from pathlib import Path


TEXT_SUFFIXES = {".py", ".md", ".txt", ".json", ".toml", ".yaml", ".yml"}
SKIP_DIRS = {".cache", ".git", ".venv", "__pycache__", "runtime", "finetuning", "hf_upload"}


def collect_project_context(root: str | Path, max_chars: int) -> str:
    base = Path(root).resolve()
    if not base.exists():
        raise FileNotFoundError(f"Context root does not exist: {base}")

    chunks: list[str] = []
    used = 0
    for path in sorted(base.rglob("*")):
        if path.is_dir() or path.suffix.lower() not in TEXT_SUFFIXES:
            continue
        if any(part in SKIP_DIRS for part in path.parts):
            continue
        rel = path.relative_to(base)
        try:
            text = path.read_text(encoding="utf-8", errors="ignore")
        except OSError:
            continue
        block = f"\n### {rel}\n{text[:4000]}\n"
        if used + len(block) > max_chars:
            remaining = max_chars - used
            if remaining > 200:
                chunks.append(block[:remaining])
            break
        chunks.append(block)
        used += len(block)
    return "".join(chunks)