Related
This question's answers are a community effort. Edit existing answers to improve this post. It is not currently accepting new answers or interactions.
What is this?
This is a collection of common Q&A. This is also a Community Wiki, so everyone is invited to participate in maintaining it.
Why is this?
regex is suffering from give me ze code type of questions and poor answers with no explanation. This reference is meant to provide links to quality Q&A.
What's the scope?
This reference is meant for the following languages: php, perl, javascript, python, ruby, java, .net.
This might be too broad, but these languages share the same syntax. For specific features there's the tag of the language behind it, example:
What are regular expression Balancing Groups? .net
The Stack Overflow Regular Expressions FAQ
See also a lot of general hints and useful links at the regex tag details page.
Online tutorials
RegexOne ↪
Regular Expressions Info ↪
Quantifiers
Zero-or-more: *:greedy, *?:reluctant, *+:possessive
One-or-more: +:greedy, +?:reluctant, ++:possessive
?:optional (zero-or-one)
Min/max ranges (all inclusive): {n,m}:between n & m, {n,}:n-or-more, {n}:exactly n
Differences between greedy, reluctant (a.k.a. "lazy", "ungreedy") and possessive quantifier:
Greedy vs. Reluctant vs. Possessive Quantifiers
In-depth discussion on the differences between greedy versus non-greedy
What's the difference between {n} and {n}?
Can someone explain Possessive Quantifiers to me? php, perl, java, ruby
Emulating possessive quantifiers .net
Non-Stack Overflow references: From Oracle, regular-expressions.info
Character Classes
What is the difference between square brackets and parentheses?
[...]: any one character, [^...]: negated/any character but
[^] matches any one character including newlines javascript
[\w-[\d]] / [a-z-[qz]]: set subtraction .net, xml-schema, xpath, JGSoft
[\w&&[^\d]]: set intersection java, ruby 1.9+
[[:alpha:]]:POSIX character classes
[[:<:]] and [[:>:]] Word boundaries
Why do [^\\D2], [^[^0-9]2], [^2[^0-9]] get different results in Java? java
Shorthand:
Digit: \d:digit, \D:non-digit
Word character (Letter, digit, underscore): \w:word character, \W:non-word character
Whitespace: \s:whitespace, \S:non-whitespace
Unicode categories (\p{L}, \P{L}, etc.)
Escape Sequences
Horizontal whitespace: \h:space-or-tab, \t:tab
Newlines:
\r, \n:carriage return and line feed
\R:generic newline php java-8
Negated whitespace sequences: \H:Non horizontal whitespace character, \V:Non vertical whitespace character, \N:Non line feed character pcre php5 java-8
Other: \v:vertical tab, \e:the escape character
Anchors
anchor
matches
flavors
^
Start of string
Common*
^
Start of line
Commonm
$
End of line
Commonm
$
End of text
Common* except javascript
$
Very end of string
javascript*, phpD
\A
Start of string
Common except javascript
\Z
End of text
Common except javascript python
\Z
Very end of string
python
\z
Very end of string
Common except javascript python
\b
Word boundary
Common
\B
Not a word boundary
Common
\G
End of previous match
Common except javascript, python
Term
Definition
Start of string
At the very start of the string.
Start of line
At the very start of the string, andafter a non-terminal line terminator.
Very end of string
At the very end of the string.
End of text
At the very end of the string, andat a terminal line terminator.
End of line
At the very end of the string, andat a line terminator.
Word boundary
At a word character not preceded by a word character, andat a non-word character not preceded by a non-word character.
End of previous match
At a previously set position, usually where a previous match ended.At the very start of the string if no position was set.
"Common" refers to the following: icu java javascript .net objective-c pcre perl php python swift ruby
* Default |
m Multi-line mode. |
D Dollar end only mode.
Groups
(...):capture group, (?:):non-capture group
Why is my repeating capturing group only capturing the last match?
\1:backreference and capture-group reference, $1:capture group reference
What's the meaning of a number after a backslash in a regular expression?
\g<1>123:How to follow a numbered capture group, such as \1, with a number?: python
What does a subpattern (?i:regex) mean?
What does the 'P' in (?P<group_name>regexp) mean?
