@grain/stdlib 0.6.6 → 0.7.1

package/char.gr

@@ -16,8 +16,8 @@ module Char
 from "runtime/unsafe/wasmi32" include WasmI32
 from "runtime/dataStructures" include DataStructures
 use DataStructures.{ tagSimpleNumber, tagChar, untagChar, allocateString }
-
-exception MalformedUtf8
+from "runtime/utf8" include Utf8
+use Utf8.{ usvEncodeLength, writeUtf8CodePoint }
 
 /**
  * The minimum valid Unicode scalar value.
@@ -44,9 +44,9 @@ provide let max = 0x10FFFF
  * @since v0.3.0
  */
 provide let isValid = charCode => {
-  charCode >= min &&
-    (charCode <= 0xD7FF || charCode >= 0xE000) &&
-    charCode <= max
+  charCode >= min
+    && (charCode <= 0xD7FF || charCode >= 0xE000)
+    && charCode <= max
 }
 
 /**
@@ -164,52 +164,48 @@ provide let pred = char => {
  */
 @unsafe
 provide let toString = (char: Char) => {
-  use WasmI32.{
-    (+),
-    (-),
-    (*),
-    (&),
-    (|),
-    (>>>),
-    ltU as (<),
-    gtU as (>),
-    leU as (<=),
-  }
-
+  use WasmI32.{ (+) }
   let usv = untagChar(char)
+  let byteCount = usvEncodeLength(usv)
+  let string = allocateString(byteCount)
+  writeUtf8CodePoint(string + 8n, usv)
+  WasmI32.toGrain(string): String
+}
 
-  let result = if (usv < 0x80n) {
-    let string = allocateString(1n)
-    WasmI32.store8(string, usv, 8n)
-    WasmI32.toGrain(string): String
-  } else {
-    let mut count = 0n
-    let mut offset = 0n
-    if (usv <= 0x07FFn) {
-      count = 1n
-      offset = 0xC0n
-    } else if (usv <= 0xFFFFn) {
-      count = 2n
-      offset = 0xE0n
-    } else {
-      count = 3n
-      offset = 0xF0n
-    }
-    let string = allocateString(count + 1n)
-    WasmI32.store8(string, (usv >>> (6n * count)) + offset, 8n)
-
-    let mut n = 0n
-    while (count > 0n) {
-      n += 1n
-      let temp = usv >>> (6n * (count - 1n))
-      WasmI32.store8(string + n, 0x80n | temp & 0x3Fn, 8n)
-      count -= 1n
-    }
+/**
+ * Byte encodings
+ *
+ * @since v0.7.0
+ */
+provide enum Encoding {
+  UTF8,
+  UTF16,
+  UTF32,
+}
 
-    WasmI32.toGrain(string): String
+/**
+ * Returns the byte count of a character if encoded in the given encoding.
+ *
+ * @param encoding: The encoding to check
+ * @param char: The character
+ * @returns The byte count of the character in the given encoding
+ *
+ * @example Char.encodedLength(Char.UTF8, 'a') == 1
+ * @example Char.encodedLength(Char.UTF8, '🌾') == 4
+ * @example Char.encodedLength(Char.UTF16, '©') == 1
+ *
+ * @since v0.7.0
+ */
+@unsafe
+provide let encodedLength = (encoding, char: Char) => {
+  let usv = untagChar(char)
+  let utf8ByteCount = usvEncodeLength(usv)
+  let utf8ByteCount = tagSimpleNumber(utf8ByteCount)
+  match (encoding) {
+    UTF32 => 4,
+    UTF16 => if (utf8ByteCount == 4) 2 else 1,
+    UTF8 => utf8ByteCount,
   }
-
-  result
 }
 
 /**
@@ -315,64 +311,170 @@ provide let (>=) = (x: Char, y: Char) => {
 }
 
 /**
- * Checks if the character is an ASCII digit.
+ * Utilities for working with ASCII characters.
  *
- * @param char: The character to check
- * @returns `true` if the character is an ASCII digit or `false` otherwise
- * 
- * @example assert Char.isAsciiDigit('1')
- * @example assert !Char.isAsciiDigit('a')
+ * @example Char.Ascii.isAscii('1')
  *
- * @since v0.6.0
+ * @since v0.7.0
  */
-provide let isAsciiDigit = char => char >= '0' && char <= '9'
+provide module Ascii {
+  /**
+   * The minimum valid ASCII character code.
+   *
+   * @since v0.7.0
+   */
+  provide let min = 0x00
 
-/**
- * Checks if the character is an ASCII alphabetical character.
- *
- * @param char: The character to check
- * @returns `true` if the character is an ASCII alphabetical or `false` otherwise
- *
- * @example assert Char.isAsciiAlpha('a')
- * @example assert !Char.isAsciiAlpha('1')
- *
- * @since v0.6.0
- */
-provide let isAsciiAlpha = char =>
-  char >= 'a' && char <= 'z' || char >= 'A' && char <= 'Z'
+  /**
+   * The maximum valid ASCII character code.
+   *
+   * @since v0.7.0
+   */
+  provide let max = 0x7F
 
