Struct CaptureLocations

pub struct CaptureLocations(/* private fields */);

A low level representation of the byte offsets of each capture group.

You can think of this as a lower level Captures, where this type does not support named capturing groups directly and it does not borrow the haystack that these offsets were matched on.

Primarily, this type is useful when using the lower level Regex APIs such as Regex::captures_read, which permits amortizing the allocation in which capture match offsets are stored.

In order to build a value of this type, you’ll need to call the Regex::capture_locations method. The value returned can then be reused in subsequent searches for that regex. Using it for other regexes may result in a panic or otherwise incorrect results.

Example

This example shows how to create and use CaptureLocations in a search.

use regex::Regex;

let re = Regex::new(r"(?<first>\w+)\s+(?<last>\w+)").unwrap();
let mut locs = re.capture_locations();
let m = re.captures_read(&mut locs, "Bruce Springsteen").unwrap();
assert_eq!(0..17, m.range());
assert_eq!(Some((0, 17)), locs.get(0));
assert_eq!(Some((0, 5)), locs.get(1));
assert_eq!(Some((6, 17)), locs.get(2));

// Asking for an invalid capture group always returns None.
assert_eq!(None, locs.get(3));
assert_eq!(None, locs.get(34973498648));
assert_eq!(None, locs.get(9944060567225171988));

Implementations

impl CaptureLocations

pub fn get(&self, i: usize) -> Option<(usize, usize)>

Returns the start and end byte offsets of the capture group at index i. This returns None if i is not a valid capture group or if the capture group did not match.

Example

use regex::Regex;

let re = Regex::new(r"(?<first>\w+)\s+(?<last>\w+)").unwrap();
let mut locs = re.capture_locations();
re.captures_read(&mut locs, "Bruce Springsteen").unwrap();
assert_eq!(Some((0, 17)), locs.get(0));
assert_eq!(Some((0, 5)), locs.get(1));
assert_eq!(Some((6, 17)), locs.get(2));

pub fn len(&self) -> usize

Returns the total number of capture groups (even if they didn’t match). That is, the length returned is unaffected by the result of a search.

This is always at least 1 since every regex has at least 1 capturing group that corresponds to the entire match.

Example

use regex::Regex;

let re = Regex::new(r"(?<first>\w+)\s+(?<last>\w+)").unwrap();
let mut locs = re.capture_locations();
assert_eq!(3, locs.len());
re.captures_read(&mut locs, "Bruce Springsteen").unwrap();
assert_eq!(3, locs.len());

Notice that the length is always at least 1, regardless of the regex:

use regex::Regex;

let re = Regex::new(r"").unwrap();
let locs = re.capture_locations();
assert_eq!(1, locs.len());

// [a&&b] is a regex that never matches anything.
let re = Regex::new(r"[a&&b]").unwrap();
let locs = re.capture_locations();
assert_eq!(1, locs.len());

Trait Implementations

impl Clone for CaptureLocations

fn clone(&self) -> CaptureLocations

Returns a copy of the value.

fn clone_from(&mut self, source: &Self)

Performs copy-assignment from source.

impl Debug for CaptureLocations

fn fmt(&self, f: &mut Formatter<'_>) -> Result

Formats the value using the given formatter.