(?>):atomic group or independent group, (?|):branch reset
Equivalent of branch reset in .NET/C# .net
Named capture groups:
General named capturing group reference at regular-expressions.info
java: (?<groupname>regex): Overview and naming rules (Non-Stack Overflow links)
Other languages: (?P<groupname>regex) python, (?<groupname>regex) .net, (?<groupname>regex) perl, (?P<groupname>regex) and (?<groupname>regex) php
Lookarounds
Lookaheads: (?=...):positive, (?!...):negative
Lookbehinds: (?<=...):positive, (?<!...):negative
Lookbehind limits in:
Lookbehinds need to be constant-length php, perl, python, ruby
Lookarounds of limited length {0,n} java
Variable length lookbehinds are allowed .net
Lookbehind alternatives:
Using \K php, perl (Flavors that support \K)
Alternative regex module for Python python
The hacky way
JavaScript negative lookbehind equivalents External link
Modifiers
flag
modifier
flavors
a
ASCII
python
c
current position
perl
e
expression
php perl
g
global
most
i
case-insensitive
most
m
multiline
php perl python javascript .net java
m
(non)multiline
ruby
o
once
perl ruby
r
non-destructive
perl
S
study
php
s
single line
ruby
U
ungreedy
php r
u
unicode
most
x
whitespace-extended
most
y
sticky ↪
javascript
How to convert preg_replace e to preg_replace_callback?
What are inline modifiers?
What is '?-mix' in a Ruby Regular Expression
Other:
|:alternation (OR) operator, .:any character, [.]:literal dot character
What special characters must be escaped?
Control verbs (php and perl): (*PRUNE), (*SKIP), (*FAIL) and (*F)
php only: (*BSR_ANYCRLF)
Recursion (php and perl): (?R), (?0) and (?1), (?-1), (?&groupname)
Common Tasks
Get a string between two curly braces: {...}
Match (or replace) a pattern except in situations s1, s2, s3...
How do I find all YouTube video ids in a string using a regex?
Validation:
Internet: email addresses, URLs (host/port: regex and non-regex alternatives), passwords
Numeric: a number, min-max ranges (such as 1-31), phone numbers, date
Parsing HTML with regex: See "General Information > When not to use Regex"
Advanced Regex-Fu
Strings and numbers:
Regular expression to match a line that doesn't contain a word
How does this PCRE pattern detect palindromes?
Match strings whose length is a fourth power
How does this regex find triangular numbers?
How to determine if a number is a prime with regex?
How to match the middle character in a string with regex?
Other:
How can we match a^n b^n?
Match nested brackets
Using a recursive pattern php, perl
Using balancing groups .net
“Vertical” regex matching in an ASCII “image”
List of highly up-voted regex questions on Code Golf
How to make two quantifiers repeat the same number of times?
An impossible-to-match regular expression: (?!a)a
Match/delete/replace this except in contexts A, B and C
Match nested brackets with regex without using recursion or balancing groups?
Flavor-Specific Information
(Except for those marked with *, this section contains non-Stack Overflow links.)
Java
Official documentation: Pattern Javadoc ↪, Oracle's regular expressions tutorial ↪
The differences between functions in java.util.regex.Matcher:
matches()): The match must be anchored to both input-start and -end
find()): A match may be anywhere in the input string (substrings)
lookingAt(): The match must be anchored to input-start only
(For anchors in general, see the section "Anchors")
The only java.lang.String functions that accept regular expressions: matches(s), replaceAll(s,s), replaceFirst(s,s), split(s), split(s,i)
*An (opinionated and) detailed discussion of the disadvantages of and missing features in java.util.regex
.NET
How to read a .NET regex with look-ahead, look-behind, capturing groups and back-references mixed together?
Official documentation:
Boost regex engine: General syntax, Perl syntax (used by TextPad, Sublime Text, UltraEdit, ...???)
JavaScript general info and RegExp object
.NET MySQL Oracle Perl5 version 18.2
PHP: pattern syntax, preg_match
Python: Regular expression operations, search vs match, how-to
Rust: crate regex, struct regex::Regex
Splunk: regex terminology and syntax and regex command
Tcl: regex syntax, manpage, regexp command
Visual Studio Find and Replace
General information
(Links marked with * are non-Stack Overflow links.)
Other general documentation resources: Learning Regular Expressions, *Regular-expressions.info, *Wikipedia entry, *RexEgg, Open-Directory Project
DFA versus NFA
Generating Strings matching regex
Books: Jeffrey Friedl's Mastering Regular Expressions
When to not use regular expressions:
Some people, when confronted with a problem, think "I know, I'll use regular expressions." Now they have two problems. (blog post written by Stack Overflow's founder)*
Do not use regex to parse HTML:
Don't. Please, just don't
Well, maybe...if you're really determined (other answers in this question are also good)
Examples of regex that can cause regex engine to fail
Why does this regular expression kill the Java regex engine?
