* bump apple utils
* Add App Clip support to iOS bundle ID registration
Detects App Clips via entitlements and registers App Clip bundle identifiers with parent app relationships. Updates related interfaces and logic to handle App Clip-specific requirements during bundle ID creation.
* Update CHANGELOG.md
* Update ensureAppExists.ts
## Why
The build-and-deploy-worker workflow had three separate jobs (setup, build-workers, deploy) with complex inter-job dependencies and redundant conditional logic. I think if we put it all in one job it should make stuff faster.
## How
Consolidated all three jobs into a single `build-and-deploy` job with conditional steps:
1. **Check**: Verify if tarballs for the current SHA exist in GCS (when deploying)
2. **Build**: Execute EAS workflow if needed (PR testing or cache miss)
3. **Upload**: Upload tarballs to GCS with SHA-based naming
4. **Deploy**: Copy tarballs to environment-specific names (when deploying)
Key simplifications:
- **Assumption**: Main branch always specifies an environment (staging/production), so `environment == ''` only occurs on PR builds
- **SHA-based caching**: Check cache by SHA regardless of branch - if the code is identical (same SHA), reuse existing tarballs
- **Clearer conditions**: Each step has a single clear purpose without overlapping branch-specific logic
I also added back building Linux tarball on Linux not to mix CPU architectures accidentally.
## Test Plan
[Build on PR](https://github.com/expo/eas-cli/actions/runs/20920297113/job/60104039095?pr=3303). [Deployment from a branch](https://github.com/expo/eas-cli/actions/runs/20921082109/job/60106516235).
Co-authored-by: Claude <[email protected]>
# Why
- Build Android and iOS worker tarballs together in a single EAS workflow run
- Avoid split builds across different machines and simplify uploads
# How
- Update the EAS workflow to build and upload both iOS and Android tarballs
- Update the GitHub Actions workflow to run the EAS workflow once and upload both artifacts
- Cancel the EAS workflow run when the GitHub Actions job is canceled
# Test plan
- Verified auto-cancel behavior: https://github.com/expo/eas-cli/actions/runs/20849238272/job/59900072125
# Why
- Canceling the GitHub Actions workflow should cancel the underlying EAS build/workflow runs to avoid wasting resources.
# How
- Start the iOS build and Android workflow with `--no-wait` and capture their IDs from JSON.
- Poll the EAS build/workflow statuses to determine success/failure.
- Persist IDs to `GITHUB_ENV` and cancel them in a `if: cancelled()` step.
# Test Plan
https://github.com/expo/eas-cli/actions/runs/20859556010/job/59935355509
* Handle missing Apple metadata gracefully in tasks
Updated Apple metadata tasks to skip operations when app version or info is not available, instead of throwing errors. This improves support for cases where the app or version has not been set up yet, particularly for metadata:pull. Tests have been updated to reflect the new behavior.
* Update CHANGELOG.md
This PR adds system tests that run after staging deployment and gate production deployment.
## Changes
- Add `system-tests.yml` workflow (can be dispatched manually or called from worker.yml)
- Add EAS workflow fixture for Android build
- Gate production deployment on system tests passing
- Run iOS build (`eas build`) and Android build (`eas workflow:run`) concurrently
## Testing
Once merged, the workflow can be manually dispatched to test, or it will run automatically on the next push to main that touches worker files.
<!-- If this PR requires a changelog entry, add it by commenting the PR with the command `/changelog-entry [breaking-change|new-feature|bug-fix|chore] [message]`. -->
<!-- You can skip the changelog check by labeling the PR with "no changelog". -->
# Why
Gives us more concurrencies for building worker (which is not a huge problem, but may be nice). Also, IF we botched worker staging, we won't break the whole pipeline (system tests won't let worker deploy to production so worker in production will still work and let us rebuild worker).
# How
Changed app ID and token to expo.dev ones.
# Test Plan
CI ran ok.
# Why
The worker integration tests were slow and required platform-specific CI runners (macOS for iOS, Linux for Android) because they attempted to build real React Native apps. This made the CI pipeline complex, slow, and resource-intensive.
# How
- **Replace full-build tests with fast mocked tests**: Instead of running actual Android/iOS builds, we now mock `@expo/build-tools` at the library boundary. This lets us test the worker's WebSocket communication, job dispatch, state sync, abort handling, and hanging worker detection without the overhead of real builds.
- **Add new `jobExecution.test.ts`**: A new integration test that verifies the worker accepts jobs, executes them (via mocked builders), and returns success/error responses correctly.
- **Update existing tests**: `abortBuild.test.ts`, `stateSync.test.ts`, and `possiblyHangingWorker.test.ts` now mock `@expo/build-tools` instead of the worker's internal `build.ts` module, allowing more of the real code path to execute.
- **Simplify CI**: Consolidated the separate `worker-android-checks` and `worker-ios-checks` jobs into a single `worker-checks` job running on `ubuntu-latest`. Removed dependencies on `direnv`, `setup-gcloud`, and platform-specific scripts.
- **Cleanup**: Deleted unused files including `__mocks__/build.ts`, macOS-specific Jest config, CI shell scripts, and the `direnv` internal action.
The build-build part is going to be tested in system tests running in CI on `main`.
# Test Plan
cd packages/worker
yarn test:integrationAll 19 tests pass in ~6 seconds (down from minutes with real builds).
```bash
cd packages/worker
yarn test:integration
```
All 19 tests pass in ~6 seconds (down from minutes with real builds).
## Why
The `worker.yml` GitHub Actions workflow was recently moved from a private repository to this public one. This means we lost access to private GitHub Actions runners and now have to wait in long queues for public `macos-latest` runners when building the iOS worker tarball.
## How
Instead of building the iOS worker tarball on GitHub Actions `macos-latest`, we now:
1. **Trigger an EAS Workflow** from GitHub Actions that builds the tarball on EAS's macOS runners
2. **Upload the tarball as an EAS artifact** using `eas/upload_artifact`
3. **Uploads the artifact to tarball bucket** from GitHub Actions
### Changes
- Added `packages/worker/app.json` - minimal Expo app config required for EAS
- Added `packages/worker/eas.json` - EAS configuration
- Added `packages/worker/.eas/workflows/build-worker.yml` - EAS Workflow that builds the iOS worker tarball and uploads it as an artifact
- Updated `.github/workflows/build-and-deploy-worker.yml`:
- Split the matrix build into separate `build-android` and `build-ios` jobs
- `build-ios` now runs on `ubuntu-latest` and triggers the EAS Workflow using `eas workflow:run --ref --wait --json`
- Extracts the artifact download URL from the workflow output and streams it directly to GCS
## Test Plan
https://github.com/expo/eas-cli/actions/runs/20796579975 Don't mind the CI failure on the PR, I dispatched the workflow manually later.
<!-- If this PR requires a changelog entry, add it by commenting the PR with the command `/changelog-entry [breaking-change|new-feature|bug-fix|chore] [message]`. -->
<!-- You can skip the changelog check by labeling the PR with "no changelog". -->
# Why
We want to streamline DX of the worker part of EAS. Part of it is colocating all code in one repository.
# How
Copied code over to `eas-cli`. Adjusted code — each commit represents a single piece of work towards adjusting code to this repository. The main changes are around removing dependency on `turtle-v2` internal libraries.
One thing to figure out still is deployment process. I've added `build-and-deploy-worker` which is capable of deploying to staging and production. The question is how to automate it. In `turtle-v2` we have system tests. Should we just copy them over to here? (Don't like that.) Should we streamline system tests now and instead of having this weird `turtle-v2-example` repo, create project on the go? Should we deploy worker manually for now?
I think I'd rather deploy manually for now?
# Test Plan
Deployed worker to staging. Ran system tests. They succeeded!! https://github.com/expo/turtle-v2/actions/runs/20042956545/job/57535011098
# Why
The CLI does not validate the project name and slug before creating the project on the server, leading to unnecessary GraphQL errors.
# How
Copy the server's self-contained validation code to the CLI. Eventually, this code should move to `eas-build` repo for use by both CLI and server. Add code to run the same validation checks that the server GraphQL resolver uses, and add a method to strip invalid characters from the project name before suggesting it at the prompt.
# Test Plan
The server's validation unit tests have been added here as well.
# Why
The command `eas credentials` does not provide a way to exit other than control-C, if no platform flag is passed in.
# How
Add an "Exit" choice to the platform prompt.
# Test Plan
Tested locally.
<!-- If this PR requires a changelog entry, add it by commenting the PR with the command `/changelog-entry [breaking-change|new-feature|bug-fix|chore] [message]`. -->
<!-- You can skip the changelog check by labeling the PR with "no changelog". -->
# Why
Node 24 is in LTS, it's time to officially support it and test for it.
# How
Add 24 to CI test matrix.
# Test Plan
CI
# Why
GraphQL test started failing today after a recent server change in updates GraphQL queries. Failed test run at https://github.com/expo/eas-cli/actions/runs/19529909252
# How
- Updated GraphQL generated code
- Fixed a TS type error in one test
# Test Plan
- CI should pass
# Why
In `workflow:status`, the argument name for the passed-in workflow run ID is incorrect. Instead of showing the status of that run, it shows the dialog to select from recent runs.
# How
Fixed the argument name.
# Test Plan
Works when testing locally.
# Why
Shows the status of a workflow run that has already started.
If the run is in progress, and the `--wait` flag is set, the command will show periodic updates on running jobs, the current job steps, and recent log lines, similarly to `eas workflow:run --wait`.
Otherwise, the command immediately returns with information on the run result, exactly as `eas workflow:run --wait` will show at the end of a run. If the run failed, the failed jobs, failed steps, and last few log lines are shown.
```bash
USAGE
$ eas workflow:status [WORKFLOW_RUN_ID] [--non-interactive] [--wait] [--json]
ARGUMENTS
WORKFLOW_RUN_ID A workflow run ID.
FLAGS
--json Enable JSON output, non-JSON messages will be printed to stderr.
--non-interactive Run the command in non-interactive mode.
--[no-]wait Wait for workflow run to complete. Defaults to false.
DESCRIPTION
show the status of an existing workflow run. If no run ID is provided, you will be prompted to select from recent
workflow runs for the current project.
FLAG DESCRIPTIONS
--[no-]wait Wait for workflow run to complete. Defaults to false.
Exit codes: 0 = success, 11 = failure, 12 = canceled, 13 = wait aborted.
```
# How
Leverages utility code already written for `eas workflow:run` and `eas workflow:logs`.
# Test Plan
- Tested on running and completed workflow runs.
- CI should pass.
# Why
The manual release workflow is broken after #3237, since the local plugin version is no longer hardcoded, but tied to the `@expo/eas-build-job` dependency.
# How
Remove this step from the workflow.
# Test Plan
Workflow should succeed in dry run mode.
# Why
Provide a way to move the "latest-eas-build" version tag added in the EAS CLI release process. After https://github.com/expo/universe/pull/22530, this is the EAS CLI version that will be used by EAS workflows.
# Test Plan
Test with different version inputs, ensure failures are handled correctly.
# Why
Currently, the CLI is using an old version of `eas-cli-local-build-plugin` for builds with `eas build --local`. This is not ideal, as this means that local builds are not using the most recent `build-tools` and other packages.
# How
- Modify `local.ts` so that the version of the plugin matches the version of `@expo/eas-build-job` used by the CLI
- Add some missing environment variables that are required in the most recent releases of `@expo/build-tools`
# Test Plan
- Testing with local builds on my dev machine
- CI should pass
# Why
Customers would like to wait for a workflow run to complete, and get JSON output.
# How
Remove the restriction on --wait and --json flags together, and ensure that when --json is enabled, all messages other than the final JSON result will go to stderr.
# Test Plan
- Tested locally by running workflows
- CI should pass
# Why
If a user executes `eas workflow:run --wait`, it would be desirable to show status updates for the individual jobs and steps in the workflow while the workflow run is in progress.
# How
Added code that leverages some of the same utilities used by `workflow:logs`.
During the run, the logs for the various workflow jobs are retrieved and parsed, and the progress of each job is shown in the text of a spinner, showing the current step and the last few lines of the step's logs.
If the workflow run fails, the output will show jobs that failed, along with the failed step for each, and any logs for the failed steps.
Other changes:
- Bumped the "ora" package to the latest version, to pick up bug fixes
- Removed deprecated Node 18 from test runs
- Refactored the code in `workflow:run` that parses inputs, moving that code to a file in `commandUtils`
- Modified the Zod type for inputs to align with server side changes in https://github.com/expo/universe/pull/21950
# Test Plan
- Testing with different workflows
- CI should pass
# Why
Build and repack jobs do not have logs at the same URL as other workflow jobs, so no logs were showing for these job types.
# How
- Update BuildFragment to include the log file URL
- Check for build jobs, and add the correct queries and fetch calls for those
- Update and fix tests
- Bump `@expo/eas-build-job` to 1.0.239 to fix TS error in environment type
# Test Plan
- Tested with various workflow runs, including ones with custom, build, repack, and update jobs.
- CI should pass.
