prism 0.22.0 → 0.23.0

data/lib/prism/translation.rb

@@ -2,10 +2,10 @@
 
 module Prism
   # This module is responsible for converting the prism syntax tree into other
-  # syntax trees. At the moment it only supports converting to the
-  # whitequark/parser gem's syntax tree, but support is planned for the
-  # seattlerb/ruby_parser gem's syntax tree as well.
+  # syntax trees.
   module Translation
     autoload :Parser, "prism/translation/parser"
+    autoload :Ripper, "prism/translation/ripper"
+    autoload :RubyParser, "prism/translation/ruby_parser"
   end
 end

data/lib/prism.rb

@@ -22,7 +22,6 @@ module Prism
   autoload :LexRipper, "prism/lex_compat"
   autoload :MutationCompiler, "prism/mutation_compiler"
   autoload :NodeInspector, "prism/node_inspector"
-  autoload :RipperCompat, "prism/ripper_compat"
   autoload :Pack, "prism/pack"
   autoload :Pattern, "prism/pattern"
   autoload :Serialize, "prism/serialize"

data/prism.gemspec

@@ -2,7 +2,7 @@
 
 Gem::Specification.new do |spec|
   spec.name = "prism"
-  spec.version = "0.22.0"
+  spec.version = "0.23.0"
   spec.authors = ["Shopify"]
   spec.email = ["ruby@shopify.com"]
 
@@ -36,6 +36,7 @@ Gem::Specification.new do |spec|
     "docs/releasing.md",
     "docs/ripper.md",
     "docs/ruby_api.md",
+    "docs/ruby_parser_translation.md",
     "docs/serialization.md",
     "docs/testing.md",
     "ext/prism/api_node.c",
@@ -83,13 +84,14 @@ Gem::Specification.new do |spec|
     "lib/prism/parse_result/comments.rb",
     "lib/prism/parse_result/newlines.rb",
     "lib/prism/pattern.rb",
-    "lib/prism/ripper_compat.rb",
     "lib/prism/serialize.rb",
     "lib/prism/translation.rb",
     "lib/prism/translation/parser.rb",
     "lib/prism/translation/parser/compiler.rb",
     "lib/prism/translation/parser/lexer.rb",
     "lib/prism/translation/parser/rubocop.rb",
+    "lib/prism/translation/ripper.rb",
+    "lib/prism/translation/ruby_parser.rb",
     "lib/prism/visitor.rb",
     "src/diagnostic.c",
     "src/encoding.c",

