# Is there a bug in the ActiveRecord method validates\_format\_o **URL:** \ **Category:** Rails\ **Created:** [April 22, 2007, 11:04pm UTC](https://www.ruby-forum.com/t/is-there-a-bug-in-the-activerecord-method-validates-format-o/92672 "2007-04-22T23:04:49Z")\ **Posts on this page:** 5\ **Page:** 1
**Author:** ![dparkmit](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/d/cdc98d/32.png) [@dparkmit](https://www.ruby-forum.com/u/dparkmit)\ **Post date:** [April 22, 2007, 11:04pm UTC](https://www.ruby-forum.com/t/is-there-a-bug-in-the-activerecord-method-validates-format-o/92672/1 "2007-04-22T23:04:49Z")
Hi All, I think that I might have found a bug in the validates\_format\_of method. Below is my code and the test case that threw an error. Please let me know if there is a bug in my code that I just didn’t catch. Here is the code for my User model. As you can see, it has one attribute called name. I wanted to make sure that the names had to be at least 6 characters long and consisted only of letters, numbers, dots (.), underscores (\_) and at symbols (@). class User \< ActiveRecord::Base validates\_presence\_of :name validates\_uniqueness\_of :name validates\_format\_of :name, :with =\> /\A[.\w@]{6,}\z/ end Here is the code for a unit test that I wrote. This unit test tries to create a fairly comprehensive set of invalid names by creating several tests for each ASCII character that shouldn’t be part of a valid name such as ~ or `. None of the bad names that are created should be valid. def test\_invalid\_name bad\_characters = [] for i in 0…255 bad\_characters \<\< i.chr unless i.chr =~ /[.\w@]/ end ``` bad_names = [] bad_characters.each do |c| bad_names << (c + "abcdef") bad_names << ("abcdef" + c) bad_names << ("abc" + c + "def") bad_names << (c + "abc" + c + "def" + c) end bad_names.each do |name| user = User.new(:name => name, :password => ``` “password”, :password\_confirmation =\> “password”) assert !user.valid? assert user.errors.invalid?(:name), “Name:” + name end end When I ran this test, it failed for bad characters that are extended ASCII characters (i.e., characters that have an ASCII value between 128 and 255). For example, user.valid? had a value of true when the name was Çabcdef The ASCII value of Ç is 128. User.valid? should have had a value of false because the name doesn’t match the regex that I used in the validates\_format\_of method call. In fact, the result of user.name =~ /\A[.\w@]{6,}\z/ is false. Can anyone tell me if there is a bug in my code or not? Any help would be appreciated! ---
**Author:** ![dparkmit](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/d/cdc98d/32.png) [@dparkmit](https://www.ruby-forum.com/u/dparkmit)\ **Post date:** [April 22, 2007, 11:27pm UTC](https://www.ruby-forum.com/t/is-there-a-bug-in-the-activerecord-method-validates-format-o/92672/2 "2007-04-22T23:27:58Z")
From the documentation on Ruby regular expressions, I find: /[single character] Set the character encoding, character should be one of ‘neus’: none, EUC, UTF-8 or SJIS. So, perhaps what you’re looking for is of the syntax: /expr/u See if it works. ---
**Author:** ![dparkmit](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/d/cdc98d/32.png) [@dparkmit](https://www.ruby-forum.com/u/dparkmit)\ **Post date:** [April 22, 2007, 11:34pm UTC](https://www.ruby-forum.com/t/is-there-a-bug-in-the-activerecord-method-validates-format-o/92672/3 "2007-04-22T23:34:58Z")
Hi, Thanks for the advice. I actually figured out how to make it work. I changed the regex for the validates\_format\_of code. Here’s the original code: validates\_format\_of :name, :with =\> /\A[.\w@]{6,}\z/ Here’s the new code: validates\_format\_of :name, :with =\> /\A[A-Za-z0-9\_.@]{6,}\z/ When I used the new code, all tests passed. It’s really strange though because [\w] is supposed to equal [A-Za-z0-9\_] ---
**Author:** ![dparkmit](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/d/cdc98d/32.png) [@dparkmit](https://www.ruby-forum.com/u/dparkmit)\ **Post date:** [April 23, 2007, 7:58am UTC](https://www.ruby-forum.com/t/is-there-a-bug-in-the-activerecord-method-validates-format-o/92672/4 "2007-04-23T07:58:35Z")
Thanks for the info! I think that there must be a problem with the UnitTest code then. I tried to force all RoR environments to use a $KCODE value of “n” by setting that value in the environment.rb file. But the tests still fail when I use [\w@.]. In my particular case, it’s not a big deal. But it is sort of annoying to think that the test harness may have some bugs in it since it’s supposed to help you catch bugs in your own code! ---
**Author:** ![dparkmit](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/d/cdc98d/32.png) [@dparkmit](https://www.ruby-forum.com/u/dparkmit)\ **Post date:** [April 23, 2007, 4:21am UTC](https://www.ruby-forum.com/t/is-there-a-bug-in-the-activerecord-method-validates-format-o/92672/5 "2007-04-23T04:21:17Z")
Not when $KCODE == “UTF8”, your tests must be running in a different environment somehow… irb(main):001:0\> $KCODE = “u” =\> “u” irb(main):002:0\> (0…255).select { |c| c.chr =~ /[\w@.]/ }.map { |c| c.chr } =\> [".", “0”, “1”, “2”, “3”, “4”, “5”, “6”, “7”, “8”, “9”, “@”, “A”, “B”, “C”, “D”, “E”, “F”, “G”, “H”, “I”, “J”, “K”, “L”, “M”, “N”, “O”, “P”, “Q”, “R”, “S”, “T”, “U”, “V”, “W”, “X”, “Y”, “Z”, “\_”, “a”, “b”, “c”, “d”, “e”, “f”, “g”, “h”, “i”, “j”, “k”, “l”, “m”, “n”, “o”, “p”, “q”, “r”, “s”, “t”, “u”, “v”, “w”, “x”, “y”, “z”, “\302”, “\303”, “\304”, “\305”, “\306”, “\307”, “\310”, “\311”, “\312”, “\313”, “\314”, “\315”, “\316”, “\317”, “\320”, “\321”, “\322”, “\323”, “\324”, “\325”, “\326”, “\327”, “\330”, “\331”, “\332”, “\333”, “\334”, “\335”, “\336”, “\337”, “\340”, “\341”, “\342”, “\343”, “\344”, “\345”, “\346”, “\347”, “\350”, “\351”, “\352”, “\353”, “\354”, “\355”, “\356”, “\357”, “\360”, “\361”, “\362”, “\363”, “\364”, “\365”, “\366”, “\367”, “\370”, “\371”, “\372”, “\373”, “\374”, “\375”]