Error Handling Reference

Error Handling Reference

Comprehensive error handling guide for the Market Sentiment Analyzer.

API Errors

Fear & Greed Index Unavailable

Error: Cannot fetch Fear & Greed Index from Alternative.me API

Symptoms:

Fear & Greed Index unavailable
Using cached/fallback value

Causes:

• API rate limiting (>10 requests/minute)
• API server downtime
• Network connectivity issues

Solutions:

1. Automatic: Uses cached value with "stale" warning
2. Manual: Wait 1-2 minutes and retry
3. Persistent: Check https://alternative.me/crypto/fear-and-greed-index/ status

Impact: Composite score uses cached or neutral (50) value

CoinGecko API Rate Limit

Error: Market momentum data unavailable

Symptoms:

API request failed: 429 Too Many Requests
Market data unavailable

Causes:

• Exceeded free tier limit (~10-30 calls/minute)
• Too many concurrent requests

Solutions:

1. Automatic: Falls back to cached data
2. Manual: Wait 60 seconds before retry
3. Upgrade: CoinGecko Pro API for higher limits

Impact: Momentum component uses neutral (50) value

News Feed Unavailable

Error: Cannot fetch news from RSS feeds

Symptoms:

Failed to fetch [feed_url]: Connection timed out
No articles available

Causes:

• RSS feed URL changed
• Feed server downtime
• Network firewall blocking

Solutions:

1. Automatic: Skips unavailable feeds, uses others
2. Check: Verify feed URLs in settings
3. Network: Test connectivity to feed domains

Impact: News sentiment based on fewer articles; score may be less representative

Data Parsing Errors

Invalid JSON Response

Error: Failed to parse API response

Symptoms:

json.JSONDecodeError: Expecting value
Failed to parse response: [error details]

Causes:

• API returning HTML error page instead of JSON
• Malformed API response
• Incomplete response due to timeout

Solutions:

1. Check API status pages
2. Verify API endpoint URLs
3. Increase timeout in settings

Missing Data Fields

Error: Expected field not in response

Symptoms:

KeyError: 'price_change_percentage_24h'

Causes:

• API schema changed
• New coin not fully supported
• Delisted coin

Solutions:

1. Update to latest script version
2. Use market-wide analysis instead of coin-specific
3. Report issue if persistent

Configuration Errors

Invalid Weights

Error: Custom weights don't sum to 1.0

Note: This is auto-corrected, not a fatal error

Causes:

• User-provided weights like "news:0.5,fng:0.5,momentum:0.5" (sum > 1)

Solutions:

1. Automatic: Weights are normalized
2. Manual: Ensure weights sum to 1.0

Unknown Coin Symbol

Error: Coin not found in known mappings

Symptoms:

No data for [COIN], using market analysis

Causes:

• Coin symbol not in known mappings
• New coin not yet added
• Typo in symbol

Solutions:

1. Use common symbols (BTC, ETH, SOL, etc.)
2. Try full CoinGecko ID instead of symbol
3. Fall back to market-wide analysis

Graceful Degradation

The analyzer implements graceful degradation:

Full Analysis (all 3 components)
     │
     ├─► Fear & Greed fails → Use cached/neutral (50)
     │
     ├─► News fails → Use remaining components with adjusted weights
     │
     ├─► Momentum fails → Use remaining components with adjusted weights
     │
     └─► All fail → Return neutral (50) with error warnings

Component Failure Handling

Diagnostic Commands

Test Individual Components

# Test Fear & Greed
python fear_greed.py -v

# Test News Sentiment
python news_sentiment.py -v

# Test Market Momentum
python market_momentum.py -v

# Full analysis with verbose
python sentiment_analyzer.py -v

Check Cache Files

# List cache files
ls -la scripts/.*.json

# View cached data
cat scripts/.fng_cache.json | python -m json.tool

Clear Caches

rm scripts/.fng_cache.json
rm scripts/.news_cache.json
rm scripts/.momentum_cache.json

Common Issues

Issue: Score Always Neutral (50)

Causes:

• All APIs unavailable
• Network connectivity issues
• VPN blocking API access

Diagnosis:

# Run with verbose
python sentiment_analyzer.py -v

# Check network
curl -I https://api.alternative.me/fng/
curl -I https://api.coingecko.com/api/v3/ping

Issue: News Sentiment Skewed

Causes:

• Only one RSS feed responding
• Feed returns old articles
• Keyword bias in recent news

Diagnosis:

python news_sentiment.py -v
# Check "articles_analyzed" count
# Should be 15-50 for balanced sentiment

Issue: Slow Response Time

Causes:

• Cache expired, fetching fresh data
• Network latency to APIs
• API rate limiting delays

Solutions:

1. Increase cache TTL in settings
2. Use --verbose to see timing
3. Consider local caching proxy