prism 0.18.0 → 0.19.0

data/src/util/pm_buffer.c

@@ -138,7 +138,7 @@ pm_buffer_append_byte(pm_buffer_t *buffer, uint8_t value) {
  * Append a 32-bit unsigned integer to the buffer as a variable-length integer.
  */
 void
-pm_buffer_append_varint(pm_buffer_t *buffer, uint32_t value) {
+pm_buffer_append_varuint(pm_buffer_t *buffer, uint32_t value) {
     if (value < 128) {
         pm_buffer_append_byte(buffer, (uint8_t) value);
     } else {
@@ -151,6 +151,15 @@ pm_buffer_append_varint(pm_buffer_t *buffer, uint32_t value) {
     }
 }
 
+/**
+ * Append a 32-bit signed integer to the buffer as a variable-length integer.
+ */
+void
+pm_buffer_append_varsint(pm_buffer_t *buffer, int32_t value) {
+    uint32_t unsigned_int = ((uint32_t)(value) << 1) ^ ((uint32_t)(value >> 31));
+    pm_buffer_append_varuint(buffer, unsigned_int);
+}
+
 /**
  * Concatenate one buffer onto another.
  */

data/src/util/pm_memchr.c

@@ -8,7 +8,7 @@
  * of a multibyte character.
  */
 void *
-pm_memchr(const void *memory, int character, size_t number, bool encoding_changed, pm_encoding_t *encoding) {
+pm_memchr(const void *memory, int character, size_t number, bool encoding_changed, const pm_encoding_t *encoding) {
     if (encoding_changed && encoding->multibyte && character >= PRISM_MEMCHR_TRAILING_BYTE_MINIMUM) {
         const uint8_t *source = (const uint8_t *) memory;
         size_t index = 0;

data/src/util/pm_strpbrk.c

@@ -4,7 +4,7 @@
  * This is the slow path that does care about the encoding.
  */
 static inline const uint8_t *
-pm_strpbrk_multi_byte(pm_parser_t *parser, const uint8_t *source, const uint8_t *charset, size_t maximum) {
+pm_strpbrk_multi_byte(const pm_parser_t *parser, const uint8_t *source, const uint8_t *charset, size_t maximum) {
     size_t index = 0;
 
     while (index < maximum) {
@@ -12,7 +12,7 @@ pm_strpbrk_multi_byte(pm_parser_t *parser, const uint8_t *source, const uint8_t
             return source + index;
         }
 
-        size_t width = parser->encoding.char_width(source + index, (ptrdiff_t) (maximum - index));
+        size_t width = parser->encoding->char_width(source + index, (ptrdiff_t) (maximum - index));
         if (width == 0) {
             return NULL;
         }
@@ -61,10 +61,10 @@ pm_strpbrk_single_byte(const uint8_t *source, const uint8_t *charset, size_t max
  * need to take a slower path and iterate one multi-byte character at a time.
  */
 const uint8_t *
-pm_strpbrk(pm_parser_t *parser, const uint8_t *source, const uint8_t *charset, ptrdiff_t length) {
+pm_strpbrk(const pm_parser_t *parser, const uint8_t *source, const uint8_t *charset, ptrdiff_t length) {
     if (length <= 0) {
         return NULL;
-    } else if (parser->encoding_changed && parser->encoding.multibyte) {
+    } else if (parser->encoding_changed && parser->encoding->multibyte) {
         return pm_strpbrk_multi_byte(parser, source, charset, (size_t) length);
     } else {
         return pm_strpbrk_single_byte(source, charset, (size_t) length);

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: prism
 version: !ruby/object:Gem::Version
-  version: 0.18.0
+  version: 0.19.0
 platform: ruby
 authors:
 - Shopify
 autorequire:
 bindir: bin
 cert_chain: []
-date: 2023-11-21 00:00:00.000000000 Z
+date: 2023-12-14 00:00:00.000000000 Z
 dependencies: []
 description:
 email:
@@ -33,6 +33,7 @@ files:
 - docs/fuzzing.md
 - docs/heredocs.md
 - docs/javascript.md
+- docs/local_variable_depth.md
 - docs/mapping.md
 - docs/releasing.md
 - docs/ripper.md
@@ -48,7 +49,7 @@ files:
 - include/prism/ast.h
 - include/prism/defines.h
 - include/prism/diagnostic.h
-- include/prism/enc/pm_encoding.h
+- include/prism/encoding.h
 - include/prism/node.h
 - include/prism/options.h
 - include/prism/pack.h
@@ -94,14 +95,7 @@ files:
 - sig/prism.rbs
 - sig/prism_static.rbs
 - src/diagnostic.c
-- src/enc/pm_big5.c
-- src/enc/pm_cp51932.c
-- src/enc/pm_euc_jp.c
-- src/enc/pm_gbk.c
-- src/enc/pm_shift_jis.c
-- src/enc/pm_tables.c
-- src/enc/pm_unicode.c
-- src/enc/pm_windows_31j.c
+- src/encoding.c
 - src/node.c
 - src/options.c
 - src/pack.c