Tools: Testers and Explainers
(This section contains non-Stack Overflow links.)
Online (* includes replacement tester, + includes split tester):
Debuggex (Also has a repository of useful regexes) javascript, python, pcre
*Regular Expressions 101 php, pcre, python, javascript, java
Regex Pal, regular-expressions.info javascript
Rubular ruby RegExr Regex Hero dotnet
*+ regexstorm.net .net
*RegexPlanet: Java java, Go go, Haskell haskell, JavaScript javascript, .NET dotnet, Perl perl php PCRE php, Python python, Ruby ruby, XRegExp xregexp
freeformatter.com xregexp
*+regex.larsolavtorvik.com php PCRE and POSIX, javascript
Offline:
Microsoft Windows: RegexBuddy (analysis), RegexMagic (creation), Expresso (analysis, creation, free)
MySQL 8.0: Various syntax changes were made. Note especially the doubling of backslashes in some contexts. (This Answer need further editing to reflect the differences.)
This question's answers are a community effort. Edit existing answers to improve this post. It is not currently accepting new answers or interactions.
What is this?
This is a collection of common Q&A. This is also a Community Wiki, so everyone is invited to participate in maintaining it.
Why is this?
regex is suffering from give me ze code type of questions and poor answers with no explanation. This reference is meant to provide links to quality Q&A.
What's the scope?
This reference is meant for the following languages: php, perl, javascript, python, ruby, java, .net.
This might be too broad, but these languages share the same syntax. For specific features there's the tag of the language behind it, example:
What are regular expression Balancing Groups? .net
The Stack Overflow Regular Expressions FAQ
See also a lot of general hints and useful links at the regex tag details page.
Online tutorials
RegexOne ↪
Regular Expressions Info ↪
Quantifiers
Zero-or-more: *:greedy, *?:reluctant, *+:possessive
One-or-more: +:greedy, +?:reluctant, ++:possessive
?:optional (zero-or-one)
Min/max ranges (all inclusive): {n,m}:between n & m, {n,}:n-or-more, {n}:exactly n
Differences between greedy, reluctant (a.k.a. "lazy", "ungreedy") and possessive quantifier:
Greedy vs. Reluctant vs. Possessive Quantifiers
In-depth discussion on the differences between greedy versus non-greedy
What's the difference between {n} and {n}?
Can someone explain Possessive Quantifiers to me? php, perl, java, ruby
Emulating possessive quantifiers .net
Non-Stack Overflow references: From Oracle, regular-expressions.info
Character Classes
What is the difference between square brackets and parentheses?
[...]: any one character, [^...]: negated/any character but
[^] matches any one character including newlines javascript
[\w-[\d]] / [a-z-[qz]]: set subtraction .net, xml-schema, xpath, JGSoft
[\w&&[^\d]]: set intersection java, ruby 1.9+
[[:alpha:]]:POSIX character classes
[[:<:]] and [[:>:]] Word boundaries
Why do [^\\D2], [^[^0-9]2], [^2[^0-9]] get different results in Java? java
Shorthand:
Digit: \d:digit, \D:non-digit
Word character (Letter, digit, underscore): \w:word character, \W:non-word character
Whitespace: \s:whitespace, \S:non-whitespace
Unicode categories (\p{L}, \P{L}, etc.)
Escape Sequences
Horizontal whitespace: \h:space-or-tab, \t:tab
Newlines:
\r, \n:carriage return and line feed
\R:generic newline php java-8
Negated whitespace sequences: \H:Non horizontal whitespace character, \V:Non vertical whitespace character, \N:Non line feed character pcre php5 java-8
Other: \v:vertical tab, \e:the escape character
Anchors
anchor
matches
flavors
^
Start of string
Common*
^
Start of line
Commonm
$
End of line
Commonm
$
End of text
Common* except javascript
$
Very end of string
javascript*, phpD
\A
Start of string
Common except javascript
\Z
End of text
Common except javascript python
\Z
Very end of string
python
\z
Very end of string
Common except javascript python
\b
Word boundary
Common
\B
Not a word boundary
Common
\G
End of previous match
Common except javascript, python
Term
Definition
Start of string
At the very start of the string.
