Config Module (`stx.config`)

YAML-based configuration with priority resolution and path management.

Priority Resolution

All config values follow the same precedence:

direct argument > config/YAML > environment variable > default

Quick Reference

import scitex as stx

# Get global config (reads ~/.scitex/config.yaml)
config = stx.config.get_config()

# Resolve with precedence
log_level = config.resolve("logging.level", default="INFO")
# Checks: direct → YAML → SCITEX_LOGGING_LEVEL env → "INFO"

# Access nested values
config.get_nested("scholar", "crossref_email")

# Path management
paths = stx.config.get_paths()
paths.scholar        # ~/.scitex/scholar
paths.cache          # ~/.scitex/cache

YAML Configuration

# ~/.scitex/config.yaml (or ./config/*.yaml for projects)
logging:
  level: INFO

scholar:
  crossref_email: ${CROSSREF_EMAIL:-user@example.com}

Environment variable substitution with ${VAR:-default} syntax is supported in YAML files.

Key Classes

ScitexConfig – YAML-based config with env var substitution and precedence resolution
PriorityConfig – Dict-based config resolver for programmatic use
ScitexPaths – Centralized path manager (all paths derive from ~/.scitex/)
EnvVar – Environment variable definition dataclass

Path Management

ScitexPaths manages all SciTeX directories:

paths = stx.config.get_paths()
paths.base       # ~/.scitex (SCITEX_DIR)
paths.logs       # ~/.scitex/logs
paths.cache      # ~/.scitex/cache
paths.scholar    # ~/.scitex/scholar
paths.rng        # ~/.scitex/rng
paths.capture    # ~/.scitex/capture

Utility Functions

get_config(path) – Get singleton ScitexConfig instance
get_paths(base_dir) – Get singleton ScitexPaths instance
load_yaml(path) – Load YAML with env var substitution
load_dotenv(path) – Load .env file
get_scitex_dir() – Get SCITEX_DIR with precedence

API Reference

See scitex.config API Reference for the auto-generated Python API.

Config Module (stx.config)