data/src/diagnostic.c

@@ -159,7 +159,7 @@ static const pm_diagnostic_data_t diagnostic_messages[PM_DIAGNOSTIC_ID_LEN] = {
     [PM_ERR_ESCAPE_INVALID_UNICODE_LONG]        = { "invalid Unicode escape sequence; maximum length is 6 digits", PM_ERROR_LEVEL_FATAL },
     [PM_ERR_ESCAPE_INVALID_UNICODE_TERM]        = { "invalid Unicode escape sequence; needs closing `}`", PM_ERROR_LEVEL_FATAL },
     [PM_ERR_EXPECT_ARGUMENT]                    = { "expected an argument", PM_ERROR_LEVEL_FATAL },
-    [PM_ERR_EXPECT_EOL_AFTER_STATEMENT]         = { "expected a newline or semicolon after the statement", PM_ERROR_LEVEL_FATAL },
+    [PM_ERR_EXPECT_EOL_AFTER_STATEMENT]         = { "unexpected %s, expecting end-of-input", PM_ERROR_LEVEL_FATAL },
     [PM_ERR_EXPECT_EXPRESSION_AFTER_AMPAMPEQ]   = { "expected an expression after `&&=`", PM_ERROR_LEVEL_FATAL },
     [PM_ERR_EXPECT_EXPRESSION_AFTER_PIPEPIPEEQ] = { "expected an expression after `||=`", PM_ERROR_LEVEL_FATAL },
     [PM_ERR_EXPECT_EXPRESSION_AFTER_COMMA]      = { "expected an expression after `,`", PM_ERROR_LEVEL_FATAL },
@@ -184,14 +184,14 @@ static const pm_diagnostic_data_t diagnostic_messages[PM_DIAGNOSTIC_ID_LEN] = {
     [PM_ERR_FOR_IN]                             = { "expected an `in` after the index in a `for` statement", PM_ERROR_LEVEL_FATAL },
     [PM_ERR_FOR_TERM]                           = { "expected an `end` to close the `for` loop", PM_ERROR_LEVEL_FATAL },
     [PM_ERR_HASH_EXPRESSION_AFTER_LABEL]        = { "expected an expression after the label in a hash", PM_ERROR_LEVEL_FATAL },
-    [PM_ERR_HASH_KEY]                           = { "expected a key in the hash literal", PM_ERROR_LEVEL_FATAL },
+    [PM_ERR_HASH_KEY]                           = { "unexpected %s, expecting '}' or a key in the hash literal", PM_ERROR_LEVEL_FATAL },
     [PM_ERR_HASH_ROCKET]                        = { "expected a `=>` between the hash key and value", PM_ERROR_LEVEL_FATAL },
     [PM_ERR_HASH_TERM]                          = { "expected a `}` to close the hash literal", PM_ERROR_LEVEL_FATAL },
     [PM_ERR_HASH_VALUE]                         = { "expected a value in the hash literal", PM_ERROR_LEVEL_FATAL },
     [PM_ERR_HEREDOC_TERM]                       = { "could not find a terminator for the heredoc", PM_ERROR_LEVEL_FATAL },
     [PM_ERR_INCOMPLETE_QUESTION_MARK]           = { "incomplete expression at `?`", PM_ERROR_LEVEL_FATAL },
-    [PM_ERR_INCOMPLETE_VARIABLE_CLASS]          = { "incomplete class variable", PM_ERROR_LEVEL_FATAL },
-    [PM_ERR_INCOMPLETE_VARIABLE_INSTANCE]       = { "incomplete instance variable", PM_ERROR_LEVEL_FATAL },
+    [PM_ERR_INCOMPLETE_VARIABLE_CLASS]          = { "`%.*s' is not allowed as a class variable name", PM_ERROR_LEVEL_FATAL },
+    [PM_ERR_INCOMPLETE_VARIABLE_INSTANCE]       = { "`%.*s' is not allowed as an instance variable name", PM_ERROR_LEVEL_FATAL },
     [PM_ERR_INVALID_FLOAT_EXPONENT]             = { "invalid exponent", PM_ERROR_LEVEL_FATAL },
     [PM_ERR_INVALID_NUMBER_BINARY]              = { "invalid binary number", PM_ERROR_LEVEL_FATAL },
     [PM_ERR_INVALID_NUMBER_DECIMAL]             = { "invalid decimal number", PM_ERROR_LEVEL_FATAL },
@@ -202,7 +202,7 @@ static const pm_diagnostic_data_t diagnostic_messages[PM_DIAGNOSTIC_ID_LEN] = {
     [PM_ERR_INVALID_MULTIBYTE_CHARACTER]        = { "invalid multibyte character 0x%X", PM_ERROR_LEVEL_FATAL },
     [PM_ERR_INVALID_PRINTABLE_CHARACTER]        = { "invalid character `%c`", PM_ERROR_LEVEL_FATAL },
     [PM_ERR_INVALID_PERCENT]                    = { "invalid `%` token", PM_ERROR_LEVEL_FATAL }, // TODO WHAT?
-    [PM_ERR_INVALID_VARIABLE_GLOBAL]            = { "invalid global variable", PM_ERROR_LEVEL_FATAL },
+    [PM_ERR_INVALID_VARIABLE_GLOBAL]            = { "`%.*s' is not allowed as a global variable name", PM_ERROR_LEVEL_FATAL },
     [PM_ERR_IT_NOT_ALLOWED]                     = { "`it` is not allowed when an ordinary parameter is defined", PM_ERROR_LEVEL_FATAL },
     [PM_ERR_LAMBDA_OPEN]                        = { "expected a `do` keyword or a `{` to open the lambda block", PM_ERROR_LEVEL_FATAL },
     [PM_ERR_LAMBDA_TERM_BRACE]                  = { "expected a lambda block beginning with `{` to end with `}`", PM_ERROR_LEVEL_FATAL },
@@ -221,6 +221,7 @@ static const pm_diagnostic_data_t diagnostic_messages[PM_DIAGNOSTIC_ID_LEN] = {
