Spaces:
Sleeping
Sleeping
File size: 11,354 Bytes
43b3474 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 |
"""
Agent 3: Difference Analyzer (Hybrid Approach)
Uses HF Vision Model + Screenshot Analysis to detect visual differences
Detects layout, color, spacing, typography, and other visual differences
"""
from typing import Dict, Any, List
from state_schema import WorkflowState, VisualDifference
import os
from PIL import Image
import numpy as np
class ScreenshotComparator:
"""Compares Figma and website screenshots for visual differences."""
def __init__(self):
"""Initialize the comparator."""
self.differences = []
def compare_screenshots(self, figma_path: str, website_path: str, viewport: str) -> List[VisualDifference]:
"""
Compare Figma and website screenshots.
Args:
figma_path: Path to Figma screenshot
website_path: Path to website screenshot
viewport: Viewport name (desktop/mobile)
Returns:
List of detected visual differences
"""
differences = []
if not os.path.exists(figma_path) or not os.path.exists(website_path):
return differences
try:
# Load images
figma_img = Image.open(figma_path)
website_img = Image.open(website_path)
# Analyze different aspects
differences.extend(self._analyze_layout(figma_img, website_img, viewport))
differences.extend(self._analyze_colors(figma_img, website_img, viewport))
differences.extend(self._analyze_structure(figma_img, website_img, viewport))
except Exception as e:
print(f"Error comparing screenshots: {str(e)}")
return differences
def _analyze_layout(self, figma_img: Image.Image, website_img: Image.Image, viewport: str) -> List[VisualDifference]:
"""Analyze layout differences."""
differences = []
figma_size = figma_img.size
website_size = website_img.size
# Check width differences
if figma_size[0] != website_size[0]:
width_diff_percent = abs(figma_size[0] - website_size[0]) / figma_size[0] * 100
severity = "High" if width_diff_percent > 10 else "Medium"
diff = VisualDifference(
category="layout",
severity=severity,
issue_id="1.1",
title="Container width differs",
description=f"Design: {figma_size[0]}px vs Website: {website_size[0]}px ({width_diff_percent:.1f}% difference)",
design_value=str(figma_size[0]),
website_value=str(website_size[0]),
viewport=viewport,
confidence=0.95,
detection_method="screenshot_comparison"
)
differences.append(diff)
# Check height differences
if figma_size[1] != website_size[1]:
height_diff_percent = abs(figma_size[1] - website_size[1]) / figma_size[1] * 100
severity = "High" if height_diff_percent > 10 else "Medium"
diff = VisualDifference(
category="layout",
severity=severity,
issue_id="1.2",
title="Page height differs",
description=f"Design: {figma_size[1]}px vs Website: {website_size[1]}px ({height_diff_percent:.1f}% difference)",
design_value=str(figma_size[1]),
website_value=str(website_size[1]),
viewport=viewport,
confidence=0.95,
detection_method="screenshot_comparison"
)
differences.append(diff)
return differences
def _analyze_colors(self, figma_img: Image.Image, website_img: Image.Image, viewport: str) -> List[VisualDifference]:
"""Analyze color differences."""
differences = []
try:
# Convert to RGB
figma_rgb = figma_img.convert('RGB')
website_rgb = website_img.convert('RGB')
# Resize to same size for comparison (use smaller size for performance)
compare_size = (400, 300)
figma_resized = figma_rgb.resize(compare_size)
website_resized = website_rgb.resize(compare_size)
# Convert to numpy arrays
figma_array = np.array(figma_resized, dtype=np.float32)
website_array = np.array(website_resized, dtype=np.float32)
# Calculate color difference (mean absolute difference)
color_diff = np.mean(np.abs(figma_array - website_array))
# If significant color difference, flag it
if color_diff > 15:
severity = "High" if color_diff > 40 else "Medium"
diff = VisualDifference(
category="colors",
severity=severity,
issue_id="3.1",
title="Color scheme differs significantly",
description=f"Significant color difference detected (delta: {color_diff:.1f})",
design_value="Design colors",
website_value="Website colors",
viewport=viewport,
confidence=0.8,
detection_method="pixel_analysis"
)
differences.append(diff)
except Exception as e:
pass
return differences
def _analyze_structure(self, figma_img: Image.Image, website_img: Image.Image, viewport: str) -> List[VisualDifference]:
"""Analyze structural/layout differences."""
