Crate regex[−][src]

Expand description

This crate provides a library for parsing, compiling, and executing regular expressions. Its syntax is similar to Perl-style regular expressions, but lacks a few features like look around and backreferences. In exchange, all searches execute in linear time with respect to the size of the regular expression and search text.

This crate’s documentation provides some simple examples, describes Unicode support and exhaustively lists the supported syntax.

For more specific details on the API for regular expressions, please see the documentation for the Regex type.

Usage

This crate is on crates.io and can be used by adding regex to your dependencies in your project’s Cargo.toml.

[dependencies]
regex = "0.2"

and this to your crate root:

extern crate regex;

Example: find a date

General use of regular expressions in this package involves compiling an expression and then using it to search, split or replace text. For example, to confirm that some text resembles a date:

use regex::Regex;
let re = Regex::new(r"^\d{4}-\d{2}-\d{2}$").unwrap();
assert!(re.is_match("2014-01-01"));

Notice the use of the ^ and $ anchors. In this crate, every expression is executed with an implicit .*? at the beginning and end, which allows it to match anywhere in the text. Anchors can be used to ensure that the full text matches an expression.

This example also demonstrates the utility of raw strings in Rust, which are just like regular strings except they are prefixed with an r and do not process any escape sequences. For example, "\\d" is the same expression as r"\d".

Example: Avoid compiling the same regex in a loop

It is an anti-pattern to compile the same regular expression in a loop since compilation is typically expensive. (It takes anywhere from a few microseconds to a few milliseconds depending on the size of the regex.) Not only is compilation itself expensive, but this also prevents optimizations that reuse allocations internally to the matching engines.

In Rust, it can sometimes be a pain to pass regular expressions around if they’re used from inside a helper function. Instead, we recommend using the lazy_static crate to ensure that regular expressions are compiled exactly once.

For example:

#[macro_use] extern crate lazy_static;
extern crate regex;

use regex::Regex;

fn some_helper_function(text: &str) -> bool {
    lazy_static! {
        static ref RE: Regex = Regex::new("...").unwrap();
    }
    RE.is_match(text)
}

fn main() {}

Specifically, in this example, the regex will be compiled when it is used for the first time. On subsequent uses, it will reuse the previous compilation.

Example: iterating over capture groups

This crate provides convenient iterators for matching an expression repeatedly against a search string to find successive non-overlapping matches. For example, to find all dates in a string and be able to access them by their component pieces:

let re = Regex::new(r"(\d{4})-(\d{2})-(\d{2})").unwrap();
let text = "2012-03-14, 2013-01-01 and 2014-07-05";
for cap in re.captures_iter(text) {
    println!("Month: {} Day: {} Year: {}", &cap[2], &cap[3], &cap[1]);
}
// Output:
// Month: 03 Day: 14 Year: 2012
// Month: 01 Day: 01 Year: 2013
// Month: 07 Day: 05 Year: 2014

Notice that the year is in the capture group indexed at 1. This is because the entire match is stored in the capture group at index 0.

Example: replacement with named capture groups

Building on the previous example, perhaps we’d like to rearrange the date formats. This can be done with text replacement. But to make the code clearer, we can name our capture groups and use those names as variables in our replacement text:

let re = Regex::new(r"(?P<y>\d{4})-(?P<m>\d{2})-(?P<d>\d{2})").unwrap();
let before = "2012-03-14, 2013-01-01 and 2014-07-05";
let after = re.replace_all(before, "$m/$d/$y");
assert_eq!(after, "03/14/2012, 01/01/2013 and 07/05/2014");

The replace methods are actually polymorphic in the replacement, which provides more flexibility than is seen here. (See the documentation for Regex::replace for more details.)

