tomlib 0.5.0 → 0.7.0

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: 9598af874f419d93b421990cfced654f3787aa1feb30cb73c10bd42088917e9c
-  data.tar.gz: ac5980031b22fc23911e1555ce323fff74e4540071ba26960808325548b5290b
+  metadata.gz: 02730e4ee5dcfc0fe85a52e1909e42c88163e0d736545a7cf5c967dd11231b54
+  data.tar.gz: e5db101613da4dcd21db1450076df866ddf664338463f5520dbee8782bedae8d
 SHA512:
-  metadata.gz: 8eb4fe0b0054e2f3e446a62a212d211678c10a55360ea1523b38de9900e76163ec7811ba2b96a3e46db96818ca4d3c9188fe69f3c5107cead1da8fbef4360b76
-  data.tar.gz: b22db5ff36ebc41ac9a4d39798ed78a510e1793c05ec2e3c74ad91bb61eb8c92e7325902d592ce909e8d5ce289cbc9b6de624a2b471edfbb098b77252db8cf85
+  metadata.gz: 95b11c6a6f6af42b4a91710cb82fa192ac9ee1ca6b1487f2970f9694e3deae87ee4afc015947ca9be5d9fea49c32ed02acd852d18ccf98c896bae0690f8c4b01
+  data.tar.gz: 30991beebc111fef5d8c504eb42532d7ee4d7b4b09c094cd16981a28104f09e13283bad8e63a69d44a6aaad42f86d86032973a50933ff6ee5a81437c9d0067c8

data/CHANGELOG.md

@@ -1,3 +1,14 @@
+## [0.7.0] - 2024-02-10
+
+- Update tomlc99 to the latest version
+- Parse very large numbers as Infinity
+- Add support for Ruby 3.3
+- Drop support for Ruby 2.7 and 2.7
+
+## [0.6.0] - 2023-05-05
+
+- Correctly escape special characters
+
 ## [0.5.0] - 2022-08-16
 
 - Add support for Ruby 2.6

data/README.md

@@ -1,7 +1,7 @@
 # Tomlib
 
-Tomlib is a TOML parser and generator for Ruby. It is fast and standards-compliant by relying
-on native [tomlc99](https://github.com/cktan/tomlc99) parser.
+Tomlib is a TOML parser and generator for Ruby. It uses native C extension based on
+fast and standards-compliant [tomlc99](https://github.com/cktan/tomlc99) parser.
 
 ## Compliance
 
@@ -11,7 +11,7 @@ It passes both [BurntSushi/toml-test](https://github.com/BurntSushi/toml-test) a
 
 ## Installation
 
-Tomlib supports Ruby (MRI) 2.6+
+Tomlib supports Ruby (MRI) 3.0+
 
 Add this line to your application's Gemfile:
 
@@ -118,10 +118,11 @@ Tomlib.dump(hash, indent: false)
 
 ## Performance
 
-`Tomlib` parsing is ~300x faster than `toml-rb`, ~15x faster than `Tomlrb`
-and ~3x faster than `perfect_toml` for usual use case (~5KB TOML document size).
+When parsing documents, `Tomlib` is more than 600x (400x with yjit) faster than `toml-rb`,
+23x (17x with yjit) faster than `Tomlrb` and almost 5x (3.5x with yjit)
+faster than `perfect_toml` (~5KB TOML document size).
 
-Generating TOML document is about 2x faster than `toml-rb`.
+When generating TOML documents, it is about 1.5x (1.7x with yjit) faster than `toml-rb`.
 