* incorporating feedback
* small cleanup
* moved some files around
* moved tests
* remove unused line
* tweak one line
* unit tests for all the helper functions
* more tests pass
* last batch of tests for now
* update changelog
* fix changelog
* move src/new to src/commandUtils/new
* pr feedback
* update changelog
* update path in copy-new-templates command
* fix test import
* Update packages/eas-cli/package.json
Co-authored-by: Douglas Lowder <[email protected]>
* feat: add --package-manager flag to project:new command
- Add --package-manager (-p) flag with options: npm, yarn, pnpm, bun
- Default to npm when flag not provided
- Remove promptForPackageManagerAsync call
- Add corresponding tests
* feat: add --no-install flag to skip dependency installation
- Add --no-install flag to skip installing dependencies
- Update README with installation instructions when --no-install is used
- Pass skipInstall flag through to updateReadmeAsync
- Add corresponding tests
* feat: add hideOutput option to suppress dependency installation logs
- Add hideOutput parameter to runCommandAsync to use stdio: 'ignore'
- Update installDependenciesAsync to hide npm/yarn/pnpm/bun output
- Show only spinner during installation instead of verbose logs
- Apply hideOutput to all dependency installation steps
- Update tests to reflect flag definition changes
* revert: remove --no-install flag
Remove the --no-install flag and related skipInstall functionality as requested
* feat: remove prompts and use smart defaults for project name and directory
- Remove project name prompt, use smart fallback logic:
- Default: new-expo-project
- If exists: new-expo-project-USERNAME-DATE
- If exists: new-expo-project-SHORTID (using nanoid)
- Remove target directory prompt, use cwd/projectName as default
- Refactor naming to reflect generation vs prompting:
- Rename prompts.ts → configs.ts
- Combine generateDefaultProjectNameAsync + promptForProjectNameAsync → generateProjectNameAsync
- Rename promptForTargetDirectoryAsync → generateDirectoryAsync
- Rename promptForConfigsAsync → generateConfigsAsync
- Rename promptedConfigs → initialConfigs
- Add nanoid and fs-extra dependencies for smart defaults
- Update all tests to reflect new behavior (11 tests in configs.test.ts passing)
- All existing tests continue to pass
* refactor: simplify CLI args for project:new command
- Remove --name flag, always use smart generation logic
- Change --directory flag to positional 'path' argument
- Defaults to current working directory
- If no slashes: treated as subdirectory of cwd
- If relative path with slashes: resolved from cwd
- If absolute path: used as-is
- Update generateProjectNameAsync to remove optional name parameter
- Update generateDirectoryAsync to handle path logic
- Add 4 new tests for path handling scenarios
- Remove 1 test for name arg (no longer supported)
- All 17 tests passing
* more feedback
* one more big refactor
* more cleanup
* one last round of cleanup
* one last round of cleanup
* fix changelog
* fix tests
---------
Co-authored-by: Douglas Lowder <[email protected]>
# Why
The existing `workflow:create` command simply creates a very bare-bones workflow, that is not very useful to developers who need to know how to fully use workflows for managing builds, updates, and other EAS operations.
# How
- Define four different workflow templates, initialized as JSON objects.
- After the user selects which type of workflow (build, update, deploy, custom), sanity checks are made, EAS build and EAS update are configured if needed, and any missing build profiles are created in `eas.json`.
- Once the user has made their selections, the final workflow YAML is dynamically generated and written to disk with an explanatory header.
- After writing the workflow, next steps are provided for any further configuration that may be needed, especially if build credentials or app store credentials are needed.
In future, we can easily add more templates that use other workflow job types (fingerprint, repack, etc.).
# Test Plan
- Tested against different projects
- CI should pass
# Why
Add a `--ref` flag so that `eas workflow:run` can initiate a workflow run from an existing ref in the project's GitHub repository.
- Avoids the need to upload the entire project from a local machine
- Ensures that the workflow starts from a known reference point in the project's source control, with no unexpected local changes
# How
- If the ref flag is set, the run command will ensure that the ref is valid, and that the workflow can be used for that ref (using the new mutation for this from https://github.com/expo/universe/pull/22132)
- If the ref flag is set, the run command starts the run from that ref (using https://github.com/expo/universe/pull/20161)
- If the user chooses ref = HEAD, `git rev-parse` is called to convert that to a commit
- The flow is designed to work the same for running from GitHub ref as running from local, so inputs and waiting for results are supported
# Test Plan
- Tested in an existing project with workflows with and without inputs
- CI should pass
<!-- If this PR requires a changelog entry, add it by commenting the PR with the command `/changelog-entry [breaking-change|new-feature|bug-fix|chore] [message]`. -->
<!-- You can skip the changelog check by labeling the PR with "no changelog". -->
# Why
In https://github.com/expo/eas-build/pull/604 I removed `fingerprintSource` from metadata passed to a build (1). It was previously used to inform the build process of, well, the fingerprint source so it can optionally diff current fingerprint and fingerprint calculated by the `eas build` command.
In `eas build`, the fingerprint would get calculated and uploaded. Then it was included in the build mutation as `{ type: GCS, bucketKey }` (2). Worker expected that somewhere in the process of starting a job, something (`www` or `turtle-v2`) would swap out the persisted `fingerprintSource` to `{ type: URL, url }` (because worker can't download arbitrary GCS objects).
In https://github.com/expo/eas-build/pull/587 I changed the flow so that now worker does not expect any changes to the job or metadata and instead fetches the fingerprint sources straight from `www`.
This made `fingerprintSource` in `metadata` obsolete (see (1)). However, we still need to be able to send fingerprint sources information to `www` (see (2)).
# How
This pull request:
- bumps `@expo/eas-build-job` (removing `FingerprintSource` type)
- `FingerprintSource` gets replaced with `FingerprintSourceInput` generated by GraphQL schema
- this lets us remove `transformFingerprintSource` (it's being passed around in its final form)
# Test Plan
Set up my staging test app to use runtime version policy = fingerprint, installed and configured `expo-updates`, [ran a build](https://staging.expo.dev/accounts/sjchmiela/projects/staging-app/builds/7a44f4c3-e448-4657-a8f3-71e24a2a29c8).
I had a few `console.log`s in place:
<img width="943" height="779" alt="Zrzut ekranu 2025-09-30 o 17 39 14" src="https://github.com/user-attachments/assets/2b117a17-8c0a-4445-bf56-27f93a590f87" />
and the build has fingerprint:
<img width="1256" height="161" alt="Zrzut ekranu 2025-09-30 o 17 39 33" src="https://github.com/user-attachments/assets/b6133402-9910-42d2-b351-836ae9c10c95" />
Before I installed node dependencies, fingerprint was being set to `file:fingerprint` (probably from https://github.com/expo/expo/pull/26901, see https://github.com/expo/eas-cli/blob/b447194c8521270ba8b551296c4329c3edb7aa1c/packages/eas-cli/src/project/resolveRuntimeVersionAsync.ts#L88).
<!-- CURSOR_SUMMARY -->
---
> [!NOTE]
> Replaces internal FingerprintSource with GraphQL FingerprintSourceInput across build/update flows, removes transformFingerprintSource, adjusts metadata wiring, and bumps related packages.
>
> - **Build & GraphQL**:
> - Replace `Metadata.fingerprintSource` usage with GraphQL `FingerprintSourceInput`; update builders to accept `BuildMetadataInput` and pass pre-transformed metadata.
> - Remove `transformFingerprintSource` and associated imports/usages; simplify `transformMetadata` to accept `{ metadata, fingerprintSource }`.
> - Update types for fingerprint source across build paths (`computeAndMaybeUpload*`, `sendBuildRequestAsync`, Android/iOS build senders).
> - **Update/Publish**:
> - Stop transforming fingerprint source in `update publish`; pass through GraphQL `FingerprintSourceInput`.
> - Adjust runtime fingerprint computation/assembly to use `FingerprintSourceInput`.
> - **Fingerprint upload**:
> - `maybeUploadFingerprintAsync` now returns `FingerprintSourceInput`; use GraphQL enum casing (`Gcs`).
> - Mutation `createFingerprintAsync` accepts `FingerprintSourceInput`.
> - **Dependencies**:
> - Bump `@expo/eas-build-job` and `@expo/steps`; update yarn.lock accordingly.
>
> <sup>Written by [Cursor Bugbot](https://cursor.com/dashboard?tab=bugbot) for commit 5a61fdee68f3b0bbda043b8f3160c4c2f1ea3a71. This will update automatically on new commits. Configure [here](https://cursor.com/dashboard?tab=bugbot).</sup>
<!-- /CURSOR_SUMMARY -->
* incorporate pr feedback
* playing around with adding build
* create expo project
* clean up getAccountChoices
* update unit test
* removed some unused code
* wip
* working on test files
* fix previews, add more end comments
* small tweaks to tests
* update changelog
<!-- If this PR requires a changelog entry, add it by commenting the PR with the command `/changelog-entry [breaking-change|new-feature|bug-fix|chore] [message]`. -->
<!-- You can skip the changelog check by labeling the PR with "no changelog". -->
# Why
See https://github.com/expo/universe/pull/22108.
# How
- Updated GraphQL code.
- Adjusted code to the updated GraphQL code (no `isGlobal`).
- Updated types in `EnvironmentVariableMutation` so they don't re-implement what GraphQL is generating for us.
- Updated types in `env:push`.
# Test Plan
Adjusted tests.
* scaffolding out new command
* remove a newline
* incorporate pr feedback
* add unit test for package manager functions
* get rid of unnecessary lockfile git add
* remove unused test
# Why
Improve the validation of workflow files:
- Add client-side checks for valid syntax, and compatibility with the user's EAS build profiles
- Make validation usable with future improvements to `workflow:create`
# How
- New `validation.ts` in `commandUtils` with the various validation checks separated cleanly into different functions
- `workflow:validate` command refactored to use the new utility and wrap errors
- The new code reads the workflow schema from `<ExpoAPIBaseURL>/v2/workflows/schema`
- For testing purposes only, a local copy of the schema can be used by setting EXPO_TESTING_WORKFLOW_SCHEMA_PATH
- An existing check in `workflowFile.ts` has been modified to remove the requirement for a Github repository, as that is no longer required to run workflows
# Test Plan
- Tested against different workflow templates, and against workflows in a real project.
- Tested against the www server from the main branch running locally, to verify recent server side changes (e.g. https://github.com/expo/universe/pull/21950)
# Why
To allow CLI users to get artifacts of builds or workflow jobs directly using the CLI, and make the URLs available in JSON output also (for easier use by automation scripts).
# How
- Add artifact info to WorkflowJobFragment
- Add this info to the results of `workflow:view`
- For workflow jobs that are of type=BUILD, use the existing build query to get the build artifact URLs for the build and show those as well
- Make some improvements to the formatting of workflow job outputs
# Test Plan
- Tested against different EAS projects
- CI should pass
<!-- If this PR requires a changelog entry, add it by commenting the PR with the command `/changelog-entry [breaking-change|new-feature|bug-fix|chore] [message]`. -->
<!-- You can skip the changelog check by labeling the PR with "no changelog". -->
# Why
Unify packages versions across the Expo package ecosystem (Zod, eas-build-job)
# How
Copied Zod definitions from `www` to `run.ts` where we are parsing workflow to get inputs definitions.
# Test Plan
There are, fortunately, tests for this part of the code.
<!-- If this PR requires a changelog entry, add it by commenting the PR with the command `/changelog-entry [breaking-change|new-feature|bug-fix|chore] [message]`. -->
<!-- You can skip the changelog check by labeling the PR with "no changelog". -->
# Why
[exponent-internal.slack.com/archives/C011ZMXQ6SZ/p1755530483378679](https://exponent-internal.slack.com/archives/C011ZMXQ6SZ/p1755530483378679)
# How
Updated GraphQL code which removed `SubmissionAndroidTrack` enum (it's a scalar now). Adjusted code to pass track around as string.
Removed `AndroidReleaseTrack` from `eas-json`. Updated `eas.schema.json` to allow any string as `track`.
Technically, to support `track: string` we'll need to update `eas-json` in `www`. I think it's ok to merge this, publish releases and then update `eas-json` in `www`. Alternatively we could publish `eas-json` from this branch as prerelease, bump in `www` and then merge this. This would be more _correct_, but I think it's ok to have this 1 hour of unannounced discrepancy between `www` and `eas-cli` (users not being able to use custom tracks _yet_).
# Test Plan
Tested along with all the pull requests in the mix.
<img width="372" height="90" alt="Zrzut ekranu 2025-09-9 o 17 03 38" src="https://github.com/user-attachments/assets/0b260ad2-2379-4597-aa1c-dae8bf86446d" />
<img width="243" height="72" alt="Zrzut ekranu 2025-09-9 o 16 59 37" src="https://github.com/user-attachments/assets/4eef353d-ac7e-4b16-9413-a304a574b685" />
<img width="449" height="110" alt="Zrzut ekranu 2025-09-9 o 17 00 40" src="https://github.com/user-attachments/assets/58a44b7a-16bb-46b4-8d98-f233c196406a" />
<img width="366" height="85" alt="Zrzut ekranu 2025-09-9 o 17 00 35" src="https://github.com/user-attachments/assets/71013cd6-5f5e-4e38-951c-008642fbee8c" />
<!-- If this PR requires a changelog entry, add it by commenting the PR with the command `/changelog-entry [breaking-change|new-feature|bug-fix|chore] [message]`. -->
<!-- You can skip the changelog check by labeling the PR with "no changelog". -->
# Why
Preparation for merging the rest of `track`-related pull requests.
