2010-12-10 26 views
2

我是新的Ruby和Rspec。我正在写我的第一个RSpec测试,我认为我的代码不太好。但我不知道如何让它变得更好。DRS问题与Rspec我该如何解决它

在这个文件中,我将检查我的地址类。 first_name和last_name是相同的,但我有两个大块。我如何重构我的代码?什么是检查RegExp的好方法。

谢谢。

specify { Factory.build(:address).should be_valid } 


    ### first_name ### 

    it "should be invalid without an first_name" do 
    Factory.build(:address, :first_name => nil).should_not be_valid 
    end 


    context "first_name" do 

    it "should be invalid with more than 20 chars" do 
     Factory.build(:address, :first_name => "#{'b'*21}").should_not be_valid 
    end 

    it "should be invalid with less than 3 chars" do 
     Factory.build(:address, :first_name => "ll").should_not be_valid 
    end 

    it "should be valid with an valid first_name" do  
     valid_names.each do |name| 
     Factory.build(:address, :first_name => name).should be_valid 
     end 
    end 

    it "should be invalid with an invalid first_name" do 
     invalid_names.each do |name| 
     Factory.build(:address, :first_name => name).should_not be_valid 
     end 
    end  
    end 


    ### last_name ### 

    it "should be invalid without an last_name" do 
    Factory.build(:address, :last_name => nil).should_not be_valid 
    end 

    context "last_name" do 
    it "should be invalid with more than 20 chars" do 
     Factory.build(:address, :last_name => "#{'b'*21}").should_not be_valid 
    end 

    it "should be invalid with less than 3 chars" do 
     Factory.build(:address, :last_name => "ll").should_not be_valid 
    end 

    it "should be valid with an valid last_name" do  
     valid_names.each do |name| 
     Factory.build(:address, :last_name => name).should be_valid 
     end 
    end 

    it "should be invalid with an invalid last_name" do 
     invalid_names.each do |name| 
     Factory.build(:address, :last_name => name).should_not be_valid 
     end 
    end  
    end 
def valid_names  
    ["Kai","Ülück's","Schmeißtzs","Rald","Dr. Franzen","rolfes","Lars Michael","Öcück","Mark-Anthony"] 
    end 

    def invalid_names  
    ["-#+*32","   ","a& &lkdf","_-_.l##df"," aaadsa","M€lzer"] 
    end 

回答

1

所以这里的我有时会做这种事情的方式:

describe Address do 
    describe "validations" do 
    before do 
     @address = Factory(:address) 
    end 
    describe "#first_name" do 
     #prove that your factory is correct 
     it "should be valid" do 
     @address.should be_valid 
     end 
     it "should be less than 20 chars" do 
     @address.name = "0" * 20 
     @address.should_not be_valid 
     end 
     it "should be more than 3 chars" do 
     @address.name = "000" 
     @address.should_not be_valid 
     end 
    end 
    end 
end 
+1

我会用本发明的方法对于这一点,就像它在这里完成:https://github.com/rspec/rspec-expectations/blob/master/features/expectations/attribute_of_subject.feature – iain 2010-12-10 23:43:44

+0

谢谢您为您的帮助 – ThreeFingerMark 2010-12-13 22:37:19

1

记住,你的测试并不需要是真干。不要为此牺牲可读性。

将您的规格视为例子:哪些示例应该有效,哪些不应该?这也是测试正则表达式的线索:提供一些例子,通过和一些不。

对于验证,我做了一些自定义匹配器,可以找到here。例如:

describe Address do 
    it { should deny(:last_name).to_be(nil, "", "1", "br", "a& &lkdf","_-_.l##df", "lzer") } 
    it { should allow(:last_name).to_be("Kai","Ülück's","Schmeißtzs","Rald","Dr. Franzen","rolfes","Lars Michael","Öcück","Mark-Anthony") } 
end 
+0

感谢您的帮助 – ThreeFingerMark 2010-12-13 22:29:28