     [PM_ERR_MODULE_NAME]                        = { "expected a constant name after `module`", PM_ERROR_LEVEL_FATAL },
     [PM_ERR_MODULE_TERM]                        = { "expected an `end` to close the `module` statement", PM_ERROR_LEVEL_FATAL },
     [PM_ERR_MULTI_ASSIGN_MULTI_SPLATS]          = { "multiple splats in multiple assignment", PM_ERROR_LEVEL_FATAL },
+    [PM_ERR_MULTI_ASSIGN_UNEXPECTED_REST]       = { "unexpected '%.*s' resulting in multiple splats in multiple assignment", PM_ERROR_LEVEL_FATAL },
     [PM_ERR_NOT_EXPRESSION]                     = { "expected an expression after `not`", PM_ERROR_LEVEL_FATAL },
     [PM_ERR_NO_LOCAL_VARIABLE]                  = { "%.*s: no such local variable", PM_ERROR_LEVEL_FATAL },
     [PM_ERR_NUMBER_LITERAL_UNDERSCORE]          = { "number literal ending with a `_`", PM_ERROR_LEVEL_FATAL },
@@ -274,7 +275,8 @@ static const pm_diagnostic_data_t diagnostic_messages[PM_DIAGNOSTIC_ID_LEN] = {
     [PM_ERR_STATEMENT_UNDEF]                    = { "unexpected an `undef` at a non-statement position", PM_ERROR_LEVEL_FATAL },
     [PM_ERR_STRING_CONCATENATION]               = { "expected a string for concatenation", PM_ERROR_LEVEL_FATAL },
     [PM_ERR_STRING_INTERPOLATED_TERM]           = { "expected a closing delimiter for the interpolated string", PM_ERROR_LEVEL_FATAL },
-    [PM_ERR_STRING_LITERAL_TERM]                = { "expected a closing delimiter for the string literal", PM_ERROR_LEVEL_FATAL },
+    [PM_ERR_STRING_LITERAL_EOF]                 = { "unterminated string meets end of file", PM_ERROR_LEVEL_FATAL },
+    [PM_ERR_STRING_LITERAL_TERM]                = { "unexpected %s, expected a string literal terminator", PM_ERROR_LEVEL_FATAL },
     [PM_ERR_SYMBOL_INVALID]                     = { "invalid symbol", PM_ERROR_LEVEL_FATAL }, // TODO expected symbol? prism.c ~9719
     [PM_ERR_SYMBOL_TERM_DYNAMIC]                = { "expected a closing delimiter for the dynamic symbol", PM_ERROR_LEVEL_FATAL },
     [PM_ERR_SYMBOL_TERM_INTERPOLATED]           = { "expected a closing delimiter for the interpolated symbol", PM_ERROR_LEVEL_FATAL },
@@ -282,17 +284,14 @@ static const pm_diagnostic_data_t diagnostic_messages[PM_DIAGNOSTIC_ID_LEN] = {
     [PM_ERR_TERNARY_EXPRESSION_FALSE]           = { "expected an expression after `:` in the ternary operator", PM_ERROR_LEVEL_FATAL },
     [PM_ERR_TERNARY_EXPRESSION_TRUE]            = { "expected an expression after `?` in the ternary operator", PM_ERROR_LEVEL_FATAL },
     [PM_ERR_UNDEF_ARGUMENT]                     = { "invalid argument being passed to `undef`; expected a bare word, constant, or symbol argument", PM_ERROR_LEVEL_FATAL },
-    [PM_ERR_UNARY_RECEIVER_BANG]                = { "expected a receiver for unary `!`", PM_ERROR_LEVEL_FATAL },
-    [PM_ERR_UNARY_RECEIVER_MINUS]               = { "expected a receiver for unary `-`", PM_ERROR_LEVEL_FATAL },
-    [PM_ERR_UNARY_RECEIVER_PLUS]                = { "expected a receiver for unary `+`", PM_ERROR_LEVEL_FATAL },
+    [PM_ERR_UNARY_RECEIVER]                     = { "unexpected %s, expected a receiver for unary `%c`", PM_ERROR_LEVEL_FATAL },
     [PM_ERR_UNEXPECTED_TOKEN_CLOSE_CONTEXT]     = { "unexpected %s, assuming it is closing the parent %s", PM_ERROR_LEVEL_FATAL },
     [PM_ERR_UNEXPECTED_TOKEN_IGNORE]            = { "unexpected %s, ignoring it", PM_ERROR_LEVEL_FATAL },
-    [PM_ERR_UNARY_RECEIVER_TILDE]               = { "expected a receiver for unary `~`", PM_ERROR_LEVEL_FATAL },
     [PM_ERR_UNTIL_TERM]                         = { "expected an `end` to close the `until` statement", PM_ERROR_LEVEL_FATAL },
     [PM_ERR_VOID_EXPRESSION]                    = { "unexpected void value expression", PM_ERROR_LEVEL_FATAL },
     [PM_ERR_WHILE_TERM]                         = { "expected an `end` to close the `while` statement", PM_ERROR_LEVEL_FATAL },
     [PM_ERR_WRITE_TARGET_IN_METHOD]             = { "dynamic constant assignment", PM_ERROR_LEVEL_FATAL },
-    [PM_ERR_WRITE_TARGET_READONLY]              = { "immutable variable as a write target", PM_ERROR_LEVEL_FATAL },
+    [PM_ERR_WRITE_TARGET_READONLY]              = { "Can't set variable %.*s", PM_ERROR_LEVEL_FATAL },
     [PM_ERR_WRITE_TARGET_UNEXPECTED]            = { "unexpected write target", PM_ERROR_LEVEL_FATAL },
     [PM_ERR_XSTRING_TERM]                       = { "expected a closing delimiter for the `%x` or backtick string", PM_ERROR_LEVEL_FATAL },
 