Currently, the only place fetching `track` from `www` for display is `submit:internal` command. There, we fetch the submission config to know what parameters to supply to `fastlane supply` in the Submit workflow job.
Currently, we're using a mapping of `{ PRODUCTION: production, INTERNAL: internal, ... }` to transform the GraphQL value (enum key) to config value ("enum value").
# How
Since we're planning to make `androidConfig.track` be a plain string (e.g. `internal`), we should fall back to the actual value if the mapping does not produce a match.
This currently should not change anything (`track` is guaranteed to be a GraphQL enum value). This is a preparation for landing of https://github.com/expo/universe/pull/21827 which (somewhat breakily, but we expect to be the only consumers of our GraphQL API) changes `track` to be any string.
This `track` value is being used [here](https://github.com/expo/universe/blob/2f143aa8d51821c7e1c306768787c46b18bdd544/server/www/src/data/entities/workflow/job/WorkflowSubmitJob.ts#L399-L403).
# Test Plan
This is currently a no-op. Tested along with all the pull requests in the mix.
<img width="372" height="90" alt="Zrzut ekranu 2025-09-9 o 17 03 38" src="https://github.com/user-attachments/assets/0b260ad2-2379-4597-aa1c-dae8bf86446d" />
<img width="243" height="72" alt="Zrzut ekranu 2025-09-9 o 16 59 37" src="https://github.com/user-attachments/assets/4eef353d-ac7e-4b16-9413-a304a574b685" />
<img width="449" height="110" alt="Zrzut ekranu 2025-09-9 o 17 00 40" src="https://github.com/user-attachments/assets/58a44b7a-16bb-46b4-8d98-f233c196406a" />
<img width="366" height="85" alt="Zrzut ekranu 2025-09-9 o 17 00 35" src="https://github.com/user-attachments/assets/71013cd6-5f5e-4e38-951c-008642fbee8c" />
# Why
It would be great if we could use the environment setting `DEBUG=eas:*` to enable debug logs when running the EAS CLI, similarly to the way `DEBUG=expo:*` is used in the Expo CLI.
This is a first step towards implementing that. Ideally, different commands and utility modules would set different tags for the debug log messages, but this will at least allow the debug logs to appear for existing `Log.debug()` calls.
# How
Modify `src/log.ts` to write debug logs if DEBUG is set to a regex that matches `eas:log:debug`.
# Test Plan
CI should pass.
# Why
Sync up the schema to reflect recent additions to the server GraphQL schema
# How
`yarn --cwd packages/eas-cli verify-graphql-code`, then inspect changes in schema and generated code.
# Test Plan
CI should pass
# Why
Make it easier for customers to use `eas update:rollback` in CI.
# How
Add description text on the underlying commands that do support the non-interactive flag.
# Why
The existing release trigger workflow does not obtain the GItHub history and tags, so that Lerna cannot correctly determine which package versions to bump.
# How
Set `fetch-depth: 0` in the checkout action, and do an explicit fetch of all tags.
# Test Plan
- After merge, a dry run of the release trigger should correctly bump only changed packages.
# Why
The new release GitHub workflow has an error because the release script needs to be run without prompts:
```
Checking if eas-cli-local-build-plugin upgrade is available
eas-cli-local-build-plugin is already on the latest version - 1.0.171
lerna notice cli v8.2.3
lerna info ci enabled
lerna info current version 16.17.3
lerna info Assuming all packages changed
Changes:
- eas-build-cache-provider: 16.4.2 => 16.17.4
- eas-cli: 16.17.3 => 16.17.4
- @expo/eas-json: 16.13.2 => 16.17.4
error Command failed with exit code 130.
? Are you sure you want to create these versions? (ynH) info Visit https://yarnpkg.com/en/docs/cli/run for documentation about this command.
Error: Process completed with exit code 130.
```
# How
- Added the `--yes` flag to the script commands
- Instead of failing when run locally, the script now executes the release flow in "dry run" mode, without committing, tagging, or pushing changes.
- Add workflow input for dry run mode (true by default, must be manually set to false to actually release)
- Get full Git history before running release script, as required for correct Lerna version check
# Test Plan
- Test locally
- Run the workflow in dry run mode after merging
- Attempt a release after this PR is merged
# Why
The current release script requires that the user executing the script have permission to push changes to main without any PR requirement or checks.
# How
To allow us to add PR requirements and checks, I have added a manual GitHub workflow so that the existing script is executed by the expo-bot user, which is exempted from PR requirements.
I have also made a few adjustments to the `release.yml` workflow (remove incorrect release name, update Node version to 22).
# Test Plan
- CI should pass
- Try a release after this PR is merged
Summary:
I'm working in a weird setup where the browser won't launch from cli.
This required me to visit the url manually from my browser.
To get the url, I had to add a `console.log` to the cli's code directly.
Figured this change will make it easier for similar users to login, without exposing sensitive data
Test Plan:
Tested manually on the compiled code, not the best
* Workaround early stream termination using filehandle.read
* Add length check to createReadStreamAsync
* Align asset and multipart entry file types
* Recalculate and check hash during upload
* Update error messages and add comment
* Emit errors earlier and refactor loop
* Use same reader function and disable objectMode
* Add CHANGELOG entry
* Apply lints
# Why
Adding the new `workflow:logs` command (see the [workflow CLI design document](https://www.notion.so/expo/EAS-CLI-workflow-commands-8bc5ccc4f9f949c8adfec1347817113c)) and the related command `workflow:view`, to provide more details on workflow runs, jobs, and their associated logs.
```bash
$ eas workflow:view --help
view details for a workflow run, including jobs. If no run ID is provided, you will be prompted to select
from recent workflow runs for the current project.
USAGE
$ eas workflow:view [ID] [--json] [--non-interactive]
ARGUMENTS
ID ID of the workflow run to view
FLAGS
--json Enable JSON output, non-JSON messages will be printed to stderr.
--non-interactive Run the command in non-interactive mode.
DESCRIPTION
view details for a workflow run, including jobs. If no run ID is provided, you will be prompted to select
from recent
workflow runs for the current project.
```
```bash
$ eas workflow:logs --help
view logs for a workflow run, selecting a job and step to view. You can pass in either a workflow run ID or a job ID.
If no ID is passed in, you will be prompted to select from recent workflow runs for the current project.
USAGE
$ eas workflow:logs [ID] [--json] [--non-interactive] [--allSteps]
ARGUMENTS
ID ID of the workflow run or workflow job to view logs for
FLAGS
--allSteps Print all logs, rather than prompting for a specific step. This will be automatically set when in
non-interactive mode.
--json Enable JSON output, non-JSON messages will be printed to stderr.
--non-interactive Run the command in non-interactive mode.
DESCRIPTION
view logs for a workflow run, selecting a job and step to view. You can pass in either a workflow run ID or a job ID.
If no ID is passed in, you will be prompted to select from recent workflow runs for the current project.
```
# How
Implemented the new commands. New queries and and a new fragment for workflow jobs have been added.
A state machine has been implemented to ensure correct flow for the `workflow:logs` command, allowing the user to drill down from workflow runs, to jobs, to viewing logs for an individual step. Where it makes sense, the selection prompts allow backtracking to search for a different run, job, etc.
# Test Plan
- CI should pass
- Tested locally with different inputs, and with different flag values
- New unit tests added for `workflow:logs`
# Why
The current implementation of capabilities syncing can lead to provisioning profiles becoming invalid. This PR refactors the syncing process to avoid this issue by making the following improvements:
1. Preventing unnecessary creation of capability identifiers that already exist
2. Only updating bundle capabilities when necessary
closes https://github.com/expo/eas-cli/issues/1069
closes https://github.com/expo/eas-cli/issues/1445
# How
I refactored the capability syncing logic to:
- Add test coverage for faulty behaviors. More tests are [up the stack](https://github.com/expo/eas-cli/pull/3086/files#diff-09c5aad6e72a7a336e9783710d3899ff2ea432146929bc07ba8bfec7c7bbd2a8).
- Only create capability identifiers that don't already exist on the server
- Skip updating bundle capabilities when no changes are needed
# Test Plan
`easd` command below is from https://github.com/expo/eas-cli/blob/main/CONTRIBUTING.md#development.
To best test this locally, have an iOS app with
(1) no capabilities enabled. Enable some capabilities in `app.json` and run prebuild. [iCloud](https://docs.expo.dev/versions/latest/sdk/document-picker/#example-appjson-with-config-plugin), Apple Sign In, app groups, or `"com.apple.developer.default-data-protection": "NSFileProtectionComplete / ..."` are of special interest, but any will do.
OR
(2) a buch of capabilities already enabled
using this app, run `easd build -p ios` TWO times. Confirm that the first build creates new capabilities on your app bundle (https://developer.apple.com/account/resources/identifiers/list) and provisioning profiles (if any). Confirm the second build passes and doesn't require any provisioning profile changes.
Finally, change the capabilities, and again run `easd build -p ios` twice to confirm no capabilities / provisioning profiles are changed on the second run (that's what should happen on the first run).
- green CI
- local tests
# Why
This PR refactors the iOS capability synchronization logic to make it more "linear" and easier to maintain
# How
- Encapsulating capability-specific logic with the capability definition
- Making the operation decision process more explicit and easier to follow
# Test Plan
- green CI
- local tests
# Why
This PR refactors the iOS bundle ID capabilities management to improve handling of capabilities with complex settings. The current implementation doesn't properly handle capabilities that have settings - Sign In with Apple, iCloud, and Data Protection.
# How
- Added test cases that would previously fail
- Extracted capability list to a separate file (`capabilityList.ts`) to improve code organization
- Fixed synchronization logic to make tests pass (in a dirty way, refactored later)
- Fixed handling of MDM Managed Associated Domains to prevent disabling when Associated Domains is enabled
- Improved test assertions to use non-deprecated matchers (`toHaveBeenCalledWith` instead of `toBeCalled`)
# Test Plan
- green CI
- local testing
* Add custom multipart format
* Reuse list of files for upload params
* Remove unused gzip compression
* Clean up UploadParams
* Update file crawling to include size and mime type
* Move file-size check to assets.ts
* Refactor upload functions
* Update upload API function to support multipart uploads
* Update uploadAsync params usage in deploy command
* Upload as multipart batches, if instructed
* Count batch length against total
* Add CHANGELOG
* Fix signal not being used
* Add new assets.json format
* Skip empty assets
* Fix lint error
* Adjust concurrency
Beyond 10 we do risk some CPU contention in rare cases for small files
(especially on Windows)
However, we can now run more requests concurrently, since they're slower
each and hence are less likely to run into network issues due to lower
rate.
* Revert "Skip empty assets"
This reverts commit 18434cdb48.
* feat(eas-cli): Implement new asset upload progress bar for `deploy` command (#3099)
* Update comment
* Add CHANGELOG entry for progress bar change
* [eas-cli] update apple team info as needed
* Temporary Commit at 7/7/2025, 2:01:08 PM
* Temporary Commit at 7/7/2025, 2:12:28 PM
* Temporary Commit at 7/8/2025, 11:28:08 AM
<!-- If this PR requires a changelog entry, add it by commenting the PR with the command `/changelog-entry [breaking-change|new-feature|bug-fix|chore] [message]`. -->
<!-- You can skip the changelog check by labeling the PR with "no changelog". -->
# Why
While creating a previous PR, I moved a graphql query into a directory not explicitly specified in `graphql-codegen.yml`. It took me a while to figure out what was going on (only error was `tsc`).
To avoid other devs running into this, just generate it for all files in src.
# How
1. Update paths
2. Fix `fetchUserAsync` to be typed and have a unique query name.
# Test Plan
```
neas logout
neas login
neas whoami
```
<!-- If this PR requires a changelog entry, add it by commenting the PR with the command `/changelog-entry [breaking-change|new-feature|bug-fix|chore] [message]`. -->
<!-- You can skip the changelog check by labeling the PR with "no changelog". -->
# Why
The docs for ending an update-based rollout are incorrect: https://docs.expo.dev/eas-update/rollouts/#ending-a-rollout
In order to revert a update-based rollout, it currently says:
> To accomplish this end state, republish the previous update by using the `eas update:republish` command.
But this is incorrect since one is not allowed to publish a new update until the control update is finished (can't have multiple roll-outs).
What actually needs to happen under the hood is one of two options:
- When on a branch with a rollout with a control update
1. Delete the rollout update
2. Republish the control update
- When on a branch with a rollout with only a rollout update
1. Delete the rollout update
2. Publish a roll-back-to-embedded (after warning since this may not be intended behavior).
This PR adds a new command, `eas update:revert-update-rollout` that does this.
# How
Add command, refactoring shared code where necessary.
This handles a ton of cases. Essentially the cross product of:
- Rollout published on empty branch or branch with existing updates. (rollout has control)
- Rollout published per-platform or single group with both platforms,
- Control published per-platform or single group with both platforms.
So, 8 total cases.
# Test Plan
Tested most of the 8 cases.
