I have a string: "y, i agree with u."
And I have array dictionary [(word_will_replace, [word_will_be_replaced])]:
[('yes', ['y', 'ya', 'ye']), ('you', ['u', 'yu'])]
i want to replace 'y' with 'yes' and 'u' with 'you' according to the array dictionary.
So the result i want: "yes, i agree with you."
I want to keep the punctuation there.
import re
s="y, i agree with u. yu."
l=[('yes', ['y', 'ya', 'ye']), ('you', ['u', 'yu'])]
d={ k : "\\b(?:" + "|".join(v) + ")\\b" for k,v in l}
for k,r in d.items(): s = re.sub(r, k, s)
print s
Output
yes, i agree with you. you.
key:re.compile(...)
Extending @gnibbler's answer from Replacing substrings given a dictionary of strings-to-be-replaced as keys and replacements as values. Python with the tips implemented from Raymond Hettinger in the comments.
import re
text = "y, i agree with u."
replacements = [('yes', ['y', 'ya', 'ye']), ('you', ['u', 'yu'])]
d = {w: repl for repl, words in replacements for w in words}
def fn(match):
return d[match.group()]
print re.sub('|'.join(r'\b{0}\b'.format(re.escape(k)) for k in d), fn, text)
>>>
yes, i agree with you.
That's not a dictionary -- it's a list, but it could be converted to a dict pretty easily. In this case, however, I would make it a little more explicit:
d = {}
replacements = [('yes', ['y', 'ya', 'ye']), ('you', ['u', 'yu'])]
for value,words in replacements:
for word in words:
d[word] = value
Now you have the dictionary mapping responses to what you want to replace them with:
{'y':'yes', 'ya':'yes', 'ye':'yes',...}
once you have that, you can pop in my answer from here using regular expressions: https://stackoverflow.com/a/15324369/748858
\bmeta character matches on word boundaries, as in, between a word and space or word and symbol. e.g.\by\bwill match ONLY the wordyon its own.{abc <-- ab, abcd <-- abc}. If you don't have this assumption, then only jamylak's solution is correct.