ElasticResults

Store the results of your test runs in Elasticsearch via custom formatters.

Installation

Add this line to your application's Gemfile:

gem 'elastic_results'

And then execute:

$ bundle

Or install it yourself as:

$ gem install elastic_results

Cucumber Usage

Add this line to your features/support/env.rb file:

require 'elastic_results/cucumber'

Tell cucumber to use elastic_results:

bundle exec cucumber  --format ElasticResults::Cucumber::Formatter

Tell cucumber to use elastic_results but still give normal output:

bundle exec cucumber --format pretty --format ElasticResults::Cucumber::Formatter -o /dev/null

Make this the default by adding the following to config/cucumber.yml

default: --format pretty --format ElasticResults::Cucumber::Formatter -o /dev/null

Cucumber 1.X

Elastic results supports older versions of cucumber through the legacy formatter.

Add this line to your features/support/env.rb file:

require 'elastic_results/legacy_cucumber'

Tell cucumber to use elastic_results:

bundle exec cucumber  --format ElasticResults::Cucumber::LegacyFormatter

RSpec Usage

Add this line to your spec_helper.rb file:

require 'elastic_results/rspec'

Tell rspec to use elastic_results:

bundle exec rspec -r 'elastic_results/rspec'  --format ElasticResults::RSpec::Formatter

Tell rspec to use elastic_results but still give normal output:

bundle exec rspec -r 'elastic_results/rspec'  --format ElasticResults::RSpec::Formatter --format progress

Make this the default by adding the following to spec_helper.rb

RSpec.configure do |config|
  config.add_formatter 'ElasticResults::RSpec::Formatter'
  config.add_formatter 'progress'
  # the rest of your rspec configuration
end

Then run rspec as normal.

SimpleCov Usage

Add this lines to your spec_helper.rb file:

require 'elastic_results/simplecov'
SimpleCov.formatters = [SimpleCov::Formatter::HTMLFormatter, ElasticResults::SimpleCov::Formatter]

Then run simplecov as normal.

Configuration

elastic_results exposes several configuration points that can be set via environment variables or by accssing them on the ElasticResults module:

ENV	ElasticResults	Default	Notes
ES_URL	es_url	http://localhost	The URL where elasticsearch can be reached
ES_INDEX_RESULT	es_index_result	test_results-YYYY-MM-DD	The index to post test results to. The default uses a prefix of test_results and a suffix of today's date
ES_TYPE_RESULT	es_type_result	test_result	The data type to use for test results.
ES_INDEX_COVERAGE	es_index_coverage	coverage-YYYY-MM-DD	The index to post coverage data to. The default uses a prefix of coverage and a suffix of today's date
ES_TYPE_COVERAGE	es_type_coverage	simplecov	The data type to use for coverage data.
ES_LOG	es_log	ENV['DEBUG']	Log calls to elasticsearch to STDOUT? (useful for debugging elastic_results)
SUITE_NAME	suite_name	Dir.pwd	The name of your suite. If not given, it will be guessed using the name of the current folder when you launch your tests.
SUITE_TYPE	suite_type	integration	The type of suite you're running. i.e. unit, integration, regression, etc
GIT_COMMIT		Output of: git rev-parse HEAD	The git revision being tested. Jenkins will set this in CI, otherwise it's pulled from git.
GIT_URL		Output of: git config --get remote.origin.url	The url to the git repo being used. Jenkins will set this in CI, otherwise it's pulled from git.
GIT_BRANCH		Output of: git rev-parse --abbrev-ref HEAD	The git branch being used. Jenkins will set this in CI, otherwise it's pulled from git.
TEAM_NAME	team_name	???	The name of your team. This makes it possible to slice your results easier.
BUILD_URL			The url to your Jenkins build if any. Set by Jenkins.
NODE_NAME		Socket.gethostname	The node that ran the test. Set by Jenkins, or pulled from the machine.
BUILD_NUMBER		MMDDYYHHMMSSUU	The jenkins build number. If not present, a fake build number will be built using the current date/time down to the millisecond.
JOB_NAME			The jenkins job name.
BUILD_TAG			The jenkins build tag.
TEST_ENV/RAILS_ENV			The enivironment the tests were run against. Will use TEST_ENV if present, otherwise will use RAILS_ENV
	use_unsafe_index	true if WebMock is defined, otherwise false	If true, validation of the SSL cert for the elasticsearch host will be skipped. Useful when your testing tools break things.

Example config block:

ElasticResults.team_name = 'MyTeam - MySubTeam'
ElasticResults.suite_type = 'unit'
ElasticResults.kibana_url = 'https://kibana.mycompany.com'
ElasticResults.es_url =  %w(https://es1.mycompany.net:9200 https://es1.mycompany.net:9200).sample

Development

After checking out the repo, run bin/setup to install dependencies. Then, run rake false to run the tests. You can also run bin/console for an interactive prompt that will allow you to experiment.

To install this gem onto your local machine, run bundle exec rake install. To release a new version, update the version number in version.rb, and then run bundle exec rake release, which will create a git tag for the version, push git commits and tags, and push the .gem file to rubygems.org.

Contributing

Bug reports and pull requests are welcome on GitHub at https://github.com/covermymeds/elastic_results