style_code_lisad/config_loader_reference.py

"""Centralized gin-config loader for multi-config pipelines.

Reference implementation of the recommended pattern:
- One function `load_all_configs()` parses ALL .gin files at once
- `gin.clear_config()` resets global state before loading
- Individual `get_*_cfg()` kept ONLY for isolated testing

This file serves as a TEMPLATE — copy and adapt for each project.
"""

from __future__ import annotations

import logging
from pathlib import Path
from typing import Any

import gin

logger = logging.getLogger(__name__)


# ---------------------------------------------------------------------------
# Import config classes (adapt to your project)
# ---------------------------------------------------------------------------
# from conf.pipeline_conf import PipelineConfig
# from conf.hardware_conf import HardwareConfig
# from conf.models_conf import ModelsConfig
# from conf.input_conf import InputConfig
# from conf.seg_conf import SegConfig


# ---------------------------------------------------------------------------
# Central loader (PRODUCTION USE)
# ---------------------------------------------------------------------------

def load_all_configs(path2cfg: str) -> dict[str, Any]:
    """Parse ALL .gin files at once and instantiate all config objects.

    This is the **only** function that should call gin.parse in production.
    It clears gin global state first, then loads all .gin files in sorted
    order, then creates config instances from the fully-populated state.

    Args:
        path2cfg: Path to config directory (WITH trailing slash).
            Example: ``"/home/servml/project/in/config_files/"``

    Returns:
        Dictionary with config objects keyed by short name::

            {
                "pipeline": PipelineConfig(),
                "hardware": HardwareConfig(),
                "models":   ModelsConfig(),
                "input":    InputConfig(),
                "seg":      SegConfig(),
            }

    Raises:
        FileNotFoundError: If path2cfg does not exist or contains no .gin files.

    Example::

        configs = load_all_configs("/home/servml/project/in/config_files/")
        run_pipeline(configs["pipeline"], configs["hardware"], ...)
    """
    cfg_dir = Path(path2cfg)
    if not cfg_dir.is_dir():
        raise FileNotFoundError(f"Config directory not found: {cfg_dir}")

    gin_files = sorted(cfg_dir.glob("*.gin"))
    if not gin_files:
        raise FileNotFoundError(f"No .gin files in {cfg_dir}")

    # CRITICAL: clear global gin state before loading.
    # Without this, parameters from previous calls accumulate silently.
    gin.clear_config()

    # Parse ALL .gin files in one call — order is alphabetical (sorted).
    gin.parse_config_files_and_bindings(
        config_files=[str(f) for f in gin_files],
        bindings=[],  # No CLI overrides; add if needed.
    )
    logger.info("Loaded %d gin files from %s", len(gin_files), cfg_dir)

    # Instantiate all configs from the fully-populated gin state.
    # Each class picks its parameters from gin via @gin.configurable.
    configs = {
        "pipeline": PipelineConfig(),
        "hardware": HardwareConfig(),
        "models": ModelsConfig(),
        "input": InputConfig(),
        "seg": SegConfig(),
    }

    logger.info("Created %d config objects: %s", len(configs), list(configs.keys()))
    return configs


# ---------------------------------------------------------------------------
# Individual loaders (TESTING / DEBUGGING ONLY)
# ---------------------------------------------------------------------------
# Use these when you need a single config in isolation (unit tests, notebooks).
# ALWAYS call gin.clear_config() first to avoid state leaks.

def get_pipeline_cfg(path2cfg: str) -> Any:  # -> PipelineConfig
    """Load ONLY pipeline config (for isolated testing)."""
    gin.clear_config()
    gin.parse_config_file(f"{path2cfg}pipeline.gin")
    return PipelineConfig()


def get_hardware_cfg(path2cfg: str) -> Any:  # -> HardwareConfig
    """Load ONLY hardware config (for isolated testing)."""
    gin.clear_config()
    gin.parse_config_file(f"{path2cfg}hardware.gin")
    return HardwareConfig()


def get_models_cfg(path2cfg: str) -> Any:  # -> ModelsConfig
    """Load ONLY models config (for isolated testing)."""
    gin.clear_config()
    gin.parse_config_file(f"{path2cfg}models.gin")
    return ModelsConfig()


def get_input_cfg(path2cfg: str) -> Any:  # -> InputConfig
    """Load ONLY input config (for isolated testing)."""
    gin.clear_config()
    gin.parse_config_file(f"{path2cfg}input.gin")
    return InputConfig()


def get_seg_cfg(path2cfg: str) -> Any:  # -> SegConfig
    """Load ONLY segmentation config (for isolated testing)."""
    gin.clear_config()
    gin.parse_config_file(f"{path2cfg}segmentation.gin")
    return SegConfig()


# ---------------------------------------------------------------------------
# Project root discovery
# ---------------------------------------------------------------------------

# Markers that indicate the project root directory.
# Checked in order: first match wins.
_ROOT_MARKERS = ("pyproject.toml", ".git", "in")


def get_proj_dir() -> str:
    """Return project root directory with trailing slash.

    Walks up from this file's location until a directory containing
    one of the root markers is found.  More robust than the legacy
    ``split('src')[0]`` approach which breaks when ``src`` appears
    elsewhere in the path (e.g. ``/home/user/resources/project/src/``).

    Root markers (checked in order): ``pyproject.toml``, ``.git``, ``in/``.

    Returns:
        Absolute path to project root with trailing ``/``.

    Raises:
        RuntimeError: If no marker is found within 10 parent levels.
    """
    current = Path(__file__).resolve().parent
    for _ in range(10):
        if any((current / marker).exists() for marker in _ROOT_MARKERS):
            return str(current) + "/"
        current = current.parent
    raise RuntimeError(
        f"Cannot find project root: none of {_ROOT_MARKERS} found "
        f"in ancestors of {Path(__file__).resolve()}"
    )


def main() -> None:
    """Template entry point: load all configs -> run pipeline."""
    proj_dir = get_proj_dir()
    path2cfg = f"{proj_dir}in/config_files/"

    configs = load_all_configs(path2cfg)

    # Unpack and pass explicitly to pipeline:
    # run_pipeline(
    #     configs["pipeline"],
    #     configs["hardware"],
    #     configs["models"],
    #     configs["input"],
    #     configs["seg"],
    # )

    # Print loaded configs for verification:
    for name, cfg in configs.items():
        print(f"\n[{name}]")
        for k, v in vars(cfg).items():
            print(f"  {k} = {v}")


if __name__ == "__main__":
    main()