Start of line
At the very start of the string, andafter a non-terminal line terminator.
Very end of string
At the very end of the string.
End of text
At the very end of the string, andat a terminal line terminator.
End of line
At the very end of the string, andat a line terminator.
Word boundary
At a word character not preceded by a word character, andat a non-word character not preceded by a non-word character.
End of previous match
At a previously set position, usually where a previous match ended.At the very start of the string if no position was set.
"Common" refers to the following: icu java javascript .net objective-c pcre perl php python swift ruby
* Default |
m Multi-line mode. |
D Dollar end only mode.
Groups
(...):capture group, (?:):non-capture group
Why is my repeating capturing group only capturing the last match?
\1:backreference and capture-group reference, $1:capture group reference
What's the meaning of a number after a backslash in a regular expression?
\g<1>123:How to follow a numbered capture group, such as \1, with a number?: python
What does a subpattern (?i:regex) mean?
What does the 'P' in (?P<group_name>regexp) mean?
(?>):atomic group or independent group, (?|):branch reset
Equivalent of branch reset in .NET/C# .net
Named capture groups:
General named capturing group reference at regular-expressions.info
java: (?<groupname>regex): Overview and naming rules (Non-Stack Overflow links)
Other languages: (?P<groupname>regex) python, (?<groupname>regex) .net, (?<groupname>regex) perl, (?P<groupname>regex) and (?<groupname>regex) php
Lookarounds
Lookaheads: (?=...):positive, (?!...):negative
Lookbehinds: (?<=...):positive, (?<!...):negative
Lookbehind limits in:
Lookbehinds need to be constant-length php, perl, python, ruby
Lookarounds of limited length {0,n} java
Variable length lookbehinds are allowed .net
Lookbehind alternatives:
Using \K php, perl (Flavors that support \K)
Alternative regex module for Python python
The hacky way
JavaScript negative lookbehind equivalents External link
Modifiers
flag
modifier
flavors
a
ASCII
python
c
current position
perl
e
expression
php perl
g
global
most
i
case-insensitive
most
m
multiline
php perl python javascript .net java
m
(non)multiline
ruby
o
once
perl ruby
r
non-destructive
perl
S
study
php
s
single line
ruby
U
ungreedy
php r
u
unicode
most
x
whitespace-extended
most
y
sticky ↪
javascript
How to convert preg_replace e to preg_replace_callback?
What are inline modifiers?
What is '?-mix' in a Ruby Regular Expression
Other:
|:alternation (OR) operator, .:any character, [.]:literal dot character
What special characters must be escaped?
Control verbs (php and perl): (*PRUNE), (*SKIP), (*FAIL) and (*F)
php only: (*BSR_ANYCRLF)
Recursion (php and perl): (?R), (?0) and (?1), (?-1), (?&groupname)
Common Tasks
Get a string between two curly braces: {...}
Match (or replace) a pattern except in situations s1, s2, s3...
How do I find all YouTube video ids in a string using a regex?
Validation:
Internet: email addresses, URLs (host/port: regex and non-regex alternatives), passwords
Numeric: a number, min-max ranges (such as 1-31), phone numbers, date
Parsing HTML with regex: See "General Information > When not to use Regex"
Advanced Regex-Fu
Strings and numbers:
Regular expression to match a line that doesn't contain a word
How does this PCRE pattern detect palindromes?
Match strings whose length is a fourth power
How does this regex find triangular numbers?
How to determine if a number is a prime with regex?
How to match the middle character in a string with regex?
Other:
How can we match a^n b^n?
Match nested brackets
Using a recursive pattern php, perl
Using balancing groups .net
“Vertical” regex matching in an ASCII “image”
List of highly up-voted regex questions on Code Golf
How to make two quantifiers repeat the same number of times?
An impossible-to-match regular expression: (?!a)a
Match/delete/replace this except in contexts A, B and C
Match nested brackets with regex without using recursion or balancing groups?
Flavor-Specific Information
(Except for those marked with *, this section contains non-Stack Overflow links.)