differences = []
try:
# Convert to grayscale for edge detection
figma_gray = figma_img.convert('L')
website_gray = website_img.convert('L')
# Resize to same size
compare_size = (400, 300)
figma_resized = figma_gray.resize(compare_size)
website_resized = website_gray.resize(compare_size)
# Convert to numpy arrays
figma_array = np.array(figma_resized, dtype=np.float32)
website_array = np.array(website_resized, dtype=np.float32)
# Calculate structural difference (MSE)
mse = np.mean((figma_array - website_array) ** 2)
# Normalize MSE to 0-100 scale
structural_diff = min(100, mse / 255)
if structural_diff > 10:
severity = "High" if structural_diff > 30 else "Medium"
diff = VisualDifference(
category="layout",
severity=severity,
issue_id="1.3",
title="Layout structure differs",
description=f"Visual structure difference detected (score: {structural_diff:.1f})",
design_value="Design layout",
website_value="Website layout",
viewport=viewport,
confidence=0.75,
detection_method="structural_analysis"
)
differences.append(diff)
except Exception as e:
pass
return differences
class DifferenceAnalyzer:
"""
Agent 3: Difference Analyzer
Analyzes visual differences between Figma designs and website implementations
"""
def __init__(self):
"""Initialize the analyzer."""
self.comparator = ScreenshotComparator()
def analyze_differences(self, state: WorkflowState) -> WorkflowState:
"""
Analyze visual differences between Figma and website screenshots.
Args:
state: Current workflow state
Returns:
Updated state with analysis results
"""
print("\n🔍 Agent 3: Difference Analyzer - Analyzing Visual Differences...")
try:
all_differences = []
# Compare screenshots for each viewport
for viewport in ["desktop", "mobile"]:
figma_key = f"{viewport}"
website_key = f"{viewport}"
figma_path = state.get("figma_screenshots", {}).get(figma_key)
website_path = state.get("website_screenshots", {}).get(website_key)
if figma_path and website_path:
print(f" 📊 Comparing {viewport} screenshots...")
differences = self.comparator.compare_screenshots(
figma_path,
website_path,
viewport
)
all_differences.extend(differences)
print(f" ✓ Found {len(differences)} differences")
else:
print(f" ⚠️ Missing screenshots for {viewport}")
# Calculate similarity score
total_differences = len(all_differences)
high_severity = len([d for d in all_differences if d.severity == "High"])
medium_severity = len([d for d in all_differences if d.severity == "Medium"])
low_severity = len([d for d in all_differences if d.severity == "Low"])
# Similarity score: 100 - (differences weighted by severity)
severity_weight = (high_severity * 10) + (medium_severity * 5) + (low_severity * 1)
similarity_score = max(0, 100 - severity_weight)
state["visual_differences"] = [d.to_dict() if hasattr(d, "to_dict") else d for d in all_differences]
state["similarity_score"] = similarity_score
state["status"] = "analysis_complete"
print(f"\n 📈 Analysis Summary:")
print(f" - Total differences: {total_differences}")
print(f" - High severity: {high_severity}")
print(f" - Medium severity: {medium_severity}")
print(f" - Low severity: {low_severity}")
print(f" - Similarity score: {similarity_score:.1f}/100")
return state
except Exception as e:
print(f" ❌ Error analyzing differences: {str(e)}")
import traceback
traceback.print_exc()
state["status"] = "analysis_failed"
state["error_message"] = f"Agent 3 Error: {str(e)}"
return state
def agent_3_node(state: Dict) -> Dict:
"""
LangGraph node for Agent 3 (Difference Analyzer).
Args:
state: Current workflow state
Returns:
Updated state
"""
# Convert dict to WorkflowState if needed
if isinstance(state, dict):
workflow_state = WorkflowState(**state)
else:
workflow_state = state
# Create analyzer and analyze differences
analyzer = DifferenceAnalyzer()
updated_state = analyzer.analyze_differences(workflow_state)
# Convert back to dict for LangGraph
return updated_state.__dict__
|