How to replace two things at once in a string?

import string
"abaababb".translate(string.maketrans("ab", "ba"))
# result: 'babbabaa'

Note that this only works for one-character substitutions.

For longer substrings or substitutions, this is a bit complex, but might work:

import re

def replace_all(repls, str):
    # return re.sub('|'.join(repls.keys()), lambda k: repls[k.group(0)], str)                                     
    return re.sub('|'.join(re.escape(key) for key in repls.keys()),
                  lambda k: repls[k.group(0)], str)                                     


text =  "i like apples, but pears scare me"
print replace_all({"apple": "pear", "pear": "apple"}, text)

Unfortunately this won’t work if you include any regexp special characters you can’t use regexps this way 🙁

(Thanks @TimPietzcker)

the_string="ab"
new_string=""

for x in range(len(the_string)):
    if the_string[x]=='a':
        new_string+='b'
        continue
    if the_string[x]=='b':
        new_string+='a'
        continue
    new_string+=the_string[x]

the_string=new_string

print the_string

If you are OK with two lines, this is more elegant.

d={'a':'b','b':'a'}
''.join(d[s] for s in "abaababbd" if s in d.keys())

Your example is a little bit abstract but in the past I’ve used this recipe which builds a regular expression to do single-pass multiple replace. Here’s my tweaked version of it:

import re 

def multiple_replace(dict, text): 
  regex = re.compile("|".join(map(re.escape, dict.keys())))
  return regex.sub(lambda mo: dict[mo.group(0)], text) 

Note that the keys (searchstrings) are re.escaped.

In your case it would be:

from utils import multiple_replace

print multiple_replace({
    "a": "b",
    "b": "a"
}, "ab") 

UPDATE:

By now this is basically the same as Amadan’s answer

>>> import re
>>> re.sub('.', lambda m: {'a':'b', 'b':'a'}.get(m.group(), m.group()), 'abc')
'bac'