vlib: remove ustring usage (#10618)

doc/docs.md

@@ -505,6 +505,18 @@ s := '42'
 n := s.int() // 42
 ```
 
+### Runes
+A `rune` represents a unicode character and is an alias for `u32`. Runes can be created like this:
+```v
+x := `🚀`
+```
+
+A string can be converted to runes by the `.runes()` method.
+```v
+hello := 'Hello World 👋'
+hello_runes := hello.runes() // [`H`, `e`, `l`, `l`, `o`, ` `, `W`, `o`, `r`, `l`, `d`, ` `, `👋`]
+```
+
 ### String interpolation
 
 Basic interpolation syntax is pretty simple - use `$` before a variable name.

vlib/builtin/rune.v

@@ -28,6 +28,15 @@ pub fn (c rune) str() string {
 	*/
 }
 
+// string converts a rune array to a string
+pub fn (ra []rune) string() string {
+	mut res := ''
+	for r in ra {
+		res += r.str()
+	}
+	return res
+}
+
 // Define this on byte as well, so that we can do `s[0].is_capital()`
 pub fn (c byte) is_capital() bool {
 	return c >= `A` && c <= `Z`
@@ -42,7 +51,7 @@ pub fn (b []byte) clone() []byte {
 	return res
 }
 
-// TODO remove this once runes are implemented
+// TODO: remove this once runes are implemented
 pub fn (b []byte) bytestr() string {
 	unsafe {
 		buf := malloc_noscan(b.len + 1)

vlib/builtin/string.v

@@ -73,7 +73,8 @@ pub fn (s string) runes() []rune {
 	for i := 0; i < s.len; i++ {
 		char_len := utf8_char_len(unsafe { s.str[i] })
 		if char_len > 1 {
-			mut r := unsafe { s[i..i + char_len] }
+			end := if s.len - 1 >= i + char_len { i + char_len } else { s.len }
+			mut r := unsafe { s[i..end] }
 			runes << r.utf32_code()
 			i += char_len - 1
 		} else {

vlib/os/notify/notify_test.v

@@ -1,6 +1,5 @@
 import os
 import os.notify
-import time
 
 // make a pipe and return the (read, write) file descriptors
 fn make_pipe() ?(int, int) {

vlib/os/os.v

@@ -90,18 +90,6 @@ pub fn read_lines(path string) ?[]string {
 	return res
 }
 
-// read_ulines reads the file in `path` into an array of ustring lines.
-fn read_ulines(path string) ?[]ustring {
-	lines := read_lines(path) ?
-	// mut ulines := new_array(0, lines.len, sizeof(ustring))
-	mut ulines := []ustring{}
-	for myline in lines {
-		// ulines[i] = ustr
-		ulines << myline.ustring()
-	}
-	return ulines
-}
-
 // sigint_to_signal_name will translate `si` signal integer code to it's string code representation.
 pub fn sigint_to_signal_name(si int) string {
 	// POSIX signals:

vlib/readline/readline.v

@@ -33,13 +33,13 @@ pub struct Readline {
 mut:
 	is_raw            bool
 	orig_termios      Termios // Linux
-	current           ustring // Line being edited
+	current           []rune  // Line being edited
 	cursor            int     // Cursor position
 	overwrite         bool
 	cursor_row_offset int
 	prompt            string
 	prompt_offset     int
-	previous_lines    []ustring
+	previous_lines    [][]rune
 	search_index      int
 	is_tty            bool
 }

vlib/readline/readline_default.c.v

@@ -17,58 +17,58 @@ import os
 // read_line_utf8 blocks execution in a loop and awaits user input
 // characters from a terminal until `EOF` or `Enter` key is encountered
 // in the input stream.
-// read_line_utf8 returns the complete input line as an UTF-8 encoded `ustring` or
+// read_line_utf8 returns the complete input line as an UTF-8 encoded `[]rune` or
 // an error if the line is empty.
 // The `prompt` `string` is output as a prefix text for the input capturing.
