||||
# browser_evaluate returns raw result from bridge
|
||||
assert "result" in result
|
||||
assert result["result"]["value"] == tweets_loaded
|
||||
|
||||
@pytest.mark.asyncio
|
||||
async def test_youtube_video_player_interaction(self, mcp: FastMCP, mock_bridge: MagicMock):
|
||||
"""Test YouTube video player controls and state management."""
|
||||
player_state = {"playing": False, "currentTime": 0, "duration": 300}
|
||||
|
||||
mock_bridge.evaluate = AsyncMock(return_value={"result": {"value": player_state}})
|
||||
mock_bridge.click = AsyncMock(return_value={"ok": True})
|
||||
|
||||
register_advanced_tools(mcp)
|
||||
register_interaction_tools(mcp)
|
||||
|
||||
browser_evaluate = mcp._tool_manager._tools["browser_evaluate"].fn
|
||||
|
||||
with patch("gcu.browser.tools.advanced.get_bridge", return_value=mock_bridge):
|
||||
with patch(
|
||||
"gcu.browser.tools.advanced._get_context",
|
||||
return_value={"groupId": 1, "activeTabId": 100},
|
||||
):
|
||||
# Interact with YouTube player
|
||||
play_script = """
|
||||
document.querySelector('.ytp-play-button')?.click();
|
||||
return true;
|
||||
"""
|
||||
result = await browser_evaluate(script=play_script)
|
||||
|
||||
# browser_evaluate returns raw result from bridge
|
||||
assert "result" in result
|
||||
|
||||
@pytest.mark.asyncio
|
||||
async def test_youtube_comments_expansion(self, mcp: FastMCP, mock_bridge: MagicMock):
|
||||
"""Test YouTube comments section expansion and loading."""
|
||||
comments = ["comment_1", "comment_2", "comment_3"]
|
||||
|
||||
mock_bridge.evaluate = AsyncMock(return_value={"result": {"value": comments}})
|
||||
mock_bridge.scroll = AsyncMock(return_value={"ok": True})
|
||||
mock_bridge.click = AsyncMock(return_value={"ok": True})
|
||||
|
||||
register_advanced_tools(mcp)
|
||||
|
||||
browser_evaluate = mcp._tool_manager._tools["browser_evaluate"].fn
|
||||
|
||||
with patch("gcu.browser.tools.advanced.get_bridge", return_value=mock_bridge):
|
||||
with patch(
|
||||
"gcu.browser.tools.advanced._get_context",
|
||||
return_value={"groupId": 1, "activeTabId": 100},
|
||||
):
|
||||
# Scroll to comments and expand
|
||||
expand_script = """
|
||||
const commentsSection = document.querySelector('ytd-comments#comments');
|
||||
if (commentsSection) {
|
||||
commentsSection.scrollIntoView();
|
||||
return true;
|
||||
}
|
||||
return false;
|
||||
"""
|
||||
result = await browser_evaluate(script=expand_script)
|
||||
|
||||
# browser_evaluate returns raw result from bridge
|
||||
assert "result" in result
|
||||
|
||||
@pytest.mark.asyncio
|
||||
async def test_complex_form_filling_linkedin(self, mcp: FastMCP, mock_bridge: MagicMock):
|
||||
"""Test complex form filling on LinkedIn with dynamic fields."""
