Equality for regexes, defined by having equal string representations and flags (including flags that cannot be embedded).

Notes:

• Functionally equivalent regexes (e.g. #"..." and #".{3}" are not considered ='.
• Some regexes may not be =' initially due to differing flag sets, but after being run through embed-flags may become =', due to non-embeddable flags being silently dropped (see embed-flags for details).

Returns a regex that will match any one of res, via alternation:

• re|re|re|...

Returns an empty regex (#"") if no res are provided, or they’re all empty?’.

Notes:

• Duplicate elements in res will only appear once in the result. This equality comparison occurs after each re is run through embed-flags.
• Does not wrap the result in a group, which, because alternation has the lowest precedence in regexes, runs the risk of behaving unexpectedly if the result is then combined with further regexes. tl;dr - one of the grouping variants should almost always be preferred.

alt then cg:

• (re|re|re|...)

alt then fgrp:

• (?flgs:re|re|re|...)

alt then grp:

• (?:re|re|re|...)

alt then ncg:

• (?<nm>re|re|re|...)

Returns an ‘and’ regex that will match a and b in any order, and with the separator regex s (if provided) between them:

• asb|bsa

Returns an empty regex (#"") if re is empty?’.

Notes:

• a and b must be distinct (must not match the same text) or else the resulting regex will be logically inconsistent (will not be an ‘and’)
• May optimise the expression (via de-duplication in alt).
• Does not wrap the result in a group, which, because alternation has the lowest precedence in regexes, runs the risk of behaving unexpectedly if the result is then combined with further regexes. tl;dr - one of the grouping variants should almost always be preferred.

and’ then cg:

• (asb|bsa)

Notes:

• Unlike most other -cg fns, this one does not accept any number of res.
• May optimise the expression (via de-duplication in alt).

and’ then fgrp:

• (?flgs:asb|bsa)

Notes:

• Unlike most other -fgrp fns, this one does not accept any number of res.
• May optimise the expression (via de-duplication in alt).

and’ then grp:

• (?:asb|bsa)

Notes:

• Unlike most other -grp fns, this one does not accept any number of res.
• May optimise the expression (via de-duplication in alt).

and’ then ncg:

• (?<nm>asb|bsa)

Notes:

• Unlike most other -ncg fns, this one does not accept any number of res.
• May optimise the expression (via de-duplication in alt).

As for grp, but emits a capturing group:

• (res)

Returns an empty capturing group (#"()") if no res are provided, or they’re all empty?’. It does this to ensure that capturing groups are preserved during composition, even if they’re empty (since not doing so will break code that uses indexes to access matched group content).

As for join, but encloses the joined res into a character class:

• [res]

Returns an empty regex (#"") if no res are provided, or they’re all empty?’.

Notes:

• ⚠️ On ClojureScript nested character classes don’t work as one might expect, even though they will compile just fine. For example, this code matches as expected on ClojureJVM, but does not on ClojureScript (despite the regex compiling): (re-matches #"[[a-m][o-z]]+" "az").

Embeds any programmatic or ungrouped flags found in re. It does this by removing all flags from re then wrapping it in a flag group containing those flags that are embeddable (non-embeddable flags are silently dropped - use has-non-embeddable-flags? if you need to check for this). Returns re if re contains no flags.

For example on the JVM, both (Pattern/compile "[abc]+" Pattern/CASE_INSENSITIVE) and #"(?i)[abc]+" would become #"(?i:[abc]+)".

Similarly, on ClojureScript (doto (js/RegExp.) (.compile "[abc]+" "i")) would become #"(?i:[abc]+)".

