Always return the character width for char_is_identifier_start() and char_is_identifier_utf8()

* This is also faster than calling pm_encoding_utf_8_alpha_char/pm_encoding_utf_8_alnum_char as those compute the character width and do extra checks.

Author: eregon

include/prism/encoding.h

@@ -79,6 +79,16 @@ typedef struct {
  */
 #define PRISM_ENCODING_UPPERCASE_BIT 1 << 2
 
+/**
+ * Return the size of the next character in the UTF-8 encoding.
+ *
+ * @param b The bytes to read.
+ * @param n The number of bytes that can be read.
+ * @returns The number of bytes that the next character takes if it is valid in
+ * the encoding, or 0 if it is not.
+ */
+size_t pm_encoding_utf_8_char_width(const uint8_t *b, ptrdiff_t n);
+
 /**
  * Return the size of the next character in the UTF-8 encoding if it is an
  * alphabetical character.

src/encoding.c

@@ -2277,7 +2277,10 @@ pm_utf_8_codepoint(const uint8_t *b, ptrdiff_t n, size_t *width) {
  return 0;
 }
 
-static size_t
+/**
+ * Return the size of the next character in the UTF-8 encoding.
+ */
+size_t
 pm_encoding_utf_8_char_width(const uint8_t *b, ptrdiff_t n) {
  size_t width;
  pm_utf_8_codepoint(b, n, &width);

src/prism.c

@@ -6249,7 +6249,7 @@ char_is_identifier_start(pm_parser_t *parser, const uint8_t *b) {
  } else if (*b < 0x80) {
  return (pm_encoding_unicode_table[*b] & PRISM_ENCODING_ALPHABETIC_BIT ? 1 : 0) || (*b == '_');
  } else {
- return (size_t) (pm_encoding_utf_8_alpha_char(b, parser->end - b) || 1u);
+ return pm_encoding_utf_8_char_width(b, parser->end - b);
  }
 }
 
@@ -6262,7 +6262,7 @@ char_is_identifier_utf8(const uint8_t *b, const uint8_t *end) {
  if (*b < 0x80) {
  return (*b == '_') || (pm_encoding_unicode_table[*b] & PRISM_ENCODING_ALPHANUMERIC_BIT ? 1 : 0);
  } else {
- return (size_t) (pm_encoding_utf_8_alnum_char(b, end - b) || 1u);
+ return pm_encoding_utf_8_char_width(b, end - b);
  }
 }