 For full comparison take a look at
 [benchmarks](https://github.com/kgiszczak/tomlib/tree/master/benchmarks)

data/ext/tomlib/toml.c

@@ -45,6 +45,7 @@ void toml_set_memutil(void *(*xxmalloc)(size_t), void (*xxfree)(void *)) {
     ppfree = xxfree;
 }
 
+#define ALIGN8(sz) (((sz) + 7) & ~7)
 #define MALLOC(a) ppmalloc(a)
 #define FREE(a) ppfree(a)
 
@@ -53,7 +54,7 @@ void toml_set_memutil(void *(*xxmalloc)(size_t), void (*xxfree)(void *)) {
 #define calloc(x, y) error - forbidden - use CALLOC instead
 
 static void *CALLOC(size_t nmemb, size_t sz) {
-  int nb = sz * nmemb;
+  int nb = ALIGN8(sz) * nmemb;
   void *p = MALLOC(nb);
   if (p) {
     memset(p, 0, nb);
@@ -220,8 +221,8 @@ int toml_ucs_to_utf8(int64_t code, char buf[6]) {
      110xxxxx 10xxxxxx
   */
   if (code <= 0x000007FF) {
-    buf[0] = (unsigned char) (0xc0 | (code >> 6));
-    buf[1] = (unsigned char) (0x80 | (code & 0x3f));
+    buf[0] = (unsigned char)(0xc0 | (code >> 6));
+    buf[1] = (unsigned char)(0x80 | (code & 0x3f));
     return 2;
   }
 
@@ -229,9 +230,9 @@ int toml_ucs_to_utf8(int64_t code, char buf[6]) {
      1110xxxx 10xxxxxx 10xxxxxx
   */
   if (code <= 0x0000FFFF) {
-    buf[0] = (unsigned char) (0xe0 | (code >> 12));
-    buf[1] = (unsigned char) (0x80 | ((code >> 6) & 0x3f));
-    buf[2] = (unsigned char) (0x80 | (code & 0x3f));
+    buf[0] = (unsigned char)(0xe0 | (code >> 12));
+    buf[1] = (unsigned char)(0x80 | ((code >> 6) & 0x3f));
+    buf[2] = (unsigned char)(0x80 | (code & 0x3f));
     return 3;
   }
 
@@ -239,10 +240,10 @@ int toml_ucs_to_utf8(int64_t code, char buf[6]) {
      11110xxx 10xxxxxx 10xxxxxx 10xxxxxx
   */
   if (code <= 0x001FFFFF) {
-    buf[0] = (unsigned char) (0xf0 | (code >> 18));
-    buf[1] = (unsigned char) (0x80 | ((code >> 12) & 0x3f));
-    buf[2] = (unsigned char) (0x80 | ((code >> 6) & 0x3f));
-    buf[3] = (unsigned char) (0x80 | (code & 0x3f));
+    buf[0] = (unsigned char)(0xf0 | (code >> 18));
+    buf[1] = (unsigned char)(0x80 | ((code >> 12) & 0x3f));
+    buf[2] = (unsigned char)(0x80 | ((code >> 6) & 0x3f));
+    buf[3] = (unsigned char)(0x80 | (code & 0x3f));
     return 4;
   }
 
@@ -250,11 +251,11 @@ int toml_ucs_to_utf8(int64_t code, char buf[6]) {
      111110xx 10xxxxxx 10xxxxxx 10xxxxxx 10xxxxxx
   */
   if (code <= 0x03FFFFFF) {
-    buf[0] = (unsigned char) (0xf8 | (code >> 24));
-    buf[1] = (unsigned char) (0x80 | ((code >> 18) & 0x3f));
-    buf[2] = (unsigned char) (0x80 | ((code >> 12) & 0x3f));
-    buf[3] = (unsigned char) (0x80 | ((code >> 6) & 0x3f));
-    buf[4] = (unsigned char) (0x80 | (code & 0x3f));
+    buf[0] = (unsigned char)(0xf8 | (code >> 24));
+    buf[1] = (unsigned char)(0x80 | ((code >> 18) & 0x3f));
+    buf[2] = (unsigned char)(0x80 | ((code >> 12) & 0x3f));
+    buf[3] = (unsigned char)(0x80 | ((code >> 6) & 0x3f));
+    buf[4] = (unsigned char)(0x80 | (code & 0x3f));
     return 5;
   }
 
