RegEx排除目录，捕获用逗号分隔的文件名，排除“（number）”和扩展名

Question

I've been trying for the last three days (yeah) to make a image/short video tagging system for my own use but this has proven a challenge beyond me.

These are the strings:

d:\images\tagging 1\GIFs\kung fu panda, fight.webm
d:\images\tagging 1\GIFs\kung fu panda, fight (2).webm
d:\images\tagging 1\GIFs\kung fu panda 2, fight.webm
d:\images\tagging 1\GIFs\kung fu panda 2, fight (2).webm
d:\images\tagging 1\GIFs\pulp fiction, samuel l. jackson, angry, funny.webm

I have four things that I've tried modifying to achieve what I want with no success:

(?<=d:\\images\\tagging\s1\\GIFs\\)([\w\s])+

([a-z0-9]\s?)+

(?<=\\)[^\\]*?(?=\..*$)

[^\\/:*?"<>|\r\n]+$

1 Almost there, but it doesn't extend past the first comma.

2 This does almost everything, but I haven't found a way to exclude the directory, the (#) and the extension.

3 Taken from the internet, captures the "l." and stops there, whole filename, can't use commas as I want, captures (#).

4 Taken from regexbuddy (yes I actually bought it in my desperation), captures (#) and extension.

@timgeb

The intention is to get the filenames without the commas, the (#) and extension, so:

"kung fu panda" "fight"
"kung fu panda" "fight"
"kung fu panda 2" "fight"
"kung fu panda 2" "fight"
"pulp fiction" "samuel l. jackson" "angry" "funny"

Answer 1

Your question isn't very clear, but I think you want to parse filenames. If that's the case, I wouldn't recommend using re as your primary tool.

Instead, have a look at os.path :

import os.path  # Or `import ntpath` for Windows paths on non-Windows systems

dir, file_name = os.path.split('d:\images\tagging 1\GIFs\kung fu panda, fight (2).webm')
# dir = 'd:\images\tagging 1\GIFs'
# file_name = 'kung fu panda, fight (2).webm'

root, ext = os.path.splitext(file_name)
# root = 'kung fu panda, fight (2)'
# ext = '.webm'

Now you have a much simpler problem: removing the numbers in parentheses.

Answer 2

Get the basename, substitute integers in parentheses and the extension with the empty string and strip off the whitespace.

from ntpath import basename
import re
map(str.strip, re.sub('\(\d+\)|\.\w+$', '', basename(s)).split(','))

Demo:

>>> s = 'd:\images\tagging 1\GIFs\kung fu panda, fight.webm'
>>> map(str.strip, re.sub('\(\d+\)|\.\w+$', '', basename(s)).split(','))
['kung fu panda', 'fight']
>>> s = 'd:\images\tagging 1\GIFs\kung fu panda, fight (2).webm'
>>> map(str.strip, re.sub('\(\d+\)|\.\w+$', '', basename(s)).split(','))
['kung fu panda', 'fight']
>>> s = 'd:\images\tagging 1\GIFs\kung fu panda 2, fight.webm'
>>> map(str.strip, re.sub('\(\d+\)|\.\w+$', '', basename(s)).split(','))
['kung fu panda 2', 'fight']
>>> s = 'd:\images\tagging 1\GIFs\kung fu panda 2, fight (2).webm'
>>> map(str.strip, re.sub('\(\d+\)|\.\w+$', '', basename(s)).split(','))
['kung fu panda 2', 'fight']
>>> s = 'd:\images\tagging 1\GIFs\pulp fiction, samuel l. jackson, angry, funny.webm'
>>> map(str.strip, re.sub('\(\d+\)|\.\w+$', '', basename(s)).split(','))
['pulp fiction', 'samuel l. jackson', 'angry', 'funny']

Answer 3

If I got you, you want last tags (kung fu panda, fight.webm) that is after 1\\GIFs\\ - If you add more content-string then I can normalize code for you. This code just extracts tags and generates a regular list. import re

s="""d:\images\tagging 1\GIFs\kung fu panda, fight.webm
d:\images\tagging 1\GIFs\kung fu panda, fight (2).webm
d:\images\tagging 1\GIFs\kung fu panda 2, fight.webm
d:\images\tagging 1\GIFs\kung fu panda 2, fight (2).webm
d:\images\tagging 1\GIFs\pulp fiction, samuel l. jackson, angry, funny.webm"""

lines = s.split('\n')# Just generate a list of lines
for t in lines:
    data = re.search(r'1\\GIFs\\(.+$)',t)
    print data.group(1).split(',')

Output-

['kung fu panda', ' fight.webm']
['kung fu panda', ' fight (2).webm']
['kung fu panda 2', ' fight.webm']
['kung fu panda 2', ' fight (2).webm']
['pulp fiction', ' samuel l. jackson', ' angry', ' funny.webm']

Expression 1\\\\GIFs\\\\(.+$) will capture last tags that is after 1\\\\GIFs

SEE LIVE-DEMO