Java
Official documentation: Pattern Javadoc ↪, Oracle's regular expressions tutorial ↪
The differences between functions in java.util.regex.Matcher:
matches()): The match must be anchored to both input-start and -end
find()): A match may be anywhere in the input string (substrings)
lookingAt(): The match must be anchored to input-start only
(For anchors in general, see the section "Anchors")
The only java.lang.String functions that accept regular expressions: matches(s), replaceAll(s,s), replaceFirst(s,s), split(s), split(s,i)
*An (opinionated and) detailed discussion of the disadvantages of and missing features in java.util.regex
.NET
How to read a .NET regex with look-ahead, look-behind, capturing groups and back-references mixed together?
Official documentation:
Boost regex engine: General syntax, Perl syntax (used by TextPad, Sublime Text, UltraEdit, ...???)
JavaScript general info and RegExp object
.NET MySQL Oracle Perl5 version 18.2
PHP: pattern syntax, preg_match
Python: Regular expression operations, search vs match, how-to
Rust: crate regex, struct regex::Regex
Splunk: regex terminology and syntax and regex command
Tcl: regex syntax, manpage, regexp command
Visual Studio Find and Replace
General information
(Links marked with * are non-Stack Overflow links.)
Other general documentation resources: Learning Regular Expressions, *Regular-expressions.info, *Wikipedia entry, *RexEgg, Open-Directory Project
DFA versus NFA
Generating Strings matching regex
Books: Jeffrey Friedl's Mastering Regular Expressions
When to not use regular expressions:
Some people, when confronted with a problem, think "I know, I'll use regular expressions." Now they have two problems. (blog post written by Stack Overflow's founder)*
Do not use regex to parse HTML:
Don't. Please, just don't
Well, maybe...if you're really determined (other answers in this question are also good)
Examples of regex that can cause regex engine to fail
Why does this regular expression kill the Java regex engine?
Tools: Testers and Explainers
(This section contains non-Stack Overflow links.)
Online (* includes replacement tester, + includes split tester):
Debuggex (Also has a repository of useful regexes) javascript, python, pcre
*Regular Expressions 101 php, pcre, python, javascript, java
Regex Pal, regular-expressions.info javascript
Rubular ruby RegExr Regex Hero dotnet
*+ regexstorm.net .net
*RegexPlanet: Java java, Go go, Haskell haskell, JavaScript javascript, .NET dotnet, Perl perl php PCRE php, Python python, Ruby ruby, XRegExp xregexp
freeformatter.com xregexp
*+regex.larsolavtorvik.com php PCRE and POSIX, javascript
Offline:
Microsoft Windows: RegexBuddy (analysis), RegexMagic (creation), Expresso (analysis, creation, free)
MySQL 8.0: Various syntax changes were made. Note especially the doubling of backslashes in some contexts. (This Answer need further editing to reflect the differences.)
Is there any differences in invoking variables with syntax ${var} and $(var)? For instance, in the way the variable will be expanded or anything?
There's no difference – they mean exactly the same (in GNU Make and in POSIX make).
I think that $(round brackets) look tidier, but that's just personal preference.
(Other answers point to the relevant sections of the GNU Make documentation, and note that you shouldn't mix the syntaxes within a single expression)
The Basics of Variable References section from the GNU make documentation state no differences:
To substitute a variable's value, write a dollar sign followed by the
name of the variable in parentheses or braces: either $(foo) or
${foo} is a valid reference to the variable foo.
As already correctly pointed out, there is no difference but be be wary not to mix the two kind of delimiters as it can lead to cryptic errors like in the GNU make example by unomadh.
From the GNU make manual on the Function Call Syntax (emphasis mine):
[…] If the arguments themselves contain other function calls or variable references, it is wisest to use the same kind of delimiters for all the references; write $(subst a,b,$(x)), not $(subst a,b,${x}). This is because it is clearer, and because only one type of delimiter is matched to find the end of the reference.
The ${} style lets you test the make rules in the shell, if you have the corresponding environment variables set, since that is compatible with bash.
Actually, it seems to be fairly different:
, = ,
list = a,b,c
$(info $(subst $(,),-,$(list))_EOL)
$(info $(subst ${,},-,$(list))_EOL)
outputs
a-b-c_EOL
md/init-profile.md:4: *** unterminated variable reference. Stop.
But so far I only found this difference when the variable name into ${...} contains itself a comma. I first thought ${...} was expanding the comma not as part as the value, but it turns out i'm not able to hack it this way. I still don't understand this... If anyone had an explanation, I'd be happy to know !
