Im trying to do something fairly simple with regular expression in python... thats what i thought at least.
What i want to do is matching words from a string if its preceded and followed by a whitespace. If its at the beginning of the string there is no whitespace required before - if its at the end, dont't search for whitespace either.
Example:
"WordA WordB WordC-WordD WordE"
I want to match WordA WordB WordE.
I only came up with overcomplicated way of doing this...
(?<=(?<=^)|(?<=\s))\w+(?=(?=\s)|(?=$))
It seems to me there has to be a simple way for such a simple problem....
I figured i can just start with (?<=\s|^) but that doesnt seem possible because "look-behind requires fixed-width pattern".
You seem to work in Python as (?<=^|\s) is perfectly valid in PCRE, Java and Ruby (and .NET regex supports infinite width lookbehind patterns).
Use
(?<!\S)\w+(?!\S)
It will match 1 or more word chars that are enclosed with whitespace or start/end of string.
See the regex demo.
Pattern details:
(?<!\S) - a negative lookbehind that fails the match once the engine finds a non-whitespace char immediately to the left of the current location
\w+ - 1 or more word chars
(?!\S) - a negative lookahead that fails the match once the engine finds a non-whitespace char immediately to the right of the current location.
Related
I have a string. The end is different, such as index.php?test=1&list=UL or index.php?list=UL&more=1. The one thing I'm looking for is &list=.
How can I match it, whether it's in the middle of the string or it's at the end? So far I've got [&|\?]list=.*?([&|$]), but the ([&|$]) part doesn't actually work; I'm trying to use that to match either & or the end of the string, but the end of the string part doesn't work, so this pattern matches the second example but not the first.
Use:
/(&|\?)list=.*?(&|$)/
Note that when you use a bracket expression, every character within it (with some exceptions) is going to be interpreted literally. In other words, [&|$] matches the characters &, |, and $.
In short
Any zero-width assertions inside [...] lose their meaning of a zero-width assertion. [\b] does not match a word boundary (it matches a backspace, or, in POSIX, \ or b), [$] matches a literal $ char, [^] is either an error or, as in ECMAScript regex flavor, any char. Same with \z, \Z, \A anchors.
You may solve the problem using any of the below patterns:
[&?]list=([^&]*)
[&?]list=(.*?)(?=&|$)
[&?]list=(.*?)(?![^&])
If you need to check for the "absolute", unambiguous string end anchor, you need to remember that is various regex flavors, it is expressed with different constructs:
[&?]list=(.*?)(?=&|$) - OK for ECMA regex (JavaScript, default C++ `std::regex`)
[&?]list=(.*?)(?=&|\z) - OK for .NET, Go, Onigmo (Ruby), Perl, PCRE (PHP, base R), Boost, ICU (R `stringr`), Java/Andorid
[&?]list=(.*?)(?=&|\Z) - OK for Python
Matching between a char sequence and a single char or end of string (current scenario)
The .*?([YOUR_SINGLE_CHAR_DELIMITER(S)]|$) pattern (suggested by João Silva) is rather inefficient since the regex engine checks for the patterns that appear to the right of the lazy dot pattern first, and only if they do not match does it "expand" the lazy dot pattern.
In these cases it is recommended to use negated character class (or bracket expression in the POSIX talk):
[&?]list=([^&]*)
See demo. Details
[&?] - a positive character class matching either & or ? (note the relationships between chars/char ranges in a character class are OR relationships)
list= - a substring, char sequence
([^&]*) - Capturing group #1: zero or more (*) chars other than & ([^&]), as many as possible
Checking for the trailing single char delimiter presence without returning it or end of string
Most regex flavors (including JavaScript beginning with ECMAScript 2018) support lookarounds, constructs that only return true or false if there patterns match or not. They are crucial in case consecutive matches that may start and end with the same char are expected (see the original pattern, it may match a string starting and ending with &). Although it is not expected in a query string, it is a common scenario.
In that case, you can use two approaches:
A positive lookahead with an alternation containing positive character class: (?=[SINGLE_CHAR_DELIMITER(S)]|$)
A negative lookahead with just a negative character class: (?![^SINGLE_CHAR_DELIMITER(S)])
The negative lookahead solution is a bit more efficient because it does not contain an alternation group that adds complexity to matching procedure. The OP solution would look like
[&?]list=(.*?)(?=&|$)
or
[&?]list=(.*?)(?![^&])
See this regex demo and another one here.
