[Bug #22423] Preserve encoding with String capacity

Merged: https://github.com/ruby/ruby/pull/19297

string.c

@@ -2194,7 +2194,10 @@ rb_str_init(int argc, VALUE *argv, VALUE str)
             TERM_FILL(&RSTRING(str)->as.heap.ptr[len], termlen);
             if (n == 1) {
                 memcpy(RSTRING(str)->as.heap.ptr, RSTRING_PTR(orig), len);
-                rb_enc_cr_str_exact_copy(str, orig);
+                if (!enc || rb_enc_get(orig) == enc) {
+                    rb_enc_cr_str_exact_copy(str, orig);
+                    enc = 0;
+                }
             }
             FL_SET(str, STR_NOEMBED);
             RSTRING(str)->as.heap.aux.capa = capa;
@@ -2203,7 +2206,8 @@ rb_str_init(int argc, VALUE *argv, VALUE str)
             rb_str_replace(str, orig);
         }
         if (enc) {
-            rb_enc_associate(str, enc);
+            str_modifiable(str);
+            rb_enc_raw_set(str, enc);
             ENC_CODERANGE_CLEAR(str);
         }
     }

test/ruby/test_string.rb

@@ -1,5 +1,6 @@
 # frozen_string_literal: false
 require 'test/unit'
+require '-test-/string'
 
 class TestString < Test::Unit::TestCase
   WIDE_ENCODINGS = [
@@ -4649,4 +4650,38 @@ class TestString2 < TestString
     super
     @cls = S2
   end
+
+  def assert_capacity(capa, str, msg)
+    assert_equal capa, Bug::String.capacity(str), msg
+  end
+
+  def test_capacity_inflation
+    orig = ("\u{30AF}" * 5)
+    all_assertions_foreach(nil, Encoding::UTF_8, *WIDE_ENCODINGS) do |enc|
+      assert_capacity(200, S2.new(orig.encode(enc), encoding: "UTF-8", capacity: 200), enc.name)
+    end
+  end
+
+  def test_capacity_no_inflation
+    orig = "hello"
+    all_assertions_foreach(nil, Encoding::UTF_8, *WIDE_ENCODINGS) do |enc|
+      assert_capacity(100, S2.new(orig, encoding: enc, capacity: 100), enc.name)
+    end
+  end
+
+  def test_capacity_no_overflow_on_append
+    orig = ("\u{30AF}" * 5)
+    all_assertions_foreach(nil, *WIDE_ENCODINGS) do |enc|
+      encoded = orig.encode(enc)
+      size = encoded.bytesize
+
+      s = S2.new(encoded, encoding: "UTF-8", capacity: 200)
+      s << "x" * 191
+      assert_equal(size + 191, s.bytesize, enc.name)
+
+      s = S2.new(encoded, encoding: "UTF-8", capacity: 207)
+      s << "y" * 207
+      assert_equal(size + 207, s.bytesize, enc.name)
+    end
+  end
 end