On 2010-08-11 12:29 PM, Craig Demyanovich wrote:
On Wed, Aug 11, 2010 at 12:53 PM, Nadal <node.j...@gmail.com
<mailto:node.j...@gmail.com>> wrote:
I posted a question on shoulda forum. It is under moderation so I
don't have a link yet.
To the people who do not use shoulda: How would you write a test for
above case. I assume it would require set up a subject, set nil value,
and then see if there is an error message. I still think that for such
cases should macro is succint and does it job well unless I am missing
some new feature of rspec2.
I don't yet use RSpec 2, but I don't think you're missing anything.
Shoulda may very well be a good fit for these kinds of examples.
Since you asked, here's how I currently write examples like the ones
you posted.
require 'spec_helper'
describe User, "being valid" do
it "requires an email" do
user = User.new(:email => nil)
user.should_not be_valid
user.should have(1).error_on(:email)
end
it "requires a name" do
user = User.new(:name => nil)
user.should_not be_valid
user.should have(1).error_on(:name)
end
end
Regards,
Craig
That's exactly what I used to do, also, until I switched to Remarkable.
Now it's:
before :each do
@user = User.create(valid_user_hash)
end
should_validate_presence_of :first_name
should_validate_presence_of :last_name
should_validate_presence_of :email
should_validate_uniqueness_of :email, :case_sensitive => false
should_validate_format_of_email
One thing I have come to appreciate about this is being able to copy the
should_* statements to my model and, using TextMate's column editing
mode, remove the 'should_' and add an 's' at the end of 'validate' and
I'm done. For all of the validations that Remarkable supports, the same
options are supported as well. That's actually why I switch from Shoulda
to Remarkable a year or so ago (things might have changed in Shoulda by
now, though): Not all validation options were supported in Shoulda.
Peace,
Phillip
_______________________________________________
rspec-users mailing list
rspec-users@rubyforge.org
http://rubyforge.org/mailman/listinfo/rspec-users
