bootstrap.py

from pathlib import Path
import json
import os
from huggingface_hub import hf_hub_download

DATASET_REPO = "noodbox/sv_territorial_master_dataset_v1"
BOOTSTRAP_FLAG = Path(".bootstrap_complete")
RUNTIME_DIR = Path("runtime_data")
MANIFEST_PATH = RUNTIME_DIR / "manifest.json"

REQUIRED_FILES = [
    "data/geo/sv_geo_44.geojson",
    "data/geo/sv_geo_distritos.geojson",
    "data/territorial/territories_master.csv",
    "data/territorial/capitals.csv",
    "data/socioeconomic/indicators.csv",
    "data/electoral/elections_results.csv",
    "data/political/parties.csv",
    "data/political/candidates.csv",
    "data/game/campaign_effects.csv",
]


def run_bootstrap(force: bool = False):
    RUNTIME_DIR.mkdir(exist_ok=True)

    if MANIFEST_PATH.exists() and BOOTSTRAP_FLAG.exists() and not force:
        return

    token = os.getenv("HF_TOKEN")
    downloaded = {}

    for file_path in REQUIRED_FILES:
        local_path = hf_hub_download(
            repo_id=DATASET_REPO,
            repo_type="dataset",
            filename=file_path,
            token=token,
        )
        downloaded[file_path] = local_path

    manifest = {
        "dataset_repo": DATASET_REPO,
        "files": downloaded,
        "status": "ready",
    }

    with open(MANIFEST_PATH, "w", encoding="utf-8") as f:
        json.dump(manifest, f, ensure_ascii=False, indent=2)

    BOOTSTRAP_FLAG.write_text("ok", encoding="utf-8")