builtin: implement a JS version of string.split_any (#21612)

vlib/builtin/js/jsfns.js.v

@@ -19,6 +19,12 @@ pub interface JS.Number {
 	JS.Any
 }
 
+pub interface JS.RegExp {
+	JS.Any
+}
+
+pub type SplitSeparator = JS.RegExp | JS.String
+
 @[single_impl]
 pub interface JS.String {
 	JS.Any
@@ -32,7 +38,7 @@ pub interface JS.String {
 	endsWith(substr JS.String) JS.Boolean
 	startsWith(substr JS.String) JS.Boolean
 	slice(a JS.Number, b JS.Number) JS.String
-	split(dot JS.String) JS.Array
+	split(delim SplitSeparator) JS.Array
 	indexOf(needle JS.String) JS.Number
 	lastIndexOf(needle JS.String) JS.Number
 }

vlib/builtin/js/string.js.v

@@ -85,6 +85,46 @@ pub fn (s string) split(dot string) []string {
 	return arr
 }
 
+pub fn (s string) split_any(delim string) []string {
+	if delim.len == 0 {
+		return s.split(delim)
+	}
+
+	mut pattern := delim
+
+	// we use a regex with a bracket expression to match any of the characters in delim
+	// so we need to prevent the caller from escaping the regex
+	// to do this we escape any `]`, and remove all `\` while adding an escaped `\\`
+	// back if the original string contained any
+	if pattern.contains('\\') {
+		pattern = pattern.replace('\\', '')
+		pattern = '${pattern}\\\\'
+	}
+	pattern = pattern.replace(']', '\\]')
+
+	mut regexp := JS.RegExp{}
+	#regexp = new RegExp('[' + pattern.str + ']', 'g')
+
+	tmparr := s.str.split(regexp).map(fn (it JS.Any) JS.Any {
+		res := ''
+		#res.str = it
+
+		return res
+	})
+	_ := tmparr
+
+	mut arr := []string{}
+	#arr = new array(new array_buffer({arr: tmparr,index_start: new int(0),len: new int(tmparr.length)}))
+
+	// FIXME: ugly hack to handle edge case where the last character in the string is
+	// one of the delimiters to match V behavior
+	#if (s.len > 0 && pattern.str.includes(s.str[s.len - 1])) {
+	arr.pop()
+	#}
+
+	return arr
+}
+
 pub fn (s string) bytes() []u8 {
 	sep := ''
 	tmparr := s.str.split(sep.str).map(fn (it JS.Any) JS.Any {

vlib/builtin/js/string_test.js.v

@@ -232,6 +232,51 @@ fn test_split() {
 	assert vals[1] == ''
 }
 
+fn test_split_any() {
+	mut s := 'aaa'
+	mut a := s.split_any('')
+	assert a.len == 3
+	assert a[0] == 'a'
+	assert a[1] == 'a'
+	assert a[2] == 'a'
+	s = ''
+	a = s.split_any('')
+	assert a.len == 0
+	s = '12131415'
+	a = s.split_any('1')
+	assert a.len == 5
+	assert a[0] == ''
+	assert a[1] == '2'
+	assert a[2] == '3'
+	assert a[3] == '4'
+	assert a[4] == '5'
+	s = '12131415'
+	a = s.split_any('2345')
+	assert a.len == 4
+	assert a[0] == '1'
+	assert a[1] == '1'
+	assert a[2] == '1'
+	assert a[3] == '1'
+	s = 'a,b,c'
+	a = s.split_any('],')
+	assert a.len == 3
+	assert a[0] == 'a'
+	assert a[1] == 'b'
+	assert a[2] == 'c'
+	s = 'a]b]c'
+	a = s.split_any('],')
+	assert a.len == 3
+	assert a[0] == 'a'
+	assert a[1] == 'b'
+	assert a[2] == 'c'
+	s = 'a]b]c'
+	a = s.split_any('],\\')
+	assert a.len == 3
+	assert a[0] == 'a'
+	assert a[1] == 'b'
+	assert a[2] == 'c'
+}
+
 /*
 fn test_trim_space() {
 	a := ' a '