Certainly, in case the trailing delimiters are multichar sequences, only a positive lookahead solution will work since [^yes] does not negate a sequence of chars, but the chars inside the class (i.e. [^yes] matches any char but y, e and s).
I am getting a weird situation in Python 3. Attempting to find some characters (a-z) that have one period and space(s) in between followed by a section sign (0-2 times) followed by a space, followed by digits (1 or more times).
When running this in Python, I don't get a result. It just get hung up.
import re
string = "Guidelines on International Protection: Membership in a “Particular Social Group” Within the Context of Article 1A(2) of the 1951 Convention and/or its 1967 Protocol Relating to the Status
...: of Refugees (UNHCR Guidelines), U.N. Doc. HCR/GIP/02/02 (May 7, 2002)"
regex = r'((?:[a-z]+\.{0,1}\s*)+)(§{0,2})\s+(\d+)'
compile = re.compile(regex, flags=re.IGNORECASE)
re.findall(compile,string)
In https://regexr.com/, it actually gives me correct matches: https://regexr.com/53394
What is happening?
If you test your regex in a Python-compatible regex tester, you will see it takes a huge amount of steps to complete matching and ends in catastrophic backtracking. This is caused by the (?:[a-z]+\.{0,1}\s*)+ pattern that is not at the pattern end. It forces the regex engine to backtrack heavily when the consequent (§{0,2})\s+(\d+) is not found.
The fix is to make the pattern match so that each subsequent pattern could not match the same text at the same location:
([a-z]+(?:(?:\s*\.\s*|\s+)[a-z]+)*)(§{0,2})\s+(\d+)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
See the regex demo
Here, (?:[a-z]+\.{0,1}\s*)+ is replaced with [a-z]+(?:(?:\s*\.\s*|\s+)[a-z]+)*:
[a-z]+ - 1+ ASCII letters (mind the re.IGNORECASE modifier you are using)
(?:(?:\s*\.\s*|\s+)[a-z]+)* - 0 or more occurrences of
(?:\s*\.\s*|\s+) - a . enclosed with 0+ whitespaces or 1+ whitespaces
[a-z]+ - 1+ ASCII letters.
An effective solution to your problem of catastrophic backtracking is to simulate an atomic group (not supported by Python's re module) by using a positive lookahead with a capture group:
(?=((?:[a-z]+\.?\s*)+))\1§{0,2}\s*\d+
with multiline and case-indifferent flags set. This avoids catastrophic backtracking as the lookahead is not subject to giving up characters.
Demo
A cleaner way would be to use a possessive quantifier when matching one or more occurrences of a letter. One could use the regular expression
(?:[a-z]++\.?\s*)+§{0,2}\s*\d+
with multiline and case-indifferent flags set. Here the possessive modifier is indicated by ++.
Demo
Unfortunately, Python's re module does not support possessive modifiers, but its alternative PyPi module does.
Suppose we had used the regex (?:[a-z]+\.?\s*)+§{0,2}\s*\d+, which does not have a possessive modifier.
For the string
Within the Context. of Article 1A(2)
[a-z]+ matches Within the Context, then \.?\s* matches .. The regex engine then continues trying to find a match for the entire regex. If it fails it will backtrack to consider alternatives and may return to the match of [a-z]+, at which point it would give back the character t (making that match Within the Contex), reset the regex engine's internal pointer right before the t, and once again continue forward.
The possessive modifier ++ in prevents the regex engine from giving up any characters in its original match of Within the Context, thereby avoid the catastrophic backtracking problem. Even where catastrophic backtracking is not an issue, possessive modifiers, where they can be used, can greatly improve efficiency.
I am trying to create a regex expression in Python for non-hyphenated words but I am unable to figure out the right syntax.
The requirements for the regex are:
It should not contain hyphens AND
It should contain atleast 1 number
The expressions that I tried are:=
^(?!.*-)
This matches all non-hyphenated words but I am not able to figure out how to additionally add the second condition.
