#!/usr/bin/env python3 """ migrate-tasklist.py — One-shot: convert _rjj/tasklist/INDEX.md table rows to individual .md files with YAML frontmatter under .meshkore/modules//{tasks,log}/ (standard §2 canonical layout). Each row in INDEX.md becomes a single .md file: - DONE rows → .meshkore/modules//log/{YYYY-MM}/-.md (status: done) - NEXT → .meshkore/modules//tasks/-.md (status: next) - UPCOMING → .meshkore/modules//tasks/-.md (status: next) - PLANNED → .meshkore/modules//tasks/-.md (status: backlog) - FUTURE → .meshkore/modules//tasks/-.md (status: backlog) - ECOSYSTEM → .meshkore/modules/ecosystem/tasks/-.md (status: backlog) Continuation rows (`| | text | | | |`) are appended as the body of the task. Usage: python3 migrate-tasklist.py [--dry-run] The argument is the `.meshkore/` directory of the target repo. Module folders are created under `/modules//`. """ from __future__ import annotations import argparse import re import sys from datetime import date from pathlib import Path ID_RE = re.compile(r"^\|\s*(?PD\d+|N\d+|U\d+|P\d+|F\d+|E\d+)\s*\|", re.IGNORECASE) SECTION_RE = re.compile(r"^##\s+(?PDONE|NEXT|UPCOMING|PLANNED|FUTURE|ECOSYSTEM)", re.IGNORECASE) STATUS_BY_SECTION = { "done": "done", "next": "next", "upcoming": "next", "planned": "backlog", "future": "backlog", "ecosystem": "backlog", } def slugify(text: str) -> str: t = text.lower() t = re.sub(r"[^a-z0-9\s-]", "", t) t = re.sub(r"\s+", "-", t) t = re.sub(r"-+", "-", t) return t.strip("-")[:60] or "task" def parse_row(line: str) -> list[str] | None: """Parse a markdown table row into cells. Returns None if not a row.""" line = line.rstrip("\n") if not line.startswith("|"): return None if "---" in line: return None # Split on | but keep cells parts = line.split("|") # First and last are empty cells = [c.strip() for c in parts[1:-1]] if len(parts) >= 3 else [] return cells def is_continuation(cells: list[str]) -> bool: return cells and cells[0] == "" def section_to_folder(section: str, module: str, when: str | None = None) -> tuple[str, str]: """Returns (subdir-path relative to .meshkore/, status). Output follows the standard §2 canonical layout: modules//{tasks,log}/, NOT the pre-v3 roadmap/{tasks,log}/ layout.""" section = section.lower() status = STATUS_BY_SECTION[section] if section == "ecosystem": return "modules/ecosystem/tasks", status mod = (module or "general").strip().lower() mod = re.sub(r"[^a-z0-9_-]", "-", mod) or "general" if section == "done": # modules//log// ym = "2026-04" # default heuristic if when: w = when.lower() if "mar" in w: ym = "2026-03" elif "apr" in w or "abr" in w: ym = "2026-04" elif "may" in w: ym = "2026-05" return f"modules/{mod}/log/{ym}", status return f"modules/{mod}/tasks", status def main(): ap = argparse.ArgumentParser() ap.add_argument("index_md", type=Path) ap.add_argument("meshkore_dir", type=Path, help=".meshkore/ root of the target repo (paths are built as /modules//...)") ap.add_argument("--dry-run", action="store_true") args = ap.parse_args() if not args.index_md.exists(): print(f"ERROR: {args.index_md} not found", file=sys.stderr) sys.exit(1) text = args.index_md.read_text(encoding="utf-8") lines = text.splitlines() section: str | None = None rows: list[dict] = [] # accumulated rows current: dict | None = None for line in lines: sm = SECTION_RE.match(line.strip()) if sm: if current: rows.append(current) current = None section = sm.group("name").lower() continue if not section: continue cells = parse_row(line) if cells is None: # Not a table row — flush current if current: rows.append(current) current = None continue if is_continuation(cells): if current is None: continue # Append non-empty cells as body body_cells = [c for c in cells[1:] if c] if body_cells: current.setdefault("body", []).append(" ".join(body_cells)) continue # New row — flush previous if current: rows.append(current) current = None idm = ID_RE.match(line) if not idm: continue task_id = idm.group("id").upper() # Parse cells based on section # Common: cells[0]=id, cells[1]=what what = cells[1] if len(cells) >= 2 else "" if section == "done": # | # | What | When | Details | when = cells[2] if len(cells) >= 3 else "" details = cells[3] if len(cells) >= 4 else "" module = "" current = { "id": task_id, "section": section, "what": what, "when": when, "module": module, "body": [details] if details else [], } elif section == "ecosystem": # | # | What | Who builds | Depends on | Module | module = cells[4] if len(cells) >= 5 else "ecosystem" current = { "id": task_id, "section": section, "what": what, "module": module, "depends_on": cells[3] if len(cells) >= 4 else "", "body": [], } else: # | # | What | Depends on | Effort | Module | depends = cells[2] if len(cells) >= 3 else "" effort = cells[3] if len(cells) >= 4 else "" module = cells[4] if len(cells) >= 5 else "uncategorized" current = { "id": task_id, "section": section, "what": what, "module": module, "depends_on": depends, "effort": effort, "body": [], } if current: rows.append(current) print(f"Parsed {len(rows)} rows from {args.index_md.name}", file=sys.stderr) args.meshkore_dir.mkdir(parents=True, exist_ok=True) today = date.today().isoformat() for r in rows: section = r["section"] what = r.get("what", "") if not what or what.startswith("~~"): # Skip strikethroughs (cancelled / superseded entries) continue slug = slugify(what) subpath, status = section_to_folder(section, r.get("module", ""), r.get("when")) out_dir = args.meshkore_dir / subpath out_path = out_dir / f"{r['id']}-{slug}.md" # Build frontmatter title = what.replace('"', "'") # Strip markdown formatting from title for frontmatter cleanliness title_clean = re.sub(r"\*+", "", title).strip() tags = [r.get("module", "")] if r.get("module") else [] tags = [t for t in tags if t and t not in ("uncategorized", "—")] fm = [ "---", f"id: {r['id']}", f'title: "{title_clean[:120]}"', f"status: {status}", f"category: {r.get('module', 'uncategorized') or 'uncategorized'}", f"updated: {today}", "owner: rjj", f"tags: {tags}", ] if r.get("depends_on") and r["depends_on"] not in ("—", "-", ""): fm.append(f'depends_on: "{r["depends_on"]}"') if r.get("effort"): fm.append(f'effort: "{r["effort"]}"') if r.get("when"): fm.append(f'when: "{r["when"]}"') fm.append("---") fm.append("") body = "\n\n".join(r.get("body", [])).strip() if not body: body = f"Migrated from `_rjj/tasklist/INDEX.md` ({r['id']})." content = "\n".join(fm) + "\n# " + title_clean + "\n\n" + body + "\n" if args.dry_run: print(f"[DRY] {out_path.relative_to(args.roadmap_dir.parent.parent)}") continue out_dir.mkdir(parents=True, exist_ok=True) out_path.write_text(content, encoding="utf-8") if not args.dry_run: print(f"Wrote {len(rows)} task files under {args.roadmap_dir}") if __name__ == "__main__": main()