Normal update group rollout (both platforms), normal control on same update group:
```
(branch with an existing notmal update group)
$ neas update --rollout-percentage=50
$ neas update:revert-update-rollout
✔ Find update by branch or channel? › Branch
✔ Select branch from which to choose update › main
✔ Select a target runtime › 1.0.0
✔ Are you sure you want to revert the rollout update group with ID "b0db8e15-d3d0-41db-985a-d74f0bd8fb00"? This will delete the rollout update group and republish the control update group (ID: "7df732c7-ccad-4755-8b80-acd1c03c5a81"). … yes
✔ Provide an update message. … watwatrep
✔ Republished update group
```
Normal update group rollout (both platforms), controls in different update groups (latest updates for each platform at time of rollout publish were in different groups):
```
$ neas update --platform android
$ neas update --platform ios
$ neas update --rollout-percentage=50
$ neas update:revert-update-rollout
✔ Find update by branch or channel? › Branch
✔ Select branch from which to choose update › main
✔ Select a target runtime › 1.0.0
✔ Are you sure you want to revert the rollout update group with ID "47ef8fe7-6f9b-4c59-bdd5-cdcc12eef0f2"? This will delete the rollout update group and republish the control update groups (IDs: "a821e0be-04af-455b-8b75-ceabea3afc09", "5a07a994-7a71-4d4e-a188-83df9e456e2c"). … yes
✔ Provide an update message. … Republish "ioswat" - group: a821e0be-04af-455b-8b75-ceabea3afc09
✔ Provide an update message. … Republish "watandroid" - group: 5a07a994-7a71-4d4e-a188-83df9e456e2c
✔ Republished update group
Branch main
Runtime version 1.0.0
Platform ios
Update group ID c1d05402-4391-4cd8-9eb0-ab0c7958bce5
iOS update ID 06b57ea9-4016-40b7-962b-756d463f2ba9
Message Republish "ioswat" - group: a821e0be-04af-455b-8b75-ceabea3afc09
EAS Dashboard https://staging.expo.dev/accounts/wschurman/projects/test-ojdasksajajaja/updates/c1d05402-4391-4cd8-9eb0-ab0c7958bce5
✔ Republished update group
Branch main
Runtime version 1.0.0
Platform android
Update group ID 56d10192-9bbd-4557-ab25-44250b3ccf6b
Android update ID eb5b1a25-cbba-4f2a-afe9-d698068369ed
Message Republish "watandroid" - group: 5a07a994-7a71-4d4e-a188-83df9e456e2c
EAS Dashboard https://staging.expo.dev/accounts/wschurman/projects/test-ojdasksajajaja/updates/56d10192-9bbd-4557-ab25-44250b3ccf6b
```
Two single platform rollouts:
```
$ neas update --platform android
$ neas update --platform ios
$ neas update --rollout-percentage=50 --platform android
$ neas update --rollout-percentage=50 --platform ios
$ neas update:revert-update-rollout
✔ Find update by branch or channel? › Branch
✔ Select branch from which to choose update › main
✔ Select a target runtime › 1.0.0
✔ Which rollout update group would you like to revert? › Rollout update group ID: 38369dcd-fe24-40c5-a85d-c7d4c03c9230, Platform: android, Rollout Percentage: 50
✔ Are you sure you want to revert the rollout update group with ID "38369dcd-fe24-40c5-a85d-c7d4c03c9230"? This will delete the rollout update group and republish the control update group (ID: "5a07a994-7a71-4d4e-a188-83df9e456e2c"). … yes
✔ Provide an update message. … Republish "watandroid" - group: 5a07a994-7a71-4d4e-a188-83df9e456e2c
✔ Republished update group
$ neas update:revert-update-rollout
(node:96817) [DEP0040] DeprecationWarning: The `punycode` module is deprecated. Please use a userland alternative instead.
(Use `node --trace-deprecation ...` to show where the warning was created)
✔ Find update by branch or channel? › Branch
✔ Select branch from which to choose update › main
✔ Select a target runtime › 1.0.0
✔ Are you sure you want to revert the rollout update group with ID "ae7aaa74-06cc-43b1-98dd-63e019cb4a64"? This will delete the rollout update group and republish the control update group (ID: "a821e0be-04af-455b-8b75-ceabea3afc09"). … yes
✔ Provide an update message. … Republish "ioswat" - group: a821e0be-04af-455b-8b75-ceabea3afc09
✔ Republished update group
```
Empty branch rollout (both platforms), no control:
```
(empty branch)
$ neas update --rollout-percentage=50
$ neas update:revert-update-rollout
✔ Find update by branch or channel? › Branch
✔ Select branch from which to choose update › main
✔ Select a target runtime › 1.0.0
✔ Are you sure you want to revert the rollout update group with ID "1797d1fe-cf58-4134-abe1-b0333250da55"? This will delete the rollout update group and publish a new roll-back-to-embedded update (no control update to roll back to), whose behavior may not be a true revert depending on the previous state of the branch. … yes
✔ Provide an update message: … geageawga
✔ Published!
```
Empty branch rollout (just one platform), no control:
```
(empty branch)
$ neas update --rollout-percentage=50 --platform android
$ neas update:revert-update-rollout
(node:87664) [DEP0040] DeprecationWarning: The `punycode` module is deprecated. Please use a userland alternative instead.
(Use `node --trace-deprecation ...` to show where the warning was created)
✔ Find update by branch or channel? › Branch
✔ Select branch from which to choose update › main
✔ Select a target runtime › 1.0.0
✔ Are you sure you want to revert the rollout update group with ID "b2115695-ba9f-4f2d-a901-9f147d09e113"? This will delete the rollout update group and publish a new roll-back-to-embedded update (no control update to roll back to), whose behavior may not be a true revert depending on the previous state of the branch. … yes
✔ Provide an update message: … dwada
✔ Published!
Branch main
Runtime version 1.0.0
Platform android
```
# Why
Now that #3091 is resolved, we need to add some CI to guard against similar future issues with GraphQL breakage.
# How
Add a new workflow that
- Executes GraphQL generation, and fails if there is an error.
- Verifies that for a given PR, any changes in GraphQL generated code or schema files have been checked in as part of the PR. If not, the workflow fails.
The new workflow will execute on pushes, and on a daily schedule to guard against breakage due to changes on the server side.
# Test Plan
- CI should pass
- Push a source change in the GraphQL code, but do not regenerate GraphQL, and verify that the new workflow fails
Why
===
The message about queuing builds faster mentions the Production and Enterprise plans. The Starter plan is also a good choice.
How
===
This PR changes the message just to say "paid plan" which is simple and understandable.
Test Plan
===
Just automated tests since this is a copy change.
<!-- If this PR requires a changelog entry, add it by commenting the PR with the command `/changelog-entry [breaking-change|new-feature|bug-fix|chore] [message]`. -->
<!-- You can skip the changelog check by labeling the PR with "no changelog". -->
# Why
The previously recommended xcode-select command which was `sudo xcode-select -s /Applications/Xcode.app` did not work. Updated to suggest the correct fix which is running `sudo xcode-select -s /Applications/Xcode.app/Contents/Developer`
# How
Attempted to use `eas build:dev` after installing Xcode beta to use iOS 26, got error message which suggested the `sudo xcode-select -s /Applications/Xcode.app` command. It didn't work, but `sudo xcode-select -s /Applications/Xcode.app/Contents/Developer` did.
# Test Plan
Not quite sure how to reproduce the scenario in order to test this, but I can verify that this is what the correct suggestion should be as it fixed the problem which triggered the error message, and that's the location where it should look for the Simulator.
<!-- If this PR requires a changelog entry, add it by commenting the PR with the command `/changelog-entry [breaking-change|new-feature|bug-fix|chore] [message]`. -->
<!-- You can skip the changelog check by labeling the PR with "no changelog". -->
# Why
Since https://linear.app/expo/issue/ENG-14634/add-support-for-groups-option-in-fastlane-pilot-call we support passing `--groups` parameter to `eas submit` and configuring submission profile with `groups` parameter ("undocumentedly").
I need to add support for this property to `submit:internal`. We're going to source the `groups` to pass to `fastlane pilot` from either:
- submit profile (this pull request adds this)
- or workflow config (in which case we'll don't use this output).
# How
Added `groups` output to the `submit:internal` JSON object. We're going to parse it in https://github.com/expo/universe/blob/5dcea35fb50bf9fc8e3ae4ffe445f958bb5789e6/server/www/src/data/entities/workflow/job/WorkflowSubmitJob.ts#L192-L200.
# Test Plan
Ran `easd submit:internal --id 60a2d86a-b6d8-4ba2-b7dd-bf05bdd78791 --platform ios` in a test project of mine after adding submit profile like
```
"submit": {
"production": {
"ios": {
"ascAppId": "6769090000",
"groups": ["family", "friends"]
}
}
}
```
<img width="253" alt="Zrzut ekranu 2025-06-25 o 17 39 22" src="https://github.com/user-attachments/assets/584e1bea-a905-4ba3-a68e-5df4a4d3477e" />
<!-- If this PR requires a changelog entry, add it by commenting the PR with the command `/changelog-entry [breaking-change|new-feature|bug-fix|chore] [message]`. -->
<!-- You can skip the changelog check by labeling the PR with "no changelog". -->
# Why
Update group background job polling (and all background job polling in the CLI) is broken due to query caching in the client. Not sure why this worked when I added it as the caching has been present for a while.
# How
Disable cache for the poll request.
This also inverts the confirmation on `update:delete` to be affirmative as most of our other confirm questions are (default to yes). The reason for this change is that I found it non-intuitive while testing this to have "yes" cancel the command. I think the reason we use to have this inverted is it was dangerous, but IMO it's already enough intent by running the command to just have the default be yes. Happy to revert this back if reviewers disagree.
# Test Plan
```
eas update
neas update:delete <update_group>
```
See it now works.
<!-- If this PR requires a changelog entry, add it by commenting the PR with the command `/changelog-entry [breaking-change|new-feature|bug-fix|chore] [message]`. -->
<!-- You can skip the changelog check by labeling the PR with "no changelog". -->
# Why
Noticed this condition was inverted during creation of a new related command.
# How
Correct the if statement. It should warn when platform is not `all`.
# Test Plan
Run `neas update:republish --platform ios`, see logged message:
```
$ eas update:republish --platform ios
✔ Find update by branch or channel? › Branch
✔ Select branch from which to choose update › main
✔ Load more update groups? › [Jun 23 11:14 by wschurman, runtimeVersion: 1.0.0] Initial commit
Generated by create-expo-app 3.4.2.
You are republishing an update group that wasn't published for all platforms.
✔ The republished update group will appear on the same platforms it was originally published on: ios
✔ Provide an update message. … Republish "Initial commit
Generated by create-expo-app 3.4.2." - group: 8e1b11be-ef82-4ccd-8294-a4e7075849d1
✔ Republished update group
```
* [eas-cli] align text of messages from build and workflow:run
* Apply suggestions from code review
Co-authored-by: Stanisław Chmiela <[email protected]>
---------
Co-authored-by: Stanisław Chmiela <[email protected]>
# Why
I tried to use eas dev:build with --build-profile parameter, but I got eas build with different profile
# How
I tested locally that changes in this branch fixes the problem for my case
# Test Plan
I have run following command (easd linked to locally build eas-cli, as described in CONTRIBUTING.md)
```
EXPO_DEBUG=1 easd build:dev -e preview -p android
```
<!-- If this PR requires a changelog entry, add it by commenting the PR with the command `/changelog-entry [breaking-change|new-feature|bug-fix|chore] [message]`. -->
<!-- You can skip the changelog check by labeling the PR with "no changelog". -->
# Why
6636f76cff had a logical merge conflict in the changelog since a new version was released.
# How
Move the entry to the correct release.
# Test Plan
Inspect.
<!-- If this PR requires a changelog entry, add it by commenting the PR with the command `/changelog-entry [breaking-change|new-feature|bug-fix|chore] [message]`. -->
<!-- You can skip the changelog check by labeling the PR with "no changelog". -->
# Why
This is the eas-cli component of https://github.com/expo/universe/pull/20493, which allows creating a rollout update on a branch with no existing updates.
# How
Remove error when no existing update exists.
# Test Plan
```
yarn create expo-app test-kjhsadjhkkjhafskjhjkhhhh
cd test-kjhsadjhkkjhafskjhjkhhhh
neas init
neas update:configure
neas update --rollout-percentage=60
```
View published update in db, see it has 60% rollout percentage but no control
# Why
since the onboarding repack is not available now, let's remove the code for onboarding repack. we can later use `BuildMode.Repack` for workflow repack job.
# How
remove repack relevant code
# Test Plan
ci passed
<!-- If this PR requires a changelog entry, add it by commenting the PR with the command `/changelog-entry [breaking-change|new-feature|bug-fix|chore] [message]`. -->
<!-- You can skip the changelog check by labeling the PR with "no changelog". -->
# Why
https://exponent-internal.slack.com/archives/C089ED313PV/p1748545207578949?thread_ts=1748543276.564549&cid=C089ED313PV
# How
Added support for `EAS_SKIP_CLI_VERSION_CHECK` environment variable.
# Test Plan
Added tests.