-/**
- * Converts the character to ASCII lowercase if it is an ASCII uppercase character.
- * 
- * @param char: The character to convert
- * @returns The lowercased character
- * 
- * @example assert Char.toAsciiLowercase('B') == 'b'
- * 
- * @since v0.6.0
- */
-provide let toAsciiLowercase = char => {
-  if (char >= 'A' && char <= 'Z') {
-    fromCode(code(char) + 0x20)
-  } else {
-    char
+  /**
+   * Checks if the character is a valid ASCII character.
+   *
+   * @param char: The character to check
+   * @returns `true` if the character is an ASCII character or `false` otherwise
+   *
+   * @example assert Char.Ascii.isValid('1')
+   * @example assert Char.Ascii.isValid('a')
+   * @example assert !Char.Ascii.isValid('🌾')
+   *
+   * @since v0.7.0
+   */
+  provide let isValid = char => char <= '\u{007F}'
+
+  /**
+   * Checks if the character is an ASCII digit.
+   *
+   * @param char: The character to check
+   * @returns `true` if the character is an ASCII digit or `false` otherwise
+   *
+   * @example assert Char.Ascii.isDigit('1')
+   * @example assert !Char.Ascii.isDigit('a')
+   *
+   * @since v0.7.0
+   * @history v0.6.0: Originally `Char.isAsciiDigit`
+   */
+  provide let isDigit = char => char >= '0' && char <= '9'
+
+  /**
+   * Checks if the character is an ASCII alphabetical character.
+   *
+   * @param char: The character to check
+   * @returns `true` if the character is an ASCII alphabetical or `false` otherwise
+   *
+   * @example assert Char.Ascii.isAlpha('a')
+   * @example assert !Char.Ascii.isAlpha('1')
+   *
+   * @since v0.7.0
+   * @history v0.6.0: Originally `Char.isAsciiAlpha`
+   */
+  provide let isAlpha = char =>
+    char >= 'a' && char <= 'z' || char >= 'A' && char <= 'Z'
+
+  /**
+  * Checks if the character is an ASCII control character.
+  *
+  * @param char: The character to check
+  * @returns `true` if the character is an ASCII control character or `false` otherwise
+  *
+  * @example assert Char.Ascii.isControl('\t')
+  * @example assert Char.Ascii.isControl('\n')
+  * @example assert !Char.Ascii.isControl('1')
+  * @example assert !Char.Ascii.isControl('a')
+  *
+  * @since v0.7.0
+  */
+  provide let isControl = char => char <= '\u{001F}' || char == '\u{007F}'
+
+  /**
+   * Checks if the character is an ASCII whitespace character.
+   *
+   * @param char: The character to check
+   * @returns `true` if the character is an ASCII whitespace character or `false` otherwise
+   *
+   * @example assert Char.isWhitespace('\t')
+   * @example assert Char.isWhitespace('\n')
+   * @example assert !Char.isWhitespace('1')
+   * @example assert !Char.isWhitespace('a')
+   *
+   * @since v0.7.0
+   */
+  provide let isWhitespace = char => {
+    match (char) {
+      '\t' | '\n' | '\x0C' | '\r' | ' ' => true,
+      _ => false,
+    }
   }
-}
 
-/**
- * Converts the character to ASCII uppercase if it is an ASCII lowercase character.
- * 
- * @param char: The character to convert
- * @returns The uppercased character
- * 
- * @example assert Char.toAsciiUppercase('b') == 'B'
- * 
- * @since v0.6.0
- */
-provide let toAsciiUppercase = char => {
-  if (char >= 'a' && char <= 'z') {
-    fromCode(code(char) - 0x20)
-  } else {
-    char
+  /**
+   * Checks if the character is an ASCII punctuation character.
+   *
+   * @param char: The character to check
+   * @returns `true` if the character is an ASCII punctuation character or `false` otherwise
+   *
+   * @example assert Char.Ascii.isPunctuation('!')
+   * @example assert !Char.Ascii.isPunctuation('1')
+   *
+   * @since v0.7.0
+   */
+  provide let isPunctuation = char =>
+    char >= '!' && char <= '/'
+      || char >= ':' && char <= '@'
+      || char >= '[' && char <= '`'
+      || char >= '{' && char <= '~'
+
+  /**
+   * Checks if the character is an ASCII graphic character.
+   *
+   * @param char: The character to check
+   * @returns `true` if the character is an ASCII graphic character or `false` otherwise
+   *
+   * @example assert Char.Ascii.isGraphic('!')
+   * @example assert !Char.Ascii.isGraphic('\t')
+   *
+   * @since v0.7.0
+   */
+  provide let isGraphic = char => char >= '!' && char <= '~'
+
+  /**
+   * Converts the character to ASCII lowercase if it is an ASCII uppercase character.
+   *
+   * @param char: The character to convert
+   * @returns The lowercased character
+   *
+   * @example assert Char.Ascii.toLowercase('B') == 'b'
+   *
+   * @since v0.7.0
+   * @history v0.6.0: Originally `Char.toAsciiLowercase`
+   */
+  provide let toLowercase = char => {
+    if (char >= 'A' && char <= 'Z') {
+      fromCode(code(char) + 0x20)
+    } else {
+      char
+    }
+  }
+
+  /**
+   * Converts the character to ASCII uppercase if it is an ASCII lowercase character.
+   *
+   * @param char: The character to convert
+   * @returns The uppercased character
+   *
+   * @example assert Char.Ascii.toUppercase('b') == 'B'
+   *
+   * @since v0.7.0
+   * @history v0.6.0: Originally `Char.toAsciiUppercase`
+   */
+  provide let toUppercase = char => {
+    if (char >= 'a' && char <= 'z') {
+      fromCode(code(char) - 0x20)
+    } else {
+      char
+    }
   }
 }