|
||||
filled_fields = {}
|
||||
|
||||
async def mock_type_text(tab_id: int, selector: str, text: str, **kwargs) -> dict:
|
||||
filled_fields[selector] = text
|
||||
return {"ok": True}
|
||||
|
||||
async def mock_select_option(tab_id: int, selector: str, values: list, **kwargs) -> dict:
|
||||
filled_fields[selector] = values
|
||||
return {"ok": True, "selected": values}
|
||||
|
||||
mock_bridge.type_text = AsyncMock(side_effect=mock_type_text)
|
||||
mock_bridge.select_option = AsyncMock(side_effect=mock_select_option)
|
||||
|
||||
register_interaction_tools(mcp)
|
||||
|
||||
browser_type = mcp._tool_manager._tools["browser_type"].fn
|
||||
browser_select = mcp._tool_manager._tools["browser_select"].fn
|
||||
|
||||
with patch("gcu.browser.tools.interactions.get_bridge", return_value=mock_bridge):
|
||||
with patch(
|
||||
"gcu.browser.tools.interactions._get_context",
|
||||
return_value={"groupId": 1, "activeTabId": 100},
|
||||
):
|
||||
# Fill out a LinkedIn job application form
|
||||
await browser_type(selector="#first-name", text="John")
|
||||
await browser_type(selector="#last-name", text="Doe")
|
||||
await browser_type(selector="#email", text="john.doe@example.com")
|
||||
await browser_select(selector="#experience-level", values=["5-10 years"])
|
||||
|
||||
assert filled_fields.get("#first-name") == "John"
|
||||
assert filled_fields.get("#last-name") == "Doe"
|
||||
assert filled_fields.get("#email") == "john.doe@example.com"
|
||||
|
||||
|
||||
class TestTabLifecycle:
|
||||
"""Tests for tab lifecycle management."""
|
||||
|
||||
@pytest.mark.asyncio
|
||||
async def test_create_and_close_tab(self, mcp: FastMCP, mock_bridge: MagicMock):
|
||||
"""Test creating and closing a tab."""
|
||||
mock_bridge.create_tab = AsyncMock(return_value={"tabId": 123})
|
||||
mock_bridge.close_tab = AsyncMock(return_value={"ok": True})
|
||||
|
||||
register_tab_tools(mcp)
|
||||
browser_open = mcp._tool_manager._tools["browser_open"].fn
|
||||
browser_close = mcp._tool_manager._tools["browser_close"].fn
|
||||
|
||||
with patch("gcu.browser.tools.tabs.get_bridge", return_value=mock_bridge):
|
||||
with patch(
|
||||
"gcu.browser.tools.tabs._get_context",
|
||||
return_value={"groupId": 1, "activeTabId": 100},
|
||||
):
|
||||
open_result = await browser_open(url="https://example.com")
|
||||
assert open_result.get("ok") is True
|
||||
|
||||
close_result = await browser_close(tab_id=123)
|
||||
assert close_result.get("ok") is True
|
||||
|
||||
@pytest.mark.asyncio
|
||||
async def test_tab_focus_switching(self, mcp: FastMCP, mock_bridge: MagicMock):
|
||||
"""Test switching focus between tabs."""
|
||||
mock_bridge.activate_tab = AsyncMock(return_value={"ok": True})
|
||||
|
||||
register_tab_tools(mcp)
|
||||
browser_focus = mcp._tool_manager._tools["browser_focus"].fn
|
||||
|
||||
with patch("gcu.browser.tools.tabs.get_bridge", return_value=mock_bridge):
|
||||
with patch(
|
||||
"gcu.browser.tools.tabs._get_context",
|
||||
return_value={"groupId": 1, "activeTabId": 100},
|
||||
):
|
||||
result = await browser_focus(tab_id=200)
|
||||
|
||||
assert result.get("ok") is True
|
||||
mock_bridge.activate_tab.assert_awaited_once_with(200)
|
||||
|
||||
|
||||
class TestNavigation:
|
||||
"""Tests for navigation tools."""
|
||||
|
||||
@pytest.mark.asyncio
|
||||
async def test_navigate_with_wait_until(self, mcp: FastMCP, mock_bridge: MagicMock):
|
||||
"""Test navigation with different wait_until options."""
|
||||
mock_bridge.navigate = AsyncMock(return_value={"ok": True, "url": "https://example.com"})
|
||||
|
||||
register_navigation_tools(mcp)
|
||||
browser_navigate = mcp._tool_manager._tools["browser_navigate"].fn
|
||||
|
||||
with patch("gcu.browser.tools.navigation.get_bridge", return_value=mock_bridge):
|
||||
with patch(
|
||||
"gcu.browser.tools.navigation._get_context",
|
||||
return_value={"groupId": 1, "activeTabId": 100},
|
||||
):
|
||||
result = await browser_navigate(
|
||||
url="https://example.com", wait_until="networkidle"
|
||||
)
|
||||
|
||||
assert result.get("ok") is True
|
||||
# The bridge.navigate is called with wait_until as keyword argument
|
||||
mock_bridge.navigate.assert_awaited_once_with(100, "https://example.com", wait_until="networkidle")
|
||||
|
||||
@pytest.mark.asyncio
|
||||
async def test_navigation_history(self, mcp: FastMCP, mock_bridge: MagicMock):
|
||||
"""Test back/forward navigation."""