data/include/prism/enc/pm_encoding.h

@@ -1,227 +0,0 @@
-/**
- * @file pm_encoding.h
- *
- * The encoding interface and implementations used by the parser.
- */
-#ifndef PRISM_ENCODING_H
-#define PRISM_ENCODING_H
-
-#include "prism/defines.h"
-
-#include <assert.h>
-#include <stdbool.h>
-#include <stddef.h>
-#include <stdint.h>
-
-/**
- * This struct defines the functions necessary to implement the encoding
- * interface so we can determine how many bytes the subsequent character takes.
- * Each callback should return the number of bytes, or 0 if the next bytes are
- * invalid for the encoding and type.
- */
-typedef struct {
-    /**
-     * Return the number of bytes that the next character takes if it is valid
-     * in the encoding. Does not read more than n bytes. It is assumed that n is
-     * at least 1.
-     */
-    size_t (*char_width)(const uint8_t *b, ptrdiff_t n);
-
-    /**
-     * Return the number of bytes that the next character takes if it is valid
-     * in the encoding and is alphabetical. Does not read more than n bytes. It
-     * is assumed that n is at least 1.
-     */
-    size_t (*alpha_char)(const uint8_t *b, ptrdiff_t n);
-
-    /**
-     * Return the number of bytes that the next character takes if it is valid
-     * in the encoding and is alphanumeric. Does not read more than n bytes. It
-     * is assumed that n is at least 1.
-     */
-    size_t (*alnum_char)(const uint8_t *b, ptrdiff_t n);
-
-    /**
-     * Return true if the next character is valid in the encoding and is an
-     * uppercase character. Does not read more than n bytes. It is assumed that
-     * n is at least 1.
-     */
-    bool (*isupper_char)(const uint8_t *b, ptrdiff_t n);
-
-    /**
-     * The name of the encoding. This should correspond to a value that can be
-     * passed to Encoding.find in Ruby.
-     */
-    const char *name;
-
-    /**
-     * Return true if the encoding is a multibyte encoding.
-     */
-    bool multibyte;
-} pm_encoding_t;
-
-/**
- * All of the lookup tables use the first bit of each embedded byte to indicate
- * whether the codepoint is alphabetical.
- */
-#define PRISM_ENCODING_ALPHABETIC_BIT 1 << 0
-
-/**
- * All of the lookup tables use the second bit of each embedded byte to indicate
- * whether the codepoint is alphanumeric.
- */
-#define PRISM_ENCODING_ALPHANUMERIC_BIT 1 << 1
-
-/**
- * All of the lookup tables use the third bit of each embedded byte to indicate
- * whether the codepoint is uppercase.
- */
-#define PRISM_ENCODING_UPPERCASE_BIT 1 << 2
-
-/**
- * Return the size of the next character in the ASCII encoding if it is an
- * alphabetical character.
- *
- * @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_ascii_alpha_char(const uint8_t *b, PRISM_ATTRIBUTE_UNUSED ptrdiff_t n);
-
-/**
- * Return the size of the next character in the ASCII encoding if it is an
- * alphanumeric character.
- *
- * @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_ascii_alnum_char(const uint8_t *b, PRISM_ATTRIBUTE_UNUSED ptrdiff_t n);
-
-/**
- * Return true if the next character in the ASCII encoding if it is an uppercase
- * character.
- *
- * @param b The bytes to read.
- * @param n The number of bytes that can be read.
- * @returns True if the next character is valid in the encoding and is an
- *     uppercase character, or false if it is not.
- */
-bool pm_encoding_ascii_isupper_char(const uint8_t *b, PRISM_ATTRIBUTE_UNUSED ptrdiff_t n);
-
-/**
- * Return the size of the next character in the UTF-8 encoding if it is an
- * alphabetical character.
- *
- * @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_alpha_char(const uint8_t *b, ptrdiff_t n);
-
-/**
- * Return the size of the next character in the UTF-8 encoding if it is an
- * alphanumeric character.
- *
- * @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_alnum_char(const uint8_t *b, ptrdiff_t n);
-
-/**