Note that if your regex gets complicated, you can use the x flag to enable insigificant whitespace mode, which also lets you write comments:

let re = Regex::new(r"(?x)
  (?P<y>\d{4}) # the year
  -
  (?P<m>\d{2}) # the month
  -
  (?P<d>\d{2}) # the day
").unwrap();
let before = "2012-03-14, 2013-01-01 and 2014-07-05";
let after = re.replace_all(before, "$m/$d/$y");
assert_eq!(after, "03/14/2012, 01/01/2013 and 07/05/2014");

If you wish to match against whitespace in this mode, you can still use \s, \n, \t, etc. For escaping a single space character, you can use its hex character code \x20 or temporarily disable the x flag, e.g., (?-x: ).

Example: match multiple regular expressions simultaneously

This demonstrates how to use a RegexSet to match multiple (possibly overlapping) regular expressions in a single scan of the search text:

use regex::RegexSet;

let set = RegexSet::new(&[
    r"\w+",
    r"\d+",
    r"\pL+",
    r"foo",
    r"bar",
    r"barfoo",
    r"foobar",
]).unwrap();

// Iterate over and collect all of the matches.
let matches: Vec<_> = set.matches("foobar").into_iter().collect();
assert_eq!(matches, vec![0, 2, 3, 4, 6]);

// You can also test whether a particular regex matched:
let matches = set.matches("foobar");
assert!(!matches.matched(5));
assert!(matches.matched(6));

Pay for what you use

With respect to searching text with a regular expression, there are three questions that can be asked:

Does the text match this expression?
If so, where does it match?
Where did the capturing groups match?

Generally speaking, this crate could provide a function to answer only #3, which would subsume #1 and #2 automatically. However, it can be significantly more expensive to compute the location of capturing group matches, so it’s best not to do it if you don’t need to.

Therefore, only use what you need. For example, don’t use find if you only need to test if an expression matches a string. (Use is_match instead.)

Unicode

This implementation executes regular expressions only on valid UTF-8 while exposing match locations as byte indices into the search string.

Only simple case folding is supported. Namely, when matching case-insensitively, the characters are first mapped using the “simple” case folding rules defined by Unicode.

Regular expressions themselves are only interpreted as a sequence of Unicode scalar values. This means you can use Unicode characters directly in your expression:

let re = Regex::new(r"(?i)Δ+").unwrap();
let mat = re.find("ΔδΔ").unwrap();
assert_eq!((mat.start(), mat.end()), (0, 6));

Most features of the regular expressions in this crate are Unicode aware. Here are some examples:

. will match any valid UTF-8 encoded Unicode scalar value except for \n. (To also match \n, enable the s flag, e.g., (?s:.).)
\w, \d and \s are Unicode aware. For example, \s will match all forms of whitespace categorized by Unicode.
\b matches a Unicode word boundary.
Negated character classes like [^a] match all Unicode scalar values except for a.
^ and $ are not Unicode aware in multi-line mode. Namely, they only recognize \n and not any of the other forms of line terminators defined by Unicode.

Unicode general categories, scripts, script extensions, ages and a smattering of boolean properties are available as character classes. For example, you can match a sequence of numerals, Greek or Cherokee letters:

let re = Regex::new(r"[\pN\p{Greek}\p{Cherokee}]+").unwrap();
let mat = re.find("abcΔᎠβⅠᏴγδⅡxyz").unwrap();
assert_eq!((mat.start(), mat.end()), (3, 23));

For a more detailed breakdown of Unicode support with respect to UTS#18, please see the UNICODE document in the root of the regex repository.

Opt out of Unicode support

The bytes sub-module provides a Regex type that can be used to match on &[u8]. By default, text is interpreted as UTF-8 just like it is with the main Regex type. However, this behavior can be disabled by turning off the u flag, even if doing so could result in matching invalid UTF-8. For example, when the u flag is disabled, . will match any byte instead of any Unicode scalar value.

Disabling the u flag is also possible with the standard &str-based Regex type, but it is only allowed where the UTF-8 invariant is maintained. For example, (?-u:\w) is an ASCII-only \w character class and is legal in an &str-based Regex, but (?-u:\xFF) will attempt to match the raw byte \xFF, which is invalid UTF-8 and therefore is illegal in &str-based regexes.

