prism 0.30.0 → 1.0.0

data/ext/prism/extconf.rb

@@ -50,7 +50,7 @@ def make(env, target)
   Dir.chdir(File.expand_path("../..", __dir__)) do
     system(
       env,
-      RUBY_PLATFORM.include?("openbsd") ? "gmake" : "make",
+      RUBY_PLATFORM.match?(/openbsd|freebsd/) ? "gmake" : "make",
       target,
       exception: true
     )

data/ext/prism/extension.c

@@ -21,6 +21,7 @@ VALUE rb_cPrismParseError;
 VALUE rb_cPrismParseWarning;
 VALUE rb_cPrismResult;
 VALUE rb_cPrismParseResult;
+VALUE rb_cPrismLexResult;
 VALUE rb_cPrismParseLexResult;
 
 VALUE rb_cPrismDebugEncoding;
@@ -138,7 +139,13 @@ build_options_i(VALUE key, VALUE value, VALUE argument) {
     if (key_id == rb_id_option_filepath) {
         if (!NIL_P(value)) pm_options_filepath_set(options, check_string(value));
     } else if (key_id == rb_id_option_encoding) {
-        if (!NIL_P(value)) pm_options_encoding_set(options, rb_enc_name(rb_to_encoding(value)));
+        if (!NIL_P(value)) {
+            if (value == Qfalse) {
+                pm_options_encoding_locked_set(options, true);
+            } else {
+                pm_options_encoding_set(options, rb_enc_name(rb_to_encoding(value)));
+            }
+        }
     } else if (key_id == rb_id_option_line) {
         if (!NIL_P(value)) pm_options_line_set(options, NUM2INT(value));
     } else if (key_id == rb_id_option_frozen_string_literal) {
@@ -206,6 +213,7 @@ build_options(VALUE argument) {
 static void
 extract_options(pm_options_t *options, VALUE filepath, VALUE keywords) {
     options->line = 1; // default
+
     if (!NIL_P(keywords)) {
         struct build_options_data data = { .options = options, .keywords = keywords };
         struct build_options_data *argument = &data;
@@ -364,7 +372,7 @@ dump_file(int argc, VALUE *argv, VALUE self) {
  */
 static VALUE
 parser_comments(pm_parser_t *parser, VALUE source) {
-    VALUE comments = rb_ary_new();
+    VALUE comments = rb_ary_new_capa(parser->comment_list.size);
 
     for (pm_comment_t *comment = (pm_comment_t *) parser->comment_list.head; comment != NULL; comment = (pm_comment_t *) comment->node.next) {
         VALUE location_argv[] = {
@@ -386,7 +394,7 @@ parser_comments(pm_parser_t *parser, VALUE source) {
  */
 static VALUE
 parser_magic_comments(pm_parser_t *parser, VALUE source) {
-    VALUE magic_comments = rb_ary_new();
+    VALUE magic_comments = rb_ary_new_capa(parser->magic_comment_list.size);
 
     for (pm_magic_comment_t *magic_comment = (pm_magic_comment_t *) parser->magic_comment_list.head; magic_comment != NULL; magic_comment = (pm_magic_comment_t *) magic_comment->node.next) {
         VALUE key_loc_argv[] = {
@@ -436,7 +444,7 @@ parser_data_loc(const pm_parser_t *parser, VALUE source) {
  */
 static VALUE
 parser_errors(pm_parser_t *parser, rb_encoding *encoding, VALUE source) {
-    VALUE errors = rb_ary_new();
+    VALUE errors = rb_ary_new_capa(parser->error_list.size);
     pm_diagnostic_t *error;
 
     for (error = (pm_diagnostic_t *) parser->error_list.head; error != NULL; error = (pm_diagnostic_t *) error->node.next) {
@@ -479,7 +487,7 @@ parser_errors(pm_parser_t *parser, rb_encoding *encoding, VALUE source) {
  */
 static VALUE
 parser_warnings(pm_parser_t *parser, rb_encoding *encoding, VALUE source) {
-    VALUE warnings = rb_ary_new();
+    VALUE warnings = rb_ary_new_capa(parser->warning_list.size);
     pm_diagnostic_t *warning;
 
     for (warning = (pm_diagnostic_t *) parser->warning_list.head; warning != NULL; warning = (pm_diagnostic_t *) warning->node.next) {
@@ -556,9 +564,10 @@ static void
 parse_lex_token(void *data, pm_parser_t *parser, pm_token_t *token) {
     parse_lex_data_t *parse_lex_data = (parse_lex_data_t *) parser->lex_callback->data;
 
-    VALUE yields = rb_ary_new_capa(2);
-    rb_ary_push(yields, pm_token_new(parser, token, parse_lex_data->encoding, parse_lex_data->source));
-    rb_ary_push(yields, INT2FIX(parser->lex_state));
+    VALUE yields = rb_assoc_new(
+        pm_token_new(parser, token, parse_lex_data->encoding, parse_lex_data->source),
+        INT2FIX(parser->lex_state)
+    );
 
     rb_ary_push(parse_lex_data->tokens, yields);
 }
@@ -599,7 +608,7 @@ parse_lex_input(pm_string_t *input, const pm_options_t *options, bool return_nod
     pm_parser_register_encoding_changed_callback(&parser, parse_lex_encoding_changed_callback);
 