- * Return true if the next character in the UTF-8 encoding if it is an uppercase
- * character.
- *
- * @param b The bytes to read.
- * @param n The number of bytes that can be read.
- * @returns True if the next character is valid in the encoding and is an
- *     uppercase character, or false if it is not.
- */
-bool pm_encoding_utf_8_isupper_char(const uint8_t *b, ptrdiff_t n);
-
-/**
- * This lookup table is referenced in both the UTF-8 encoding file and the
- * parser directly in order to speed up the default encoding processing. It is
- * used to indicate whether a character is alphabetical, alphanumeric, or
- * uppercase in unicode mappings.
- */
-extern const uint8_t pm_encoding_unicode_table[256];
-
-// Below are the encodings that are supported by the parser. They are defined in
-// their own files in the src/enc directory.
-
-extern pm_encoding_t pm_encoding_ascii;
-extern pm_encoding_t pm_encoding_ascii_8bit;
-extern pm_encoding_t pm_encoding_big5;
-extern pm_encoding_t pm_encoding_big5_hkscs;
-extern pm_encoding_t pm_encoding_big5_uao;
-extern pm_encoding_t pm_encoding_cp51932;
-extern pm_encoding_t pm_encoding_cp850;
-extern pm_encoding_t pm_encoding_cp852;
-extern pm_encoding_t pm_encoding_cp855;
-extern pm_encoding_t pm_encoding_euc_jp;
-extern pm_encoding_t pm_encoding_gb1988;
-extern pm_encoding_t pm_encoding_gbk;
-extern pm_encoding_t pm_encoding_ibm437;
-extern pm_encoding_t pm_encoding_ibm720;
-extern pm_encoding_t pm_encoding_ibm737;
-extern pm_encoding_t pm_encoding_ibm775;
-extern pm_encoding_t pm_encoding_ibm852;
-extern pm_encoding_t pm_encoding_ibm855;
-extern pm_encoding_t pm_encoding_ibm857;
-extern pm_encoding_t pm_encoding_ibm860;
-extern pm_encoding_t pm_encoding_ibm861;
-extern pm_encoding_t pm_encoding_ibm862;
-extern pm_encoding_t pm_encoding_ibm863;
-extern pm_encoding_t pm_encoding_ibm864;
-extern pm_encoding_t pm_encoding_ibm865;
-extern pm_encoding_t pm_encoding_ibm866;
-extern pm_encoding_t pm_encoding_ibm869;
-extern pm_encoding_t pm_encoding_iso_8859_1;
-extern pm_encoding_t pm_encoding_iso_8859_2;
-extern pm_encoding_t pm_encoding_iso_8859_3;
-extern pm_encoding_t pm_encoding_iso_8859_4;
-extern pm_encoding_t pm_encoding_iso_8859_5;
-extern pm_encoding_t pm_encoding_iso_8859_6;
-extern pm_encoding_t pm_encoding_iso_8859_7;
-extern pm_encoding_t pm_encoding_iso_8859_8;
-extern pm_encoding_t pm_encoding_iso_8859_9;
-extern pm_encoding_t pm_encoding_iso_8859_10;
-extern pm_encoding_t pm_encoding_iso_8859_11;
-extern pm_encoding_t pm_encoding_iso_8859_13;
-extern pm_encoding_t pm_encoding_iso_8859_14;
-extern pm_encoding_t pm_encoding_iso_8859_15;
-extern pm_encoding_t pm_encoding_iso_8859_16;
-extern pm_encoding_t pm_encoding_koi8_r;
-extern pm_encoding_t pm_encoding_mac_cent_euro;
-extern pm_encoding_t pm_encoding_mac_croatian;
-extern pm_encoding_t pm_encoding_mac_cyrillic;
-extern pm_encoding_t pm_encoding_mac_greek;
-extern pm_encoding_t pm_encoding_mac_iceland;
-extern pm_encoding_t pm_encoding_mac_roman;
-extern pm_encoding_t pm_encoding_mac_romania;
-extern pm_encoding_t pm_encoding_mac_thai;
-extern pm_encoding_t pm_encoding_mac_turkish;
-extern pm_encoding_t pm_encoding_mac_ukraine;
-extern pm_encoding_t pm_encoding_shift_jis;
-extern pm_encoding_t pm_encoding_tis_620;
-extern pm_encoding_t pm_encoding_utf_8;
-extern pm_encoding_t pm_encoding_utf8_mac;
-extern pm_encoding_t pm_encoding_windows_1250;
-extern pm_encoding_t pm_encoding_windows_1251;
-extern pm_encoding_t pm_encoding_windows_1252;
-extern pm_encoding_t pm_encoding_windows_1253;
-extern pm_encoding_t pm_encoding_windows_1254;
-extern pm_encoding_t pm_encoding_windows_1255;
-extern pm_encoding_t pm_encoding_windows_1256;
-extern pm_encoding_t pm_encoding_windows_1257;
-extern pm_encoding_t pm_encoding_windows_1258;
-extern pm_encoding_t pm_encoding_windows_31j;
-extern pm_encoding_t pm_encoding_windows_874;
-
-#endif