<!-- If this PR requires a changelog entry, add it by commenting the PR with the command `/changelog-entry [breaking-change|new-feature|bug-fix|chore] [message]`. -->
<!-- You can skip the changelog check by labeling the PR with "no changelog". -->
# Why
As requested during feature review, these should live in the top-level "cli" key of eas.json.
Closes ENG-15925.
# How
Change location.
# Test Plan
`yarn tsc`
<!-- If this PR requires a changelog entry, add it by commenting the PR with the command `/changelog-entry [breaking-change|new-feature|bug-fix|chore] [message]`. -->
<!-- You can skip the changelog check by labeling the PR with "no changelog". -->
# Why
`eas.json` is not _required_ to run workflows. Builds will fail, but we can warn about it and also the workflow error makes sense.
# How
Check before uploading the non-required files.
# Test Plan
<img width="946" alt="Zrzut ekranu 2025-06-2 o 21 45 31" src="https://github.com/user-attachments/assets/b6b4aad1-608f-471c-b541-302850744062" />
<!-- If this PR requires a changelog entry, add it by commenting the PR with the command `/changelog-entry [breaking-change|new-feature|bug-fix|chore] [message]`. -->
<!-- You can skip the changelog check by labeling the PR with "no changelog". -->
# Why
These fields were missing in republished updates.
Closes ENG-15846.
# How
Copy these fields over as well when republishing.
# Test Plan
`neas update` with dirty working tree.
`neas update:republish`
Verify republished update has same dirty working tree status.
<!-- If this PR requires a changelog entry, add it by commenting the PR with the command `/changelog-entry [breaking-change|new-feature|bug-fix|chore] [message]`. -->
<!-- You can skip the changelog check by labeling the PR with "no changelog". -->
# Why
CLI change corresponding to https://github.com/expo/universe/pull/20131.
This adds the ability to override the manifest and asset hosts for EAS Update such that a set of proxy hosts can be used.
Closes ENG-15804.
# How
Considered putting these in flags, but it turned out that many many commands need to detect EAS Update configuration so they need access to these as well. Therefore, I added the new fields to `eas-json`.
The fields are:
- `manifestHostOverride` - a host to replace `u.expo.dev` in the URL.
- `assetHostOverride` - a host to replace `assets.eascdn.net` in asset URLs in the generated update manifests.
Then, in every place in EAS CLI when we generate a manifest URL, instead use the overridden URL.
Then, supply the manifest URL and asset URL overrides when creating updates (or republishes).
# Test Plan
Lengthy, but here it goes:
1. Create two proxies, one for assets and one for manifests. All these proxies do is proxy the requests.
- https://github.com/wschurman/updates-test-proxy is a proxy for the manifest request built using expo router API routes and EAS Hosting. It is hosted at https://updates-test-proxy.expo.app and just fetches the URL input but at `u.expo.dev` and returns the response.
- https://github.com/wschurman/updates-test-asset-proxy is the same but for assets. https://updates-test-asset-proxy.expo.app
2. Create an app on staging
3. Add the following to eas.json:
```
"update": {
"assetHostOverride": "updates-test-asset-proxy.expo.app",
"manifestHostOverride": "updates-test-proxy.expo.app"
}
```
4. run `neas update:configure`, see overridden URL in app.json
5. run `neas update`, see update is published, visit overridden manifest URL, see an update!
6. Visit one of the asset URLs (supplying authorization header) and see it correctly proxies.
7. configure code signing
8. publish again, see code signing works (downloads from `manifestPermalink`)
9. republish the last update, see it works and sets the overrides correctly.
# Why
[ENG-15618: Support different suffixes of custom gradle commands](https://linear.app/expo/issue/ENG-15618/support-different-suffixes-of-custom-gradle-commands)
Currently when user specifies a custom gradle command that does not end with `*Release` or `*Debug` the command is not parsed correctly. Build type and flavour are not parsed, which causes problems with setting code version and app version down the line.
# How
- Changed the regexp that parses gradle command to be more flexible, allowing any suffix that starts with a capital letter.
# Test Plan
- add a custom gradle command to `eas.json`, for example
```
"development": {
"android": {
"gradleCommand": ":app:assembleAbcStaging"
}
},
```
- add a `buildType` and packageFlavor to `build.gradle` file:
```
productFlavors {
abc { // New flavor
versionCode 3
versionName "abc"
}
}
buildTypes {
staging {}
debug {... }
release {...}
}
```
- start an android build from cli.
- started build should not warn about invalid configuration
- on website, build version should be set to the value assigned in flavour.
# Why
When publishing updates, users need to know if there are compatible builds for their fingerprints. This PR prints this information at the end of `eas update` to warn users, if they have updates being published that don't have corresponding builds, causing potential compatibility issues.
<img width="1536" alt="Screenshot 2025-04-01 at 5 09 12 PM" src="https://github.com/user-attachments/assets/400ac0d9-c39e-4aee-9793-465c6acb634e" />
# How
Added functionality to check for compatible builds when publishing updates:
1. Enhanced `fetchBuildsAsync` to support filtering by `fingerprintHash`
2. Implemented `findCompatibleBuildsAsync` to find builds matching fingerprints
3. Added warnings in the update publish command when no compatible builds are found for fingerprints
4. Included links to the fingerprint details on the Expo website for easier debugging
# Test Plan
Run `eas update` with a fingerprint that doesn't have a compatible build
# Why
This PR refactors the `UpdatePublish` command to improve code readability and flow by removing a nested `else` block and using an early return pattern.
before
```
if (jsonFlag) {
printJsonOnlyOutput(getUpdateJsonInfosForUpdates(newUpdates));
} else {
// all the logic
}
```
after
```
if (jsonFlag) {
printJsonOnlyOutput(getUpdateJsonInfosForUpdates(newUpdates));
return;
}
// all the logic
```
# How
- Removed the large `else` block that contained most of the code in the `UpdatePublish` command
- Added an early return after handling the JSON flag case
- Kept all the existing functionality intact while making the code more maintainable
- Improved the logical flow by flattening the code structure
# Test Plan
Manually tested the `eas update` command to ensure the code path works correctly
<!-- If this PR requires a changelog entry, add it by commenting the PR with the command `/changelog-entry [breaking-change|new-feature|bug-fix|chore] [message]`. -->
<!-- You can skip the changelog check by labeling the PR with "no changelog". -->
# Why
I'm going to move Workflow and GitHub builds to scoped token soon. If users depend on the automatically provided `EXPO_TOKEN` _and_ the action they're going to try to do has not been whitelisted with scoped rules yet, they're going to get an `UnauthorizedError` mentioning `ScopedAccountActorViewerContext`.
On top of logs annotations (working on them [here](https://github.com/expo/universe/pull/19494)) I think we can make the error message a bit better.
# How
Special-case `ScopedAccountActorViewerContext` error in `process.env.EAS_BUILD` and print message asking the user to contact us.
Alongside this, I think https://github.com/expo/universe/pull/19500 may be a good idea.
# Test Plan
With `readRules` in `UserPrivacyPolicy` cleared out:
<img width="1440" alt="Zrzut ekranu 2025-03-31 o 23 43 00" src="https://github.com/user-attachments/assets/89245ccc-21e7-4041-98ce-21a517f932d4" />
<!-- If this PR requires a changelog entry, add it by commenting the PR with the command `/changelog-entry [breaking-change|new-feature|bug-fix|chore] [message]`. -->
<!-- You can skip the changelog check by labeling the PR with "no changelog". -->
# Why
When we move Workflows to scoped token which has at most DEVELOPER permission, `type: deploy, prod: true` jobs will fail with an ominous
```
You don't have the required permissions to perform this operation.
This can sometimes happen if you are logged in as incorrect user.
Run eas whoami to check the username you are logged in as.
Run eas login to change the account.
Original error message: Entity not authorized: AppDevDomainNameEntity[00000000-0000-0000-0000-000000000000] (viewer = ScopedAccountActorViewerContext[16d62ef3-a219-47ca-936b-89958f6cb1f8], action = CREATE, ruleIndex = -1)
Request ID: 863a1317-882b-4cab-b8f2-c5f5d63b6f29
```
# How
Special-cased the `--prod` && `AppDevDomainEntity` && `CREATE` && `UNAUTHORIZED_ERROR` error.
Not sure if we should call out "as an app admin" in the error message. Open to suggestions.
# Test Plan
Ran `eas deploy --prod` as a developer-permission-only user. I've updated message in code since taking the screenshot.
<img width="836" alt="Zrzut ekranu 2025-03-28 o 21 21 02" src="https://github.com/user-attachments/assets/fc3a8b08-ae2f-4630-9e88-ccb48151d3e2" />
<!-- If this PR requires a changelog entry, add it by commenting the PR with the command `/changelog-entry [breaking-change|new-feature|bug-fix|chore] [message]`. -->
<!-- You can skip the changelog check by labeling the PR with "no changelog". -->
# Why
Closes https://github.com/expo/eas-cli/issues/2401
# How
Add `corepack` field to `eas.json`
# Test Plan
Tests, tested manually
<!-- If this PR requires a changelog entry, add it by commenting the PR with the command `/changelog-entry [breaking-change|new-feature|bug-fix|chore] [message]`. -->
<!-- You can skip the changelog check by labeling the PR with "no changelog". -->
# Why
I was thinking we can offer `EAS_NO_VCS=1` as a simple escape hatch for people that are getting weird Git-related issues. One of them may be an old version of Git that does not support our `git clone` properly.
# How
In `getRootPath` of no-VCS mode, return first of those values (in order):
- `EAS_PROJECT_ROOT`
- Git root path
- `process.cwd()`
# Test Plan
Added a basic test.
Co-authored-by: Szymon Dziedzic <[email protected]>
<!-- If this PR requires a changelog entry, add it by commenting the PR with the command `/changelog-entry [breaking-change|new-feature|bug-fix|chore] [message]`. -->
<!-- You can skip the changelog check by labeling the PR with "no changelog". -->
# Why
Thanks to [a conversation on Bluesky](https://bsky.app/profile/sjchmiela.bsky.social/post/3lj6k2xo7rk2g) it turned out [this check](https://github.com/expo/eas-cli/blob/3a434940eb44dfa8dddc42af055793134ca9016f/packages/eas-cli/src/vcs/clients/git.ts#L255-L257) is invalid (always true), because `Ignore` has `DEFAULT_IGNORE`, so even if a user had no `.git` in `.easignore`, especially if they had `requireCommit: true`, `.git` would be removed.
# How
- Added a check to disallow `.easignore` if one has `requireCommit: true`.
- Changed logic around `Ignore` so we have separate `createForCopying` (ignores `.git` and `node_modules` always) and `createForChecking` (does not ignore anything by default). I don't like it, but it should work and is the best of what I could come up with without further refactoring.
# Test Plan
Added more tests.
* [eas-cli] skip creation of testflight group when there are already exisitng testflight groups + allow to opt out of the behavior by setting env var
* rename env var
* add changelog
<!-- If this PR requires a changelog entry, add it by commenting the PR with the command `/changelog-entry [breaking-change|new-feature|bug-fix|chore] [message]`. -->
<!-- You can skip the changelog check by labeling the PR with "no changelog". -->
# Why
I struggled with adding a new workflow within `.eas/build` since I had `build/` in my `.gitignore`. I removed the line but it was still an issue. Turns out I had the same thing in `.easignore`. I think adding this to the error message could be helpful for others
# How
Not tested
<!-- If this PR requires a changelog entry, add it by commenting the PR with the command `/changelog-entry [breaking-change|new-feature|bug-fix|chore] [message]`. -->
<!-- You can skip the changelog check by labeling the PR with "no changelog". -->
# Why
When we `git clone` the source repository and then copy worktree changes over, we add and modify files. However, we do not _delete_ files that have been deleted in worktree.
# How
When we operate in `requireCommit: false` mode, we add `--no-checkout` to the `git clone` command. This way the destination directory only has `.git` directory when we start copying files over. Thus, files deleted from worktree will not suddenly appear in tarball.
In `requireCommit: true` mode we don't add `--no-checkout`, because we don't copy files.
# Test Plan
Added test. Tested manually.
Side effect of this is that `git status` reports that all files have been deleted and there's a bunch of untracked files (which, `git add`-ed, make the repository appear as the source repo). I don't know how to copy the Git index 1-1…
<!-- If this PR requires a changelog entry, add it by commenting the PR with the command `/changelog-entry [breaking-change|new-feature|bug-fix|chore] [message]`. -->
<!-- You can skip the changelog check by labeling the PR with "no changelog". -->
# Why
With `toNamespacedPath` I was finally able to make ignoring files work properly.
# How
Removed `if (__NORMALIZE)` checks.
# Test Plan
Ran `eas build:inspect` on Windows, confirmed no `node_modules` was copied. Also ran `eas build:inspect` on macOS, worked too.
<!-- If this PR requires a changelog entry, add it by commenting the PR with the command `/changelog-entry [breaking-change|new-feature|bug-fix|chore] [message]`. -->
<!-- You can skip the changelog check by labeling the PR with "no changelog". -->
# Why
`fs` adds a [Win32 File Namespace](https://learn.microsoft.com/en-us/windows/win32/fileio/naming-a-file#win32-file-namespaces) to the beginning of the path which breaks `path.relative` if `src` does not have the namespace.
