From a34a57a498acf377a01dd3db2f43dec9168e97b8 Mon Sep 17 00:00:00 2001
From: Prem Sichanugrist <s@sikac.hu>
Date: Tue, 24 Nov 2015 14:14:31 -0500
Subject: [PATCH] Introduce String#squish and String#squish!

This is a port of methods that are part of Active Support[1]. Both of
these methods are useful when you are dealing with heredoc and would
like to ignore all the whitespaces before, after, and in-between the
string.

Example:

    sql = <<-SQL.squish
      SELECT *
      FROM users
      WHERE users.username = 'sikachu'
    SQL
    #=> "SELECT * FROM users WHERE users.username='sikachu'"

[1]: http://api.rubyonrails.org/v4.2.5/classes/String.html#method-i-squish
---
 string.c                 | 58 ++++++++++++++++++++++++++++++++++++++++++++++++
 test/ruby/test_string.rb | 18 +++++++++++++++
 2 files changed, 76 insertions(+)

diff --git a/string.c b/string.c
index d834754..90b32a0 100644
--- a/string.c
+++ b/string.c
@@ -8803,6 +8803,62 @@ str_scrub_bang(int argc, VALUE *argv, VALUE str)
     return str;
 }
 
+/*
+ *  call-seq:
+ *    str.squish! -> str
+ *
+ *  Performs a destructive squish. See String#squish.
+ *
+ *     str = " foo   bar    \n   \t   boo"
+ *     str.squish!                         # => "foo bar boo"
+ *     str                                 # => "foo bar boo"
+ */
+static VALUE
+rb_str_squish_bang(VALUE str)
+{
+    static const char before_regex_source[] = "\\A[[:space:]]+";
+    static const char after_regex_source[] = "[[:space:]]+\\z";
+    static const char between_regex_source[] = "[[:space:]]+";
+    VALUE before_argv[] = {
+        rb_reg_new(before_regex_source, sizeof before_regex_source - 1, 0),
+        rb_str_new_cstr("")
+    };
+    VALUE after_argv[] = {
+        rb_reg_new(after_regex_source, sizeof after_regex_source - 1, 0),
+        rb_str_new_cstr("")
+    };
+    VALUE between_argv[] = {
+        rb_reg_new(between_regex_source, sizeof between_regex_source - 1, 0),
+        rb_str_new_cstr(" ")
+    };
+
+    rb_str_gsub_bang(2, before_argv, str);
+    rb_str_gsub_bang(2, after_argv, str);
+    rb_str_gsub_bang(2, between_argv, str);
+    return str;
+}
+
+/*
+ *  call-seq:
+ *    str.squish -> str
+ *
+ *  Returns the string, first removing all whitespace on both ends of the
+ *  string, and then changing remaining consecutive whitespace groups into one
+ *  space each.
+ *
+ *  Note that it handles both ASCII and Unicode whitespace.
+ *
+ *     %{ Multi-line
+ *        string }.squish                   # => "Multi-line string"
+ *     " foo   bar    \n   \t   boo".squish # => "foo bar boo"
+ */
+static VALUE
+rb_str_squish(VALUE str)
+{
+    str = rb_str_dup(str);
+    return rb_str_squish_bang(str);
+}
+
 /**********************************************************************
  * Document-class: Symbol
  *
@@ -9328,6 +9384,7 @@ Init_String(void)
     rb_define_method(rb_cString, "strip", rb_str_strip, 0);
     rb_define_method(rb_cString, "lstrip", rb_str_lstrip, 0);
     rb_define_method(rb_cString, "rstrip", rb_str_rstrip, 0);
+    rb_define_method(rb_cString, "squish", rb_str_squish, 0);
 
     rb_define_method(rb_cString, "sub!", rb_str_sub_bang, -1);
     rb_define_method(rb_cString, "gsub!", rb_str_gsub_bang, -1);
@@ -9336,6 +9393,7 @@ Init_String(void)
     rb_define_method(rb_cString, "strip!", rb_str_strip_bang, 0);
     rb_define_method(rb_cString, "lstrip!", rb_str_lstrip_bang, 0);
     rb_define_method(rb_cString, "rstrip!", rb_str_rstrip_bang, 0);
+    rb_define_method(rb_cString, "squish!", rb_str_squish_bang, 0);
 
     rb_define_method(rb_cString, "tr", rb_str_tr, 2);
     rb_define_method(rb_cString, "tr_s", rb_str_tr_s, 2);
diff --git a/test/ruby/test_string.rb b/test/ruby/test_string.rb
index ce21db9..e62ca67 100644
--- a/test/ruby/test_string.rb
+++ b/test/ruby/test_string.rb
@@ -2304,6 +2304,24 @@ def test_LSHIFT_neary_long_max
     end;
   end if [0].pack("l!").bytesize < [nil].pack("p").bytesize
   # enable only when string size range is smaller than memory space
+
+  def test_string_squish
+    original = %{\u205f\u3000 A string surrounded by various unicode spaces,
+      with tabs(\t\t), newlines(\n\n), unicode nextlines(\u0085\u0085) and many spaces(  ). \u00a0\u2007}
+
+    expected = "A string surrounded by various unicode spaces, " +
+      "with tabs( ), newlines( ), unicode nextlines( ) and many spaces( )."
+
+    # Make sure squish returns what we expect:
+    assert_equal expected, original.squish
+    # But doesn't modify the original string:
+    assert_not_equal expected, original
+
+    # Make sure squish! returns what we expect:
+    assert_equal expected, original.squish!
+    # And changes the original string:
+    assert_equal expected, original
+  end
 end
 
 class TestString2 < TestString
-- 
2.6.3