Re: partial / wildcard string match in 'in' and 'list.index()'
From: Josiah Carlson (jcarlson_at_uci.edu)
Date: 06/04/04
- Next message: Jacek Generowicz: "Re: Optimizing multiple dispatch"
- Previous message: Matteo Dell'Amico: "Re: variables in string.count"
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
Date: Fri, 04 Jun 2004 00:25:07 -0700
> These would work, but I was wondering if there was some compact
> way to get 'in' and lists.index() to do wildcard matching as opposed
> to exact matching.
Short answer: no.
Long answer:
There is a regular expression library, aptly called 're', for non-exact
string searching.
The standard string methods only do exact searching likely because there
are algorithms that make it fast (linear in the length of the searched
string), that are also quite small.
There exist very simple looking regular expressions that force searches
to take exponential (in the length of the pattern) time to search, and
are generally fairly large.
An example of an exponentially slow regular expression is contained in
the following example...
>>> import re
>>> import time
>>> def test(n):
... a = 100*'0'
... t = time.time()
... re.search('(0+)'*n, a)
... return time.time()-t
...
>>> for i in xrange(18, 23):
... print i, test(i)
...
18 0.141000032425
19 0.266000032425
20 0.530999898911
21 1.07899999619
22 2.1400001049
>>>
Trust me, you don't want that kind of slow down when you are searching
in strings with "if i in 'somestring':".
- Josiah
- Next message: Jacek Generowicz: "Re: Optimizing multiple dispatch"
- Previous message: Matteo Dell'Amico: "Re: variables in string.count"
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
Relevant Pages
|