data/src/enc/pm_big5.c

@@ -1,116 +0,0 @@
-#include "prism/enc/pm_encoding.h"
-
-static size_t
-pm_encoding_big5_char_width(const uint8_t *b, ptrdiff_t n) {
-    // These are the single byte characters.
-    if (*b < 0x80) {
-        return 1;
-    }
-
-    // These are the double byte characters.
-    if ((n > 1) && (b[0] >= 0xA1 && b[0] <= 0xFE) && (b[1] >= 0x40 && b[1] <= 0xFE)) {
-        return 2;
-    }
-
-    return 0;
-}
-
-static size_t
-pm_encoding_big5_star_char_width(const uint8_t *b, ptrdiff_t n) {
-    // These are the single byte characters.
-    if (*b < 0x80) {
-        return 1;
-    }
-
-    // These are the double byte characters.
-    if ((n > 1) && (b[0] >= 0x87 && b[0] <= 0xFE) &&
-                   ((b[1] >= 0x40 && b[1] <= 0x7E) || (b[1] >= 0xA1 && b[1] <= 0xFE))) {
-        return 2;
-    }
-
-    return 0;
-}
-
-static size_t
-pm_encoding_big5_alpha_char(const uint8_t *b, ptrdiff_t n) {
-    if (pm_encoding_big5_char_width(b, n) == 1) {
-        return pm_encoding_ascii_alpha_char(b, n);
-    } else {
-        return 0;
-    }
-}
-
-static size_t
-pm_encoding_big5_star_alpha_char(const uint8_t *b, ptrdiff_t n) {
-    if (pm_encoding_big5_star_char_width(b, n) == 1) {
-        return pm_encoding_ascii_alpha_char(b, n);
-    } else {
-        return 0;
-    }
-}
-
-static size_t
-pm_encoding_big5_alnum_char(const uint8_t *b, ptrdiff_t n) {
-    if (pm_encoding_big5_char_width(b, n) == 1) {
-        return pm_encoding_ascii_alnum_char(b, n);
-    } else {
-        return 0;
-    }
-}
-
-static size_t
-pm_encoding_big5_star_alnum_char(const uint8_t *b, ptrdiff_t n) {
-    if (pm_encoding_big5_star_char_width(b, n) == 1) {
-        return pm_encoding_ascii_alnum_char(b, n);
-    } else {
-        return 0;
-    }
-}
-
-static bool
-pm_encoding_big5_isupper_char(const uint8_t *b, ptrdiff_t n) {
-    if (pm_encoding_big5_char_width(b, n) == 1) {
-        return pm_encoding_ascii_isupper_char(b, n);
-    } else {
-        return false;
-    }
-}
-
-static bool
-pm_encoding_big5_star_isupper_char(const uint8_t *b, ptrdiff_t n) {
-    if (pm_encoding_big5_star_char_width(b, n) == 1) {
-        return pm_encoding_ascii_isupper_char(b, n);
-    } else {
-        return false;
-    }
-}
-
-/** Big5 encoding */
-pm_encoding_t pm_encoding_big5 = {
-    .name = "big5",
-    .char_width = pm_encoding_big5_char_width,
-    .alnum_char = pm_encoding_big5_alnum_char,
-    .alpha_char = pm_encoding_big5_alpha_char,
-    .isupper_char = pm_encoding_big5_isupper_char,
-    .multibyte = true
-};
-
-/** Big5-HKSCS encoding */
-pm_encoding_t pm_encoding_big5_hkscs = {
-    .name = "big5-hkscs",
-    .char_width = pm_encoding_big5_star_char_width,
-    .alnum_char = pm_encoding_big5_star_alnum_char,
-    .alpha_char = pm_encoding_big5_star_alpha_char,
-    .isupper_char = pm_encoding_big5_star_isupper_char,
-    .multibyte = true
-};
-
-/** Big5-UAO encoding */
-pm_encoding_t pm_encoding_big5_uao = {
-    .name = "big5-uao",
-    .char_width = pm_encoding_big5_star_char_width,
-    .alnum_char = pm_encoding_big5_star_alnum_char,
-    .alpha_char = pm_encoding_big5_star_alpha_char,
-    .isupper_char = pm_encoding_big5_star_isupper_char,
-    .multibyte = true
-};

