feat(stdlib): Add Char.encodedLength (#2238)

Co-authored-by: Oscar Spencer <[email protected]>
grain-lang · Feb 16, 2025 · c549fac · c549fac
1 parent bdb7f7c
commit c549fac
Show file tree

Hide file tree

Showing 3 changed files with 111 additions and 0 deletions.
diff --git a/compiler/test/stdlib/char.test.gr b/compiler/test/stdlib/char.test.gr
@@ -40,6 +40,20 @@ assert Char.code(Char.pred(Char.fromCode(0xE000))) == 0xD7FF
 assert Char.toString('f') == "f"
 assert Char.toString('💯') == "💯"
 
+// byteCount
+assert Char.encodedLength(Char.UTF8, 'a') == 1
+assert Char.encodedLength(Char.UTF8, '©') == 2
+assert Char.encodedLength(Char.UTF8, '☃') == 3
+assert Char.encodedLength(Char.UTF8, '🌾') == 4
+assert Char.encodedLength(Char.UTF16, 'a') == 1
+assert Char.encodedLength(Char.UTF16, '©') == 1
+assert Char.encodedLength(Char.UTF16, '☃') == 1
+assert Char.encodedLength(Char.UTF16, '🌾') == 2
+assert Char.encodedLength(Char.UTF32, 'a') == 4
+assert Char.encodedLength(Char.UTF32, '©') == 4
+assert Char.encodedLength(Char.UTF32, '☃') == 4
+assert Char.encodedLength(Char.UTF32, '🌾') == 4
+
 // issue #927
 let chars = [> '\u{1F3F4}', '\u{E0067}']
 let mut charPosition = 0

diff --git a/stdlib/char.gr b/stdlib/char.gr
@@ -172,6 +172,42 @@ provide let toString = (char: Char) => {
   WasmI32.toGrain(string): String
 }
 
+/**
+ * Byte encodings
+ *
+ * @since v0.7.0
+ */
+provide enum Encoding {
+  UTF8,
+  UTF16,
+  UTF32,
+}
+
+/**
+ * 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,
+  }
+}
+
 /**
  * Checks if the first character is less than the second character by Unicode scalar value.
  *

diff --git a/stdlib/char.md b/stdlib/char.md
@@ -27,6 +27,27 @@ from "char" include Char
 '🌾'
 ```
 
+## Types
+
+Type declarations included in the Char module.
+
+### Char.**Encoding**
+
+<details disabled>
+<summary tabindex="-1">Added in <code>next</code></summary>
+No other changes yet.
+</details>
+
+```grain
+enum Encoding {
+  UTF8,
+  UTF16,
+  UTF32,
+}
+```
+
+Byte encodings
+
 ## Values
 
 Functions and constants included in the Char module.
@@ -285,6 +306,46 @@ Char.toString('a') == "a"
 Char.toString('🌾') == "🌾"
 ```
 
+### Char.**encodedLength**
+
+<details disabled>
+<summary tabindex="-1">Added in <code>next</code></summary>
+No other changes yet.
+</details>
+
+```grain
+encodedLength : (encoding: Encoding, char: Char) => Number
+```
+
+Returns the byte count of a character if encoded in the given encoding.
+
+Parameters:
+
+|param|type|description|
+|-----|----|-----------|
+|`encoding`|`Encoding`|The encoding to check|
+|`char`|`Char`|The character|
+
+Returns:
+
+|type|description|
+|----|-----------|
+|`Number`|The byte count of the character in the given encoding|
+
+Examples:
+
+```grain
+Char.encodedLength(Char.UTF8, 'a') == 1
+```
+
+```grain
+Char.encodedLength(Char.UTF8, '🌾') == 4
+```
+
+```grain
+Char.encodedLength(Char.UTF16, '©') == 1
+```
+
 ### Char.**(<)**
 
 <details disabled>