Created
May 30, 2012 10:37
-
-
Save lvictorino/2835406 to your computer and use it in GitHub Desktop.
A collegue asked for some help for her job. She needed a script to apply specific French langage needs (a nbsp and not a classical space after a number) to a large amount of files in several directories.
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
| #!/usr/bin/python | |
| import os | |
| import re | |
| import sys | |
| if sys.version_info < (3,2): | |
| print("Error: Python version 3.2 is required.") | |
| if len(sys.argv) <= 1: | |
| print("Not enough arguments.") | |
| print("usage: nbsp-dealer path_to_directory") | |
| exit() | |
| #PATH definition | |
| p=sys.argv[1] | |
| PATH = p if p[-1] != '/' else p[:-1] | |
| BACKUP_PATH=PATH+"_nbsp_backup" | |
| #REGEXP definition | |
| NUMBER_MATCH = r'([0-9]) ' | |
| NUMBER_SUB = "\\1\xa0" | |
| SPEC_MATCH = r' ([^\s\d\w<])' | |
| SPEC_SUB = "\xa0\\1" | |
| LANG_MATCH = r'<\s*html\s+lang="en"\s*>' | |
| LANG_SUB = '<html lang="fr">' | |
| def get_file_path(path, filename): | |
| return path + "/" + filename | |
| def apply_nbsp(path, filename): | |
| filepath = get_file_path(path,filename) | |
| if os.path.exists(filepath) is False: | |
| return | |
| d = open(filepath, "r", -1, "utf-8").read() | |
| open(filepath.replace(PATH,BACKUP_PATH),"w",-1,"utf-8").write(d) | |
| r = d | |
| r = re.sub(NUMBER_MATCH, NUMBER_SUB, r) | |
| r = re.sub(SPEC_MATCH, SPEC_SUB, r) | |
| r = re.sub(LANG_MATCH, LANG_SUB, r) | |
| os.remove(filepath) | |
| open(filepath, "w", -1, "utf-8").write(r) | |
| # open(get_file_path("./results", filename), "w", -1, "utf-8").write(r) | |
| return 1 if r != d else 0 | |
| def search_for_files(path): | |
| if os.path.exists(path) is False: | |
| return 0 | |
| change_counter = 0 | |
| for f in os.listdir(path): | |
| filepath = get_file_path(path, f) | |
| if os.path.isdir(filepath): | |
| backup_dir_path = filepath.replace(PATH,BACKUP_PATH) | |
| if os.path.exists(backup_dir_path) is False: | |
| os.mkdir(backup_dir_path, 0o755) | |
| change_counter += search_for_files(filepath) | |
| else: | |
| change_counter += apply_nbsp(path, f) | |
| return change_counter | |
| if os.path.isdir(PATH) is False: | |
| print("Error: Path parameter has to be a directory (don't play with me, I am no toy!).") | |
| exit() | |
| if os.path.exists(BACKUP_PATH) is False: | |
| os.mkdir(BACKUP_PATH, 0o755) | |
| count = search_for_files(PATH) | |
| if count > 0: | |
| print("Lucky you! Changes are now applied and world has not been destroyed.") | |
| print(str(count)+" file"+("s have" if count > 1 else " has")+" been modified.") | |
| else: | |
| print("Do you think I work for free?!") | |
| print("Nothing has change.") |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment