I have a regular expression like this:
regexp = u'ba[r|z|d]'
Function must return True if word contains bar, baz or bad. In short, I need regexp analog for Python's
'any-string' in 'text'
How can I realize it? Thanks!
5 Answers
import re
word = 'fubar'
regexp = re.compile(r'ba[rzd]')
if regexp.search(word):
print('matched')
3 Comments
bawejakunal
I am working on a similar case where I want to search for an exact string (
xyz) and want to know which is a more efficient way to do this, should I use python's 'xyz' in given_text or use re.compile(r'xyz').search(given_text) ?
radtek
the
[] brackets contain a character class, so your re also matches: >>> word = 'ba|'; regexp.search(word) <_sre.SRE_Match object at 0x101030b28> . You can drop all the pipe symbols.
Charlie Parker
why do you need the
r in front of your string? i.e. what is the difference between re.compile(r'x[\d+]') vs re.compile('x[\d+]')?The best one by far is
bool(re.search('ba[rzd]', 'foobarrrr'))
Returns True
4 Comments
Kresten
Why is this one better than the other solutions?
bballdave025
For one thing, it returns a
bool. OP: "must return True if word contains bar, baz or bad." Other answers use the behavior of if - auto-converting the expression to its right to a bool. e.g. import re; rgx=re.compile(r'ba[rzd]'); rgx.search('foobar') => <re.Match object; span=(2, 5), match='bar'>, but if(rgx.search(w)): print('y') => y. Closest to documentation of auto-convert I could find(archived)Charlie Parker
why did you prefer
.search vs .match?Charlie Parker
also why did you not use
r infront of the regex e.g. re.compile(r'x[\d+]') vs re.compile('x[\d+]')?Match objects are always true, and None is returned if there is no match. Just test for trueness.
Code:
>>> st = 'bar'
>>> m = re.match(r"ba[r|z|d]",st)
>>> if m:
... m.group(0)
...
'bar'
Output = bar
If you want search functionality
>>> st = "bar"
>>> m = re.search(r"ba[r|z|d]",st)
>>> if m is not None:
... m.group(0)
...
'bar'
and if regexp not found than
>>> st = "hello"
>>> m = re.search(r"ba[r|z|d]",st)
>>> if m:
... m.group(0)
... else:
... print "no match"
...
no match
As @bukzor mentioned if st = foo bar than match will not work. So, its more appropriate to use re.search.
4 Comments
ruakh
As I understand the question, OP actually wants
search rather than match. (See docs.python.org/library/re.html#matching-vs-searching.) Also, I think it would be helpful if you showed actual possible arguments, in the correct order, rather than just ....
bukzor
if you change
st to "foo bar", the match method will not work here. You want search.
freeforall tousez
@ruakh that link no longer auto scroll to that part of the doc, now the link is docs.python.org/2/library/re.html#search-vs-match
Piyush S. Wanare
@RanRag what is the complexity difference in searching substring with
in and regex?Here's a function that does what you want:
import re
def is_match(regex, text):
pattern = re.compile(regex)
return pattern.search(text) is not None
The regular expression search method returns an object on success and None if the pattern is not found in the string. With that in mind, we return True as long as the search gives us something back.
Examples:
>>> is_match('ba[rzd]', 'foobar')
True
>>> is_match('ba[zrd]', 'foobaz')
True
>>> is_match('ba[zrd]', 'foobad')
True
>>> is_match('ba[zrd]', 'foobam')
False
Comments
You can do something like this:
Using search will return a SRE_match object, if it matches your search string.
>>> import re
>>> m = re.search(u'ba[r|z|d]', 'bar')
>>> m
<_sre.SRE_Match object at 0x02027288>
>>> m.group()
'bar'
>>> n = re.search(u'ba[r|z|d]', 'bas')
>>> n.group()
If not, it will return None
Traceback (most recent call last):
File "<pyshell#17>", line 1, in <module>
n.group()
AttributeError: 'NoneType' object has no attribute 'group'
And just to print it to demonstrate again:
>>> print n
None
bool(re.search('ba[rzd]', 'sometext')).bool?boolfor typing. e.g.def f(x: bool) -> bool: ... body etccheers!