OCR-MCP

""" OCR-MCP Configuration Management """ import os from pathlib import Path from typing import Optional from pydantic import BaseModel, Field class OCRConfig(BaseModel): """Configuration for OCR-MCP server.""" # Cache and model directories cache_dir: Path = Field(default_factory=lambda: Path.home() / ".cache" / "ocr-mcp") model_cache_dir: Path = Field(default_factory=lambda: Path.home() / ".cache" / "ocr-mcp" / "models") # Device configuration device: str = Field(default="auto") # "auto", "cuda", "cpu" max_memory_gb: Optional[float] = Field(default=None) # Default backend settings default_backend: str = Field(default="auto") # "auto", "got-ocr", "tesseract", etc. # Processing settings batch_size: int = Field(default=4) max_concurrent_jobs: int = Field(default=4) # Backend-specific settings got_ocr_model_size: str = Field(default="base") # "base" or "large" tesseract_languages: str = Field(default="eng") easyocr_languages: list = Field(default_factory=lambda: ["en"]) def __init__(self, **data): # Set default values first default_cache_dir = Path.home() / ".cache" / "ocr-mcp" # Override defaults with environment variables data.setdefault("cache_dir", Path(os.getenv("OCR_CACHE_DIR", str(default_cache_dir)))) data.setdefault("device", os.getenv("OCR_DEVICE", "auto")) data.setdefault("max_memory_gb", float(os.getenv("OCR_MAX_MEMORY", 0)) if os.getenv("OCR_MAX_MEMORY") else None) data.setdefault("default_backend", os.getenv("OCR_DEFAULT_BACKEND", "auto")) data.setdefault("batch_size", int(os.getenv("OCR_BATCH_SIZE", 4))) data.setdefault("max_concurrent_jobs", int(os.getenv("OCR_MAX_CONCURRENT", 4))) super().__init__(**data) # Ensure directories exist self.cache_dir.mkdir(parents=True, exist_ok=True) self.model_cache_dir.mkdir(parents=True, exist_ok=True) # Global config instance config = OCRConfig()