Note:

• fgrp is almost always a better choice than this function! embed-flags is primarily intended for internal use by wreck, but may be useful in those rare cases where Clojure(Script) code receives a 3rd party regex, wishes to use it as part of composing a larger regex, doesn’t know if it contains flags or not, and doesn’t care that non-embeddable flags will be silently dropped.
• ⚠️ On the JVM, ungrouped embedded flags in the middle of re will be moved to the beginning of the regex. This may alter the semantics of the regex - for example a(?i)b will become (?i:ab), which means that a will be matched case-insensitively by the result, which is not the same as the original (which matches lower-case a only). This is an unavoidable consequence of how the JVM regex engine reports flags. If you really need to use embedded flag(s) midway through a regex, use fgrp to ensure proper scoping of the flag(s).
• ⚠️ On the JVM, the programmatic flags LITERAL and CANON_EQ have no embeddable equivalent, and will be silently dropped by this function.
• ⚠️ On JavaScript, only the flags i, m, and s can be embedded. All other flags will be silently dropped by this function.

Is re nil or (=' #"")?

Notes:

• Takes flags (if any) into account.

Escapes s (a String) for use in a regex, returning a String. Returns nil if s is nil.

Notes:

• unlike most other fns in this namespace, this one does not support a regex as an input, nor return a regex as an output

Returns a regex where re will match exactly n times:

• re{n}

Returns an empty regex (#"") if re is empty?’.

cg then exn:

• (res){n}

chcl then exn:

• [res]{n}

fgrp then exn:

• (?flgs:res){n}

grp then exn:

• (?:res){n}

ncg then exn:

• (?<nm>res){n}

As for grp, but emits an embedded flag group with flgs (a String):

• (?flgs:res)

Devolves to grp if flgs is blank. Throws if flgs contains an invalid flag character, including those that (ClojureScript only) cannot be embedded.

Notes:

• If you must use regex flags, it is STRONGLY RECOMMENDED that you use this function! Programmatically set flags and ungrouped embedded flags (e.g. (?i)) have no explicit scope and so cannot be reliably used to compose larger regexes. wreck makes a best effort to always convert such ‘unscoped’ flags into their embedded (scoped) equivalents (using embed-flags) when composing larger regexes , but using fgrp voids potential footguns.
• Removes any ungrouped embedded flags in re (e.g. (?i)ab), but does not add them to flgs if they aren’t already there.
• ⚠️ On the JVM, ungrouped embedded flags in the middle of re (e.g. a(?i)b) will also be removed, which may alter the semantics of the regex.
• ⚠️ On JavaScript, only the flags i, m and s can be embedded (this is a limitation of the JavaScript regex engine). Other flags will result in a js/SyntaxError being thrown.
• For the JVM, see the ‘special constructs’ section of the java.util.regex.Pattern JavaDoc for the set of valid flag characters.
• For JavaScript, see the RegExp flags reference for the set of valid flag characters (while keeping in mind most of them can’t be embedded).

See fgrp

As for join, but encloses the joined res into a single non-capturing group:

• (?:res)

Returns an empty regex (#"") if no res are provided, or they’re all empty?’.

Does re have non-embeddable flags?

Notes:

• On the JVM, the only non-embeddable flags are the programmatic flags LITERAL and CANON_EQ.
• On JavaScript, this is every flag except i, m, and s.

Returns a regex that is all of the res joined together. Each element in res can be a regex, a String or something that can be turned into a String (including numbers, etc.). Returns an empty regex (#"") if no res are provided, or they’re all empty?’.

Notes:

• ⚠️ In ClojureScript be cautious about using numbers in these calls, since JavaScript’s number handling is a 🤡show. See the unit tests for examples.

Returns a regex where re will match from n to m times:

• re{n,m}

Returns an empty regex (#"") if re is empty?’.

cg then n2m:

• (res){n,m}

chcl then n2m:

• [res]{n,m}

fgrp then n2m:

• (?flgs:res){n,m}

grp then n2m:

• (?:res){n,m}

ncg then n2m:

• (?<nm>res){n,m}

As for grp, but emits a named capturing group named nm:

• (?<nm>res)

Devolves to cg if nm is blank. Throws if nm is an invalid name for a named capturing group (alphanumeric only, must start with an alphabetical character, must be unique within the regex).

