/ruby-api-framework

An example ruby API test framework

Primary LanguageRuby

Ruby API Automation Example

Repository: git@github.com:isonic1/ruby-api-framework.git

Install Dependencies:

Mac: >>>>>> Never EVER EVER sudo install anything!!! <<<<<<

  1. Install RVM and Ruby: \curl -sSL https://get.rvm.io | bash -s stable –ruby

    a. Run: rvm list

    b. Run: rvm --default use 2.2.? to set the version as default.

  2. Install bundler: gem install bundler

  3. Clone the repo: git clone git@github.com:isonic1/ruby-api-framework.git

  4. Go into repo: cd ruby-api-framework

  5. Run: ruby dependencies.rb

  6. The above bundle install should install all framework dependencies.

Windows:

  1. Download and install Ruby 64-bit or 32-bit

    a. Make note of the installation/destination path! e.g. c:\Ruby22-x64\bin

    b. Goto Control Panel\System and Security\System > Advanced system > Edit Variables...

    c. Under "User variables for your_user_name"

    d. Add path for ruby location. e.g. c:\Ruby22-x64\bin

    e. Click OK

  2. Download and install Ruby Developer Kit 32-bit or 64-bit

    a. Extract the files to c:\ruby-dev-kit

    b. Open a power shell terminal and run as administrator.

    c. goto cd c:\ruby-dev-kit

    d. Run: ruby dk.rb init

    e. Run: ruby dk.rb install

    f. Test the installation by running: gem install json --platform=ruby --source http://rubygems.org in CMD. You should see "1 gem installed" if all goes well.

    g. The above json gem should install successfully.

  3. Install bundler: gem install bundler

  4. Clone the repo: git clone git@github.com:isonic1/ruby-api-framework.git

  5. Go into repo: cd ruby-api-framework

  6. Run: ruby dependencies.rb

  7. The above ruby script will install all framework dependencies.

Run a test:

  • Note: This example uses THIS EXAMPLE API. The POSTs, DELETE and PUTS just simulate these actions and do not work. So that is why you will see failures for the tests on these actions.
  • Look at the comments in each spec to give you more specifics on what things do and recommendations...

  1. Go into repo: cd ruby-api-framework

  2. Run all tests: rspec spec

  3. Run individual test: rspec spec/posts_spec.rb

  • You should see tests running after.

*** PREFERRED RUN METHOD ***

  1. Use the runner.rb script.
    • $ ruby runner.rb specs (list all specs)
    • $ ruby runner.rb (Run all specs)
    • $ ruby runner.rb posts (runs all posts tests)
    • $ ruby runner.rb parallel (runs all specs in parallel using parallel_rspec gem. Merges html reports into one afterward!)
    • $ ruby runner.rb failures (Reruns failures from previous test run via failures.txt)

Test Results:

  1. Html report: ruby-api-framework/output/rspec.html

  2. Json report: ruby-api-framework/output/rspec.json (You can use this json output to create your own reports or feed it into an existing reporting framework)

  3. When tests are run in parallel, by default parallel_rspec generates a report for every process. e.g. rspec.html ... rspec2.html etc... I created a script "merge_reports.rb" to combine all reports into one "rspec.html" after the tests complete.

  4. Rspec has a bug when using --color with adding ascii characters to the html report. I created a new html formatter to fix this issue. Additionally, I capture the full request path on every api call and print it inside the report on failures. I also modified the html snippet extractor to capture the entire test so you dont have to open the framework to see all the logic of what the test was doing.

    • See the .rspec, .rspec_parallel, and spec_helper.rb files for rspec settings.

Debugging with IRB: (Interactive Ruby REPL)

  • Use pry "binding.pry" in the code to put the framework into IRB with loaded classes and variables.
  • See spec/post_specs.rb for examples. Though this can be placed in any ruby file to set a break point at any point.

Adding a Test

  • Add new tests in the spec folder with _spec.rb. e.g. mytest_spec.rb
  • I like to use the it { expect() matcher } technique. Since you can have multiple assertions not tied to a single context. If one assertion fails then the others are still checked. Otherwise, if one expect (assertion) fails the entier test fails without ever checking the other assertions.

TODO:

  • Show examples on how to set an API Key.
  • Show examples on how to use OAUTH.
  • Show examples on how to add Headers.