An easy to use gem that provides datasets that can be used by your application and tests. The data is stored in yaml files.
In order to use data_magic you will have to inform the gem where it can find the yaml files. You can do this with the following code:
DataMagic.yml_directory = 'data/yml'
If you do not specify a directory the gem will default to using a directory named config/data.
After setting the directory you must load a file. This can be accomplished by calling the load method.
DataMagic.load 'filename.yml'
Another way to specify the file to load is to use a tag in a cucumber scenario. You tag should take the
form of @datamagic_FILENAME
where FILENAME
is replaced with the file you wish to load. For example,
if you add the tag @datamagic_foo
then the file foo.yml
will be loaded. If you want
to use the tags you simply have to add the following code in a hook:
Before do |scenario|
DataMagic.load_for_scenario(scenario)
end
If you do not specify a filename the gem will attempt to use a file named default.yml. If you are using this for testing you will more than likely want to call load before each test to load the proper data for the specific test, or use the namespaced keys method, detailed below.
Another option is to set an environment variable DATA_MAGIC_FILE. When this is set it will be used instead of the default.yml file.
The final thing to do is use the data. The gem has a data_for
method that will return the data for a specific key. The most common way to use this is to include the DataMagic module in a page-object and then populate a page with the data. Here's an example:
class MyPage
include PageObject
include DataMagic
...
def populate_page
populate_page_with data_for :my_page
end
end
Notice that I am including the module on line 3. On lin 8 I am calling the data_for method passing the key :my_page. The populate_page_with method is a part of the page-object gem.
To organize your data into namespaces, and load that data just in time for testing, use namespaced keys instead:
page.populate_page_with data_for "user_form/valid"
This will load user_form.yml
, and populate the page with the valid:
record therein.
Your data might look something like this:
my_page:
name: Cheezy
address: 123 Main Street
email: cheezy@example.com
pay_type: 'Credit card'
In order to access the data directly you can just call the method on the module like this:
page = MyPage.new
my_data = page.data_for :my_test
You can call one of many built-in methods in your yaml file to randomize the data. Here is an example of how you would randomize the above yaml:
my_page:
name: ~full_name
address: ~street_address
email: ~email_address
pay_type: ~randomize ['Credit card', 'Purchase order', 'Check']
Here is a list of the built-in methods:
built-in methods | built-in methods |
---|---|
first_name last_name | |
last_name | full_name |
name_prefix | name_suffix |
title | street_address(include_secondary=false) |
secondary_address | city |
state | state_abbr |
zip_code | country |
company_name | catch_phrase |
words(number = 3) | sentence(min_word_count = 4) |
sentences(sentence_count = 3) | paragraphs(paragraph_count = 3) |
characters(character_count = 255) | email_address(name = nil) |
domain_name | url |
user_name | |
phone_number | cell_phone |
randomize([]) | randomize(1..4) |
mask - #=num a=lower A=upper | |
today(format = '%D') | tomorrow(format = '%D') |
yesterday(format = '%D') | |
3.days_from_today(format = '%D') | 3.days_ago(format = '%D') |
month | month_abbr |
day_of_week | day_of_week_abbr |
sequential([]) | sequential(1..4) |
If you wish to add your own built-in methods you can simply pass a module to DataMagic and all of the methods will be available.
module MyData
def abc
'abc'
end
end
DataMagic.add_translator MyData # this line must go in the same file as the module
# can now use ~abc in my yml files
The rdocs for this project can be found at rubydoc.info.
To see the changes from release to release please look at the ChangeLog
This gem only works with Ruby >= 2.2. Use versions 1.1 or below if you have an older Ruby.
See http://github.com/cheezy/data_magic/issues
Please ensure all contributions contain proper tests.
- Fork it
- Create your feature branch (
git checkout -b my-new-feature
) - Commit your changes (
git commit -am 'Added some feature'
) - Push to the branch (
git push origin my-new-feature
) - Create new Pull Request
Copyright (c) 2012-2013 Jeffrey S. Morgan. See LICENSE for details.