data/src/enc/pm_cp51932.c

@@ -1,57 +0,0 @@
-#include "prism/enc/pm_encoding.h"
-
-static size_t
-pm_encoding_cp51932_char_width(const uint8_t *b, ptrdiff_t n) {
-    // These are the single byte characters.
-    if (*b < 0x80) {
-        return 1;
-    }
-
-    // These are the double byte characters.
-    if (
-        (n > 1) &&
-        ((b[0] >= 0xa1 && b[0] <= 0xfe) || (b[0] == 0x8e)) &&
-        (b[1] >= 0xa1 && b[1] <= 0xfe)
-    ) {
-        return 2;
-    }
-
-    return 0;
-}
-
-static size_t
-pm_encoding_cp51932_alpha_char(const uint8_t *b, ptrdiff_t n) {
-    if (pm_encoding_cp51932_char_width(b, n) == 1) {
-        return pm_encoding_ascii_alpha_char(b, n);
-    } else {
-        return 0;
-    }
-}
-
-static size_t
-pm_encoding_cp51932_alnum_char(const uint8_t *b, ptrdiff_t n) {
-    if (pm_encoding_cp51932_char_width(b, n) == 1) {
-        return pm_encoding_ascii_alnum_char(b, n);
-    } else {
-        return 0;
-    }
-}
-
-static bool
-pm_encoding_cp51932_isupper_char(const uint8_t *b, ptrdiff_t n) {
-    if (pm_encoding_cp51932_char_width(b, n) == 1) {
-        return pm_encoding_ascii_isupper_char(b, n);
-    } else {
-        return 0;
-    }
-}
-
-/** cp51932 encoding */
-pm_encoding_t pm_encoding_cp51932 = {
-    .name = "cp51932",
-    .char_width = pm_encoding_cp51932_char_width,
-    .alnum_char = pm_encoding_cp51932_alnum_char,
-    .alpha_char = pm_encoding_cp51932_alpha_char,
-    .isupper_char = pm_encoding_cp51932_isupper_char,
-    .multibyte = true
-};

data/src/enc/pm_euc_jp.c

@@ -1,69 +0,0 @@
-#include "prism/enc/pm_encoding.h"
-
-static size_t
-pm_encoding_euc_jp_char_width(const uint8_t *b, ptrdiff_t n) {
-    // These are the single byte characters.
-    if (*b < 0x80) {
-        return 1;
-    }
-
-    // These are the double byte characters.
-    if (
-        (n > 1) &&
-        (
-            ((b[0] == 0x8E) && (b[1] >= 0xA1 && b[1] <= 0xFE)) ||
-            ((b[0] >= 0xA1 && b[0] <= 0xFE) && (b[1] >= 0xA1 && b[1] <= 0xFE))
-        )
-    ) {
-        return 2;
-    }
-
-    // These are the triple byte characters.
-    if (
-        (n > 2) &&
-        (b[0] == 0x8F) &&
-        (b[1] >= 0xA1 && b[2] <= 0xFE) &&
-        (b[2] >= 0xA1 && b[2] <= 0xFE)
-    ) {
-        return 3;
-    }
-
-    return 0;
-}
-
-static size_t
-pm_encoding_euc_jp_alpha_char(const uint8_t *b, ptrdiff_t n) {
-    if (pm_encoding_euc_jp_char_width(b, n) == 1) {
-        return pm_encoding_ascii_alpha_char(b, n);
-    } else {
-        return 0;
-    }
-}
-
-static size_t
-pm_encoding_euc_jp_alnum_char(const uint8_t *b, ptrdiff_t n) {
-    if (pm_encoding_euc_jp_char_width(b, n) == 1) {
-        return pm_encoding_ascii_alnum_char(b, n);
-    } else {
-        return 0;
-    }
-}
-
-static bool
-pm_encoding_euc_jp_isupper_char(const uint8_t *b, ptrdiff_t n) {
-    if (pm_encoding_euc_jp_char_width(b, n) == 1) {
-        return pm_encoding_ascii_isupper_char(b, n);
-    } else {
-        return 0;
-    }
-}
-
-/** EUC-JP encoding */
-pm_encoding_t pm_encoding_euc_jp = {
-    .name = "euc-jp",
-    .char_width = pm_encoding_euc_jp_char_width,
-    .alnum_char = pm_encoding_euc_jp_alnum_char,
-    .alpha_char = pm_encoding_euc_jp_alpha_char,
-    .isupper_char = pm_encoding_euc_jp_isupper_char,
-    .multibyte = true
-};