<img width="899" alt="Zrzut ekranu 2025-02-9 o 12 15 45" src="https://github.com/user-attachments/assets/942b28d1-881f-4e5d-861e-297bb5eddbf2" />
# How
Ensured that running `toNamespacedPath` on an already namespaced path does not add a _second_ namespace:
```
> path.win32.toNamespacedPath("D:\\abc")
'\\\\?\\D:\\abc'
> path.win32.toNamespacedPath(path.win32.toNamespacedPath("D:\\abc"))
'\\\\?\\D:\\abc'
```
Wrapped `src` and `srcFilePath` (two arguments to `path.relative`) with `path.toNamespacedPath` so `path.relative` works.
# Test Plan
Will release and test.
<!-- If this PR requires a changelog entry, add it by commenting the PR with the command `/changelog-entry [breaking-change|new-feature|bug-fix|chore] [message]`. -->
<!-- You can skip the changelog check by labeling the PR with "no changelog". -->
# Why
Since v15 files that should be ignored get added to the tarball on Windows.
# How
After adding some debugging I was able to find that `src`, `srcFilePath` don't match on path separators which breaks `ignore.ignores`.
<img width="827" alt="Zrzut ekranu 2025-02-7 o 21 57 02" src="https://github.com/user-attachments/assets/9ed3b166-dbc0-4287-87a1-0009769e0d4d" />
Added `path.normalize` if `__NORMALIZE` is `1`.
---
Also noticed we used wrong `cwd` for is-tracked check.
# Test Plan
Going to release this and test.
<!-- If this PR requires a changelog entry, add it by commenting the PR with the command `/changelog-entry [breaking-change|new-feature|bug-fix|chore] [message]`. -->
<!-- You can skip the changelog check by labeling the PR with "no changelog". -->
# Why
`requireCommit: true` setups expect worktree to be clean (e.g. encrypted files to be locked).
# How
Made it so if `requireCommit` is true we don't copy files over.
# Test Plan
I have enabled `git-crypt` in my test repository, added a secret file, ran `easd build:inspect -p android -s archive -o ~/test-production --force`. The clone did contain secret file in plaintext.
Then I enabled `requireCommit: true` and repeated the command. The clone did contain the file encrypted.
<!-- If this PR requires a changelog entry, add it by commenting the PR with the command `/changelog-entry [breaking-change|new-feature|bug-fix|chore] [message]`. -->
<!-- You can skip the changelog check by labeling the PR with "no changelog". -->
# Why
`git ls-files` returns submodules as a single filepath entry, so `rm` must remove it with `-rf`. Fixes https://github.com/expo/eas-cli/issues/2883.
# How
Added `{ recursive: true, force: true }` to `fs.rm`.
# Test Plan
I tried `easd build:inspect -p android -s archive -o ~/testtest --force` on a repo with a submodule in `.easignore`. It succeeded and the submodule was not included in the folder.
<!-- If this PR requires a changelog entry, add it by commenting the PR with the command `/changelog-entry [breaking-change|new-feature|bug-fix|chore] [message]`. -->
<!-- You can skip the changelog check by labeling the PR with "no changelog". -->
# Why
Adding `.git` directory to the tarball makes it bigger. Users notice this warning and want it to be actionable, but currently adding `.git` to `.easignore` does nothing.
# How
For now, raise threshold for warning from 100 MiB to 150 MiB. In the next pull request I'm going to fix `.git` exclusion via `.easignore`.
# Test Plan
CI should pass.
<!-- If this PR requires a changelog entry, add it by commenting the PR with the command `/changelog-entry [breaking-change|new-feature|bug-fix|chore] [message]`. -->
<!-- You can skip the changelog check by labeling the PR with "no changelog". -->
# Why
Currently it's not possible to exclude `.git` from being uploaded in project tarball.
# How
In `makeShallowCopyAsync()` we first clone the repository, then we delete files ignored by `.easignore` and then we copy working directory over.
The "delete files ignored by `.easignore`" step is governed by `git ls-files --exclude-from=.easignore`. `git ls-files` is never going to consider `.git` directory to be excluded, even if specified in `.easignore`, so we need to special-case the `.git` directory and remove it manually if specified.
# Test Plan
Added `.git` to a test `.easignore`, ran `eas build:inspect` and confirmed the copied directory did _not_ contain `.git` folder.
<!-- If this PR requires a changelog entry, add it by commenting the PR with the command `/changelog-entry [breaking-change|new-feature|bug-fix|chore] [message]`. -->
<!-- You can skip the changelog check by labeling the PR with "no changelog". -->
# Why
`fs-extra` fails to copy symbolic links as one would expect. https://github.com/jprichardson/node-fs-extra/issues/708
# How
We could:
- copy and dereference symbolic links — make it so that the clone of the original directory has two copies of the files -- original and linked
- copy symbolic links without re-resolving, assuming any links will be relative and should work when unpacked on a different machine.
I think we can count on the latter.
# Test Plan
Created a link in my test repository
```
drwxr-xr-x 7 sjchmiela staff 224B Jan 28 16:14 assets
lrwxr-xr-x 1 sjchmiela staff 6B Feb 4 22:12 assets-linked -> assets
```
ran
```
easd build:inspect -p android -s archive -o ~/testtest --force
```
confirmed I see
```
drwxr-xr-x 7 sjchmiela staff 224B Feb 4 22:20 assets
lrwxr-xr-x 1 sjchmiela staff 6B Feb 4 22:20 assets-linked -> assets
```
in the result directory.
<!-- If this PR requires a changelog entry, add it by commenting the PR with the command `/changelog-entry [breaking-change|new-feature|bug-fix|chore] [message]`. -->
<!-- You can skip the changelog check by labeling the PR with "no changelog". -->
# Why
`[email protected]` doesn't interpret `.easignore` right because I haven't initialized the `Ignore` class properly.
# How
Make it so it's not possible to use `Ignore` without initializing it.
# Test Plan
Tests should pass.
<!-- If this PR requires a changelog entry, add it by commenting the PR with the command `/changelog-entry [breaking-change|new-feature|bug-fix|chore] [message]`. -->
<!-- You can skip the changelog check by labeling the PR with "no changelog". -->
# Why
If you had an `.easignore`, `isFileIgnoredAsync` was always thinking all files are NOT ignored which was causing it to think all builds are in `.GENERIC` mode (with native code) which was not true, because ACTUALLY the files WERE ignored.
All mostly because of forgotten `ignore.initIgnoreAsync()`…
# How
Added `ignore.initIgnoreAsync()`. Also realized whether the file is tracked / not-tracked may affect how actually is it going to behave, so added it to the mix.
# Test Plan
Added tests that I think test all cases.
# Why
receiving an issue for a customer who uses eas environment and continuous-deploy-fingerprint github action. they have build variants setup in their app.config.ts. the app id, scheme and some other properties are determined by their `APP_VARIANT` variable. the `APP_VARIANT` is from eas server environment.
when continuous-deploy-fingerprint first calculates fingerprint, it uses `eas env:exec` for the `npx expo-updates fingerprint:generate` call. the `APP_VARIANT` is correctly passed.
then for `eas update` call, it doesn't pass the `APP_VARIANT` env and generates an inconsistent fingerprint.
# How
pass server env to the `getRuntimeVersionInfoObjectsAsync()` (`npx expo-updates runtimeversion:resolve`)
<!-- If this PR requires a changelog entry, add it by commenting the PR with the command `/changelog-entry [breaking-change|new-feature|bug-fix|chore] [message]`. -->
<!-- You can skip the changelog check by labeling the PR with "no changelog". -->
# Why
We don't need to redownload app every single time in `eas build:dev`. We can cache it locally. This functionality already exists for `eas build:run`.
# How
Use caching the same way we do for `eas build:run`
# Test Plan
Tested manually
<!-- If this PR requires a changelog entry, add it by commenting the PR with the command `/changelog-entry [breaking-change|new-feature|bug-fix|chore] [message]`. -->
<!-- You can skip the changelog check by labeling the PR with "no changelog". -->
# Why
@brentvatne suggested that it would be useful to let people know how long the previous build took when rebuilding in `eas build:dev`
# How
Get the information about how long previous build took and show it to the user
# Test Plan
Tested manually
<!-- If this PR requires a changelog entry, add it by commenting the PR with the command `/changelog-entry [breaking-change|new-feature|bug-fix|chore] [message]`. -->
<!-- You can skip the changelog check by labeling the PR with "no changelog". -->
# Why
Currently, when `eas-cli` create a project tarball for EAS Build or Workflows, unless the user has `requireCommit: true` in `eas.json`, the project is being packed with a plain `makeShallowCopyAsync` which copies all the files that are deemed ignored by `ignore` package and `.easignore` or `.gitignore` files.
This is suboptimal for `eas workflow:run` experience — a developer with a project with Git expects the repository to exist on EAS.
# How
This was a big pain.
Previously, we had `GitClient` which required commit and its subclass, `GitNoCommitClient` which overrode some functions to allow uncommitted changes. On top of that, `GitClient` copied project with `git clone`, while `GitNoCommitClient` copied it with `makeShallowCopyAsync`. This meant `GitClient`'s copy would include `.git`, whereas `GitNoCommitClient` would not. Moreover, thus, `GitClient` would adhere to `.gitignore` and `GitNoCommitClient` would _not_ — only to `.easignore`.
This pull request:
- merges the two Git clients into one with `requireCommit` option
- this causes `.git` directory to exist in copies with `requireCommit = false`
- changes how we log changes caused by Git case sensitivity change (we need to handle the case where there is a preexisting dirty tree)
- after we clone the repository, we delete files that should have been ignored according to `.easignore`
- afterwards, we make a shallow copy of all the files over to the clone. This copies live changes to the working tree over to the clean clone. (They may only exist in `requireCommit = false` situation.)
- adds better support for `.easignore` to `isFileIgnoredAsync`.
As far as I can see the behavior looks like this:

The breaking change in here is that we're going to start adhering to `.easignore` even if someone has `requireCommit = true`.
# Test Plan
I used `easd build:inspect` a bunch to verify it works like I expect. I encourage fellow reviewers to do the same.
<!-- If this PR requires a changelog entry, add it by commenting the PR with the command `/changelog-entry [breaking-change|new-feature|bug-fix|chore] [message]`. -->
<!-- You can skip the changelog check by labeling the PR with "no changelog". -->
# Why
https://exponent-internal.slack.com/archives/C017N0N99RA/p1738630158821649
# How
Print warn message only once
# Test Plan
Tests
# Why
Add support for users accidentally passing in update group id into `fingerprint:compare` instead of update id.
# How
Added support for handling update group IDs in the fingerprint comparison command by:
- Attempting to fetch updates using the provided ID as an update group ID
- If successful and there's only one update, using that update directly
- If multiple updates exist in the group, prompting the user to select the specific platform/update
- Falling back to the original update ID logic if the group ID lookup fails
- Adding helpful error messages that include the direct URL to find the correct update ID
# Test Plan
- [ ] `EXPO_STAGING=1 ~/Documents/eas-cli/packages/eas-cli/bin/run fingerprint:compare --update-id ed36a46b-6cc7-4f49-88df-e6ffb66f22d2`
- [ ] `EXPO_STAGING=1 ~/Documents/eas-cli/packages/eas-cli/bin/run fingerprint:compare --update-id ed36a46b-6cc7-4f49-88df-e6ffb66f22d2 --non-interactive`
# Why
Add support for comparing fingerprints between updates using the `--update-id` flag in the `eas fingerprint compare` command. This allows developers to compare fingerprints between updates, similar to how they can already compare fingerprints between builds.
# How
- Added `updateId` flag to the `FingerprintCompare` command
- Extended the fingerprint comparison logic to handle updates as a fingerprint source
- Added `UpdateQuery.viewByUpdateAsync` to fetch update details including fingerprint information
- Updated the fingerprint origin type system to include updates as a source
- Enhanced the pretty printing of fingerprint sources to show platform information for updates
# Test Plan
- [ ] `EXPO_STAGING=1 ~/Documents/eas-cli/packages/eas-cli/bin/run fingerprint:compare --update-id 087ccb84-5b64-4285-aebf-e890cb95cd20`
* [eas-cli] add eas dev command
* update logs
* rename command do `build:dev`
* filter by dev client
* apply suggested changes
* use real vs virtaul build profiles
* make it hidden
* prompt to commit after adding new build profile
<!-- If this PR requires a changelog entry, add it by commenting the PR with the command `/changelog-entry [breaking-change|new-feature|bug-fix|chore] [message]`. -->
<!-- You can skip the changelog check by labeling the PR with "no changelog". -->
# Why
[ENG-14307: move environment to job object from metadata](https://linear.app/expo/issue/ENG-14307/move-environment-to-job-object-from-metadata)
`Environment` should be considering a first class citizen in job object instead of being stored in the metadata.
# How
* Pass `environment` in `job` object
* remove `environment` from `metadata`.
# Test Plan
1. Start a build specifying `environment` in `eas.json`.
2. Build should use the specified environment (check on website in `Build details`)
* Replace preview with help
* Remove redundant alias
* Update description
* Rename top level command from worker -> deploy
* Add changelog
* End sentence with a .