 // read_line_utf8 is the main method of the `readline` module and `Readline` struct.
-pub fn (mut r Readline) read_line_utf8(prompt string) ?ustring {
-	r.current = ''.ustring()
+pub fn (mut r Readline) read_line_utf8(prompt string) ?[]rune {
+	r.current = []rune{}
 	r.cursor = 0
 	r.prompt = prompt
 	r.search_index = 0
 	if r.previous_lines.len <= 1 {
-		r.previous_lines << ''.ustring()
-		r.previous_lines << ''.ustring()
+		r.previous_lines << []rune{}
+		r.previous_lines << []rune{}
 	} else {
-		r.previous_lines[0] = ''.ustring()
+		r.previous_lines[0] = []rune{}
 	}
 	print(r.prompt)
 	line := os.get_raw_line()
 	if line.len >= 0 {
-		r.current = line.ustring()
+		r.current = line.runes()
 	}
-	r.previous_lines[0] = ''.ustring()
+	r.previous_lines[0] = []rune{}
 	r.search_index = 0
-	if r.current.s == '' {
+	if r.current.len == 0 {
 		return error('empty line')
 	}
 	return r.current
 }
 
 // read_line does the same as `read_line_utf8` but returns user input as a `string`.
-// (As opposed to `ustring` returned by `read_line_utf8`).
+// (As opposed to `[]rune` returned by `read_line_utf8`).
 pub fn (mut r Readline) read_line(prompt string) ?string {
 	s := r.read_line_utf8(prompt) ?
-	return s.s
+	return s.string()
 }
 
 // read_line_utf8 blocks execution in a loop and awaits user input
 // characters from a terminal until `EOF` or `Enter` key is encountered
 // in the input stream.
-// read_line_utf8 returns the complete input line as an UTF-8 encoded `ustring` or
+// read_line_utf8 returns the complete input line as an UTF-8 encoded `[]rune` or
 // an error if the line is empty.
 // The `prompt` `string` is output as a prefix text for the input capturing.
 // read_line_utf8 is the main method of the `readline` module and `Readline` struct.
 // NOTE that this version of `read_line_utf8` is a standalone function without
 // persistent functionalities (e.g. history).
-pub fn read_line_utf8(prompt string) ?ustring {
+pub fn read_line_utf8(prompt string) ?[]rune {
 	mut r := Readline{}
 	s := r.read_line_utf8(prompt) ?
 	return s
 }
 
 // read_line does the same as `read_line_utf8` but returns user input as a `string`.
-// (As opposed to `ustring` as returned by `read_line_utf8`).
+// (As opposed to `[]rune` as returned by `read_line_utf8`).
 // NOTE that this version of `read_line` is a standalone function without
 // persistent functionalities (e.g. history).
 pub fn read_line(prompt string) ?string {

vlib/readline/readline_linux.c.v

@@ -101,21 +101,21 @@ pub fn (r Readline) read_char() int {
 // read_line_utf8 blocks execution in a loop and awaits user input
 // characters from a terminal until `EOF` or `Enter` key is encountered
 // in the input stream.
-// read_line_utf8 returns the complete input line as an UTF-8 encoded `ustring` or
+// read_line_utf8 returns the complete input line as an UTF-8 encoded `[]rune` or
 // an error if the line is empty.
 // The `prompt` `string` is output as a prefix text for the input capturing.
 // read_line_utf8 is the main method of the `readline` module and `Readline` struct.