Syntax

The syntax supported in this crate is documented below.

Note that the regular expression parser and abstract syntax are exposed in a separate crate, regex-syntax.

Matching one character

.             any character except new line (includes new line with s flag)
\d            digit (\p{Nd})
\D            not digit
\pN           One-letter name Unicode character class
\p{Greek}     Unicode character class (general category or script)
\PN           Negated one-letter name Unicode character class
\P{Greek}     negated Unicode character class (general category or script)

Character classes

[xyz]         A character class matching either x, y or z (union).
[^xyz]        A character class matching any character except x, y and z.
[a-z]         A character class matching any character in range a-z.
[[:alpha:]]   ASCII character class ([A-Za-z])
[[:^alpha:]]  Negated ASCII character class ([^A-Za-z])
[x[^xyz]]     Nested/grouping character class (matching any character except y and z)
[a-y&&xyz]    Intersection (matching x or y)
[0-9&&[^4]]   Subtraction using intersection and negation (matching 0-9 except 4)
[0-9--4]      Direct subtraction (matching 0-9 except 4)
[a-g~~b-h]    Symmetric difference (matching `a` and `h` only)
[\[\]]        Escaping in character classes (matching [ or ])

Any named character class may appear inside a bracketed [...] character class. For example, [\p{Greek}[:digit:]] matches any Greek or ASCII digit. [\p{Greek}&&\pL] matches Greek letters.

Precedence in character classes, from most binding to least:

Ranges: a-cd == [a-c]d
Union: ab&&bc == [ab]&&[bc]
Intersection: ^a-z&&b == ^[a-z&&b]
Negation

Composites

xy    concatenation (x followed by y)
x|y   alternation (x or y, prefer x)

Repetitions

x*        zero or more of x (greedy)
x+        one or more of x (greedy)
x?        zero or one of x (greedy)
x*?       zero or more of x (ungreedy/lazy)
x+?       one or more of x (ungreedy/lazy)
x??       zero or one of x (ungreedy/lazy)
x{n,m}    at least n x and at most m x (greedy)
x{n,}     at least n x (greedy)
x{n}      exactly n x
x{n,m}?   at least n x and at most m x (ungreedy/lazy)
x{n,}?    at least n x (ungreedy/lazy)
x{n}?     exactly n x

Empty matches

^     the beginning of text (or start-of-line with multi-line mode)
$     the end of text (or end-of-line with multi-line mode)
\A    only the beginning of text (even with multi-line mode enabled)
\z    only the end of text (even with multi-line mode enabled)
\b    a Unicode word boundary (\w on one side and \W, \A, or \z on other)
\B    not a Unicode word boundary

Grouping and flags

(exp)          numbered capture group (indexed by opening parenthesis)
(?P<name>exp)  named (also numbered) capture group (allowed chars: [_0-9a-zA-Z])
(?:exp)        non-capturing group
(?flags)       set flags within current group
(?flags:exp)   set flags for exp (non-capturing)

Flags are each a single character. For example, (?x) sets the flag x and (?-x) clears the flag x. Multiple flags can be set or cleared at the same time: (?xy) sets both the x and y flags and (?x-y) sets the x flag and clears the y flag.

All flags are by default disabled unless stated otherwise. They are:

i     case-insensitive: letters match both upper and lower case
m     multi-line mode: ^ and $ match begin/end of line
s     allow . to match \n
U     swap the meaning of x* and x*?
u     Unicode support (enabled by default)
x     ignore whitespace and allow line comments (starting with `#`)

Flags can be toggled within a pattern. Here’s an example that matches case-insensitively for the first part but case-sensitively for the second part:

let re = Regex::new(r"(?i)a+(?-i)b+").unwrap();
let cap = re.captures("AaAaAbbBBBb").unwrap();
assert_eq!(&cap[0], "AaAaAbb");

Notice that the a+ matches either a or A, but the b+ only matches b.

