Speed of many regular expressions in python - python

I'm writing a python program that deals with a fair amount of strings/files. My problem is that I'm going to be presented with a fairly short piece of text, and I'm going to need to search it for instances of a fairly broad range of words/phrases.
I'm thinking I'll need to compile regular expressions as a way of matching these words/phrases in the text. My concern, however, is that this will take a lot of time.
My question is how fast is the process of repeatedly compiling regular expressions, and then searching through a small body of text to find matches? Would I be better off using some string method?
Edit: So, I guess an example of my question would be: How expensive would it be to compile and search with one regular expression versus say, iterating 'if "word" in string' say, 5 times?

You should try to compile all your regexps into a single one using the | operator. That way, the regexp engine will do most of the optimizations for you. Use the grouping operator () to determine which regexp matched.

If speed is of the essence, you are better off running some tests before you decide how to code your production application.
First of all, you said that you are searching for words which suggests that you may be able to do this using split() to break up the string on whitespace. And then use simple string comparisons to do your search.
Definitely do compile your regular expressions and do a timing test comparing that with the plain string functions. Check the documentation for the string class for a full list.

Your requirement appears to be searching a text for the first occurrence of any one of a collection of strings. Presumably you then wish to restart the search to find the next occurrence, and so on until the searched string is exhausted. Only plain old string comparison is involved.
The classic algorithm for this task is Aho-Corasick for which there is a Python extension (written in C). This should beat the socks off any alternative that's using the re module.

If you like to know how does it fast during compiling regex patterns, you need to benchmark it.
Here is how I do that. Its compile 1 Million time each patterns.
import time,re
def taken(f):
def wrap(*arg):
t1,r,t2=time.time(),f(*arg),time.time()
print t2-t1,"s taken"
return r
return wrap
#taken
def regex_compile_test(x):
for i in range(1000000):
re.compile(x)
print "for",x,
#sample tests
regex_compile_test("a")
regex_compile_test("[a-z]")
regex_compile_test("[A-Z0-9._%+-]+#[A-Z0-9.-]+\.[A-Z]{2,4}")
Its took around 5 min for each patterns in my computer.
for a 4.88999986649 s taken
for [a-z] 4.70300006866 s taken
for [A-Z0-9._%+-]+#[A-Z0-9.-]+\.[A-Z]{2,4} 4.78200006485 s taken
The real Bottleneck is not in compiling patterns, its in extracting text like re.findall, replacing re.sub. If you use that against Several MB texts, Its quite slow.
If your text is fixed, use normal str.find, its faster than regex.
Actually, If you give your text samples, and your regex patterns samples, we could give you better idea, there is many many great regex, and python guys out there.
Hope this help, sorry If my answer couldn't help you.

When you compile the regexp, it is converted into a state machine representation. Provided the regexp is efficiently expressed, it should still be very fast to match. Compiling the regexp can be expensive though, so you will want to do that up front, and as infrequently as possible. Ultimately though, only you can answer if it is fast enough for your requirements.
There are other string searching approaches, such as the Boyer-Moore algorithm. But I'd wager the complexity of searching for multiple separate strings is much higher than a regexp that can switch off each successive character.

This is a question that can readily be answered by just trying it.
>>> import re
>>> import timeit
>>> find = ['foo', 'bar', 'baz']
>>> pattern = re.compile("|".join(find))
>>> with open('c:\\temp\\words.txt', 'r') as f:
words = f.readlines()
>>> len(words)
235882
>>> timeit.timeit('r = filter(lambda w: any(s for s in find if w.find(s) >= 0), words)', 'from __main__ import find, words', number=30)
18.404569854548527
>>> timeit.timeit('r = filter(lambda w: any(s for s in find if s in w), words)', 'from __main__ import find, words', number=30)
10.953313759150944
>>> timeit.timeit('r = filter(lambda w: pattern.search(w), words)', 'from __main__ import pattern, words', number=30)
6.8793022576891758
It looks like you can reasonably expect regular expressions to be faster than using find or in. Though if I were you I'd repeat this test with a case that was more like your real data.

If you're just searching for a particular substring, use str.find() instead.

Depending on what you're doing it might be better to use a tokenizer and loop through the tokens to find matches.
However, when it comes to short pieces of text regexes have incredibly good performance. Personally I remember only coming into problems when text sizes became ridiculous like 100k words or something like that.
Furthermore, if you are worried about the speed of actual regex compilation rather than matching, you might benefit from creating a daemon that compiles all the regexes then goes through all the pieces of text in a big loop or runs as a service. This way you will only have to compile the regexes once.

in general case, you can use "in" keyword
for line in open("file"):
if "word" in line:
print line.rstrip()
regex is usually not needed when you use Python :)

Related

How can I speed up an email-finding regular expression when searching through a massive string?