-pub fn (mut r Readline) read_line_utf8(prompt string) ?ustring {
-	r.current = ''.ustring()
+pub fn (mut r Readline) read_line_utf8(prompt string) ?[]rune {
+	r.current = []rune{}
 	r.cursor = 0
 	r.prompt = prompt
 	r.search_index = 0
 	r.prompt_offset = get_prompt_offset(prompt)
 	if r.previous_lines.len <= 1 {
-		r.previous_lines << ''.ustring()
-		r.previous_lines << ''.ustring()
+		r.previous_lines << []rune{}
+		r.previous_lines << []rune{}
 	} else {
-		r.previous_lines[0] = ''.ustring()
+		r.previous_lines[0] = []rune{}
 	}
 	if !r.is_raw {
 		r.enable_raw_mode()
@@ -129,39 +129,39 @@ pub fn (mut r Readline) read_line_utf8(prompt string) ?ustring {
 			break
 		}
 	}
-	r.previous_lines[0] = ''.ustring()
+	r.previous_lines[0] = []rune{}
 	r.search_index = 0
 	r.disable_raw_mode()
-	if r.current.s == '' {
+	if r.current.len == 0 {
 		return error('empty line')
 	}
 	return r.current
 }
 
 // read_line does the same as `read_line_utf8` but returns user input as a `string`.
-// (As opposed to `ustring` returned by `read_line_utf8`).
+// (As opposed to `[]rune` returned by `read_line_utf8`).
 pub fn (mut r Readline) read_line(prompt string) ?string {
 	s := r.read_line_utf8(prompt) ?
-	return s.s
+	return s.string()
 }
 
 // read_line_utf8 blocks execution in a loop and awaits user input
 // characters from a terminal until `EOF` or `Enter` key is encountered
 // in the input stream.
-// read_line_utf8 returns the complete input line as an UTF-8 encoded `ustring` or
+// read_line_utf8 returns the complete input line as an UTF-8 encoded `[]rune` or
 // an error if the line is empty.
 // The `prompt` `string` is output as a prefix text for the input capturing.
 // read_line_utf8 is the main method of the `readline` module and `Readline` struct.
 // NOTE that this version of `read_line_utf8` is a standalone function without
 // persistent functionalities (e.g. history).
-pub fn read_line_utf8(prompt string) ?ustring {
+pub fn read_line_utf8(prompt string) ?[]rune {
 	mut r := Readline{}
 	s := r.read_line_utf8(prompt) ?
 	return s
 }
 
 // read_line does the same as `read_line_utf8` but returns user input as a `string`.
-// (As opposed to `ustring` as returned by `read_line_utf8`).
+// (As opposed to `[]rune` as returned by `read_line_utf8`).
 // NOTE that this version of `read_line` is a standalone function without
 // persistent functionalities (e.g. history).
 pub fn read_line(prompt string) ?string {
@@ -373,14 +373,14 @@ fn (mut r Readline) refresh_line() {
 	mut end_of_input := [0, 0]
 	end_of_input = calculate_screen_position(r.prompt.len, 0, get_screen_columns(), r.current.len,
 		end_of_input)
-	end_of_input[1] += r.current.count('\n'.ustring())
+	end_of_input[1] += r.current.filter(it == `\n`).len
 	mut cursor_pos := [0, 0]
 	cursor_pos = calculate_screen_position(r.prompt.len, 0, get_screen_columns(), r.cursor,
 		cursor_pos)
 	shift_cursor(0, -r.cursor_row_offset)
 	term.erase_toend()
 	print(r.prompt)
-	print(r.current)
+	print(r.current.string())
 	if end_of_input[0] == 0 && end_of_input[1] > 0 {
 		print('\n')
 	}
@@ -401,11 +401,9 @@ fn (mut r Readline) eof() bool {
 // insert_character inserts the character `c` at current cursor position.
 fn (mut r Readline) insert_character(c int) {
 	if !r.overwrite || r.cursor == r.current.len {
-		r.current = r.current.left(r.cursor).ustring() + utf32_to_str(u32(c)).ustring() +
-			r.current.right(r.cursor).ustring()
+		r.current.insert(r.cursor, c)
 	} else {
-		r.current = r.current.left(r.cursor).ustring() + utf32_to_str(u32(c)).ustring() +
-			r.current.right(r.cursor + 1).ustring()
+		r.current[r.cursor] = rune(c)
 	}
 	r.cursor++
 	// Refresh the line to add the new character
@@ -420,7 +418,7 @@ fn (mut r Readline) delete_character() {
 		return
 	}
 	r.cursor--
-	r.current = r.current.left(r.cursor).ustring() + r.current.right(r.cursor + 1).ustring()
+	r.current.delete(r.cursor)
 	r.refresh_line()
 }
 
@@ -429,15 +427,14 @@ fn (mut r Readline) suppr_character() {
 	if r.cursor > r.current.len {
 		return
 	}
-	r.current = r.current.left(r.cursor).ustring() + r.current.right(r.cursor + 1).ustring()
+	r.current.delete(r.cursor)
 	r.refresh_line()
 }
 
 // commit_line adds a line break and then stops the main loop.
