simple-nanoshare/cli/nanoshare_client/sync.py

from __future__ import annotations

import hashlib
import json
import os
import shutil
import sys
import time
from pathlib import Path

from .client import delete_remote, download, list_remote, upload

SYNC_DIR_NAME = '.nanoshare-sync'
STATE_FILE_NAME = 'state.json'

def sha256(path: Path) -> str:
  digest = hashlib.sha256()
  with path.open('rb') as handle:
    for chunk in iter(lambda: handle.read(1024 * 1024), b''):
      digest.update(chunk)
  return digest.hexdigest()

def state_path(root: Path) -> Path:
  return root / SYNC_DIR_NAME / STATE_FILE_NAME

def load_state(root: Path) -> dict:
  path = state_path(root)
  if not path.is_file():
    return {'version': 1, 'files': {}}
  try:
    data = json.loads(path.read_text())
  except (OSError, ValueError):
    return {'version': 1, 'files': {}}
  if not isinstance(data, dict):
    return {'version': 1, 'files': {}}
  data.setdefault('version', 1)
  data.setdefault('files', {})
  return data

def save_state(root: Path, state: dict) -> None:
  path = state_path(root)
  path.parent.mkdir(parents=True, exist_ok=True)
  tmp = path.with_name(f'{path.name}.{os.getpid()}.tmp')
  tmp.write_text(json.dumps(state, indent=2, sort_keys=True))
  os.replace(tmp, path)

def iter_local_files(root: Path):
  sync_dir = root / SYNC_DIR_NAME
  for path in sorted(root.rglob('*')):
    if not path.is_file():
      continue
    if path == state_path(root) or sync_dir in path.parents:
      continue
    yield path

def relative(root: Path, path: Path) -> str:
  return path.relative_to(root).as_posix()

def safe_remote_name(name: str, file_id: str) -> str:
  candidate = Path(name or file_id).name
  return candidate or file_id

def unique_path(path: Path) -> Path:
  if not path.exists():
    return path
  stamp = time.strftime('%Y%m%d-%H%M%S')
  suffix = ''.join(path.suffixes)
  stem = path.name[:-len(suffix)] if suffix else path.name
  candidate = path.with_name(f'{stem}.conflict-{stamp}{suffix}')
  counter = 2
  while candidate.exists():
    candidate = path.with_name(f'{stem}.conflict-{stamp}-{counter}{suffix}')
    counter += 1
  return candidate

def remote_by_id(remote_files: list[dict]) -> dict[str, dict]:
  return {item['file_id']: item for item in remote_files if isinstance(item, dict) and item.get('file_id')}

def remote_by_name(remote_files: list[dict]) -> dict[str, dict]:
  result = {}
  for item in remote_files:
    if isinstance(item, dict) and item.get('file_name') and item.get('file_id'):
      result.setdefault(item['file_name'], item)
  return result

def sync_once(args, client) -> int:
  root = Path(args.folder).expanduser().resolve()
  root.mkdir(parents=True, exist_ok=True)
  state = load_state(root)
  known = state.setdefault('files', {})
  try:
    remote_files = list_remote(client, args.node)
    remote_ids = remote_by_id(remote_files)
    remote_names = remote_by_name(remote_files)
    local_paths = {relative(root, path): path for path in iter_local_files(root)}

    for rel, path in local_paths.items():
      current_hash = sha256(path)
      entry = known.get(rel)
      remote_id = entry.get('file_id') if isinstance(entry, dict) else None
      old_hash = entry.get('sha256') if isinstance(entry, dict) else None

      if old_hash == current_hash and remote_id in remote_ids:
        continue

      if remote_id and remote_id not in remote_ids and old_hash != current_hash:
        conflict_path = unique_path(path)
        shutil.copy2(path, conflict_path)
        print(f'conflict: kept changed local copy at {conflict_path}')

      print(f'upload: {rel}')
      result = upload(client, args.node, path, rel, args.note or '', args.expires or '')
      new_file_id = result.get('file_id') or result.get('id')
      if not new_file_id:
        matching = remote_names.get(rel)
        refreshed = list_remote(client, args.node)
        new_file_id = (remote_by_name(refreshed).get(rel) or matching or {}).get('file_id')
        remote_files = refreshed
        remote_ids = remote_by_id(remote_files)
        remote_names = remote_by_name(remote_files)
      if not new_file_id:
        print(f'warning: upload succeeded but file_id is unknown for {rel}', file=sys.stderr)
        continue
      if remote_id and remote_id in remote_ids and remote_id != new_file_id:
        try:
          delete_remote(client, args.node, remote_id)
          print(f'delete old remote: {rel}')
          remote_files = [item for item in remote_files if item.get('file_id') != remote_id]
          remote_ids = remote_by_id(remote_files)
          remote_names = remote_by_name(remote_files)
        except Exception as exc:
          print(f'warning: could not delete old remote {remote_id}: {exc}', file=sys.stderr)
      known[rel] = {
        'file_id': new_file_id,
        'sha256': current_hash,
        'local_mtime': path.stat().st_mtime,
        'file_name': rel,
      }

    known_ids = {entry.get('file_id') for entry in known.values() if isinstance(entry, dict)}
    for remote in remote_files:
      file_id = remote.get('file_id')
      if not file_id or file_id in known_ids:
        continue
      name = safe_remote_name(remote.get('file_name') or '', file_id)
      path = unique_path(root / name)
      print(f'download: {file_id} -> {relative(root, path)}')
      download(client, args.node, file_id, path)
      known[relative(root, path)] = {
        'file_id': file_id,
        'sha256': sha256(path),
        'local_mtime': path.stat().st_mtime,
        'file_name': relative(root, path),
      }

    if args.delete:
      for rel, entry in list(known.items()):
        if rel in local_paths:
          continue
        file_id = entry.get('file_id') if isinstance(entry, dict) else None
        if file_id and file_id in remote_ids:
          print(f'delete remote missing locally: {rel}')
          delete_remote(client, args.node, file_id)
        known.pop(rel, None)

    save_state(root, state)
  except Exception as exc:
    print(f'error: {exc}', file=sys.stderr)
    return 1
  return 0