@@ -262,12 +263,12 @@ int toml_ucs_to_utf8(int64_t code, char buf[6]) {
      1111110x 10xxxxxx 10xxxxxx 10xxxxxx 10xxxxxx 10xxxxxx
   */
   if (code <= 0x7FFFFFFF) {
-    buf[0] = (unsigned char) (0xfc | (code >> 30));
-    buf[1] = (unsigned char) (0x80 | ((code >> 24) & 0x3f));
-    buf[2] = (unsigned char) (0x80 | ((code >> 18) & 0x3f));
-    buf[3] = (unsigned char) (0x80 | ((code >> 12) & 0x3f));
-    buf[4] = (unsigned char) (0x80 | ((code >> 6) & 0x3f));
-    buf[5] = (unsigned char) (0x80 | (code & 0x3f));
+    buf[0] = (unsigned char)(0xfc | (code >> 30));
+    buf[1] = (unsigned char)(0x80 | ((code >> 24) & 0x3f));
+    buf[2] = (unsigned char)(0x80 | ((code >> 18) & 0x3f));
+    buf[3] = (unsigned char)(0x80 | ((code >> 12) & 0x3f));
+    buf[4] = (unsigned char)(0x80 | ((code >> 6) & 0x3f));
+    buf[5] = (unsigned char)(0x80 | (code & 0x3f));
     return 6;
   }
 
@@ -411,8 +412,10 @@ static void *expand(void *p, int sz, int newsz) {
   if (!s)
     return 0;
 
-  memcpy(s, p, sz);
-  FREE(p);
+  if (p) {
+    memcpy(s, p, sz);
+    FREE(p);
+  }
   return s;
 }
 
@@ -422,8 +425,10 @@ static void **expand_ptrarr(void **p, int n) {
     return 0;
 
   s[n] = 0;
-  memcpy(s, p, n * sizeof(void *));
-  FREE(p);
+  if (p) {
+    memcpy(s, p, n * sizeof(void *));
+    FREE(p);
+  }
   return s;
 }
 
@@ -2198,7 +2203,7 @@ int toml_rtod_ex(toml_raw_t src, double *ret_, char *buf, int buflen) {
 }
 
 int toml_rtod(toml_raw_t src, double *ret_) {
-  char buf[100];
+  char buf[400];
   return toml_rtod_ex(src, ret_, buf, sizeof(buf));
 }
 
@@ -2211,6 +2216,7 @@ int toml_rtos(toml_raw_t src, char **ret) {
   if (!src)
     return -1;
 
+  // for strings, first char must be a s-quote or d-quote
   int qchar = src[0];
   int srclen = strlen(src);
   if (!(qchar == '\'' || qchar == '"')) {
@@ -2219,12 +2225,14 @@ int toml_rtos(toml_raw_t src, char **ret) {
 
   // triple quotes?
   if (qchar == src[1] && qchar == src[2]) {
-    multiline = 1;
-    sp = src + 3;
-    sq = src + srclen - 3;
-    /* last 3 chars in src must be qchar */
-    if (!(sp <= sq && sq[0] == qchar && sq[1] == qchar && sq[2] == qchar))
+    multiline = 1;         // triple-quote implies multiline
+    sp = src + 3;          // first char after quote
+    sq = src + srclen - 3; // first char of ending quote
+
+    if (!(sp <= sq && sq[0] == qchar && sq[1] == qchar && sq[2] == qchar)) {
+      // last 3 chars in src must be qchar
       return -1;
+    }
 
     /* skip new line immediate after qchar */
     if (sp[0] == '\n')
@@ -2233,13 +2241,18 @@ int toml_rtos(toml_raw_t src, char **ret) {
       sp += 2;
 
   } else {
-    sp = src + 1;
-    sq = src + srclen - 1;
-    /* last char in src must be qchar */
-    if (!(sp <= sq && *sq == qchar))
+    sp = src + 1;          // first char after quote
+    sq = src + srclen - 1; // ending quote
+    if (!(sp <= sq && *sq == qchar)) {
+      /* last char in src must be qchar */
       return -1;
+    }
   }
 
