Python正则表达式从列表中删除除字符串外的所有内容

Question

I have string:

bdv. mot. g. vns. kilm.

And knowing list of strings like

important_strings_lst=['bdv.', 'dktv.', 'mot. g.', 'vyr. g.']

I want to get regex selection like:

bdv. mot. g.

I joined list and tried: idea from here

regex = re.compile(r'\b(?!bdv.|dktv.|mot. g.|vyr. g.)\w+', re.UNICODE)
regex.sub("", 'bdv. mot. g. vns. kilm.')

Got

'bdv. mot. . . .'

Changing places in regex with \\s also didn't work out. How to do it?

I could use something like [x for x in important_strings_lst if x in my_string] but I need good performance as this will be used with million rows of pandas dataframe with str.replace

Answer 1

The . character has special meaning in regular expressions. You can use re.escape to make a string "safe" for use in a regular expression.

>>> import re
... important_strings=['bdv.', 'dktv.', 'mot. g.', 'vyr. g.']
... regex = re.compile('|'.join(re.escape(s) for s in important_strings))
... regex.findall('bdv. mot. g. vns. kilm.')
['bdv.', 'mot. g.']

Pandas has its own findall which should work like re.findall

Answer 2

Maybe split string

    bdv. mot. g. vns. kilm.

using your list and remove from oryginal string what left after spliting.