Last active
February 2, 2019 07:12
-
-
Save humpydonkey/a0d98092a6f6d7146b2996cd0e16b5dd to your computer and use it in GitHub Desktop.
[check missing files]check missing files, if missing then copy them to somewhere #python #os #tool #file
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
import os | |
from typing import Set | |
import shutil | |
def check_existence_or_copy(source_dir: str, check_against_dir: str, copy_dst_dir: str) -> None: | |
missing_files = check_missing(source_dir, check_against_dir) | |
copy_missing_files(source_dir, copy_dst_dir, missing_files) | |
def check_missing(source_dir: str, check_against_dir: str) -> Set[str]: | |
source_files: Set = _collect_files(source_dir) | |
check_files: Set = _collect_files(check_against_dir) | |
diff = source_files - check_files | |
print(f'Found {len(diff)} missing files.') | |
if diff: | |
print(f'Missing file names: {diff}.') | |
return diff | |
def copy_missing_files(source_dir: str, destination_dir: str, missing_file_names: Set[str]) -> None: | |
if not os.path.exists(destination_dir): | |
os.mkdir(destination_dir) | |
for curr_dir, subdirList, fileList in os.walk(source_dir): | |
curr_missing = set(fileList) & missing_file_names | |
if not curr_missing: | |
continue | |
print(f'Start copying {len(curr_missing)} missing files under {curr_dir}.') | |
_batch_copy(curr_dir, destination_dir, curr_missing) | |
def _batch_copy(src_dir, dst_dir, file_names) -> None: | |
for f in file_names: | |
copy_src_file = os.path.join(src_dir, f) | |
assert os.path.isfile(copy_src_file) | |
copy_dst_file = os.path.join(dst_dir, f) | |
if os.path.exists(copy_dst_file): | |
continue | |
# copy2() also attempts to preserve file metadata. | |
shutil.copy2(copy_src_file, copy_dst_file) | |
def _collect_files(source: str) -> Set[str]: | |
source_files: Set = set() | |
for curr_dir, subdirList, fileList in os.walk(source): | |
source_files.update(fileList) | |
print(f'Found {len(source_files)} files under {source}.') | |
return source_files | |
if __name__ == '__main__': | |
source_dir = '/Users/yacao/Pictures/2018/' | |
check_against_dir = '/Volumes/private_admin/photos/' | |
copy_dst_dir = '/Users/yacao/Downloads/missing_photos' | |
# check_existence_or_copy(source_dir, check_against_dir, copy_dst_dir) | |
check_missing(source_dir, check_against_dir) | |
source_dir = '/Volumes/private_admin/photos/Lightroom CC/12b6e686c06f48658af72171d6e84674/originals/2015' | |
check_against_dir = '/Volumes/private_admin/photos/past/2015' | |
copy_dst_dir = '/Users/yacao/Downloads/missing_photos_2015' | |
# check_missing(source_dir, check_against_dir) | |
# check_existence_or_copy(source_dir, check_against_dir, copy_dst_dir) |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment