Python Regex Sub: Using Dictionary with Regex Expressions

Question

I am using a dictionary that contain regular expressions to substitute portions of different strings, as elegantly described in a previous SO question by @roippi. The first 're.sub' expression works perfectly. However, whenever my code actually involves regex expressions (the second 're.sub' expression), the substitutions don't work.

I am very confused as to why this is the case. I have tried both using and taking out the 'r' as well as incorporating the lookahead/lookbehind expressions, nothing seems to work. Any help would be greatly appreciated!

test_dict = {r'(\d+)': 'THIS IS A NUMBER', 'john_doe':'THIS IS A NAME'}

re.sub('(john_doe)', lambda x: test_dict.get(x.group(1),x.group(1)),'john_doe_jr')

re.sub(r'(\d+)', lambda x: test_dict.get(x.group(1), x.group(1)), '999la')

Got problems to understand what this achieves. Care to explain? — Patrick Artner
– Patrick Artner, Commented Jan 3, 2018 at 21:17
Hi @WiktorStribiżew, thank you for your response. I had incorrectly added the html tags, which I have removed. However, this expression should work now (I have tried it in Pythex and it's fine) and it still does not when using the "re.sub" expression. Any thoughts as to why? — tinydanc3r212
– tinydanc3r212, Commented Jan 3, 2018 at 21:45
The regex matches 999, which becomes x.group(1). 999 is not a key in test_dict, which causes the .get method to return 999, replacing 999 with 999. — L3viathan
– L3viathan, Commented Jan 4, 2018 at 9:23

L3viathan · Accepted Answer · 2018-01-04 09:25:19Z

2

match.group(n) does not return the regular expression that was used to match the nth group, but the nth group itself.

The lambda therefore returns test_dict.get('999', '999'), which returns '999', because '999' is not a key in your dictionary.

You could iterate over the keys of the dictionary and check if any key matches your capture group, and then replace it, but that has O(n) time complexity (in the size of the dictionary).

def replacement(match, d, group=1):
    for key in d:
        if re.match(key, match.group(group)):
            return d[key]
    return match.group(group)

re.sub(r'(\d+)', lambda x: replacement(x, test_dict), '999la')

answered Jan 4, 2018 at 9:25

L3viathan

27.5k2 gold badges63 silver badges84 bronze badges

Sign up to request clarification or add additional context in comments.

1 Comment

tinydanc3r212 Over a year ago

this function addresses my dilemma perfectly!! Thank you so much.

Collectives™ on Stack Overflow

Python Regex Sub: Using Dictionary with Regex Expressions

1 Answer 1

1 Comment

Your Answer

Linked

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

1 Comment

Your Answer

Sign up or log in

Post as a guest

Linked

Related