 fn (mut r Readline) commit_line() bool {
 	r.previous_lines.insert(1, r.current)
-	a := '\n'.ustring()
-	r.current += a
+	r.current << `\n`
 	r.cursor = r.current.len
 	if r.is_tty {
 		r.refresh_line()
@@ -483,9 +480,9 @@ fn (r Readline) is_break_character(c string) bool {
 // move_cursor_word_left moves the cursor relative one word length worth to the left.
 fn (mut r Readline) move_cursor_word_left() {
 	if r.cursor > 0 {
-		for ; r.cursor > 0 && r.is_break_character(r.current.at(r.cursor - 1)); r.cursor-- {
+		for ; r.cursor > 0 && r.is_break_character(r.current[r.cursor - 1].str()); r.cursor-- {
 		}
-		for ; r.cursor > 0 && !r.is_break_character(r.current.at(r.cursor - 1)); r.cursor-- {
+		for ; r.cursor > 0 && !r.is_break_character(r.current[r.cursor - 1].str()); r.cursor-- {
 		}
 		r.refresh_line()
 	}
@@ -494,9 +491,9 @@ fn (mut r Readline) move_cursor_word_left() {
 // move_cursor_word_right moves the cursor relative one word length worth to the right.
 fn (mut r Readline) move_cursor_word_right() {
 	if r.cursor < r.current.len {
-		for ; r.cursor < r.current.len && r.is_break_character(r.current.at(r.cursor)); r.cursor++ {
+		for ; r.cursor < r.current.len && r.is_break_character(r.current[r.cursor].str()); r.cursor++ {
 		}
-		for ; r.cursor < r.current.len && !r.is_break_character(r.current.at(r.cursor)); r.cursor++ {
+		for ; r.cursor < r.current.len && !r.is_break_character(r.current[r.cursor].str()); r.cursor++ {
 		}
 		r.refresh_line()
 	}

vlib/readline/readline_windows.c.v

@@ -16,55 +16,55 @@ import os
 // read_line_utf8 blocks execution in a loop and awaits user input
 // characters from a terminal until `EOF` or `Enter` key is encountered
 // in the input stream.
-// read_line_utf8 returns the complete input line as an UTF-8 encoded `ustring` or
+// read_line_utf8 returns the complete input line as an UTF-8 encoded `[]rune` or
 // an error if the line is empty.
 // The `prompt` `string` is output as a prefix text for the input capturing.
 // read_line_utf8 is the main method of the `readline` module and `Readline` struct.
-pub fn (mut r Readline) read_line_utf8(prompt string) ?ustring {
-	r.current = ''.ustring()
+pub fn (mut r Readline) read_line_utf8(prompt string) ?[]rune {
+	r.current = []rune{}
 	r.cursor = 0
 	r.prompt = prompt
 	r.search_index = 0
 	if r.previous_lines.len <= 1 {
-		r.previous_lines << ''.ustring()
-		r.previous_lines << ''.ustring()
+		r.previous_lines << []rune{}
+		r.previous_lines << []rune{}
 	} else {
-		r.previous_lines[0] = ''.ustring()
+		r.previous_lines[0] = []rune{}
 	}
 	print(r.prompt)
-	r.current = os.get_raw_line().ustring()
-	r.previous_lines[0] = ''.ustring()
+	r.current = os.get_raw_line().runes()
+	r.previous_lines[0] = []rune{}
 	r.search_index = 0
-	if r.current.s == '' {
+	if r.current.len == 0 {
 		return error('empty line')
 	}
 	return r.current
 }
 
 // read_line does the same as `read_line_utf8` but returns user input as a `string`.