|
||||
mock_bridge.go_back = AsyncMock(return_value={"ok": True})
|
||||
mock_bridge.go_forward = AsyncMock(return_value={"ok": True})
|
||||
|
||||
register_navigation_tools(mcp)
|
||||
browser_go_back = mcp._tool_manager._tools["browser_go_back"].fn
|
||||
browser_go_forward = mcp._tool_manager._tools["browser_go_forward"].fn
|
||||
|
||||
with patch("gcu.browser.tools.navigation.get_bridge", return_value=mock_bridge):
|
||||
with patch(
|
||||
"gcu.browser.tools.navigation._get_context",
|
||||
return_value={"groupId": 1, "activeTabId": 100},
|
||||
):
|
||||
back_result = await browser_go_back()
|
||||
forward_result = await browser_go_forward()
|
||||
|
||||
assert back_result.get("ok") is True
|
||||
assert forward_result.get("ok") is True
|
||||
|
||||
|
||||
class TestInteractions:
|
||||
"""Tests for interaction tools."""
|
||||
|
||||
@pytest.mark.asyncio
|
||||
async def test_click_with_different_buttons(self, mcp: FastMCP, mock_bridge: MagicMock):
|
||||
"""Test clicking with left, right, and middle buttons."""
|
||||
click_calls = []
|
||||
|
||||
async def track_click(tab_id: int, selector: str, button: str = "left", **kwargs) -> dict:
|
||||
click_calls.append((tab_id, selector, button))
|
||||
return {"ok": True}
|
||||
|
||||
mock_bridge.click = AsyncMock(side_effect=track_click)
|
||||
|
||||
register_interaction_tools(mcp)
|
||||
browser_click = mcp._tool_manager._tools["browser_click"].fn
|
||||
|
||||
with patch("gcu.browser.tools.interactions.get_bridge", return_value=mock_bridge):
|
||||
with patch(
|
||||
"gcu.browser.tools.interactions._get_context",
|
||||
return_value={"groupId": 1, "activeTabId": 100},
|
||||
):
|
||||
await browser_click(selector="button", button="left")
|
||||
await browser_click(selector="button", button="right")
|
||||
await browser_click(selector="button", button="middle")
|
||||
|
||||
assert len(click_calls) == 3
|
||||
assert [c[2] for c in click_calls] == ["left", "right", "middle"]
|
||||
|
||||
@pytest.mark.asyncio
|
||||
async def test_type_with_special_characters(self, mcp: FastMCP, mock_bridge: MagicMock):
|
||||
"""Test typing text with special characters and unicode."""
|
||||
typed_texts = []
|
||||
|
||||
async def track_type(tab_id: int, selector: str, text: str, **kwargs) -> dict:
|
||||
typed_texts.append(text)
|
||||
return {"ok": True}
|
||||
|
||||
mock_bridge.type_text = AsyncMock(side_effect=track_type)
|
||||
|
||||
register_interaction_tools(mcp)
|
||||
browser_type = mcp._tool_manager._tools["browser_type"].fn
|
||||
|
||||
with patch("gcu.browser.tools.interactions.get_bridge", return_value=mock_bridge):
|
||||
with patch(
|
||||
"gcu.browser.tools.interactions._get_context",
|
||||
return_value={"groupId": 1, "activeTabId": 100},
|
||||
):
|
||||
# Test various special characters
|
||||
special_texts = [
|
||||
"Hello, World!", # Basic punctuation
|
||||
"O'Reilly & Associates", # Quotes and ampersands
|
||||
"Price: $100 (20% off)", # Currency and parentheses
|
||||
"Email: user@example.com", # Email format
|
||||
"日本語テスト", # Japanese characters
|
||||
"Émojis: 🎉🚀💻", # Emojis
|
||||
]
|
||||
|
||||
for text in special_texts:
|
||||
result = await browser_type(selector="input", text=text)
|
||||
assert result.get("ok") is True
|
||||
|
||||
assert typed_texts == special_texts
|
||||
|
||||
@pytest.mark.asyncio
|
||||
async def test_drag_and_drop(self, mcp: FastMCP, mock_bridge: MagicMock):
|
||||
"""Test drag and drop operation."""