+  // at this point:
+  //     sp points to first valid char after quote.
+  //     sq points to one char beyond last valid char.
+  //     string len is (sq - sp).
   if (qchar == '\'') {
     *ret = norm_lit_str(sp, sq - sp, multiline, 0, 0);
   } else {

data/ext/tomlib/toml.h

@@ -26,7 +26,7 @@
 #define TOML_H
 
 #ifdef _MSC_VER
-#pragma warning(disable: 4996)
+#pragma warning(disable : 4996)
 #endif
 
 #include <stdint.h>

data/ext/tomlib/tomlib.c

@@ -187,7 +187,7 @@ static VALUE toml_table_key_to_rb_value(const toml_table_t *table, const char *k
 
   datum = toml_double_in(table, key);
 
-  if (datum.ok) {
+  if (datum.ok || datum.u.d == INFINITY || datum.u.d == -INFINITY) {
     return DBL2NUM(datum.u.d);
   }
 
@@ -320,7 +320,7 @@ static VALUE tomlib_key_type(VALUE self, VALUE rb_key) {
 
   if (str_len == 0) return sym_escape;
 
-  for(long i = 0; i < str_len; i++) {
+  for (long i = 0; i < str_len; i++) {
     const char c = *(str + i);
 
     if (c == '\n') {

data/lib/tomlib/dumper.rb

@@ -27,6 +27,22 @@ module Tomlib
     # @api private
     NAN = 'nan'.freeze
 
+    # Regex to match escape chars
+    # @api private
+    ESCAPE_CHARS_REGEX = /[\x00-\x1F\x22\x5C\x7F]/.freeze
+
+    # Escape chars mapping
+    # @api private
+    ESCAPE_CHARS = {
+      "\b" => 'b',
+      "\t" => 't',
+      "\n" => 'n',
+      "\f" => 'f',
+      "\r" => 'r',
+      '"' => '"',
+      '\\' => '\\',
+    }.freeze
+
     def initialize(use_indent: true)
       @use_indent = use_indent
     end
@@ -106,7 +122,7 @@ module Tomlib
 
       case key_type(key)
       when :quoted
-        key.inspect
+        escape_string(key)
       when :escape
         key.dump.gsub('\n', '\\\\\n')
       else
@@ -138,7 +154,7 @@ module Tomlib
     def to_toml_value(value)
       case value
       when String
-        value.inspect
+        escape_string(value)
       when Float, BigDecimal
         to_toml_float(value)
       when Time, DateTime
@@ -170,5 +186,26 @@ module Tomlib
 
       value.to_s
     end
+
+    # Escapes TOML special characters
+    #
+    # @param [String] str
+    #
+    # @return [String]
+    #
+    # @api private
+    def escape_string(str)
+      str = str.gsub(ESCAPE_CHARS_REGEX) do |chr|
+        c = ESCAPE_CHARS[chr]
+
+        if c
+          '\\' << c
+        else
+          format('\\u%04X', chr.ord)
+        end
+      end
+
+      '"' << str << '"'.freeze
+    end
   end
 end

data/lib/tomlib/version.rb

@@ -2,5 +2,5 @@
 
 module Tomlib
   # @api private
-  VERSION = '0.5.0'
+  VERSION = '0.7.0'
 end

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: tomlib
 version: !ruby/object:Gem::Version
-  version: 0.5.0
+  version: 0.7.0
 platform: ruby
 authors:
 - Kamil Giszczak
 autorequire:
 bindir: bin
 cert_chain: []
-date: 2022-08-16 00:00:00.000000000 Z
+date: 2024-02-10 00:00:00.000000000 Z
 dependencies: []
 description: Fast TOML parser and generator with native extension.
 email:
@@ -54,7 +54,7 @@ required_rubygems_version: !ruby/object:Gem::Requirement
     - !ruby/object:Gem::Version
       version: '0'
 requirements: []
-rubygems_version: 3.3.7
+rubygems_version: 3.5.3
 signing_key:
 specification_version: 4
 summary: Fast TOML parser and generator with native extension.