Issue #8206 has been updated by sam.saffron (Sam Saffron).


This is a MASSIVE improvement:

#!/usr/bin/env ruby
$: << File.dirname(__FILE__)+'/lib'
require 'benchmark'
require 'fast_blank'

class String
  # active support implementation
  def slow_blank?
    self !~ /[^[:space:]]/
  end
end


n = 1000000


strings = [
  "",
  "\r\n\r\n  ",
  "this is a test",
  "   this is a longer test",
  "   this is a longer test
      this is a longer test
      this is a longer test
      this is a longer test
      this is a longer test"
]

strings.each do |s|
  raise "failed on #{s.inspect}" if s.blank? != s.slow_blank?
end

Benchmark.bmbm  do |x|
  strings.each do |s|
    x.report("Fast Blank #{s.length}    :") do  n.times { s.blank? }  end
    x.report("Fast Blank (Active Support)  #{s.length}    :") do  n.times { s.blank_as? }  end
    x.report("Slow Blank #{s.length}    :") do  n.times { s.slow_blank? }  end
    x.report("include? #{s.length}    :") do  n.times { !s.include?(/[^[:space]]/) }  end
  end
end


                                            user     system      total        real
Fast Blank 0    :                       0.080000   0.000000   0.080000 (  0.077008)
Fast Blank (Active Support)  0    :     0.080000   0.000000   0.080000 (  0.076362)
Slow Blank 0    :                       0.380000   0.000000   0.380000 (  0.378698)
include? 0    :                         0.180000   0.000000   0.180000 (  0.184465)
Fast Blank 6    :                       0.180000   0.000000   0.180000 (  0.180450)
Fast Blank (Active Support)  6    :     0.210000   0.000000   0.210000 (  0.207886)
Slow Blank 6    :                       0.590000   0.000000   0.590000 (  0.588945)
include? 6    :                         0.190000   0.000000   0.190000 (  0.190898)
Fast Blank 14    :                      0.090000   0.000000   0.090000 (  0.088225)
Fast Blank (Active Support)  14    :    0.130000   0.000000   0.130000 (  0.131408)
Slow Blank 14    :                      0.670000   0.000000   0.670000 (  0.674838)
include? 14    :                        0.190000   0.000000   0.190000 (  0.191627)
Fast Blank 24    :                      0.190000   0.000000   0.190000 (  0.186498)
Fast Blank (Active Support)  24    :    0.140000   0.010000   0.150000 (  0.147858)
Slow Blank 24    :                      0.770000   0.000000   0.770000 (  0.767816)
include? 24    :                        0.220000   0.000000   0.220000 (  0.220636)
Fast Blank 136    :                     0.150000   0.000000   0.150000 (  0.150967)
Fast Blank (Active Support)  136    :   0.150000   0.000000   0.150000 (  0.147665)
Slow Blank 136    :                     0.770000   0.000000   0.770000 (  0.779459)
include? 136    :                       0.200000   0.000000   0.200000 (  0.189744)


Some notes:

1. I am noticing ruby head as a 20% or so faster regex going on that 2.0 for these tests
2. the include? method is only 30% or so percent slower than hand coding, though empty strings need special casing. Essentially include? should be short cutting if the string length is zero and returning false. 
3. I love this improvement to include?, totally support it accepting regexes. Though I very much worry about consistency here. 

My suggestion would be: 

1. Amend include? to accept a regex 
2. Keep in line with the changes in https://bugs.ruby-lang.org/issues/8110 ... so for it to skip globals you MUST pass in /regx/S (a regex that skips setting globals) 

I very much worry about having a mishmash in the language where some methods avoid global settings and others do not. The cleanest way of introducing this change is simply to allow for the new rege modifier and keep all places that accept regexes in MRI consistent. 

----------------------------------------
Feature #8206: Should Ruby core implement String#blank? 
https://bugs.ruby-lang.org/issues/8206#change-38249

Author: sam.saffron (Sam Saffron)
Status: Open
Priority: Normal
Assignee: 
Category: core
Target version: 


There has been some discussion about porting the #blank? protocol over to Ruby in the past that has been rejected by Matz. 

This proposal is only about String however. 

At the moment to figure out if you have a blank string you would 

"  ".strip.length == 0

The disadvantage is that this forces unneeded allocations and does too much work: 

An optimal implementation would be:

static VALUE
rb_str_blank(VALUE str)
{
  rb_encoding *enc;
  char *s, *e;

  enc = STR_ENC_GET(str);
  s = RSTRING_PTR(str);
  if (!s || RSTRING_LEN(str) == 0) return Qtrue;

  e = RSTRING_END(str);
  while (s < e) {
	  int n;
	  unsigned int cc = rb_enc_codepoint_len(s, e, &n, enc);

	  if (!rb_isspace(cc) && cc != 0) return Qfalse;
    s += n;
  }
  return Qtrue;
}

This in turn is about 5-8x than the regex solution to the problem and way faster than allocating one massive string with strip when length is large. 

Should Ruby take on this method, to accompany #strip following its practice. 

--- 

A slight caveat though is that active support has a somewhat different definition of blank? 

const unsigned int as_blank[26] = {9, 0xa, 0xb, 0xc, 0xd,
  0x20, 0x85, 0xa0, 0x1680, 0x180e, 0x2000, 0x2001,
  0x2002, 0x2003, 0x2004, 0x2005, 0x2006, 0x2007, 0x2008,
  0x2009, 0x200a, 0x2028, 0x2029, 0x202f, 0x205f, 0x3000
};

static VALUE
rb_str_blank_as(VALUE str)
{
  rb_encoding *enc;
  char *s, *e;
  int i;
  int found;

  enc = STR_ENC_GET(str);
  s = RSTRING_PTR(str);
  if (!s || RSTRING_LEN(str) == 0) return Qtrue;

  e = RSTRING_END(str);
  while (s < e) {
	  int n;
	  unsigned int cc = rb_enc_codepoint_len(s, e, &n, enc);

    found = 0;
    for(i=0;i<26;i++){
      unsigned int current = as_blank[i];
      if(current == cc) {
        found = 1;
        break;
      }
      if(cc < current){
        break;
      }
    }

	  if (!found) return Qfalse;
    s += n;
  }
  return Qtrue;
}

Clearly it makes no sense to have such a method. 

If Ruby took over implementing String#blank? it would clash with Active Support. But imho would enforce better API consistency. 

Thoughts?


 


-- 
http://bugs.ruby-lang.org/