Regex

findRE for capturing groups across every match.

Squiggly’s regex surface is a single, versatile function — findRE — that returns every match (with capture groups) for a Java java.util.regex.Pattern (the same flavour as Scala’s Regex). Other regex idioms (full-string match, replace, split-on-pattern) compose from findRE plus the standard string / list builtins.

Function	Arity	Signature	What
`findRE`	2	`findRE pattern s`	All matches — each is a `[whole, group1, group2, …]` list
`findRE`	3	`findRE pattern s limit`	First `limit` matches only

What the result looks like

{{ findRE '(\d{4})-(\d{2})-(\d{2})' '2024-03-12 and 2025-01-09' }}

…returns [[2024-03-12, 2024, 03, 12], [2025-01-09, 2025, 01, 09]] — one list per match, with the full match first followed by each capture group’s text. Groups that didn’t participate render as an empty string.

Idioms

Validate

{{ if findRE '^[^@]+@[^@]+\.[^@]+$' .email | nonEmpty }}
  valid
{{ else }}
  invalid
{{ end }}

Extract groups

{{ ymd := findRE '^(\d{4})-(\d{2})-(\d{2})$' .date }}
{{ if ymd | nonEmpty }}
  year = {{ ymd[0][1] }}, month = {{ ymd[0][2] }}, day = {{ ymd[0][3] }}
{{ end }}

ymd[0] is the first match’s group list; ymd[0][1] is its first capture group.

Walk every match

{{ for m <- findRE '\[\[([^\]]+)\]\]' .body }}
  {{ m[1] }}        // each match's first capture group
{{ end }}

Cap the work

{{ findRE '\b\w+\b' .body 10 }}    first ten words only

Beyond `findRE`

Squiggly doesn’t ship dedicated matchRE / replaceRE / splitRE builtins yet. Common workarounds:

Full-string match: wrap the pattern in ^…$ and check findRE pattern s | nonEmpty.
Split on pattern: for literal separators, use split sep s. Splitting on a regex is on the roadmap.
Replace by pattern: combine findRE with map and string concatenation, or wait for the planned replaceRE builtin.