Using the bitrise.yml from repository

It is possible to store the build configuration ( that is, your app’s bitrise.yml file) in your repository. The approach has its pros and cons; let’s go through how to do it and why you’d want to do it!

Configuring Bitrise to run a bitrise.yml from your repository ⚓

You can, of course, simply use the Bitrise CLI to locally run your builds. You just need to set up a bitrise.yml file and use bitrise run to run the workflow or workflows you want.

In the following examples, we’ll discuss the other way: we’ll define a “wrapper” configuration which will then run the build on Bitrise, using a bitrise.yml stored in your repository.

Setting up a wrapper configuration ⚓

A wrapper config means setting up a bitrise.yml configuration on bitrise.io: this configuration defines how the actual bitrise.yml file you want to use will be retrieved from the repository.

The wrapper configuration will contain an automatically triggered workflow, and another one that will tell Bitrise to continue the build from the repository. To make this work, we take advantage of a Script Step and the $BITRISE_TRIGGERED_WORKFLOW_ID environment variable. This environment variable is automatically set to the workflow that triggered the build: if we have a workflow with the same name defined in the bitrise.yml file that is stored in the repository, we can run that workflow.

In our example, we’ll use:

A trigger map where the ci workflow is triggered by code push to the master branch.
A workflow called run_from_repo to tell Bitrise to continue the build from the repository.
A workflow called ci with an after_run attribute set to the run_from_repo workflow.

In this example, a code push will trigger the ci workflow, which in turn triggers the run_from_repo workflow. The run_from_repo runs a Script Step which runs the bitrise run ci command in the repository. Let’s do it!

Open your app on bitrise.io.
Open the Workflow Editor and go the bitrise.yml tab.

Set up a trigger map that automatically triggers a specific workflow.

---
format_version: 1.4.0
default_step_lib_source: https://github.com/bitrise-io/bitrise-steplib.git
   
trigger_map:
- push_branch: "*"
  workflow: ci
- pull_request_target_branch: "*"
  workflow: ci

Set up the workflow that is triggered by the trigger map.

This workflow must have an after_run attribute that points to another workflow.
```
ci:
  after_run:
  - run_from_repo
```

Set up the workflow that is triggered by the after_run attribute.

This workflow must have:

A Git Clone Step to clone your repository.

A Script Step with the command bitrise run "${BITRISE_TRIGGERED_WORKFLOW_ID}.

workflows:
   run_from_repo:
     steps:
     - activate-ssh-key:
         run_if: '{{getenv "SSH_RSA_PRIVATE_KEY" | ne ""}}'
     - git-clone: {}
     - script:
         title: continue from repo
         inputs:
         - content: |-
             #!/bin/bash
             set -ex
             bitrise run "${BITRISE_TRIGGERED_WORKFLOW_ID}"

Of course, this only works if the bitrise.yml file in your repository does have a ci workflow. Let’s see the details of that!

Adding a bitrise.yml to your repo ⚓

Your bitrise.yml must have a workflow that is also defined in the wrapper config. To be more precise, it must contain all the workflows that are defined in the trigger map of the wrapper config, which is stored on bitrise.io.

This bitrise.yml file does not need its own trigger map: the previously set up wrapper configuration will take care of triggering the appropriate workflows.

Create a bitrise.yml file.

No need to build it from scratch: you can edit and download your app’s bitrise.yml configuration on the bitrise.yml tab of the Workflow Editor.
Define all the workflows that are present in the trigger map of the wrapper configuration.

For example, if your wrapper configuration’s trigger map triggers a ci workflow on code push and a deploy workflow on a pull request, you need to have both in your file, with the same name.
Add all the Steps you want.

Note that at his point your repository’s SSH key would have been activated and the repository itself cloned on Bitrise. Do not add these Steps to your workflows as they are unnecessary.
Commit the file to your repository.

Potential issues of storing the bitrise.yml in the repository ⚓

The approach of storing your bitrise.yml file in your repository, and running your builds based on that configuration has its disadvantages. We’ll discuss a few potential pitfalls of this.

The trigger map is better to be managed on bitrise.io.
You can’t change the build configuration of a specific commit.
You can’t edit the configuration in the online Workflow Editor.
Any Pull Request can run builds with its own custom configuration.

The trigger map is better to be managed on bitrise.io ⚓

You can store the trigger map in the bitrise.yml of your repository - but we don’t recommend it. You’ll lose the ability to ignore patterns. On bitrise.io, the trigger map is evaluated before the repository is cloned: this way, if you set it to ignore certain patterns in code pushes or pull requests, for example, then Bitrise won’t even start a build.

However, if you store the trigger map in your repository, the only way to check it is to clone the repository first. Even if you prepare your trigger_map in your repository to ignore patterns, bitrise.io will start a build to clone the repository and then aborts it based on the stored trigger map.

You can’t change the build configuration of a commit ⚓

Storing the bitrise.yml file in your repository has a significant drawback when you want to rebuild a specific commit, it will use the same bitrise.yml every time: the one stored in the repository for that git commit.

To change the configuration, you have to check out the related branch, change the bitrise.yml, commit the changes, and then push and start a new build.

If you store your build configuration on bitrise.io, you can always rebuild any commit with a new build configuration: the configuration is not tied to the commit or the state of the repository. You can simply change a parameter and hit “rebuild”, the new build will use the latest configuration from bitrise.io.

You can’t edit the configuration in the Workflow Editor on bitrise.io ⚓

The Workflow Editor on bitrise.io can only be used to visualize and edit the configuration stored on bitrise.io.

However, the offline workflow editor can still be used.

Pull Requests can run builds with any custom configuration ⚓

When someone sends a Pull Request they can modify the bitrise.yml in your repository any way they like it. This can force your builds to queue.