|
||||
# browser_drag uses _cdp directly for DOM queries and mouse events
|
||||
mock_bridge._cdp = AsyncMock(side_effect=lambda tab_id, method, params=None: {
|
||||
"DOM.getDocument": {"root": {"nodeId": 1}},
|
||||
"DOM.querySelector": {"nodeId": 2},
|
||||
"DOM.getBoxModel": {"content": [0, 0, 100, 0, 100, 50, 0, 50]},
|
||||
"Input.dispatchMouseEvent": {},
|
||||
}.get(method, {}))
|
||||
|
||||
register_interaction_tools(mcp)
|
||||
browser_drag = mcp._tool_manager._tools["browser_drag"].fn
|
||||
|
||||
with patch("gcu.browser.tools.interactions.get_bridge", return_value=mock_bridge):
|
||||
with patch(
|
||||
"gcu.browser.tools.interactions._get_context",
|
||||
return_value={"groupId": 1, "activeTabId": 100},
|
||||
):
|
||||
result = await browser_drag(
|
||||
start_selector="#draggable",
|
||||
end_selector="#dropzone",
|
||||
)
|
||||
|
||||
assert result.get("ok") is True
|
||||
|
||||
|
||||
class TestInspection:
|
||||
"""Tests for inspection tools."""
|
||||
|
||||
@pytest.mark.asyncio
|
||||
async def test_snapshot_accessibility_tree(self, mcp: FastMCP, mock_bridge: MagicMock):
|
||||
"""Test getting accessibility tree snapshot."""
|
||||
mock_snapshot = """
|
||||
[1] document "Page Title"
|
||||
[2] button "Submit"
|
||||
[3] textbox "Search"
|
||||
"""
|
||||
mock_bridge.snapshot = AsyncMock(return_value={"tree": mock_snapshot})
|
||||
|
||||
register_inspection_tools(mcp)
|
||||
browser_snapshot = mcp._tool_manager._tools["browser_snapshot"].fn
|
||||
|
||||
with patch("gcu.browser.tools.inspection.get_bridge", return_value=mock_bridge):
|
||||
with patch(
|
||||
"gcu.browser.tools.inspection._get_context",
|
||||
return_value={"groupId": 1, "activeTabId": 100},
|
||||
):
|
||||
result = await browser_snapshot()
|
||||
|
||||
# browser_snapshot returns raw result from bridge
|
||||
assert "tree" in result
|
||||
|
||||
@pytest.mark.asyncio
|
||||
async def test_screenshot_full_page(self, mcp: FastMCP, mock_bridge: MagicMock):
|
||||
"""Test taking full page screenshot."""
|
||||
mock_bridge.screenshot = AsyncMock(
|
||||
return_value={"ok": True, "data": "base64encodedimagedata", "width": 1920, "height": 5000}
|
||||
)
|
||||
|
||||
register_inspection_tools(mcp)
|
||||
browser_screenshot = mcp._tool_manager._tools["browser_screenshot"].fn
|
||||
|
||||
with patch("gcu.browser.tools.inspection.get_bridge", return_value=mock_bridge):
|
||||
with patch(
|
||||
"gcu.browser.tools.inspection._get_context",
|
||||
return_value={"groupId": 1, "activeTabId": 100},
|
||||
):
|
||||
result = await browser_screenshot(full_page=True)
|
||||
|
||||
# browser_screenshot returns list of content blocks
|
||||
assert isinstance(result, list)
|
||||
mock_bridge.screenshot.assert_awaited_once_with(100, full_page=True)
|
||||
|
||||
|
||||
class TestAdvancedTools:
|
||||
"""Tests for advanced tools."""
