gitlab-org--gitlab-foss/.gitlab-ci.yml

image: "dev.gitlab.org:5005/gitlab/gitlab-build-images:ruby-2.3-git-2.7-phantomjs-2.1"

cache:
  key: "ruby-231"
  paths:
  - vendor/ruby

variables:
  MYSQL_ALLOW_EMPTY_PASSWORD: "1"
  # retry tests only in CI environment
  RSPEC_RETRY_RETRY_COUNT: "3"
  RAILS_ENV: "test"
  SIMPLECOV: "true"
  SETUP_DB: "true"
  USE_BUNDLE_INSTALL: "true"
  GIT_DEPTH: "20"
  PHANTOMJS_VERSION: "2.1.1"

before_script:
  - source ./scripts/prepare_build.sh
  - cp config/gitlab.yml.example config/gitlab.yml
  - bundle --version
  - '[ "$USE_BUNDLE_INSTALL" != "true" ] || retry bundle install --without postgres production --jobs $(nproc) "${FLAGS[@]}"'
  - retry gem install knapsack
  - '[ "$SETUP_DB" != "true" ] || bundle exec rake db:drop db:create db:schema:load db:migrate add_limits_mysql'

stages:
- prepare
- test
- post-test
- pages

.use-db: &use-db
  services:
    - mysql:latest
    - redis:alpine

migration paths:
  stage: test
  <<: *use-db
  variables:
    SETUP_DB: "false"
  script:
    - git fetch origin v8.5.9
    - git checkout -f FETCH_HEAD
    - cp config/resque.yml.example config/resque.yml
    - sed -i 's/localhost/redis/g' config/resque.yml
    - bundle install --without postgres production --jobs $(nproc) ${FLAGS[@]} --retry=3
    - rake db:drop db:create db:schema:load db:seed_fu
    - git checkout $CI_BUILD_REF
    - source scripts/prepare_build.sh
    - rake db:migrate
Move the images over to dev.gitlab.org 2016-09-30 11:45:03 +00:00			`image: "dev.gitlab.org:5005/gitlab/gitlab-build-images:ruby-2.3-git-2.7-phantomjs-2.1"`
More docker stuff in gitlab-ci.yml 2016-01-22 13:45:34 +00:00
Use caching, instead of haxxy /cache 2016-02-07 00:57:04 +00:00			`cache:`
Update ruby 2.3.1 2016-08-12 12:41:27 +00:00			`key: "ruby-231"`
Use caching, instead of haxxy /cache 2016-02-07 00:57:04 +00:00			`paths:`
Cache only apt and ruby from vendor Since introduction of gitignore the vendor folder contains also gitignores which affects detection when to update a cache. We explicitly cache only apt and ruby folders. 2016-06-12 21:05:19 +00:00			`- vendor/ruby`
Use caching, instead of haxxy /cache 2016-02-07 00:57:04 +00:00
More docker stuff in gitlab-ci.yml 2016-01-22 13:45:34 +00:00			`variables:`
			`MYSQL_ALLOW_EMPTY_PASSWORD: "1"`
Retry failed tests 2015-11-12 09:52:20 +00:00			`# retry tests only in CI environment`
			`RSPEC_RETRY_RETRY_COUNT: "3"`
Use build stage 2016-05-22 01:41:17 +00:00			`RAILS_ENV: "test"`
			`SIMPLECOV: "true"`
Add development environment database setup CI test 2016-09-22 10:51:06 +00:00			`SETUP_DB: "true"`
USE_BUNDLE_INSTALL 2016-06-03 21:03:58 +00:00			`USE_BUNDLE_INSTALL: "true"`
Use GIT_DEPTH when running builds 2016-06-17 08:59:48 +00:00			`GIT_DEPTH: "20"`
Install latest stable phantomjs 2016-08-09 18:25:11 +00:00			`PHANTOMJS_VERSION: "2.1.1"`
More docker stuff in gitlab-ci.yml 2016-01-22 13:45:34 +00:00
Refactor CI script Signed-off-by: Dmitriy Zaporozhets <dmitriy.zaporozhets@gmail.com> 2015-06-04 14:54:32 +00:00			`before_script:`
Speed-up docker based builds Enable to: - Use APT cache - Use Bundler cache - Use PhantomJS cache 2016-01-22 17:16:59 +00:00			`- source ./scripts/prepare_build.sh`
added ci yaml 2015-06-03 20:05:33 +00:00			`- cp config/gitlab.yml.example config/gitlab.yml`
Test 2016-05-22 05:31:38 +00:00			`- bundle --version`
Preserve knapsack state 2016-06-03 20:54:25 +00:00			`- '[ "$USE_BUNDLE_INSTALL" != "true" ] \|\| retry bundle install --without postgres production --jobs $(nproc) "${FLAGS[@]}"'`