^(?!.*-(?=/d{1,}))
I tried using double lookahead but I am not sure about the syntax to use for it. This matches ID101 but also matches STACKOVERFLOW
Sample Words Which Should Match:
1DRIVE , ID100 , W1RELESS
Sample Words Which Should Not Match:
Basically any non-numeric string (like STACK , OVERFLOW) or any hyphenated words (Test-11 , 24-hours)
Additional Info:
I am using library re and compiling the regex patterns and using re.search for matching.
Any assistance would be very helpful as I am new to regex matching and am stuck on this for quite a few hours.
Maybe,
(?!.*-)(?=.*\d)^.+$
might simply work OK.
Test
import re
string = '''
abc
abc1-
abc1
abc-abc1
'''
expression = r'(?m)(?!.*-)(?=.*\d)^.+$'
print(re.findall(expression, string))
Output
['abc1']
If you wish to simplify/modify/explore the expression, it's been explained on the top right panel of regex101.com. If you'd like, you can also watch in this link, how it would match against some sample inputs.
RegEx Circuit
jex.im visualizes regular expressions:
RegEx 101 Explanation
/
(?!.*-)(?=.*\d)^.+$
/
gm
Negative Lookahead (?!.*-)
Assert that the Regex below does not match
.* matches any character (except for line terminators)
* Quantifier — Matches between zero and unlimited times, as many times as possible, giving back as needed (greedy)
- matches the character - literally (case sensitive)
Positive Lookahead (?=.*\d)
Assert that the Regex below matches
.* matches any character (except for line terminators)
* Quantifier — Matches between zero and unlimited times, as many times as possible, giving back as needed (greedy)
\d matches a digit (equal to [0-9])
^ asserts position at start of a line
.+ matches any character (except for line terminators)
+ Quantifier — Matches between one and unlimited times, as many times as possible, giving back as needed (greedy)
$ asserts position at the end of a line
Global pattern flags
g modifier: global. All matches (don't return after first match)
m modifier: multi line. Causes ^ and $ to match the begin/end of each line (not only begin/end of string)
I came up with -
^[^-]*\d[^-]*$
so we need at LEAST one digit (\d)
We need the rest of the string to contain anything BUT a - ([^-])
We can have unlimited number of those characters, so [^-]*
but putting them together like [^-]*\d would fail on aaa3- because the - comes after a valid match- lets make sure no dashes can sneak in before or after our match ^[-]*\d$
Unfortunately that means that aaa555D fails. So we actually need to add the first group again- ^[^-]*\d[^-]$ --- which says start - any number of chars that aren't dashes - a digit - any number of chars that aren't dashes - end
Depending on style, we could also do ^([^-]*\d)+$ since the order of the digits/numbers dont matter, we can have as many of those as we want.
However, finally... this is how I would ACTUALLY solve this particular problem, since regexes may be powerful, but they tend to make the code harder to understand...
if ("-" not in text) and re.search("\d", text):
I have a string. The end is different, such as index.php?test=1&list=UL or index.php?list=UL&more=1. The one thing I'm looking for is &list=.
How can I match it, whether it's in the middle of the string or it's at the end? So far I've got [&|\?]list=.*?([&|$]), but the ([&|$]) part doesn't actually work; I'm trying to use that to match either & or the end of the string, but the end of the string part doesn't work, so this pattern matches the second example but not the first.
Use:
/(&|\?)list=.*?(&|$)/
Note that when you use a bracket expression, every character within it (with some exceptions) is going to be interpreted literally. In other words, [&|$] matches the characters &, |, and $.
In short
Any zero-width assertions inside [...] lose their meaning of a zero-width assertion. [\b] does not match a word boundary (it matches a backspace, or, in POSIX, \ or b), [$] matches a literal $ char, [^] is either an error or, as in ECMAScript regex flavor, any char. Same with \z, \Z, \A anchors.
You may solve the problem using any of the below patterns:
[&?]list=([^&]*)
[&?]list=(.*?)(?=&|$)
[&?]list=(.*?)(?![^&])
If you need to check for the "absolute", unambiguous string end anchor, you need to remember that is various regex flavors, it is expressed with different constructs:
[&?]list=(.*?)(?=&|$) - OK for ECMA regex (JavaScript, default C++ `std::regex`)
[&?]list=(.*?)(?=&|\z) - OK for .NET, Go, Onigmo (Ruby), Perl, PCRE (PHP, base R), Boost, ICU (R `stringr`), Java/Andorid
[&?]list=(.*?)(?=&|\Z) - OK for Python
Matching between a char sequence and a single char or end of string (current scenario)
The .*?([YOUR_SINGLE_CHAR_DELIMITER(S)]|$) pattern (suggested by João Silva) is rather inefficient since the regex engine checks for the patterns that appear to the right of the lazy dot pattern first, and only if they do not match does it "expand" the lazy dot pattern.