|
||||
|
||||
@pytest.mark.asyncio
|
||||
async def test_wait_for_selector_timeout(self, mcp: FastMCP, mock_bridge: MagicMock):
|
||||
"""Test wait_for_selector timeout behavior."""
|
||||
mock_bridge.wait_for_selector = AsyncMock(
|
||||
side_effect=TimeoutError("Element not found within timeout")
|
||||
)
|
||||
|
||||
register_advanced_tools(mcp)
|
||||
browser_wait = mcp._tool_manager._tools["browser_wait"].fn
|
||||
|
||||
with patch("gcu.browser.tools.advanced.get_bridge", return_value=mock_bridge):
|
||||
with patch(
|
||||
"gcu.browser.tools.advanced._get_context",
|
||||
return_value={"groupId": 1, "activeTabId": 100},
|
||||
):
|
||||
result = await browser_wait(selector=".nonexistent", timeout_ms=1000)
|
||||
|
||||
# Should return error result, not raise
|
||||
assert result.get("ok") is False
|
||||
assert "error" in result or "timed out" in str(result).lower()
|
||||
|
||||
@pytest.mark.asyncio
|
||||
async def test_evaluate_with_return_value(self, mcp: FastMCP, mock_bridge: MagicMock):
|
||||
"""Test JavaScript evaluation with return value."""
|
||||
mock_bridge.evaluate = AsyncMock(
|
||||
return_value={"result": {"value": {"status": "success", "count": 42}}}
|
||||
)
|
||||
|
||||
register_advanced_tools(mcp)
|
||||
browser_evaluate = mcp._tool_manager._tools["browser_evaluate"].fn
|
||||
|
||||
with patch("gcu.browser.tools.advanced.get_bridge", return_value=mock_bridge):
|
||||
with patch(
|
||||
"gcu.browser.tools.advanced._get_context",
|
||||
return_value={"groupId": 1, "activeTabId": 100},
|
||||
):
|
||||
result = await browser_evaluate(script="return { status: 'success', count: 42 };")
|
||||
|
||||
# browser_evaluate returns raw result from bridge
|
||||
assert "result" in result
|
||||
assert result["result"]["value"]["status"] == "success"
|
||||
|
||||
@pytest.mark.asyncio
|
||||
async def test_file_upload(self, mcp: FastMCP, mock_bridge: MagicMock):
|
||||
"""Test file upload functionality."""
|
||||
mock_bridge.upload_file = AsyncMock(return_value={"ok": True, "files": 2})
|
||||
|
||||
register_advanced_tools(mcp)
|
||||
browser_upload = mcp._tool_manager._tools["browser_upload"].fn
|
||||
|
||||
with patch("gcu.browser.tools.advanced.get_bridge", return_value=mock_bridge):
|
||||
with patch(
|
||||
"gcu.browser.tools.advanced._get_context",
|
||||
return_value={"groupId": 1, "activeTabId": 100},
|
||||
):
|
||||
result = await browser_upload(
|
||||
selector="input[type='file']",
|
||||
file_paths=["/tmp/file1.pdf", "/tmp/file2.pdf"],
|
||||
)
|
||||
|
||||
assert result.get("ok") is True
|
||||
|
||||
|
||||
class TestErrorHandling:
|
||||
"""Tests for error handling scenarios."""
|
||||
|
||||
@pytest.mark.asyncio
|
||||
async def test_bridge_not_connected(self, mcp: FastMCP):
|
||||
"""Test behavior when bridge is not connected."""
|
||||
mock_bridge = MagicMock(spec=BeelineBridge)
|
||||
mock_bridge.is_connected = False
|
||||
|
||||
register_lifecycle_tools(mcp)
|
||||
browser_start = mcp._tool_manager._tools["browser_start"].fn
|
||||
|
||||
with patch("gcu.browser.tools.lifecycle.get_bridge", return_value=mock_bridge):
|
||||
result = await browser_start(profile="test")
|
||||
|
||||
assert result.get("ok") is False
|
||||
assert "not connected" in result.get("error", "").lower()
|
||||
|
||||
@pytest.mark.asyncio
|
||||
async def test_browser_not_started(self, mcp: FastMCP, mock_bridge: MagicMock):
|
||||
"""Test behavior when browser is not started."""