data/src/encoding.c

@@ -2253,12 +2253,12 @@ static const uint8_t pm_utf_8_dfa[] = {
 static pm_unicode_codepoint_t
 pm_utf_8_codepoint(const uint8_t *b, ptrdiff_t n, size_t *width) {
     assert(n >= 0);
-    size_t maximum = (size_t) n;
 
+    size_t maximum = (n > 4) ? 4 : ((size_t) n);
     uint32_t codepoint;
     uint32_t state = 0;
 
-    for (size_t index = 0; index < 4 && index < maximum; index++) {
+    for (size_t index = 0; index < maximum; index++) {
         uint32_t byte = b[index];
         uint32_t type = pm_utf_8_dfa[byte];
 
@@ -2267,7 +2267,7 @@ pm_utf_8_codepoint(const uint8_t *b, ptrdiff_t n, size_t *width) {
             (0xffu >> type) & (byte);
 
         state = pm_utf_8_dfa[256 + (state * 16) + type];
-        if (!state) {
+        if (state == 0) {
             *width = index + 1;
             return (pm_unicode_codepoint_t) codepoint;
         }
@@ -2282,9 +2282,17 @@ pm_utf_8_codepoint(const uint8_t *b, ptrdiff_t n, size_t *width) {
  */
 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);
-    return width;
+    assert(n >= 0);
+
+    size_t maximum = (n > 4) ? 4 : ((size_t) n);
+    uint32_t state = 0;
+
+    for (size_t index = 0; index < maximum; index++) {
+        state = pm_utf_8_dfa[256 + (state * 16) + pm_utf_8_dfa[b[index]]];
+        if (state == 0) return index + 1;
+    }
+
+    return 0;
 }
 
 /**
@@ -4186,15 +4194,6 @@ pm_encoding_gbk_char_width(const uint8_t *b, ptrdiff_t n) {
     return 0;
 }
 
-/**
- * Returns the size of the next character in the KOI-8 encoding. This means
- * checking if it's a valid codepoint in KOI-8 and if it is returning 1.
- */
-static size_t
-pm_encoding_koi8_char_width(const uint8_t *b, PRISM_ATTRIBUTE_UNUSED ptrdiff_t n) {
-    return ((*b >= 0x20 && *b <= 0x7E) || (*b >= 0x80)) ? 1 : 0;
-}
-
 /**
  * Returns the size of the next character in the Shift_JIS encoding, or 0 if a
  * character cannot be decoded from the given bytes.
@@ -4652,7 +4651,7 @@ const pm_encoding_t pm_encodings[] = {
     },
     [PM_ENCODING_KOI8_R] = {
         .name = "KOI8-R",
-        .char_width = pm_encoding_koi8_char_width,
+        .char_width = pm_encoding_single_char_width,
         .alnum_char = pm_encoding_koi8_r_alnum_char,
         .alpha_char = pm_encoding_koi8_r_alpha_char,
         .isupper_char = pm_encoding_koi8_r_isupper_char,
@@ -4660,7 +4659,7 @@ const pm_encoding_t pm_encodings[] = {
     },
     [PM_ENCODING_KOI8_U] = {
         .name = "KOI8-U",
-        .char_width = pm_encoding_koi8_char_width,
+        .char_width = pm_encoding_single_char_width,
         .alnum_char = pm_encoding_koi8_u_alnum_char,
         .alpha_char = pm_encoding_koi8_u_alpha_char,
         .isupper_char = pm_encoding_koi8_u_isupper_char,

data/src/options.c

@@ -45,17 +45,22 @@ pm_options_version_set(pm_options_t *options, const char *version, size_t length
     }
 
     if (length == 5) {
-        if (strncmp(version, "3.3.0", 5) == 0) {
+        if (strncmp(version, "3.3.0", length) == 0) {
             options->version = PM_OPTIONS_VERSION_CRUBY_3_3_0;
             return true;
         }
 
-        if (strncmp(version, "latest", 6) == 0) {
+        if (strncmp(version, "3.4.0", length) == 0) {
             options->version = PM_OPTIONS_VERSION_LATEST;
             return true;
         }
     }
 
+    if (length == 6 && strncmp(version, "latest", length) == 0) {
+        options->version = PM_OPTIONS_VERSION_LATEST;
+        return true;
+    }
+
     return false;
 }