Multi-line mode means ^ and $ no longer match just at the beginning/end of the input, but at the beginning/end of lines:

let re = Regex::new(r"(?m)^line \d+").unwrap();
let m = re.find("line one\nline 2\n").unwrap();
assert_eq!(m.as_str(), "line 2");

Note that ^ matches after new lines, even at the end of input:

let re = Regex::new(r"(?m)^").unwrap();
let m = re.find_iter("test\n").last().unwrap();
assert_eq!((m.start(), m.end()), (5, 5));

Here is an example that uses an ASCII word boundary instead of a Unicode word boundary:

let re = Regex::new(r"(?-u:\b).+(?-u:\b)").unwrap();
let cap = re.captures("$$abc$$").unwrap();
assert_eq!(&cap[0], "abc");

Escape sequences

\*          literal *, works for any punctuation character: \.+*?()|[]{}^$
\a          bell (\x07)
\f          form feed (\x0C)
\t          horizontal tab
\n          new line
\r          carriage return
\v          vertical tab (\x0B)
\123        octal character code (up to three digits)
\x7F        hex character code (exactly two digits)
\x{10FFFF}  any hex character code corresponding to a Unicode code point
\u007F      hex character code (exactly four digits)
\u{7F}      any hex character code corresponding to a Unicode code point
\U0000007F  hex character code (exactly eight digits)
\U{7F}      any hex character code corresponding to a Unicode code point

Perl character classes (Unicode friendly)

These classes are based on the definitions provided in UTS#18:

\d     digit (\p{Nd})
\D     not digit
\s     whitespace (\p{White_Space})
\S     not whitespace
\w     word character (\p{Alphabetic} + \p{M} + \d + \p{Pc} + \p{Join_Control})
\W     not word character

ASCII character classes

[[:alnum:]]    alphanumeric ([0-9A-Za-z])
[[:alpha:]]    alphabetic ([A-Za-z])
[[:ascii:]]    ASCII ([\x00-\x7F])
[[:blank:]]    blank ([\t ])
[[:cntrl:]]    control ([\x00-\x1F\x7F])
[[:digit:]]    digits ([0-9])
[[:graph:]]    graphical ([!-~])
[[:lower:]]    lower case ([a-z])
[[:print:]]    printable ([ -~])
[[:punct:]]    punctuation ([!-/:-@\[-`{-~])
[[:space:]]    whitespace ([\t\n\v\f\r ])
[[:upper:]]    upper case ([A-Z])
[[:word:]]     word characters ([0-9A-Za-z_])
[[:xdigit:]]   hex digit ([0-9A-Fa-f])

Untrusted input

This crate can handle both untrusted regular expressions and untrusted search text.

Untrusted regular expressions are handled by capping the size of a compiled regular expression. (See RegexBuilder::size_limit.) Without this, it would be trivial for an attacker to exhaust your system’s memory with expressions like a{100}{100}{100}.

Untrusted search text is allowed because the matching engine(s) in this crate have time complexity O(mn) (with m ~ regex and n ~ search text), which means there’s no way to cause exponential blow-up like with some other regular expression engines. (We pay for this by disallowing features like arbitrary look-ahead and backreferences.)

When a DFA is used, pathological cases with exponential state blow up are avoided by constructing the DFA lazily or in an “online” manner. Therefore, at most one new state can be created for each byte of input. This satisfies our time complexity guarantees, but can lead to unbounded memory growth proportional to the size of the input. As a stopgap, the DFA is only allowed to store a fixed number of states. When the limit is reached, its states are wiped and continues on, possibly duplicating previous work. If the limit is reached too frequently, it gives up and hands control off to another matching engine with fixed memory requirements. (The DFA size limit can also be tweaked. See RegexBuilder::dfa_size_limit.)