Install knapsack 2016-06-07 08:45:02 +00:00			`- retry gem install knapsack`
Fix column limit on mysql 2016-10-05 23:05:16 +00:00			`- '[ "$SETUP_DB" != "true" ] \|\| bundle exec rake db:drop db:create db:schema:load db:migrate add_limits_mysql'`
updated gitlab-ci script to include builds URL 2016-02-24 21:53:13 +00:00
Try to use knapsack 2016-05-22 01:17:15 +00:00			`stages:`
Use knapsack 1.11.0 2016-06-03 15:15:00 +00:00			`- prepare`
Try to use knapsack 2016-05-22 01:17:15 +00:00			`- test`
Merge knapsack reports and upload them to external server 2016-06-03 15:35:00 +00:00			`- post-test`
Update configuration of SimpleCov 2016-06-30 17:04:00 +00:00			`- pages`
updated gitlab-ci script to include builds URL 2016-02-24 21:53:13 +00:00
Use YAML inheritance to DRY the .gitlab-ci.yml 2016-07-20 06:35:29 +00:00			`.use-db: &use-db`
Speed improvement for builds without DB Now the builds which do not use the DB or Redis won't pull the images from docker, and won't migrate the DB. This _should_ improve the build times slightly but also create a cleaner trace. 2016-06-29 14:08:20 +00:00			`services:`
			`- mysql:latest`
			`- redis:alpine`
Use YAML inheritance to DRY the .gitlab-ci.yml 2016-07-20 06:35:29 +00:00
Have CI test migration paths This commit adds a job to the pipeline to test if migration on seed data works. Even though the seed data is not perfect, it does give more guarentees that this will succeed in the real world too. seed_fu is used instead of dev:setup to save some time, as they both generate the data in the same way, this should be fine. 2016-06-23 10:35:04 +00:00			`migration paths:`
			`stage: test`
			`<<: *use-db`
Improve redis config tasks for migration paths job 2016-10-25 10:39:06 +00:00			`variables:`
			`SETUP_DB: "false"`
Have CI test migration paths This commit adds a job to the pipeline to test if migration on seed data works. Even though the seed data is not perfect, it does give more guarentees that this will succeed in the real world too. seed_fu is used instead of dev:setup to save some time, as they both generate the data in the same way, this should be fine. 2016-06-23 10:35:04 +00:00			`script:`
Fetch only one revision 2016-11-21 21:57:06 +00:00			`- git fetch origin v8.5.9`
			`- git checkout -f FETCH_HEAD`
Improve redis config tasks for migration paths job 2016-10-25 10:39:06 +00:00			`- cp config/resque.yml.example config/resque.yml`
			`- sed -i 's/localhost/redis/g' config/resque.yml`
Fix migration paths tests 2016-11-21 21:51:38 +00:00			`- bundle install --without postgres production --jobs $(nproc) ${FLAGS[@]} --retry=3`
Have CI test migration paths This commit adds a job to the pipeline to test if migration on seed data works. Even though the seed data is not perfect, it does give more guarentees that this will succeed in the real world too. seed_fu is used instead of dev:setup to save some time, as they both generate the data in the same way, this should be fine. 2016-06-23 10:35:04 +00:00			`- rake db:drop db:create db:schema:load db:seed_fu`
			`- git checkout $CI_BUILD_REF`
Improve redis config tasks for migration paths job 2016-10-25 10:39:06 +00:00			`- source scripts/prepare_build.sh`
Have CI test migration paths This commit adds a job to the pipeline to test if migration on seed data works. Even though the seed data is not perfect, it does give more guarentees that this will succeed in the real world too. seed_fu is used instead of dev:setup to save some time, as they both generate the data in the same way, this should be fine. 2016-06-23 10:35:04 +00:00			`- rake db:migrate`