# High Quality Configuration
# Optimized for accuracy and quality
server:
name: document-parser-mcp
version: 1.0.0
max_concurrent_jobs: 2
job_timeout_seconds: 1200
storage:
temp_directory: ./temp
cleanup_interval_hours: 48
max_file_size_mb: 500
allowed_schemes: [http, https, ftp]
download_timeout_seconds: 900
processing:
default_pipeline: standard
enable_pipeline_auto_detect: true
ocr:
engine: tesseract # More accurate OCR
languages: [eng, spa, fra] # Multiple languages
enable_auto_detect: true
pdf:
backend: dlparse_v4 # Highest quality backend
fallback_backend: pypdfium2
enable_table_extraction: true
table_accuracy_mode: accurate # Accurate mode
performance:
max_memory_gb: 8
enable_mlx_acceleration: true
thread_count: 4
logging:
level: INFO
format_string: '%(asctime)s - %(name)s - %(levelname)s - %(message)s'
file_path: ./logs/server.log
max_file_size_mb: 20
backup_count: 10
enable_json_logs: true
retry:
enable_pipeline_fallback: true # Enable fallbacks
enable_backend_fallback: true
max_attempts: 5 # More retries
delay_seconds: 3
