Learn R Programming

googleCloudRunner (version 0.5.0)

cr_deploy_packagetests: Deploy a cloudbuild.yml for R package tests and upload to Codecov

Description

This tests an R package each time you commit, and uploads the test coverage results to Codecov

Usage

cr_deploy_packagetests(
  steps = NULL,
  cloudbuild_file = "cloudbuild-tests.yml",
  env = c("NOT_CRAN=true"),
  test_script = NULL,
  codecov_script = NULL,
  codecov_token = "$_CODECOV_TOKEN",
  build_image = "gcr.io/gcer-public/packagetools:latest",
  create_trigger = c("file", "inline", "no"),
  trigger_repo = NULL,
  ...
)

Arguments

steps

extra steps to run before the cr_buildstep_packagetests steps run (such as decryption of auth files)

cloudbuild_file

The cloudbuild yaml file to write to. See create_trigger

env

Environment arguments to be set during the test script runs

test_script

The script that will call rcmdcheck to perform tests. If NULL a default script is used in system.file("r_buildsteps", "devtools_tests.R", package="googlecloudRunner")

codecov_script

The script that will call codecov to perform coverage. If NULL a default script is used in system.file("r_buildsteps", "codecov_tests.R", package="googleCloudRunner")

codecov_token

If using codecov, supply your codecov token here.

build_image

The docker image that will be used to run the R code for the test scripts

create_trigger

If creating a trigger, whether to create it from the cloudbuild_file or inline

trigger_repo

If not NULL, a cr_buildtrigger_repo where a buildtrigger will be created via cr_buildtrigger

...

Arguments passed on to cr_build_make

yaml

A Yaml object created from cr_build_yaml or a file location of a .yaml/.yml cloud build file

artifacts

Artifacts that may be built via cr_build_yaml_artifact

options

Options to pass to a Cloud Build

availableSecrets

Secret Manager objects built by cr_build_yaml_secrets

logsBucket

The gs:// location of a bucket to put logs in

source

A Source object specifying the location of the source files to build, usually created by cr_build_source

timeout

Amount of time that this build should be allowed to run, to second

images

A list of images to be pushed upon the successful completion of all build

substitutions

Substitutions data for `Build` resource

serviceAccount

service account email to be used for the build

Details

The trigger repository needs to hold an R package configured to do tests upon.

For GitHub, the repository will need to be linked to the project you are building within, via https://console.cloud.google.com/cloud-build/triggers/connect

If your tests need authentication details, add these via cr_buildstep_secret to the steps argument, which will prepend decrypting the authentication file before running the tests.

If you want codecov to ignore some files then also deploy a .covrignore file to your repository - see covr website at https://covr.r-lib.org/ for details.

See Also

Create your own custom deployment using cr_buildstep_packagetests which this function uses with some defaults

cr_buildstep_packagetests

Other Deployment functions: cr_deploy_docker_trigger(), cr_deploy_docker(), cr_deploy_pkgdown(), cr_deploy_run_website(), cr_deploy_run(), cr_deploy_r()

Examples

Run this code
# NOT RUN {
# create a local cloudbuild.yml file for packagetests
pd <- cr_deploy_packagetests(create_trigger = "no")
pd

# add a decryption step for an auth file
cr_deploy_packagetests(
  steps = cr_buildstep_secret("my_secret", "auth.json"),
  env = c("NOT_CRAN=true", "MY_AUTH_FILE=auth.json"),
  timeout = 1200,
  create_trigger = "no"
)


# creating a buildtrigger repo for trigger_repo
repo <- cr_buildtrigger_repo("MarkEdmondson1234/googleCloudRunner",
  branch = "master"
)
# }
# NOT RUN {
# will create the file in the repo, and point a buildtrigger at it
cr_deploy_packagetests(create_trigger = "file", trigger_repo = repo)


# will make an inline build within a buildtrigger
cr_deploy_packagetests(create_trigger = "inline", trigger_repo = repo)
# }
# NOT RUN {
unlink("cloudbuild-tests.yml")
# }

Run the code above in your browser using DataLab