MatchData#named_captures: add optional symbolize_names keyword (#6952)

Merged-By: ioquatix <samuel@codeotaku.com>

NEWS.md

@@ -30,6 +30,10 @@ Note: We're only listing outstanding class updates.
     * `Dir#chdir` added for changing the directory to the directory specified
       by the provided `Dir` object. [[Feature #19347]]
 
+* MatchData
+
+    * MatchData#named_captures now accepts optional `symbolize_names` keyword. [[Feature #19591]]
+
 * String
 
     * `String#unpack` now raises ArgumentError for unknown directives. [[Bug #19150]]

re.c

@@ -2316,7 +2316,7 @@ match_named_captures_iter(const OnigUChar *name, const OnigUChar *name_end,
 
 /*
  *  call-seq:
- *    named_captures -> hash
+ *    named_captures(symbolize_names: false) -> hash
  *
  *  Returns a hash of the named captures;
  *  each key is a capture name; each value is its captured string or +nil+:
@@ -2337,10 +2337,17 @@ match_named_captures_iter(const OnigUChar *name, const OnigUChar *name_end,
  *    # => #<MatchData "01" a:"0" a:"1">
  *    m.named_captures #=> {"a" => "1"}
  *
+ * If keyword argument +symbolize_names+ is given
+ * a true value, the keys in the resulting hash are Symbols:
+ *
+ *    m = /(?<a>.)(?<a>.)/.match("01")
+ *    # => #<MatchData "01" a:"0" a:"1">
+ *    m.named_captures(symbolize_names: true) #=> {:a => "1"}
+ *
  */
 
 static VALUE
-match_named_captures(VALUE match)
+match_named_captures(int argc, VALUE *argv, VALUE match)
 {
     VALUE hash;
     struct MEMO *memo;
@@ -2349,8 +2356,27 @@ match_named_captures(VALUE match)
     if (NIL_P(RMATCH(match)->regexp))
         return rb_hash_new();
 
+    VALUE opt;
+    VALUE symbolize_names = 0;
+
+    rb_scan_args(argc, argv, "0:", &opt);
+
+    if (!NIL_P(opt)) {
+        static ID keyword_ids[1];
+
+        VALUE symbolize_names_val;
+
+        if (!keyword_ids[0]) {
+            keyword_ids[0] = rb_intern_const("symbolize_names");
+        }
+        rb_get_kwargs(opt, keyword_ids, 0, 1, &symbolize_names_val);
+        if (!UNDEF_P(symbolize_names_val) && RTEST(symbolize_names_val)) {
+            symbolize_names = 1;
+        }
+    }
+
     hash = rb_hash_new();
-    memo = MEMO_NEW(hash, match, 0);
+    memo = MEMO_NEW(hash, match, symbolize_names);
 
     onig_foreach_name(RREGEXP(RMATCH(match)->regexp)->ptr, match_named_captures_iter, (void*)memo);
 
@@ -4754,7 +4780,7 @@ Init_Regexp(void)
     rb_define_method(rb_cMatch, "[]", match_aref, -1);
     rb_define_method(rb_cMatch, "captures", match_captures, 0);
     rb_define_alias(rb_cMatch,  "deconstruct", "captures");
-    rb_define_method(rb_cMatch, "named_captures", match_named_captures, 0);
+    rb_define_method(rb_cMatch, "named_captures", match_named_captures, -1);
     rb_define_method(rb_cMatch, "deconstruct_keys", match_deconstruct_keys, 1);
     rb_define_method(rb_cMatch, "values_at", match_values_at, -1);
     rb_define_method(rb_cMatch, "pre_match", rb_reg_match_pre, 0);

spec/ruby/core/matchdata/named_captures_spec.rb

@@ -12,4 +12,16 @@ describe 'MatchData#named_captures' do
   it 'returns the latest matched capture, even if a later one that does not match exists' do
     /\A(?<a>.)(?<b>.)(?<b>.)(?<a>.)?\z/.match('012').named_captures.should == { 'a' => '0', 'b' => '2' }
   end
+
+  ruby_version_is "3.3" do
+    it 'returns a Hash with Symbol keys when symbolize_names is provided a true value' do
+      /(?<a>.)(?<b>.)?/.match('0').named_captures(symbolize_names: true).should == { a: '0', b: nil }
+      /(?<a>.)(?<b>.)?/.match('0').named_captures(symbolize_names: "truly").should == { a: '0', b: nil }
+    end
+
+    it 'returns a Hash with String keys when symbolize_names is provided a false value' do
+      /(?<a>.)(?<b>.)?/.match('02').named_captures(symbolize_names: false).should == { 'a' => '0', 'b' => '2' }
+      /(?<a>.)(?<b>.)?/.match('02').named_captures(symbolize_names: nil).should == { 'a' => '0', 'b' => '2' }
+    end
+  end
 end

test/ruby/test_regexp.rb

@@ -304,6 +304,9 @@ class TestRegexp < Test::Unit::TestCase
     assert_equal({'a' => '1', 'b' => '2', 'c' => '3'}, /^(?<a>.)(?<b>.)(?<c>.)?/.match('123').named_captures)
     assert_equal({'a' => '1', 'b' => '2', 'c' => ''}, /^(?<a>.)(?<b>.)(?<c>.?)/.match('12').named_captures)
 
+    assert_equal({a: '1', b: '2', c: ''}, /^(?<a>.)(?<b>.)(?<c>.?)/.match('12').named_captures(symbolize_names: true))
+    assert_equal({'a' => '1', 'b' => '2', 'c' => ''}, /^(?<a>.)(?<b>.)(?<c>.?)/.match('12').named_captures(symbolize_names: false))
+
     assert_equal({'a' => 'x'}, /(?<a>x)|(?<a>y)/.match('x').named_captures)
     assert_equal({'a' => 'y'}, /(?<a>x)|(?<a>y)/.match('y').named_captures)