28 lines
855 B
YAML
28 lines
855 B
YAML
# General Scraper Settings
|
|
debug:
|
|
save_html_js: true # Set to true for debugging, false for production
|
|
headless_mode:
|
|
js: false # Set to false to run Playwright in visible mode for JS debugging
|
|
html: false # Not used yet, but we can use Playwright for HTML later if needed
|
|
|
|
|
|
|
|
scraper:
|
|
user_agent: "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/120.0.0.0 Safari/537.36"
|
|
request_timeout: 10 # Timeout in seconds
|
|
retry_attempts: 3 # Number of retries on failure
|
|
|
|
# Database Settings
|
|
database:
|
|
data_path: "data/data.db"
|
|
logs_path: "data/logs.db"
|
|
|
|
# API Keys (Optional - Can be filled later)
|
|
api_keys:
|
|
twitter: "YOUR_TWITTER_API_KEY"
|
|
discord: "YOUR_DISCORD_BOT_TOKEN"
|
|
|
|
# Logging Settings
|
|
logging:
|
|
level: "INFO" # Options: DEBUG, INFO, WARNING, ERROR
|
|
file: "data/scraper.log"
|