|
||||
register_tab_tools(mcp)
|
||||
browser_tabs = mcp._tool_manager._tools["browser_tabs"].fn
|
||||
|
||||
with patch("gcu.browser.tools.tabs.get_bridge", return_value=mock_bridge):
|
||||
with patch("gcu.browser.tools.tabs._get_context", return_value=None):
|
||||
result = await browser_tabs(profile="nonexistent")
|
||||
|
||||
assert result.get("ok") is False
|
||||
assert "not started" in result.get("error", "").lower()
|
||||
|
||||
@pytest.mark.asyncio
|
||||
async def test_cdp_command_failure(self, mcp: FastMCP, mock_bridge: MagicMock):
|
||||
"""Test handling of CDP command failures."""
|
||||
mock_bridge.click = AsyncMock(
|
||||
side_effect=RuntimeError("CDP error: Element not found")
|
||||
)
|
||||
|
||||
register_interaction_tools(mcp)
|
||||
browser_click = mcp._tool_manager._tools["browser_click"].fn
|
||||
|
||||
with patch("gcu.browser.tools.interactions.get_bridge", return_value=mock_bridge):
|
||||
with patch(
|
||||
"gcu.browser.tools.interactions._get_context",
|
||||
return_value={"groupId": 1, "activeTabId": 100},
|
||||
):
|
||||
result = await browser_click(selector=".nonexistent")
|
||||
|
||||
assert result.get("ok") is False
|
||||
assert "error" in result
|
||||
|
||||
|
||||
class TestIFWrapping:
|
||||
"""Tests for JavaScript IIFE wrapping to handle return statements."""
|
||||
|
||||
@pytest.mark.asyncio
|
||||
async def test_evaluate_with_bare_return(self, mcp: FastMCP, mock_bridge: MagicMock):
|
||||
"""Test that scripts with bare return statements are wrapped properly."""
|
||||
call_args = []
|
||||
|
||||
async def mock_evaluate_capture(tab_id: int, script: str) -> dict:
|
||||
call_args.append(script)
|
||||
return {"result": {"value": 42}}
|
||||
|
||||
mock_bridge.evaluate = AsyncMock(side_effect=mock_evaluate_capture)
|
||||
|
||||
register_advanced_tools(mcp)
|
||||
browser_evaluate = mcp._tool_manager._tools["browser_evaluate"].fn
|
||||
|
||||
with patch("gcu.browser.tools.advanced.get_bridge", return_value=mock_bridge):
|
||||
with patch(
|
||||
"gcu.browser.tools.advanced._get_context",
|
||||
return_value={"groupId": 1, "activeTabId": 100},
|
||||
):
|
||||
# Script with bare return at top level
|
||||
result = await browser_evaluate(script="return 42;")
|
||||
|
||||
# Verify the script was wrapped in IIFE
|
||||
assert len(call_args) == 1
|
||||
wrapped_script = call_args[0]
|
||||
assert wrapped_script.startswith("(function()")
|
||||
assert wrapped_script.endswith("})()")
|
||||
assert result.get("ok") is True
|
||||
|
||||
@pytest.mark.asyncio
|
||||
async def test_evaluate_complex_script(self, mcp: FastMCP, mock_bridge: MagicMock):
|
||||
"""Test complex multi-line script execution."""
|
||||
mock_bridge.evaluate = AsyncMock(
|
||||
return_value={"result": {"value": {"total": 100, "filtered": 50}}}
|
||||
)
|
||||
|
||||
register_advanced_tools(mcp)
|
||||
browser_evaluate = mcp._tool_manager._tools["browser_evaluate"].fn
|
||||
|
||||
with patch("gcu.browser.tools.advanced.get_bridge", return_value=mock_bridge):
|
||||
with patch(
|
||||
"gcu.browser.tools.advanced._get_context",
|
||||
return_value={"groupId": 1, "activeTabId": 100},
|
||||
):
|
||||
complex_script = """
|
||||
const items = document.querySelectorAll('.item');
|
||||
const filtered = Array.from(items).filter(i => i.classList.contains('active'));
|
||||
return {
|
||||
total: items.length,
|
||||
filtered: filtered.length
|
||||
};
|
||||
"""
|
||||
result = await browser_evaluate(script=complex_script)
|
||||
|
||||
assert result.get("ok") is True
|
||||
|
||||
|
||||
class TestConcurrentOperations:
|
||||
"""Tests for concurrent browser operations."""
