Strip Specific Punctuation in Python 2.x

Question

I'm using Python v2.6 and I have a string which contains a number of punctuation characters I'd like to strip out. Now I've looked at using the string.punctuation() function but unfortunately, I want to strip out all punctuation characters except fullstops and dashes. In total, there are only a total of 5 punctuation marks I'd like to strip out - ()\"'

Any suggestions? I'd like this to be the most efficient way.

Thanks

Duplicate of http://stackoverflow.com/questions/4371231/removing-punctuation-from-python-list-items/4371351#4371351 — Josh Bleecher Snyder, Jan 13 '12 at 22:14

unutbu · Answer 1 · 2012-01-14T09:37:07.360

1

Using string.translate:

s = ''' abc(de)f\gh"i' '''
print(s.translate(None, r"()\"'"))
 # abcdefghi

or re.sub:

import re
re.sub(r"[\\()'\"]",'',s)

but string.translate appears to be an order of magnitude faster:

In [148]: %timeit (s*1000).translate(None, r"()\"'")
10000 loops, best of 3: 112 us per loop

In [146]: %timeit re.sub(r"[\\()'\"]",'',s*1000)
100 loops, best of 3: 2.11 ms per loop

edited Jan 14 '12 at 09:37

answered Jan 13 '12 at 22:10

unutbu

842,883
184
1,785
1,677

No need to make a blank translation table; just use `None`. – Josh Bleecher Snyder Jan 13 '12 at 22:15
Also, your code formatting is broken. Looks like an unmatched `"`. – Josh Bleecher Snyder Jan 13 '12 at 22:15
Josh, thanks for the correction regarding `translate(None,...)`. Where is the formatting incorrect? I've run the code. It seems to work. – unutbu Jan 13 '12 at 22:21
well, `print` shows up in red. Looks like SO isn't parsing the triple quotes correctly. Maybe you need to hint to it that the code snippet is Python? I dunno. – Josh Bleecher Snyder Jan 13 '12 at 22:30

cha0site · Answer 2 · 2012-01-13T22:30:15.980

1

>>> import re
>>> r = re.compile("[\(\)\\\\'\"]")
>>> r.sub("", "\"hello\" '(world)'\\\\\\")
'hello world'

edited Jan 13 '12 at 22:30

answered Jan 13 '12 at 22:13

cha0site

10,517
3
33
51

This doesn't remove backslashes from the original string. `r.sub('', 'a\\b') --> 'a\\b'` – Andrew Clark Jan 13 '12 at 22:25
Right, there are a lot of answers on this one, but I think a compiled regexp would be the most *efficient* solution. And always remember, “Give a man a regular expression and he’ll match a string… but by teaching him how to create them, you’ve given him enough rope to hang himself” – cha0site Jan 13 '12 at 22:29
@Paulo: I don't like raw strings, they remind me of paths in Windows ;) – cha0site Jan 13 '12 at 22:31
@Paulo: I find that it depends on how much backslashes you need. `r"C:\Program Files\SomeCompany\SomeProgram Version 7\Internals\foobar.bla"` is obviously much nicer than the alternative, but a lot of times I need `"\t\n\0\xad\xde\xef\xbe"`, too... – cha0site Jan 13 '12 at 22:42
let us [continue this discussion in chat](http://chat.stackoverflow.com/rooms/6710/discussion-between-cha0site-and-paulo-scardine) – cha0site Jan 13 '12 at 22:56

Andrew Clark · Accepted Answer · 2012-01-13T22:23:16.333

1

You can use str.translate(table[, deletechars]) with table set to None, which will result in all characters from deletechars being removed from the string:

s.translate(None, r"()\"'")

Some examples:

>>> "\"hello\" '(world)'".translate(None, r"()\"'")
'hello world'
>>> "a'b c\"d e(f g)h i\\j".translate(None, r"()\"'")
'ab cd ef gh ij'

edited Jan 13 '12 at 22:23

answered Jan 13 '12 at 22:15

Andrew Clark

202,379
35
273
306

score 1 · Answer 4 · answered Jan 13 '12 at 22:16

You could make a list of all the characters you don't want:

unwanted = ['(', ')', '\\', '"', '\'']

Then you could make a function strip_punctuation(s) like so:

def strip_punctuation(s): 
    for u in unwanted: 
        s = s.replace(u, '')
    return s

score 0 · Answer 5 · answered Jan 13 '12 at 22:14

0

You can create a dict of all the characters you want to be replaced and replace them with char of your choice.

char_replace = {"'":"" , "(":"" , ")":"" , "\":"" , """:""}

for i,j in char_replace.iteritems():
        string = string.replace(i,j)

answered Jan 13 '12 at 22:14

RanRag

48,359
38
114
167

Corey Goldberg · Answer 6 · 2012-01-13T22:40:50.050

0

my_string = r'''\(""Hello ''W\orld)'''
strip_chars = r'''()\'"'''

using comprehension:

''.join(x for x in my_string if x not in strip_chars)

using filter:

''.join(filter(lambda x: x not in strip_chars, my_string))

output:

Hello World

edited Jan 13 '12 at 22:40

answered Jan 13 '12 at 22:31

Corey Goldberg

59,062
28
129
143

Strip Specific Punctuation in Python 2.x

6 Answers6

Linked