77 lines
2.7 KiB
Python
77 lines
2.7 KiB
Python
|
import re
|
||
|
|
|
||
|
|
|
||
|
|
def upgrade(file_manager):
|
||
|
|
""" Use double quote for redacted text and single quote for strings. """
|
||
|
|
# Don't use this script in production, it is broken and only serve
|
||
|
|
# as an example.
|
||
|
|
|
||
|
|
# List all the files that might need to be upgraded, here we list
|
||
|
|
# all python models.
|
||
|
|
files = [
|
||
|
|
file for file in file_manager
|
||
|
|
if 'models' in file.path.parts
|
||
|
|
if file.path.suffix == '.py'
|
||
|
|
if file.path.name != '__init__.py'
|
||
|
|
]
|
||
|
|
|
||
|
|
# Early return if case there are no file, so we don't compile
|
||
|
|
# regexps for nothing.
|
||
|
|
if not files:
|
||
|
|
return
|
||
|
|
|
||
|
|
# Python Regexp 101
|
||
|
|
#
|
||
|
|
# re.VERBOSE ignores all spaces inside the regexp so it is possible
|
||
|
|
# to indent it, it also allows for comments at the end of each line.
|
||
|
|
# to actually expect a space you have to escape it: "\ "
|
||
|
|
#
|
||
|
|
# a* vs a*? the first is greedy and the second is lazy, the first is
|
||
|
|
# gonna match as most "a" as possible, the second will stop as soon
|
||
|
|
# as possible. Lazy quantifiers are MUCH faster than greedy one.
|
||
|
|
#
|
||
|
|
# (?P<x>) it is your regular group, but you can access it via its
|
||
|
|
# name "x": match = re.match(...); match.group('x'). Much better
|
||
|
|
# than using numeric groups.
|
||
|
|
#
|
||
|
|
# (?:) it is a non-capturing group, for when you need a group inside
|
||
|
|
# the regexp but you don't need to remember what was matched inside.
|
||
|
|
# They are faster than regular (capturing) groups.
|
||
|
|
|
||
|
|
# Assume that all redacted text:
|
||
|
|
# - Start with a upper case
|
||
|
|
# - Have multiples words
|
||
|
|
# - End with a dot
|
||
|
|
# This assumption is wrong for many cases, don't use this script!
|
||
|
|
redacted_text_re = re.compile(r"""
|
||
|
|
' # Opening single quote
|
||
|
|
(?P<text>
|
||
|
|
[A-Z][^'\s]*?\ # First word
|
||
|
|
(?:[^'\s]*?\ )* # All middle words
|
||
|
|
[^'\s]*?\. # Final word
|
||
|
|
)
|
||
|
|
' # Closing single quote
|
||
|
|
""", re.VERBOSE)
|
||
|
|
|
||
|
|
# Assume that all strings:
|
||
|
|
# - Are fully lowercase
|
||
|
|
# - Have a single word
|
||
|
|
# - Have no ponctuation
|
||
|
|
# This assumption is wrong for many cases, don't use this script!
|
||
|
|
strings_re = re.compile(r'"(?P<string>[a-z]+)"')
|
||
|
|
|
||
|
|
# Iterate over all the files and run the regexps
|
||
|
|
for fileno, file in enumerate(files, start=1):
|
||
|
|
# load the content
|
||
|
|
content = file.content
|
||
|
|
|
||
|
|
# do the operations
|
||
|
|
content = redacted_text_re.sub(r'"\g<text>"', content)
|
||
|
|
content = strings_re.sub(r"'\g<string>'", content)
|
||
|
|
|
||
|
|
# write back the file, if nothing changed nothing is written
|
||
|
|
# file.content = content # uncomment this line to test the script
|
||
|
|
|
||
|
|
# have the progress bar to actually show the progression
|
||
|
|
file_manager.print_progress(fileno, len(files))
|