Returns a regex where re will match n or more times:

• re{n,}

Returns an empty regex (#"") if re is empty?’.

cg then nom:

• (res){n,}

chcl then nom:

• [res]{n,}

fgrp then nom:

• (?flgs:res){n,}

grp then nom:

• (?:res){n,}

ncg then nom:

• (?<nm>res){n,}

Returns a regex where re will match one or more times:

• re+

Returns an empty regex (#"") if re is empty?’.

cg then oom:

• (res)+

chcl then oom:

• [res]+

fgrp then oom:

• (?flgs:res)+

grp then oom:

• (?:res)+

ncg then oom:

• (?<nm>res)+

Returns a regex where re is optional:

• re?

Returns an empty regex (#"") if re is empty?’.

cg then opt:

• (res)?

chcl then opt:

• [res]?

fgrp then opt:

• (?flgs:res)?

grp then opt:

• (?:res)?

ncg then opt:

• (?<nm>res)?

Returns an ‘inclusive or’ regex that will match a or b, or both, in any order, and with the separator regex s (if provided) between them:

• asb|bsa|a|b

Returns an empty regex (#"") if re is empty?’.

Notes:

• a and b must be distinct (must not match the same text) or else the resulting regex will be logically inconsistent (will not be an ‘or’)
• May optimise the expression (via de-duplication in alt).
• Does not wrap the result in a group, which, because alternation has the lowest precedence in regexes, runs the risk of behaving unexpectedly if the result is then combined with further regexes. tl;dr - one of the grouping variants should almost always be preferred.

or’ then cg:

• (asb|bsa|a|b)

Notes:

• Unlike most other -cg fns, this one does not accept any number of res.
• May optimise the expression (via de-duplication in alt).

or’ then fgrp:

• (?flgs:asb|bsa|a|b)

Notes:

• Unlike most other -flgs fns, this one does not accept any number of res.
• May optimise the expression (via de-duplication in alt).

or’ then grp:

• (?:asb|bsa|a|b)

Notes:

• Unlike most other -grp fns, this one does not accept any number of res.
• May optimise the expression (via de-duplication in alt).

or’ then ncg:

• (?<nm>asb|bsa|a|b)

Notes:

• Unlike most other -ncg fns, this one does not accept any number of res.
• May optimise the expression (via de-duplication in alt).

Quotes re:

• \Qre\E

Returns an empty regex (#"") if re is empty?’.

Is x a regex?

Notes:

• ClojureScript already has a regexp? predicate in cljs.core, but ClojureJVM doesn’t. See this ask.clojure.org post.

Returns the String representation of x, with special handling for RegExp objects on ClojureScript in an attempt to correct JavaScript’s APPALLING default stringification.

Notes:

• Embeds flags (as per embed-flags).

Returns an ‘exclusive or’ regex that will match a or b, but not both:

• a|b

This is identical to alt called with 2 arguments, but is provided as a convenience for those who might be building up large logic based regexes and would prefer to use more easily understood logical operator names throughout.

Notes:

• May optimise the expression (via de-duplication in alt).
• Does not wrap the result in a group, which, because alternation has the lowest precedence in regexes, runs the risk of behaving unexpectedly if the result is then combined with further regexes. tl;dr - one of the grouping variants should almost always be preferred.

xor’ then cg:

• (a|b)

Notes:

• Unlike most other -cg fns, this one does not accept any number of res.
• May optimise the expression (via de-duplication in alt).

xor’ then fgrp:

• (?flgs:a|b)

Notes:

• Unlike most other -fgrp fns, this one does not accept any number of res.
• May optimise the expression (via de-duplication in alt).

xor’ then grp:

• (?:a|b)

Notes:

• Unlike most other -grp fns, this one does not accept any number of res.
• May optimise the expression (via de-duplication in alt).

xor’ then ncg:

• (?<nm>a|b)

Notes:

• Unlike most other -ncg fns, this one does not accept any number of res.
• May optimise the expression (via de-duplication in alt).

Returns a regex where re will match zero or more times:

• re*

Returns an empty regex (#"") if re is empty?’.

cg then zom:

• (res)*

chcl then zom:

• [res]*

fgrp then zom:

• (?flgs:res)*

grp then zom:

• (?:res)*

ncg then zom:

• (?<nm>res)*