     VALUE source_string = rb_str_new((const char *) pm_string_source(input), pm_string_length(input));
-    VALUE offsets = rb_ary_new();
+    VALUE offsets = rb_ary_new_capa(parser.newline_list.size);
     VALUE source = rb_funcall(rb_cPrismSource, rb_id_source_for, 3, source_string, LONG2NUM(parser.start_line), offsets);
 
     parse_lex_data_t parse_lex_data = {
@@ -628,16 +637,16 @@ parse_lex_input(pm_string_t *input, const pm_options_t *options, bool return_nod
         rb_ary_push(offsets, ULONG2NUM(parser.newline_list.offsets[index]));
     }
 
-    VALUE value;
+    VALUE result;
     if (return_nodes) {
-        value = rb_ary_new_capa(2);
+        VALUE value = rb_ary_new_capa(2);
         rb_ary_push(value, pm_ast_new(&parser, node, parse_lex_data.encoding, source));
         rb_ary_push(value, parse_lex_data.tokens);
+        result = parse_result_create(rb_cPrismParseLexResult, &parser, value, parse_lex_data.encoding, source);
     } else {
-        value = parse_lex_data.tokens;
+        result = parse_result_create(rb_cPrismLexResult, &parser, parse_lex_data.tokens, parse_lex_data.encoding, source);
     }
 
-    VALUE result = parse_result_create(rb_cPrismParseLexResult, &parser, value, parse_lex_data.encoding, source);
     pm_node_destroy(&parser, node);
     pm_parser_free(&parser);
 
@@ -646,10 +655,10 @@ parse_lex_input(pm_string_t *input, const pm_options_t *options, bool return_nod
 
 /**
  * call-seq:
- *   Prism::lex(source, **options) -> Array
+ *   Prism::lex(source, **options) -> LexResult
  *
- * Return an array of Token instances corresponding to the given string. For
- * supported options, see Prism::parse.
+ * Return a LexResult instance that contains an array of Token instances
+ * corresponding to the given string. For supported options, see Prism::parse.
  */
 static VALUE
 lex(int argc, VALUE *argv, VALUE self) {
@@ -666,10 +675,10 @@ lex(int argc, VALUE *argv, VALUE self) {
 
 /**
  * call-seq:
- *   Prism::lex_file(filepath, **options) -> Array
+ *   Prism::lex_file(filepath, **options) -> LexResult
  *
- * Return an array of Token instances corresponding to the given file. For
- * supported options, see Prism::parse.
+ * Return a LexResult instance that contains an array of Token instances
+ * corresponding to the given file. For supported options, see Prism::parse.
  */
 static VALUE
 lex_file(int argc, VALUE *argv, VALUE self) {
@@ -849,8 +858,8 @@ parse_stream_fgets(char *string, int size, void *stream) {
         return NULL;
     }
 
-    const char *cstr = StringValueCStr(line);
-    size_t length = strlen(cstr);
+    const char *cstr = RSTRING_PTR(line);
+    long length = RSTRING_LEN(line);
 
     memcpy(string, cstr, length);
     string[length] = '\0';
@@ -954,9 +963,9 @@ parse_file_comments(int argc, VALUE *argv, VALUE self) {
 
 /**
  * call-seq:
- *   Prism::parse_lex(source, **options) -> ParseResult
+ *   Prism::parse_lex(source, **options) -> ParseLexResult
  *
- * Parse the given string and return a ParseResult instance that contains a
+ * Parse the given string and return a ParseLexResult instance that contains a
  * 2-element array, where the first element is the AST and the second element is
  * an array of Token instances.
  *
@@ -981,9 +990,9 @@ parse_lex(int argc, VALUE *argv, VALUE self) {
 
 /**
  * call-seq:
- *   Prism::parse_lex_file(filepath, **options) -> ParseResult
+ *   Prism::parse_lex_file(filepath, **options) -> ParseLexResult
  *
- * Parse the given file and return a ParseResult instance that contains a
+ * Parse the given file and return a ParseLexResult instance that contains a
  * 2-element array, where the first element is the AST and the second element is
  * an array of Token instances.
  *
@@ -1124,6 +1133,7 @@ Init_prism(void) {
     rb_cPrismParseWarning = rb_define_class_under(rb_cPrism, "ParseWarning", rb_cObject);
     rb_cPrismResult = rb_define_class_under(rb_cPrism, "Result", rb_cObject);
     rb_cPrismParseResult = rb_define_class_under(rb_cPrism, "ParseResult", rb_cPrismResult);
+    rb_cPrismLexResult = rb_define_class_under(rb_cPrism, "LexResult", rb_cPrismResult);
     rb_cPrismParseLexResult = rb_define_class_under(rb_cPrism, "ParseLexResult", rb_cPrismResult);
 
     // Intern all of the IDs eagerly that we support so that we don't have to do

data/ext/prism/extension.h

@@ -1,7 +1,7 @@
 #ifndef PRISM_EXT_NODE_H
 #define PRISM_EXT_NODE_H
 
-#define EXPECTED_PRISM_VERSION "0.30.0"
+#define EXPECTED_PRISM_VERSION "1.0.0"
 
 #include <ruby.h>
 #include <ruby/encoding.h>
@@ -14,6 +14,6 @@ VALUE pm_integer_new(const pm_integer_t *integer);
 
 void Init_prism_api_node(void);
 void Init_prism_pack(void);
-PRISM_EXPORTED_FUNCTION void Init_prism(void);
+RUBY_FUNC_EXPORTED void Init_prism(void);
 
 #endif