It makes a difference if the expression contains unbalanced brackets:
${info ${subst ),(,:-)}}
$(info $(subst ),(,:-)))
->
:-(
*** insufficient number of arguments (1) to function 'subst'. Stop.
For variable references, this makes a difference for functions, or for variable names that contain brackets (bad idea)
I'm using whittle to parse a grammar, but I'm running into the classical LALR ambiguity problem. My grammar looks like this (simplified):
<comment> ::= '{' <string> '}' # string enclosed in braces
<tag> ::= '[' <name> <quoted-string> ']' # [tagname "tag value"]
<name> ::= /[A-Za-z_]+/ # subset of all printable chars
<quoted-string> ::= '"' <string> '"' # string enclosed in quotes
<string> ::= /[:print:]/ # regex for all printable chars
The problem, of course, is <string>. It contains all printable characters and is therefore very greedy. Since it's an LALR parser, it tries to parse a <name> as a <string> and everything breaks. The grammar complicates things because it uses different string delimiters for different things, which is why I tried to make the <string> rule in the first place.
Is there a canonical way to normalize this grammar to make it LALR compliant, if it's even possible?
This is not "the classical LALR ambiguity problem", whatever that might be. It is simply an error in the lexical specification of the language.
I took a quick glance at the Whittle readme, but it didn't bear any resemblance to the grammar in the OP. So I'm assuming that the text in the OP is conceptual rather than literal, and the fact that it includes the obviously incorrect
<string> ::= /[:print:]/ # regex for all printable chars
is just a typo.
Better would have been /[:print:]*/, assuming that Ruby lets you get away with [:print:] rather than the Posix-standard [[:print:]].
But that wouldn't be correct either because lexing (usually) matches the longest possible string, and consequently that will gobble up the closing quote and any following text.
So the correct solution for quoted-string is to write it out correctly:
<quoted-string> ::= /"[^"]*"/
or even
<quoted-string> :: /"([^\\"]|\\.)*"/
# any number of characters other than quote or escape, or escaped pairs
You might have other ideas about how to escape internal double quotes; those are just examples. In both cases, you need to postprocess the token in order to (at least) strip the double-quotes and possible interpret escape sequences. That's just the way it goes.
Your comment sequences present a more difficult issue, assuming that your intention was that a comment might include nested braces (eg. {This comment {with this} ends here}) because the nested brace syntax is not regular and thus cannot be matched with a regular expression. Of course, very few "regular expression" libraries are really regular these days, and I don't know if Ruby contains some sort of brace-counting extension, like for example Lua's pattern syntax. The nested brace syntax is certainly context-free but to actually parse it you need to lexically analyze the contents of the outer {...} in a different way than the rest of the program.
It is this latter observation, and not any weakness in the LALR algorithm, that is causing you pain, and I'd say that this is a weakness with the (mostly undocumented afaics) lexical analysis section of whittle. In a flex-generated lexer, for example, it would be normal to use start conditions to separate the lexical environments (program / quoted string / braced comment), and the parser would then have no ambiguity.
Hope that helps.
I'm writing a regular expression in Objective-C.
The escape sequence \w is illegal and emits a warning, so the regular expression /\w/ must be written as #"\\w"; the escape sequence \? is valid, apparently, and doesn't emit a warning, so the regular expression /\?/ must be written as #"\\?" (i.e., the backslash must be escaped).
Question marks aren't invisible like \t or \n, so why is \? a valid escape sequence?
Edit: To clarify, I'm not asking about the quantifier, I'm asking about a string escape sequence. That is, this doesn't emit a warning:
NSString *valid = #"\?";
By contrast, this does emit a warning ("Unknown escape sequence '\w'"):
NSString *invalid = #"\w";
It specifies a literal question mark. It is needed because of a little-known feature called trigraphs, where you can write a three-character sequence starting with question marks to substitute another character. If you have trigraphs enabled, in order to write "??" in a string, you need to write it as "?\?" in order to prevent the preprocessor from trying to read it as the beginning of a trigraph.
(If you're wondering "Why would anybody introduce a feature like this?": Some keyboards or character sets didn't include commonly used symbols like {. so they introduced trigraphs so you could write ??< instead.)
? in regex is a quantifier, it means 0 or 1 occurences. When appended to the + or * quantifiers, it makes it "lazy".
For example, applying the regex o? to the string foo? would match o.
However, the regex o\? in foo? would match o?, because it is searching for a literal question mark in the string, instead of an arbitrary quantifier.
Applying the regex o*? to foo? would match oo.
More info on quantifiers here.