-// (As opposed to `ustring` returned by `read_line_utf8`).
+// (As opposed to `[]rune` returned by `read_line_utf8`).
 pub fn (mut r Readline) read_line(prompt string) ?string {
 	s := r.read_line_utf8(prompt) ?
-	return s.s
+	return s.string()
 }
 
 // read_line_utf8 blocks execution in a loop and awaits user input
 // characters from a terminal until `EOF` or `Enter` key is encountered
 // in the input stream.
-// read_line_utf8 returns the complete input line as an UTF-8 encoded `ustring` or
+// read_line_utf8 returns the complete input line as an UTF-8 encoded `[]rune` or
 // an error if the line is empty.
 // The `prompt` `string` is output as a prefix text for the input capturing.
 // read_line_utf8 is the main method of the `readline` module and `Readline` struct.
 // NOTE that this version of `read_line_utf8` is a standalone function without
 // persistent functionalities (e.g. history).
-pub fn read_line_utf8(prompt string) ?ustring {
+pub fn read_line_utf8(prompt string) ?[]rune {
 	mut r := Readline{}
 	s := r.read_line_utf8(prompt) ?
 	return s
 }
 
 // read_line does the same as `read_line_utf8` but returns user input as a `string`.
-// (As opposed to `ustring` as returned by `read_line_utf8`).
+// (As opposed to `[]rune` as returned by `read_line_utf8`).
 // NOTE that this version of `read_line` is a standalone function without
 // persistent functionalities (e.g. history).
 pub fn read_line(prompt string) ?string {

vlib/v/fmt/tests/structs_expected.vv

@@ -1,6 +1,6 @@
 struct User {
-	name                  string  // name
-	name2                 ustring // name2
+	name                  string // name
+	name2                 []rune // name2
 	very_long_field       bool
 	age                   int // age
 	very_long_type_field1 very_looooog_type     // long

vlib/v/fmt/tests/structs_input.vv

@@ -1,6 +1,6 @@
 struct User {
 	name string // name
-	name2 ustring // name2
+	name2 []rune // name2
 	very_long_field bool
 	age int // age
 	very_long_type_field1 very_looooog_type // long

vlib/v/gen/c/cgen.v

@@ -3149,7 +3149,7 @@ fn (mut g Gen) expr(node ast.Expr) {
 			} else {
 				// TODO: optimize use L-char instead of u32 when possible
 				if utf8_str_len(node.val) < node.val.len {
-					g.write('((u32)0x$node.val.utf32_code().hex())')
+					g.write('((rune)0x$node.val.utf32_code().hex() /* `$node.val` */)')
 				} else {
 					g.write("'$node.val'")
 				}

vlib/v/scanner/scanner.v

@@ -1270,7 +1270,7 @@ fn (mut s Scanner) ident_char() string {
 	len--
 	c := s.text[start + 1..s.pos]
 	if len != 1 {
-		u := c.ustring()
+		u := c.runes()
 		if u.len != 1 {
 			s.error('invalid character literal (more than one character)\n' +
 				'use quotes for strings, backticks for characters')

vlib/v/tests/sumtype_equality_test.v

@@ -1,4 +1,4 @@
-type Str = string | ustring
+type Str = rune | string
 
 struct Foo {
 	v int
@@ -13,8 +13,8 @@ type FooBar = Bar | Foo
 fn test_sumtype_equality() {
 	s1 := Str('s')
 	s2 := Str('s2')
-	u1 := Str('s1'.ustring())
-	u2 := Str('s2'.ustring())
+	u1 := Str(`A`)
+	u2 := Str(`B`)
 	assert s1 == s1
 	assert u1 == u1
 	assert s1 != s2