config.py

from __future__ import annotations

from dataclasses import dataclass
import os
from typing import Tuple

DEFAULT_METRICS: Tuple[str, ...] = (
    "F1",
    "Precision",
    "Recall",
    "Mean_IoU",
    "Mean_TP_IoU",
)


def _parse_bool(value: str | None, default: bool) -> bool:
    if value is None:
        return default
    normalized = value.strip().lower()
    if normalized in {"1", "true", "yes", "y", "on"}:
        return True
    if normalized in {"0", "false", "no", "n", "off"}:
        return False
    return default


def _parse_csv(value: str | None) -> tuple[str, ...]:
    if not value:
        return tuple()
    items = [item.strip() for item in value.split(",")]
    return tuple(item for item in items if item)


@dataclass(frozen=True)
class AppConfig:
    hf_token: str | None
    hf_owner: str | None
    repo_allowlist: tuple[str, ...]
    repo_prefixes: tuple[str, ...]
    repo_id_contains: str
    discovery_enabled: bool
    max_repos: int
    include_history: bool
    max_commits_per_repo: int
    metrics_section_heading: str
    cache_ttl_seconds: int
    auto_refresh_seconds: int
    max_load_seconds: int
    metrics: tuple[str, ...]



def load_config() -> AppConfig:
    return AppConfig(
        hf_token=os.getenv("HF_TOKEN") or None,
        hf_owner=os.getenv("HF_OWNER") or None,
        repo_allowlist=_parse_csv(os.getenv("EVAL_REPO_ALLOWLIST")),
        repo_prefixes=_parse_csv(os.getenv("EVAL_REPO_PREFIXES")),
        repo_id_contains=os.getenv("EVAL_REPO_ID_CONTAINS", "eval").strip(),
        discovery_enabled=_parse_bool(os.getenv("EVAL_DISCOVERY_ENABLED"), True),
        max_repos=int(os.getenv("EVAL_MAX_REPOS", "200")),
        include_history=_parse_bool(os.getenv("EVAL_INCLUDE_HISTORY"), True),
        max_commits_per_repo=int(os.getenv("EVAL_MAX_COMMITS_PER_REPO", "20")),
        metrics_section_heading=os.getenv("EVAL_METRICS_SECTION_HEADING", "Performance Metrics").strip(),
        cache_ttl_seconds=int(os.getenv("EVAL_CACHE_TTL_SECONDS", "600")),
        auto_refresh_seconds=int(os.getenv("EVAL_AUTO_REFRESH_SECONDS", "300")),
        max_load_seconds=int(os.getenv("EVAL_MAX_LOAD_SECONDS", "90")),
        metrics=DEFAULT_METRICS,
    )