data/src/enc/pm_gbk.c

@@ -1,65 +0,0 @@
-#include "prism/enc/pm_encoding.h"
-
-static size_t
-pm_encoding_gbk_char_width(const uint8_t *b, ptrdiff_t n) {
-    // These are the single byte characters.
-    if (*b <= 0x80) {
-        return 1;
-    }
-
-    // These are the double byte characters.
-    if (
-        (n > 1) &&
-        (
-            ((b[0] >= 0xA1 && b[0] <= 0xA9) && (b[1] >= 0xA1 && b[1] <= 0xFE)) || // GBK/1
-            ((b[0] >= 0xB0 && b[0] <= 0xF7) && (b[1] >= 0xA1 && b[1] <= 0xFE)) || // GBK/2
-            ((b[0] >= 0x81 && b[0] <= 0xA0) && (b[1] >= 0x40 && b[1] <= 0xFE) && (b[1] != 0x7F)) || // GBK/3
-            ((b[0] >= 0xAA && b[0] <= 0xFE) && (b[1] >= 0x40 && b[1] <= 0xA0) && (b[1] != 0x7F)) || // GBK/4
-            ((b[0] >= 0xA8 && b[0] <= 0xA9) && (b[1] >= 0x40 && b[1] <= 0xA0) && (b[1] != 0x7F)) || // GBK/5
-            ((b[0] >= 0xAA && b[0] <= 0xAF) && (b[1] >= 0xA1 && b[1] <= 0xFE)) || // user-defined 1
-            ((b[0] >= 0xF8 && b[0] <= 0xFE) && (b[1] >= 0xA1 && b[1] <= 0xFE)) || // user-defined 2
-            ((b[0] >= 0xA1 && b[0] <= 0xA7) && (b[1] >= 0x40 && b[1] <= 0xA0) && (b[1] != 0x7F)) // user-defined 3
-        )
-    ) {
-        return 2;
-    }
-
-    return 0;
-}
-
-static size_t
-pm_encoding_gbk_alpha_char(const uint8_t *b, ptrdiff_t n) {
-    if (pm_encoding_gbk_char_width(b, n) == 1) {
-        return pm_encoding_ascii_alpha_char(b, n);
-    } else {
-        return 0;
-    }
-}
-
-static size_t
-pm_encoding_gbk_alnum_char(const uint8_t *b, ptrdiff_t n) {
-    if (pm_encoding_gbk_char_width(b, n) == 1) {
-        return pm_encoding_ascii_alnum_char(b, n);
-    } else {
-        return 0;
-    }
-}
-
-static bool
-pm_encoding_gbk_isupper_char(const uint8_t *b, ptrdiff_t n) {
-    if (pm_encoding_gbk_char_width(b, n) == 1) {
-        return pm_encoding_ascii_isupper_char(b, n);
-    } else {
-        return false;
-    }
-}
-
-/** GBK encoding */
-pm_encoding_t pm_encoding_gbk = {
-    .name = "gbk",
-    .char_width = pm_encoding_gbk_char_width,
-    .alnum_char = pm_encoding_gbk_alnum_char,
-    .alpha_char = pm_encoding_gbk_alpha_char,
-    .isupper_char = pm_encoding_gbk_isupper_char,
-    .multibyte = true
-};