#!/usr/bin/env python3
"""
Example: Using CanvasXpress Generator as a Python Tool
This shows how to import and use the generator in your own Python scripts.
"""
import os
import json
import sys
# Add src to path if running from project root
sys.path.insert(0, 'src')
from canvasxpress_generator import CanvasXpressGenerator
def example_1_basic_usage():
"""Example 1: Basic usage - generate a simple chart"""
print("=" * 60)
print("Example 1: Basic Bar Chart")
print("=" * 60)
# Initialize the generator (only do this ONCE, reuse the instance)
generator = CanvasXpressGenerator(
data_dir='data',
vector_db_path='/root/.cache/canvasxpress_mcp.db',
llm_model=os.environ.get('LLM_MODEL', 'gpt-4o-global'),
llm_environment=os.environ.get('LLM_ENVIRONMENT', 'nonprod')
)
# Generate a configuration
config = generator.generate(
description="Create a bar chart with blue bars and legend on the right",
headers="Region, Sales, Profit"
)
print("\n📊 Generated Configuration:")
print(json.dumps(config, indent=2))
return config
def example_2_multiple_charts():
"""Example 2: Generate multiple charts (reuse generator instance)"""
print("\n" + "=" * 60)
print("Example 2: Multiple Charts (Efficient)")
print("=" * 60)
# Initialize ONCE
generator = CanvasXpressGenerator(
data_dir='data',
vector_db_path='/root/.cache/canvasxpress_mcp.db',
llm_model='gpt-4o-global',
llm_environment='nonprod'
)
# Generate multiple configs
charts = [
{
"name": "Scatter Plot",
"description": "Scatter plot with red points, x-axis is time, y-axis is expression",
"headers": "Time, Expression, Gene"
},
{
"name": "Heatmap",
"description": "Heatmap with clustering, show dendrograms on both sides",
"headers": "Gene1, Gene2, Gene3, Sample"
},
{
"name": "Line Chart",
"description": "Line chart with multiple series, show all in different colors",
"headers": "Date, Series1, Series2, Series3"
}
]
results = []
for chart in charts:
print(f"\n🎨 Generating: {chart['name']}")
config = generator.generate(
description=chart['description'],
headers=chart['headers']
)
results.append({
"name": chart['name'],
"config": config
})
print(f" ✓ Generated {chart['name']}")
return results
def example_3_with_error_handling():
"""Example 3: With proper error handling"""
print("\n" + "=" * 60)
print("Example 3: With Error Handling")
print("=" * 60)
try:
generator = CanvasXpressGenerator(
data_dir='data',
vector_db_path='/root/.cache/canvasxpress_mcp.db',
llm_model='gpt-4o-global',
llm_environment='nonprod'
)
config = generator.generate(
description="Pie chart showing market share by company",
headers="Company, MarketShare",
temperature=0.0 # Deterministic output
)
print("\n✅ Success!")
print(json.dumps(config, indent=2))
return config
except json.JSONDecodeError as e:
print(f"\n❌ JSON Error: LLM returned invalid JSON")
print(f" Error: {e}")
return None
except RuntimeError as e:
print(f"\n❌ Runtime Error: {e}")
print(f" Check your API key and network connection")
return None
except Exception as e:
print(f"\n❌ Unexpected Error: {e}")
return None
def example_4_inspect_similar_examples():
"""Example 4: See what examples the RAG system finds"""
print("\n" + "=" * 60)
print("Example 4: Inspect RAG Retrieval")
print("=" * 60)
generator = CanvasXpressGenerator(
data_dir='data',
vector_db_path='/root/.cache/canvasxpress_mcp.db',
llm_model='gpt-4o-global',
llm_environment='nonprod'
)
# Get similar examples without generating
description = "Boxplot showing gene expression across samples"
similar = generator.get_similar_examples(description, num_examples=5)
print(f"\n🔍 Top 5 similar examples for: '{description}'\n")
for i, ex in enumerate(similar, 1):
print(f"{i}. Type: {ex['type']}")
print(f" Description: {ex['description'][:80]}...")
print(f" Similarity Score: {ex['score']:.4f}")
print()
return similar
def example_5_save_to_file():
"""Example 5: Generate and save config to file"""
print("\n" + "=" * 60)
print("Example 5: Generate and Save to File")
print("=" * 60)
generator = CanvasXpressGenerator(
data_dir='data',
vector_db_path='/root/.cache/canvasxpress_mcp.db',
llm_model='gpt-4o-global',
llm_environment='nonprod'
)
config = generator.generate(
description="Network graph with curved edges and labeled nodes",
headers="Source, Target, Weight"
)
# Save to file
output_file = "my_canvasxpress_config.json"
with open(output_file, 'w') as f:
json.dump(config, f, indent=2)
print(f"\n✅ Configuration saved to: {output_file}")
print(f"\n📝 Use in CanvasXpress:")
print(f" new CanvasXpress(data, {json.dumps(config)})")
return config
def main():
"""Run all examples"""
# Check environment
if not os.environ.get('AZURE_OPENAI_KEY'):
print("❌ Error: AZURE_OPENAI_KEY not set")
print("\nSet it with:")
print(" export AZURE_OPENAI_KEY=your_key_here")
print(" export LLM_MODEL=gpt-4o-global")
print(" export LLM_ENVIRONMENT=nonprod")
return 1
print("🚀 CanvasXpress Generator - Python Tool Examples")
print("=" * 60)
# Run examples
try:
# Example 1: Basic usage
example_1_basic_usage()
# Example 2: Multiple charts (comment out if you want to save API calls)
# example_2_multiple_charts()
# Example 3: Error handling
# example_3_with_error_handling()
# Example 4: Inspect RAG (no API call)
example_4_inspect_similar_examples()
# Example 5: Save to file
# example_5_save_to_file()
print("\n" + "=" * 60)
print("✅ Examples complete!")
print("=" * 60)
except KeyboardInterrupt:
print("\n\n⚠️ Interrupted by user")
return 1
return 0
if __name__ == "__main__":
sys.exit(main())