* Add worker:deploy as an alias to deploy
<!-- If this PR requires a changelog entry, add it by commenting the PR with the command `/changelog-entry [breaking-change|new-feature|bug-fix|chore] [message]`. -->
<!-- You can skip the changelog check by labeling the PR with "no changelog". -->
# Why
Allow filtering by `--fingerprint-hash` in the `eas build:list` command
# How
Update GQL schema to add a new filter field, and add a new flag to use it.
# Test Plan
Test locally
* Allow env vars to be loaded from .env and EAS env simultaneously
* Add CHANGELOG entry
* Apply lints
* Switch to new @expo/[email protected] API
* Update packages/eas-cli/src/commands/worker/deploy.ts
Co-authored-by: Kadi Kraman <[email protected]>
---------
Co-authored-by: Kadi Kraman <[email protected]>
<!-- If this PR requires a changelog entry, add it by commenting the PR with the command `/changelog-entry [breaking-change|new-feature|bug-fix|chore] [message]`. -->
<!-- You can skip the changelog check by labeling the PR with "no changelog". -->
# Why
[ENG-14518: Add `keystoreName` field to build profile to allow selecting/creating named keystore to use for a build](https://linear.app/expo/issue/ENG-14518/add-keystorename-field-to-build-profile-to-allow-selectingcreating)
Allow specifying keystore name of credentials to be used with a build.
# How
* Added `keystoreName` to eas.json schema
* Pass `name` option to `AndroidCredentialsProvider`
# Test Plan
Added test, tested manually.
1. Create a new android credentials with `eas credentials`
2. Specify `keystoreName` in build profile in `eas.json`
3. Ensure that it works only for android builds
4. Ensure it works only for `remote` `credentialsSource`
5. Ensure that the correct credentials are used
<img width="326" alt="image" src="https://github.com/user-attachments/assets/1245c459-5fa1-4253-b5fc-bdf7a68b814a" />
<img width="628" alt="image" src="https://github.com/user-attachments/assets/5bca4cf7-c895-462c-b527-ef37ce659aca" />
<!-- If this PR requires a changelog entry, add it by commenting the PR with the command `/changelog-entry [breaking-change|new-feature|bug-fix|chore] [message]`. -->
<!-- You can skip the changelog check by labeling the PR with "no changelog". -->
# Why
Thanks to @szdziedzic's changes to `www`, if we send a `package.json` along with `eas.json` when setting up a workflow run, EAS will pick the right images based on SDK and RN versions as `eas build` would.
# How
Added uploading `package.json` after uploading `eas.json`, added the new argument to the mutation.
# Test Plan
Tested manually.
<img width="874" alt="Zrzut ekranu 2024-12-20 o 22 27 20" src="https://github.com/user-attachments/assets/383c4a19-17a1-4f46-8113-bc6dc8aaba2e" />
<img width="833" alt="Zrzut ekranu 2024-12-20 o 22 27 58" src="https://github.com/user-attachments/assets/be9e2396-0779-4043-925f-d784b5168764" />
<!-- If this PR requires a changelog entry, add it by commenting the PR with the command `/changelog-entry [breaking-change|new-feature|bug-fix|chore] [message]`. -->
<!-- You can skip the changelog check by labeling the PR with "no changelog". -->
# Why
[ENG-14328: store `environment` info on update entity](https://linear.app/expo/issue/ENG-14328/store-environment-info-on-update-entity)
Pass `environment` to `PublishMutation.publishUpdateGroupAsync` to store in the db.
Waits for changes on the `www`. https://github.com/expo/universe/pull/17921
<!-- If this PR requires a changelog entry, add it by commenting the PR with the command `/changelog-entry [breaking-change|new-feature|bug-fix|chore] [message]`. -->
<!-- You can skip the changelog check by labeling the PR with "no changelog". -->
# Why
It seems like this file includes redundant entries + is not up to date
# How
Remove redundant entries, add @radoslawkrzemien to global codeowners, make @wschurman and @quinlanj codeowners of everything EAS Update related (including channel commands), remove @byCedric and @EvanBacon from EAS Update codeowners
# Test Plan
Review
<!-- If this PR requires a changelog entry, add it by commenting the PR with the command `/changelog-entry [breaking-change|new-feature|bug-fix|chore] [message]`. -->
<!-- You can skip the changelog check by labeling the PR with "no changelog". -->
# Why
https://discord.com/channels/695411232856997968/1309202098591633458/1311040361040183429
# How
Make warning up to date after recent env vars changes
# Why
Update `@expo` packages to SDK 52 versions
# How
Update `@expo` packages to SDK 52 versions
Fixes#2696
It seems like the `privacy` field was removed from the expo config in the SDK 52 update: https://expo.dev/changelog/2024/11-12-sdk-52.
I used `AppPrivay.Unlisted` for the `createApp` mutation input for which this is still required. It was an option selected for an app when `exp.privacy` was undefined.
It also seems that `findWorkspaceRoot` function in `@expo/package-manager` was changed to `resolveWorkspaceRoot`.
# Test Plan
Tests, test commands manually
<!-- If this PR requires a changelog entry, add it by commenting the PR with the command `/changelog-entry [breaking-change|new-feature|bug-fix|chore] [message]`. -->
<!-- You can skip the changelog check by labeling the PR with "no changelog". -->
# Why
https://exponent-internal.slack.com/archives/C06FK950085/p1732258126230829
# How
Improve logs
# Test Plan
Tests
<!-- If this PR requires a changelog entry, add it by commenting the PR with the command `/changelog-entry [breaking-change|new-feature|bug-fix|chore] [message]`. -->
<!-- You can skip the changelog check by labeling the PR with "no changelog". -->
# Why
Add environment to `eas.schema.json`
# How
Add environment to `eas.schema.json`
<!-- If this PR requires a changelog entry, add it by commenting the PR with the command `/changelog-entry [breaking-change|new-feature|bug-fix|chore] [message]`. -->
<!-- You can skip the changelog check by labeling the PR with "no changelog". -->
# Why
https://exponent-internal.slack.com/archives/C06FK950085/p1729212611408479?thread_ts=1729186741.835629&cid=C06FK950085
# How
Resolve the default (if not set in the build profile) `environment` to use for a build based on its configuration.
Store build -> `production`
Dev client -> `development`
Other (`internal`) -> `preview`
# Test Plan
Test manually
<!-- If this PR requires a changelog entry, add it by commenting the PR with the command `/changelog-entry [breaking-change|new-feature|bug-fix|chore] [message]`. -->
<!-- You can skip the changelog check by labeling the PR with "no changelog". -->
# Why
Make `eas env` commands visible
# How
Make `eas env` commands visible
# Test Plan
Tests
<!-- If this PR requires a changelog entry, add it by commenting the PR with the command `/changelog-entry [breaking-change|new-feature|bug-fix|chore] [message]`. -->
<!-- You can skip the changelog check by labeling the PR with "no changelog". -->
# Why
Deprecate old secret commands
# How
Mark as hidden + print warn when used
# Test Plan
Tests
# Why
This PR computes a fingerprint for each update.
- If the update does not use the fingerprint policy, it calculates the fingerprint using the exported cli in the `expo` package for projects SDK 52+.
- If the update uses the fingerprint policy, it just uses the fingerprint that was previously calculated using `expo-updates`.
# How
- Modified fingerprint computation to handle platform-specific cases
- Updated the fingerprint data structure to use `fingerprintHash` instead of the full fingerprint object
- Added support for computing fingerprints for projects without expo-updates
- Introduced new types and interfaces for handling fingerprint information groups
- Created utility functions for async map operations and fingerprint calculations
- Updated GraphQL schema to support new fingerprint-related fields
# Test Plan
- [x] on build: verify fingerprints till work
- [x] on update: verify fingerprint is calculated for new project with SDK 52
- [x] on update: verify no fingerprint is calculated for old project (< SDK 52)
- [x] on update: verify fingerprint is reused for project with fingerprint runtime policy
<!-- If this PR requires a changelog entry, add it by commenting the PR with the command `/changelog-entry [breaking-change|new-feature|bug-fix|chore] [message]`. -->
<!-- You can skip the changelog check by labeling the PR with "no changelog". -->
# Why
Make class names more consistent with other commands
Improve descriptions
# How
Make class names more consistent with other commands
Improve descriptions
# Test Plan
Tests
<!-- If this PR requires a changelog entry, add it by commenting the PR with the command `/changelog-entry [breaking-change|new-feature|bug-fix|chore] [message]`. -->
<!-- You can skip the changelog check by labeling the PR with "no changelog". -->
# Why
Stick to account-wide everywhere for now
# How
Rename shared to account-wide, fix some bugs and type issues
# Test Plan
Tests
# Why
Adds `eas workflow:create` to make it faster to create workflow files in a project. This command creates a file like **.eas/workflows/workflow.yml**. The user can choose the file name or can provide it as an arugment.
# Screenshots
<img width="552" alt="Screenshot 2024-11-08 at 8 50 58 AM" src="https://github.com/user-attachments/assets/5cf82e7e-cce6-4098-b291-fccdac25f103">
<img width="584" alt="Screenshot 2024-11-08 at 8 51 27 AM" src="https://github.com/user-attachments/assets/d6f6e7cb-18c1-420e-b52a-5bdd42446715">
<img width="617" alt="Screenshot 2024-11-08 at 8 52 04 AM" src="https://github.com/user-attachments/assets/81cc79cf-afb7-4855-bfce-0679f98655fa">
# Test Plan
Make sure that you can run the command as pictured above and that it correctly creates the workflow file. Also make sure that the unit tests pass.
# Future
Once we have docs for workflows that are public, I'd like to put a link to the docs in the default file this command creates. For now, I am just putting Hello World as a reasonable placeholder.
Co-authored-by: Szymon Dziedzic <[email protected]>
Co-authored-by: Stanisław Chmiela <[email protected]>
<!-- If this PR requires a changelog entry, add it by commenting the PR with the command `/changelog-entry [breaking-change|new-feature|bug-fix|chore] [message]`. -->
<!-- You can skip the changelog check by labeling the PR with "no changelog". -->
# Why
Going to release soon.
# How
Removed `hidden = true` overrides.
# Test Plan
`easd --help` printed out information about workflow commands.
```
~/Developer/test-local (main [$!]) easd --help
EAS command line tool
VERSION
eas-cli/13.1.1 darwin-arm64 node-v22.4.0
USAGE
$ eas [COMMAND]
TOPICS
...
workflow manage workflows
```
```
~/Developer/test-local (main [$!]) easd workflow --help
manage workflows
USAGE
$ eas workflow:COMMAND
COMMANDS
workflow:run Run an EAS workflow
workflow:validate validate a workflow configuration yaml file
```
<!-- If this PR requires a changelog entry, add it by commenting the PR with the command `/changelog-entry [breaking-change|new-feature|bug-fix|chore] [message]`. -->
<!-- You can skip the changelog check by labeling the PR with "no changelog". -->
# Why
- running workflows requires GitHub connection
- workflow config may be a valid YAML, but invalid config
We want to check both of these:
- in `workflow:validate`
- before we upload project sources to GCS in `workflow:run`.
# How
I added a `validateWorkflowRevisionYamlConfig` mutation I'm using here to validate the config _and_ GitHub connection. The GitHub requirement is checked dynamically in `www` so we can later remove it without having users upgrade to a newer `eas-cli`.
Brent also wished `eas workflow:run test.yml` resolved `test.yml` to `.eas/workflows/test.yml` automatically, so added this too.
# Test Plan
I tried validating and running invalid YAMLs, workflows and valid configs. Things seemed to work ok.
<!-- If this PR requires a changelog entry, add it by commenting the PR with the command `/changelog-entry [breaking-change|new-feature|bug-fix|chore] [message]`. -->
<!-- You can skip the changelog check by labeling the PR with "no changelog". -->
# Why
During dogfooding it came out that this flag is too long and confusing
# How
Just use `environment`
# Test Plan
Tests, types
<!-- If this PR requires a changelog entry, add it by commenting the PR with the command `/changelog-entry [breaking-change|new-feature|bug-fix|chore] [message]`. -->
<!-- You can skip the changelog check by labeling the PR with "no changelog". -->
# Why
Counterpart of https://github.com/expo/eas-cli/pull/2677, applied to `submit:internal`.
# How
Copied the solution from `build:internal`.
# Test Plan
None.
<!-- If this PR requires a changelog entry, add it by commenting the PR with the command `/changelog-entry [breaking-change|new-feature|bug-fix|chore] [message]`. -->
<!-- You can skip the changelog check by labeling the PR with "no changelog". -->
# Why
We need to use `build:internal` to resolve build configuration in GCS workflow jobs which have no Git repository.
# How
I'm going to be adding `EAS_NO_VCS=1` to GCS workflow jobs. This should prevent `build:internal` from requiring Git repository.
# Test Plan
I have verified `EAS_BUILD_ID=d7fb9f5c-dce0-40e2-b459-cbc05ba1ed77 EAS_NO_VCS=1 easd build:internal -p android` prints the JSON output on my computer.
<!-- If this PR requires a changelog entry, add it by commenting the PR with the command `/changelog-entry [breaking-change|new-feature|bug-fix|chore] [message]`. -->
<!-- You can skip the changelog check by labeling the PR with "no changelog". -->
# Why
We want to allow users to start workflows manually.
