"""Write one PDF per case from a recovered ``full_text.jsonl``. For jurisdictions whose judgment text was recovered into ``data//full_text.jsonl`` (Taiwan/Hong Kong via rendering, Serbia via scraping, India via the real PDF bundle) rather than living in the workbook's ``full_text`` column, this produces the original-language PDFs that the Harvey hand-off needs. Font is chosen per country code (Traditional Chinese for tw/hk, Devanagari for in, etc.). """ from __future__ import annotations import json from pathlib import Path from legex.pdf_export.core import write_one_pdf from legex.pdf_export.font_keys import FONT_BY_CC def export_jsonl( jsonl_path: Path, out_dir: Path, *, cc: str, min_chars: int = 1, limit: int | None = None, ) -> int: """Write PDFs from ``jsonl_path`` into ``out_dir``. Returns count written.""" font_key = FONT_BY_CC.get(cc, "default") out_dir.mkdir(parents=True, exist_ok=True) used_names: dict[str, int] = {} count = 0 with jsonl_path.open(encoding="utf-8") as f: for line in f: line = line.strip() if not line: continue try: rec = json.loads(line) except json.JSONDecodeError: continue case_id = str(rec.get("case_id") or "").strip() body = str(rec.get("full_text") or "") if not case_id or len(body) < min_chars: continue base = case_id or f"row_{count}" used_names[base] = used_names.get(base, 0) + 1 fname = f"{base}_{used_names[base]}.pdf" if used_names[base] > 1 else f"{base}.pdf" write_one_pdf( out_dir / fname, case_id, "", body, font_key, text_source="full_text.jsonl", ) count += 1 if count % 25 == 0: print(f" {cc}: {count} …", flush=True) if limit is not None and count >= limit: break print(f" {cc}: done, {count} PDFs -> {out_dir}", flush=True) return count