def file_duplicate_no(self, name: str, path: str,
remote_id: str) -> (int, DatabaseMedia):
"""
determine if there is already an entry for file. If not determine
if other entries share the same path/filename and determine a duplicate
number for providing a unique local filename suffix
Returns:
duplicate no. (zero if there are no duplicates),
Single row from the SyncRow table
"""
query = "SELECT {0} FROM SyncFiles WHERE RemoteId = ?; ". \
format(GooglePhotosRow.columns)
self.cur.execute(query, (remote_id, ))
result = self.cur.fetchone()
if result:
# return the existing file entry's duplicate no.
return result['DuplicateNo'], GooglePhotosRow(result).to_media()
self.cur.execute(
"SELECT MAX(DuplicateNo) FROM SyncFiles "
"WHERE Path = ? AND OrigFileName = ?;", (path, name))
results = self.cur.fetchone()
if results[0] is not None:
# assign the next available duplicate no.
dup = results[0] + 1
return dup, None
else:
# the file is new and has no duplicates
return 0, None
def get_extra_meta(self):
count = 0
log.warning('updating index with extra metadata for comparison '
'(may take some time) ...')
media_items = self._db.get_rows_by_search(
GooglePhotosRow, uid='ISNULL')
for item in media_items:
file_path = self._root_folder / item.relative_path
# if this item has a uid it has been scanned before
if file_path.exists():
local_file = LocalFilesMedia(file_path)
count += 1
log.info('updating metadata %d on %s', count, file_path)
item.update_extra_meta(local_file.uid,
local_file.create_date,
local_file.size)
# erm lets try some duck typing then !
# todo is the DbRow class model rubbish or brilliant Python?
# noinspection PyTypeChecker
self._db.put_row(GooglePhotosRow.from_media(item),
update=True)
if count % 2000 == 0:
self._db.store()
else:
log.debug('skipping metadata (already scanned) on %s',
file_path)
log.warning('updating index with extra metadata complete')
def get_extra_paths(self):
self.cur2.execute(Queries.pre_extra_files)
self.cur2.execute(Queries.extra_files)
while True:
records = self.cur2.fetchmany(LocalData.BLOCK_SIZE)
if not records:
break
for record in records:
r = GooglePhotosRow(record).to_media()
pth = r.relative_path.parent / r.filename
yield pth
def write_media_index(self, media: GooglePhotosMedia,
update: bool = True):
self._db.put_row(GooglePhotosRow.from_media(media), update)
if media.create_date > self.latest_download:
self.latest_download = media.create_date