I have a massive string. It looks something like this:
hej34g934gj93gh398gie foo#bar.com e34y9u394y3h4jhhrjg bar#foo.com hge98gej9rg938h9g34gug
Except that it's much longer (1,000,000+ characters).
My goal is to find all the email addresses in this string.
I've tried a number of solutions, including this one:
#matches foo#bar.com and bar#foo.com
re.findall(r'[\w\.-]{1,100}#[\w\.-]{1,100}', line)
Although the above code technically works, it takes an insane amount of time to execute. I'm not sure if it counts as catastrophic backtracking or if it's just really inefficient, but whatever the case, it's not good enough for my use case.
I suspect that there's a better way to do this. For example, if I use this regex to only search for the latter part of the email addresses:
#matches #bar.com and #foo.com
re.findall(r'#[\w-]{1,256}[\.]{1}[a-z.]{1,64}', line)
It executes in just a few milliseconds.
I'm not familiar enough with regex to write the rest, but I assume that there's some way to find the #x.x part first and then check the first part afterwards? If so, then I'm guessing that would be a lot quicker.
You can use PyPi regex module by Matthew Barnett, that is much more powerful and stable when it comes to parsing long texts. This regex library has some basic checks for pathological cases implemented. The library author mentions at his post:
The internal engine no longer interprets a form of bytecode but
instead follows a linked set of nodes, and it can work breadth-wise as
well as depth-first, which makes it perform much better when faced
with one of those 'pathological' regexes.
However, there is yet another trick you may implement in your regex: Python re (and regex, too) optimize matching at word boundary locations. Thus, if your pattern is supposed to match at a word boundary, always start your pattern with it. In your case, r'\b[\w.-]{1,100}#[\w.-]{1,100}' or r'\b\w[\w.-]{0,99}#[\w.-]{1,100}' should also work much better than the original pattern without a word boundary.
Python test:
import re, regex, timeit
text='your_long_sting'
re_pattern=re.compile(r'\b\w[\w.-]{0,99}#[\w.-]{1,100}')
regex_pattern=regex.compile(r'\b\w[\w.-]{0,99}#[\w.-]{1,100}')
timeit.timeit("p.findall(text)", 'from __main__ import text, re_pattern as p', number=100000)
# => 6034.659449000001
timeit.timeit("p.findall(text)", 'from __main__ import text, regex_pattern as p', number=100000)
# => 218.1561693
Don't use regex on the whole string. Regex are slow. Avoiding them is your best bet to better overall performance.
My first approach would look like this:
Split the string on spaces.
Filter the result down to the parts that contain #.
Create a pre-compiled regex.
Use regex on the remaining parts only to remove false positives.
Another idea:
in a loop....
use .index("#") to find the position of the next candidate
extend e.g. 100 characters to the left, 50 to the right to cover name and domain
adapt the range depending on the last email address you found so you don't overlap
check the range with a regex, if it matches, yield the match

Counting Python Regex Matches Efficiently

I need to efficiently counter Python Regex matches. Findall and finditer works, but are slow for large number of searches.
So far I have this (much simplified version):
import re
testItem = re.compile(r"var")
for matches in testItem.finditer(stringData):
counter = counter + 1
return counter
I am running multiple instances of this method. I do not care for the matches themselves; I simply wish to return the counter.
The issue is that the stringData is a very large string. The Regex itself is pretty simple.
Please advise on a more efficient way to do this.
Thanks in advance.
Since you show in your edit that you're just looking for a substring,
stringData.count('var')
should serve you well.
Of course, this does not generalize to many other uses of REs! Unfortunately, at least as of Python 3.4, re.finditer returns an iterator which does not support the "length hint" formalized by PEP 424, so there aren't many good alternatives (for the general case) to
sum(1 for _ in there.finditer(stringData))

How do I use re.search starting from a certain index in the string?

Seems like a simple thing but I'm not seeing it. How do I start the search in the middle of a string?
The re.search function doesn't take a start argument like the str methods do. But search method of a compiled re.compile/re.RegexObject pattern does take a pos argument.
This makes sense if you think about it. If you really need to use the same regular expressions over and over, you probably should be compiling them. Not so much for efficiency—the cache works nicely for most applications—but just for readability.
But what if you need to use the top-level function, because you can't pre-compile your patterns for some reason?
Well, there are plenty of third-party regular expression libraries. Some of these wrap PCRE or Google's RE2 or ICU, some implement regular expressions from scratch, and they all have at least slightly different, sometimes radically different, APIs.
But the regex module, which is being designed to be an eventual replacement for re in the stdlib (although it's been bumped a couple times now because it's not quite ready) is pretty much usable as a drop-in replacement for re, and (among other extensions) it takes pos and endpos arguments on its search function.
Normally, the most common reason you'd want to do this is to "find the next match after the one I just found", and there's a much easier way to do that: use finditer instead of search.
For example, this str-method loop:
i = 0
while True:
i = s.find(sub, i)
if i == -1:
break
do_stuff_with(s, i)
… translates to this much nicer regex loop:
for match in re.finditer(pattern, s):
do_stuff_with(match)
When that isn't appropriate, you can always slice the string:
match = re.search(pattern, s[index:])
But that makes an extra copy of half your string, which could be a problem if string is actually, say, a 12GB mmap. (Of course for the 12GB mmap case, you'd probably want to map a new window… but there are cases where that won't help.)
Finally, you can always just modify your pattern to skip over index characters:
match = re.search('.{%d}%s' % (index, pattern), s)
All I've done here is to add, e.g., .{20} to the start of the pattern, which means to match exactly 20 of any character, plus whatever else you were trying to match. Here's a simple example:
.{3}(abc)
Debuggex Demo
If I give this abcdefabcdef, it will match the first 'abc' after the 3rd character—that is, the second abc.
But notice that what it actually matches 'defabc'. Because I'm using capture groups for my real pattern, and I'm not putting the .{3} in a group, match.group(1) and so on will work exactly as I'd want them to, but match.group(0) will give me the wrong thing. If that matters, you need lookbehind.