|
||||
|
||||
@pytest.mark.asyncio
|
||||
async def test_concurrent_clicks_different_tabs(self, mcp: FastMCP, mock_bridge: MagicMock):
|
||||
"""Test clicking on multiple tabs concurrently."""
|
||||
click_order = []
|
||||
|
||||
async def mock_click(tab_id: int, selector: str, **kwargs) -> dict:
|
||||
click_order.append(tab_id)
|
||||
await asyncio.sleep(0.01) # Simulate async operation
|
||||
return {"ok": True}
|
||||
|
||||
mock_bridge.click = AsyncMock(side_effect=mock_click)
|
||||
|
||||
register_interaction_tools(mcp)
|
||||
browser_click = mcp._tool_manager._tools["browser_click"].fn
|
||||
|
||||
with patch("gcu.browser.tools.interactions.get_bridge", return_value=mock_bridge):
|
||||
with patch(
|
||||
"gcu.browser.tools.interactions._get_context",
|
||||
side_effect=lambda p: {
|
||||
"groupId": 1 if p == "agent_1" else 2 if p == "agent_2" else 3,
|
||||
"activeTabId": 101 if p == "agent_1" else 201 if p == "agent_2" else 301,
|
||||
},
|
||||
):
|
||||
# Concurrent clicks from different agents
|
||||
await asyncio.gather(
|
||||
browser_click(selector="button", profile="agent_1"),
|
||||
browser_click(selector="button", profile="agent_2"),
|
||||
browser_click(selector="button", profile="agent_3"),
|
||||
)
|
||||
|
||||
# All clicks should have been executed
|
||||
assert len(click_order) == 3
|
||||
assert set(click_order) == {101, 201, 301}
|
||||
|
||||
@pytest.mark.asyncio
|
||||
async def test_mixed_operations_same_tab(self, mcp: FastMCP, mock_bridge: MagicMock):
|
||||
"""Test mixed operations (click, type, scroll) on same tab."""
|
||||
operations = []
|
||||
|
||||
async def track_click(tab_id: int, selector: str, **kwargs) -> dict:
|
||||
operations.append("click")
|
||||
return {"ok": True}
|
||||
|
||||
async def track_type(tab_id: int, selector: str, text: str, **kwargs) -> dict:
|
||||
operations.append("type")
|
||||
return {"ok": True}
|
||||
|
||||
async def track_scroll(tab_id: int, direction: str, **kwargs) -> dict:
|
||||
operations.append("scroll")
|
||||
return {"ok": True}
|
||||
|
||||
mock_bridge.click = AsyncMock(side_effect=track_click)
|
||||
mock_bridge.type_text = AsyncMock(side_effect=track_type)
|
||||
mock_bridge.scroll = AsyncMock(side_effect=track_scroll)
|
||||
|
||||
register_interaction_tools(mcp)
|
||||
browser_click = mcp._tool_manager._tools["browser_click"].fn
|
||||
browser_type = mcp._tool_manager._tools["browser_type"].fn
|
||||
browser_scroll = mcp._tool_manager._tools["browser_scroll"].fn
|
||||
|
||||
with patch("gcu.browser.tools.interactions.get_bridge", return_value=mock_bridge):
|
||||
with patch(
|
||||
"gcu.browser.tools.interactions._get_context",
|
||||
return_value={"groupId": 1, "activeTabId": 100},
|
||||
):
|
||||
# Mix of operations
|
||||
await browser_click(selector="button")
|
||||
await browser_type(selector="input", text="hello")
|
||||
await browser_scroll(direction="down")
|
||||
|
||||
assert "click" in operations
|
||||
assert "type" in operations
|
||||
assert "scroll" in operations
|
||||
Reference in New Issue
Block a user