In these cases it is recommended to use negated character class (or bracket expression in the POSIX talk):
[&?]list=([^&]*)
See demo. Details
[&?] - a positive character class matching either & or ? (note the relationships between chars/char ranges in a character class are OR relationships)
list= - a substring, char sequence
([^&]*) - Capturing group #1: zero or more (*) chars other than & ([^&]), as many as possible
Checking for the trailing single char delimiter presence without returning it or end of string
Most regex flavors (including JavaScript beginning with ECMAScript 2018) support lookarounds, constructs that only return true or false if there patterns match or not. They are crucial in case consecutive matches that may start and end with the same char are expected (see the original pattern, it may match a string starting and ending with &). Although it is not expected in a query string, it is a common scenario.
In that case, you can use two approaches:
A positive lookahead with an alternation containing positive character class: (?=[SINGLE_CHAR_DELIMITER(S)]|$)
A negative lookahead with just a negative character class: (?![^SINGLE_CHAR_DELIMITER(S)])
The negative lookahead solution is a bit more efficient because it does not contain an alternation group that adds complexity to matching procedure. The OP solution would look like
[&?]list=(.*?)(?=&|$)
or
[&?]list=(.*?)(?![^&])
See this regex demo and another one here.
Certainly, in case the trailing delimiters are multichar sequences, only a positive lookahead solution will work since [^yes] does not negate a sequence of chars, but the chars inside the class (i.e. [^yes] matches any char but y, e and s).
>>> d = "Batman,Superman"
>>> m = re.search("(?<!Bat)\w+",d)
>>> m.group(0)
'Batman'
Why isn't group(0) matching Superman? This lookaround tutorial says:
(?<!a)b matches a "b" that is not
preceded by an "a", using negative
lookbehind
Batman isn't directly preceded by Bat, so that matches first. In fact, neither is Superman; there's a comma in-between in your string which will do just fine to allow that RE to match, but that's not matched anyway because it's possible to match earlier in the string.
Maybe this will explain better: if the string was Batman and you were starting to try to match from the m, the RE would not match until the character after (giving a match of an) because that's the only place in the string which is preceded by Bat.
At a simple level, the regex engine starts from the left of the string and moves progressively towards the right, trying to match your pattern (think of it like a cursor moving through the string). In the case of a lookaround, at each stop of the cursor, the lookaround is asserted, and if true, the engine continues trying to make a match. As soon as the engine can match your pattern, it'll return a match.
At position 0 of your string (ie. prior to the B in Batman), the assertion succeeded, as Bat is not present before the current position - thus, \w+ can match the entire word Batman (remember, regexes are inherently greedy - ie. will match as much as possible).
See this page for more information on engine internals.
To achieve what you wanted, you could instead use something like:
\b(?!Bat)\w+
In this pattern, the engine will match a word boundary (\b)1, followed by one or more word characters, with the assertion that the word characters do not start with Bat. A lookahead is used rather than a lookbehind because using a lookbehind here would have the same problem as your original pattern; it would look before the position directly following the word boundary, and since its already been determined that the position before the cursor is a word boundary, the negative lookbehind would always succeed.
1 Note that word boundaries match a boundary between \w and \W (ie. between [A-Za-z0-9_] and any other character; it also matches the ^ and $ anchors). If your boundaries need to be more complex, you'll need a different way of anchoring your pattern.
From the manual:
Patterns which start with negative
lookbehind assertions may match at the
beginning of the string being
searched.
http://docs.python.org/library/re.html#regular-expression-syntax
You're looking for the first set of one or more alphanumeric characters (\w+) that is not preceded by 'Bat'. Batman is the first such match. (Note that negative lookbehind assertions can match the start of a string.)
To do what you want, you have to constrain the regex to match 'man' specifically; otherwise, as others have pointed out, \w greedily matches anything including 'Batman'. As in:
>>> re.search("\w+(?<!Bat)man","Batman,Superman").group(0)
'Superman'