Find which lines in a file contain certain characters

Is there a way to find out if a string contains any one of the characters in a set with python?
It's straightforward to do it with a single character, but I need to check and see if a string contains any one of a set of bad characters.
Specifically, suppose I have a string:
s = 'amanaplanacanalpanama~012345'
and I want to see if the string contains any vowels:
bad_chars = 'aeiou'
and do this in a for loop for each line in a file:
if [any one or more of the bad_chars] in s:
do something
I am scanning a large file so if there is a faster method to this, that would be ideal. Also, not every bad character has to be checked---so long as one is encountered that is enough to end the search.
I'm not sure if there is a builtin function or easy way to implement this, but I haven't come across anything yet. Any pointers would be much appreciated!
any((c in badChars) for c in yourString)
or
any((c in yourString) for c in badChars) # extensionally equivalent, slower
or
set(yourString) & set(badChars) # extensionally equivalent, slower
"so long as one is encountered that is enough to end the search." - This will be true if you use the first method.
You say you are concerned with performance: performance should not be an issue unless you are dealing with a huge amount of data. If you encounter issues, you can try:
Regexes
edit Previously I had written a section here on using regexes, via the re module, programatically generating a regex that consisted of a single character-class [...] and using .finditer, with the caveat that putting a simple backslash before everything might not work correctly. Indeed, after testing it, that is the case, and I would definitely not recommend this method. Using this would require reverse engineering the entire (slightly complex) sub-grammar of regex character classes (e.g. you might have characters like \ followed by w, like ] or [, or like -, and merely escaping some like \w may give it a new meaning).
Sets
Depending on whether the str.__contains__ operation is O(1) or O(N), it may be justifiable to first convert your text/lines into a set to ensure the in operation is O(1), if you have many badChars:
badCharSet = set(badChars)
any((c in badChars) for c in yourString)
(it may be possible to make that a one-liner any((c in set(yourString)) for c in badChars), depending on how smart the python compiler is)
Do you really need to do this line-by-line?
It may be faster to do this once for the entire file O(#badchars), than once for every line in the file O(#lines*#badchars), though the asymptotic constants may be such that it won't matter.
Use python's any function.
if any((bad_char in my_string) for bad_char in bad_chars):
# do something
This should be very efficient and clear. It uses sets:
#!/usr/bin/python
bad_chars = set('aeiou')
with open('/etc/passwd', 'r') as file_:
file_string = file_.read()
file_chars = set(file_string)
if file_chars & bad_chars:
print('found something bad')
This regular expression is twice as fast as any with my minimal testing. You should try it with your own data.
r = re.compile('[aeiou]')
if r.search(s):
# do something
The following Python code should print out any character in bad_chars if it exists in s:
for i in vowels:
if i in your charset:
#do_something
You could also use the python in-built any using an example like this:
>>> any(e for e in bad_chars if e in s)
True

python regex speed

regarding regex (specifically python re), if we ignore the way the expression is written, is the length of the text the only factor for the time required to process the document? Or are there other factors (like how the text is structured) that play important roles too?
One important consideration can also be whether the text actually matches the regular expression. Take (as a contrived example) the regex (x+x+)+y from this regex tutorial.
When applied to xxxxxxxxxxy it matches, taking the regex engine 7 steps. When applied to xxxxxxxxxx, it fails (of course), but it takes the engine 2558 steps to arrive at this conclusion.
For xxxxxxxxxxxxxxy vs. xxxxxxxxxxxxxx it's already 7 vs 40958 steps, and so on exponentially...
This happens especially easily with nested repetitions or regexes where the same text can be matched by two or more different parts of the regex, forcing the engine to try all permutations before being able to declare failure. This is then called catastrophic backtracking.
Both the length of the text and its contents are important.
As an example the regular expression a+b will fail to match quickly on a string containing one million bs but more slowly on a string containing one million as. This is because more backtracking will be required in the second case.
import timeit
x = "re.search('a+b', s)"
print timeit.timeit(x, "import re;s='a'*10000", number=10)
print timeit.timeit(x, "import re;s='b'*10000", number=10)
Results:
6.85791902323
0.00795443275612
To refactor a regex to create a multi-level trie covers 95% of of the
800% increase in performance. The other 5% involves factoring to not only facilitate
the trie but to enhance it to give a possible 30x performance boost.

Categories

Resources