benchmark.log

ruby 3.2.0dev (2022-10-26T06:02:04Z master 94f3aa2126) [x86_64-linux]
Warming up --------------------------------------
              before     1.409M i/100ms
               after     1.895M i/100ms
Calculating -------------------------------------
              before     13.931M (± 0.1%) i/s -     70.435M in   5.055915s
               after     18.669M (± 0.3%) i/s -     94.769M in   5.076395s

Comparison:
               after: 18668840.0 i/s
              before: 13931312.1 i/s - 1.34x  (± 0.00) slower

benchmark.rb

require 'benchmark/ips'
require 'cgi'
require 'cgi/escape'

str = "string"

Benchmark.ips do |x|
  x.report('before') { CGI.escapeHTML(str) }
  x.report('after') { CGI.escapeHTML2(str) }
  x.compare!
end

patch.diff

diff --git a/ext/cgi/escape/escape.c b/ext/cgi/escape/escape.c
index c5b76de596..508cd36f72 100644
--- a/ext/cgi/escape/escape.c
+++ b/ext/cgi/escape/escape.c
@@ -75,6 +75,39 @@ optimized_escape_html(VALUE str)
     return escaped;
 }

+static VALUE
+optimized_escape_html2(VALUE str)
+{
+    VALUE vbuf;
+    char *buf = ALLOCV_N(char, vbuf, escaped_length(str));
+    const char *cstr = RSTRING_PTR(str);
+    const char *end = cstr + RSTRING_LEN(str);
+
+    char *dest = buf;
+    while (cstr < end) {
+        const unsigned char c = *cstr++;
+        uint8_t len = html_escape_table[c].len;
+        if (len) {
+            memcpy(dest, html_escape_table[c].str, len);
+            dest += len;
+        }
+        else {
+            *dest++ = c;
+        }
+    }
+
+    VALUE escaped;
+    if (RSTRING_LEN(str) < (dest - buf)) {
+        escaped = rb_str_new(buf, dest - buf);
+        preserve_original_state(str, escaped);
+    }
+    else {
+        escaped = str;
+    }
+    ALLOCV_END(vbuf);
+    return escaped;
+}
+
 static VALUE
 optimized_unescape_html(VALUE str)
 {
@@ -324,6 +357,19 @@ cgiesc_escape_html(VALUE self, VALUE str)
     }
 }

+static VALUE
+cgiesc_escape_html2(VALUE self, VALUE str)
+{
+    StringValue(str);
+
+    if (rb_enc_str_asciicompat_p(str)) {
+        return optimized_escape_html2(str);
+    }
+    else {
+        return rb_call_super(1, &str);
+    }
+}
+
 /*
  *  call-seq:
  *     CGI.unescapeHTML(string) -> string
@@ -456,6 +502,7 @@ InitVM_escape(void)
     rb_mEscape = rb_define_module_under(rb_cCGI, "Escape");
     rb_mUtil   = rb_define_module_under(rb_cCGI, "Util");
     rb_define_method(rb_mEscape, "escapeHTML", cgiesc_escape_html, 1);
+    rb_define_method(rb_mEscape, "escapeHTML2", cgiesc_escape_html2, 1);
     rb_define_method(rb_mEscape, "unescapeHTML", cgiesc_unescape_html, 1);
     rb_define_method(rb_mEscape, "escapeURIComponent", cgiesc_escape_uri_component, 1);
     rb_define_method(rb_mEscape, "unescapeURIComponent", cgiesc_unescape_uri_component, -1);