# How
I added a `createRunMutation` to GraphQL and used it here. We upload project archive, `eas.json` and use to start a new workflow run.
# Test Plan
```
Workflows are in beta and subject to breaking changes.
Compressing project files and uploading to EAS. Learn more: https://expo.fyi/eas-build-archive
✔ Uploaded eas.json to EAS
✔ Uploaded project archive to EAS 1s
✔ Workflow started successfully. https://staging.expo.dev/accounts/sjchmiela/projects/staging-app/workflows/019301f1-b1b4-700d-9e3b-82dc97d76a32
```
Co-authored-by: Szymon Dziedzic <[email protected]>
<!-- If this PR requires a changelog entry, add it by commenting the PR with the command `/changelog-entry [breaking-change|new-feature|bug-fix|chore] [message]`. -->
<!-- You can skip the changelog check by labeling the PR with "no changelog". -->
# Why
update expo packages to SDK 51 versions
revert https://github.com/expo/eas-cli/pull/2524
# How
update expo packages to SDK 51 versions
# Test Plan
tests
<!-- If this PR requires a changelog entry, add it by commenting the PR with the command `/changelog-entry [breaking-change|new-feature|bug-fix|chore] [message]`. -->
<!-- You can skip the changelog check by labeling the PR with "no changelog". -->
# Why
https://exponent-internal.slack.com/archives/C02123T524U/p1724970663050089?thread_ts=1724924614.773089&cid=C02123T524Uhttps://exponent-internal.slack.com/archives/C5ERY0TAR/p1725002816077089
Generally `@expo/config`, `@expo/prebuild-config`, and `@expo/config-plugins` are per SDK concept. Currently, in EAS CLI, we rely on one version of the `@expo/...` package to work well globally for all SDKs. It seems to sometimes bite us like in the case of https://exponent-internal.slack.com/archives/C02123T524U/p1724909908574979, because SDK 51 config wasn't compatible with SDK 50.
It's better to switch to using config-related functions shipped with `@expo/config`, `@expo/prebuild-config`, and `@expo/config-plugins`. I believe we can assume that `getConfig`, `getPrebuildConfig`, and `compileModsAsync` shipped with these packages have stable API https://exponent-internal.slack.com/archives/C5ERY0TAR/p1725293002128849?thread_ts=1725002816.077089&cid=C5ERY0TAR.
# How
Use `getConfig`, `getPrebuildConfig`, and `compileModsAsync` from `@expo/config`, `@expo/prebuild-config`, and `@expo/config-plugins` shipped with Expo SDK installed in the user's project. If not available, fallback to versions installed per SDK. If we can assume that these guarantee stable API we can use types of these functions from packages installed for EAS CLI as well.
# Test Plan
Run builds for repro case of https://exponent-internal.slack.com/archives/C02123T524U/p1724909908574979 with SDK 51 config packages installed for EAS CLI (that previously broke it), see that it works and capabilities are correctly synced.
Initiate a new bare RN project using community CLI, and check that it works as well using fallback config loading.
<!-- If this PR requires a changelog entry, add it by commenting the PR with the command `/changelog-entry [breaking-change|new-feature|bug-fix|chore] [message]`. -->
<!-- You can skip the changelog check by labeling the PR with "no changelog". -->
# Why
Add support for `with-eas-environment-variables-set` flag in `eas update:configure` command
# How
Use the flag the same way we do for `eas update`.
# Test Plan
Run command.
<!-- If this PR requires a changelog entry, add it by commenting the PR with the command `/changelog-entry [breaking-change|new-feature|bug-fix|chore] [message]`. -->
<!-- You can skip the changelog check by labeling the PR with "no changelog". -->
# Why
Make the `with-eas-environment-variables-set` flag reusable and improve type safety and description.
# How
Extract `with-eas-environment-variables-set` to separate file. Improve validation and type safety.
# Test Plan
Run command locally.
<!-- If this PR requires a changelog entry, add it by commenting the PR with the command `/changelog-entry [breaking-change|new-feature|bug-fix|chore] [message]`. -->
<!-- You can skip the changelog check by labeling the PR with "no changelog". -->
# Why
Add new Xcode 16.1 image and `sdk-52` tag to `eas.schema.json`
# How
Add new Xcode 16.1 image and `sdk-52` tag to `eas.schema.json`
# Test Plan
Tests
# Why
[ENG-13930: Allow passing environment without the flag](https://linear.app/expo/issue/ENG-13930/allow-passing-environment-without-the-flag)
Added capability to pass `environment` without `--environment` flag to `eas env:*` commands. This is often more intuitive.
This works only for a single environments. In case user wants to pass more than one, `--environment` flag should still be used.
# Changed commands
## `env:create`
* was only `env:create --name thisisvar --value thisinew --visibility plaintext --environment production`
* now also `env:create production --name thisisvar --value thisinew --visibility plaintext`
## `env:delete`
* was only `env:delete --variable-environment production --variable-name NEW_VAR`
* now also `env:delete production --variable-name NEW_VAR`
## `env:exec`
* was only `env:exec --environment development 'echo lol'`
* now also `eas env:exec development 'echo lol'`
## `env:get`
* was only `env:get --variable-environment development --variable-name HELLO`
* now also `env:get development --variable-name HELLO`
## `eas env:link`
* was only `eas env:link --environment preview --variable-name NAME`
* now also `eas env:link preview --variable-name NAME`
## `eas env:list`
* was only `eas env:list --environment preview`
* now also `eas env:list preview`
## `eas env:pull`
* was only `eas env:pull --environment production`
* now also `eas env:pull production`
## `eas env:push`
* was only `eas env:push --environment production`
* now also `eas env:push production`
## `eas env:unlink`
* was only `eas env:unlink --environment production --variable-name NAME`
* now also `eas env:unlink production --variable-name NAME`
## `eas env:update`
* was only `eas env:update --variable-environment production --variable-name HELLO --environment development`
* now also `eas env:update production --variable-name HELLO --environment production`
# Test Plan
Updated tests
<!-- If this PR requires a changelog entry, add it by commenting the PR with the command `/changelog-entry [breaking-change|new-feature|bug-fix|chore] [message]`. -->
<!-- You can skip the changelog check by labeling the PR with "no changelog". -->
# Why
As suggested in the base PR we need a simple context focused on only getting the project ID, without specifying the server-side environment and resolving Expo config with it.
# How
Add such context. Make the `withServerSideEnvironment` argument required for standard project config context.
# Test Plan
Tests, test manually
<!-- If this PR requires a changelog entry, add it by commenting the PR with the command `/changelog-entry [breaking-change|new-feature|bug-fix|chore] [message]`. -->
<!-- You can skip the changelog check by labeling the PR with "no changelog". -->
# Why
https://exponent-internal.slack.com/archives/C013ZK4SA12/p1725644220583309
We are planning to introduce enhancements to our env vars system soon. Currently, we only allow people to keep secrets on EAS servers. Secrets are not readable outside of EAS servers so of course EAS Update doesn't work with them. In a new system, people can also store readable env vars on our servers, which opens new possibilities about how EAS Update can interact with env vars!
In this PR I'm adding support for the `--with-eas-environment-variables-set production|preview|development` flag. If used, we use the flag's value to fetch readable env vars from our servers and use it in the `expo export` command to produce an update bundle based on these env vars. If the flag is used I'm using the `EXPO_NO_DOTENV` flag to disable Expo CLI from loading env vars from `.env` files and potentially unwillingly overwriting env vars fetched from our servers for a given environment.
Users seem to face big issues with using env vars with EAS Build & Update at the moment, originating from the fact that these 2 handle different, non-compatible ways of using env vars. Env vars in the build profile and server-side secrets vs local env vars + `.env`. With this flag it will be easy to avoid envs mismatch by doing `eas update ----with-eas-environment-variables-set environment` and `eas build` with `{ ..., environment: environment, ...} in the build profile!
# How
Add a new hidden (for now) flag `--with-eas-environment-variables-set` that indicates which environment to use. If used fetch env vars from server. Merge and overwrite `process.env` with newly fetched env vars. Use them to resolve the expo config. Don't use `.env` env var loading from Expo CLI if server-side env vars are used.
# Test Plan
Test manually. Run `eas update` with and without a new flag. Check that the expo config is resolved correctly.
# Why
File variables were not handled by the `env:pull` command. They should be downloaded just like the other values.
# How
* `env:pull` saves file type variables to `.eas/.env` directory and sets its value in `.env.local` file to its file path
* `env:push` skips file type variables
# Test Plan
Tested manually
# Why
We want to allow developers to prevent updates from being served to certain builds by "pausing" update channels.
# How
- I added two new CLI commands to toggle the channel pausing state
- I added the pause status to the channel details output in the `view` and `list` commands
# Test Plan
Build the CLI locally under the `neas` alias, and run `neas channel:pause development && neas channel:view development && neas channel:resume development && neas channel:view development`:
<img width="1364" alt="Screenshot 2024-10-17 at 05 33 36" src="https://github.com/user-attachments/assets/fd557e57-bbf7-463c-8cec-1febcb2fdde8">
# Why
[ENG-13853: Prompt user to enter the environment when not provided vs fail and rerun when using `eas env:push`](https://linear.app/expo/issue/ENG-13853/prompt-user-to-enter-the-environment-when-not-provided-vs-fail-and)
If the environment is not provided, we ask the user about it in other commands, I think we can do the same here for consistency vs throwing an error and having the user rerun it
# Test Plan
Tested manually
<!-- If this PR requires a changelog entry, add it by commenting the PR with the command `/changelog-entry [breaking-change|new-feature|bug-fix|chore] [message]`. -->
<!-- You can skip the changelog check by labeling the PR with "no changelog". -->
# Why
Rely on ExpoConfig type from the `@expo/config` package in all places for consistency
# How
Rely on ExpoConfig type from the `@expo/config` package in all places for consistency
# Test Plan
Tests
# Why
User should be able to upload files to environment variables and read content of those files
# How
* When an envvar is created with flag `--type FILE`, selected file is encoded to base64 and sent to www.
* EnvVarQueries return file contents when `includeFileContent` file is passed
* `env:get` automatically include file content, `env:list` requires `--include-file-content` flag
* Updated environment display - added `environments`, `type` and `visibility`
# Test Plan
tested manually, updated tests
<!-- If this PR requires a changelog entry, add it by commenting the PR with the command `/changelog-entry [breaking-change|new-feature|bug-fix|chore] [message]`. -->
<!-- You can skip the changelog check by labeling the PR with "no changelog". -->
# Why
[ENG-13525: Update EAS-CLI for new EnvVar features](https://linear.app/expo/issue/ENG-13525/update-eas-cli-for-new-envvar-features)
Update command to use multiple environments, polish the interface.
# How
* use `variable-name` and `variable-environment` to select the variable
* Use `multiselect` and multiple `environment` parameters to allow unlinking multiple envs
All commands that were using name parameter to identify a single variable will use variable-name and variable-environment instead. This change will enable user to distinguish between name and environment that is identifying variable that is acted upon, and name and environment that is a parameter for command.
For example, when the user wants to update variable TEST that has environment production and change its name and change its environment, they can use:
eas env:update --variable-name TEST --variable-environment production --name TEST-update --environment production --environment preview
# Test Plan
Added tests
<!-- If this PR requires a changelog entry, add it by commenting the PR with the command `/changelog-entry [breaking-change|new-feature|bug-fix|chore] [message]`. -->
<!-- You can skip the changelog check by labeling the PR with "no changelog". -->
# Why
[ENG-13525: Update EAS-CLI for new EnvVar features](https://linear.app/expo/issue/ENG-13525/update-eas-cli-for-new-envvar-features)
Allow selecting multiple environments to display variables from.
# How
* Use `multiselect`, allow specifying multiple `environment` parameters
* List variables from all environments
# Test Plan
Added tests
<!-- If this PR requires a changelog entry, add it by commenting the PR with the command `/changelog-entry [breaking-change|new-feature|bug-fix|chore] [message]`. -->
<!-- You can skip the changelog check by labeling the PR with "no changelog". -->
# Why
[ENG-13525: Update EAS-CLI for new EnvVar features](https://linear.app/expo/issue/ENG-13525/update-eas-cli-for-new-envvar-features)
Add tests, polish the interface
# How
* Use `variable-name` and `variable-environment` to select the variable
* More robust selector of variable to remove ambiguity
* Use multiselect to clearly indicate which environments are linked and which are not
* Automatic `link` and `unlink` action
All commands that were using name parameter to identify a single variable will use variable-name and variable-environment instead. This change will enable user to distinguish between name and environment that is identifying variable that is acted upon, and name and environment that is a parameter for command.
For example, when the user wants to update variable TEST that has environment production and change its name and change its environment, they can use:
eas env:update --variable-name TEST --variable-environment production --name TEST-update --environment production --environment preview
# Test Plan
Added tests
<!-- If this PR requires a changelog entry, add it by commenting the PR with the command `/changelog-entry [breaking-change|new-feature|bug-fix|chore] [message]`. -->
<!-- You can skip the changelog check by labeling the PR with "no changelog". -->
# Why
[ENG-13525: Update EAS-CLI for new EnvVar features](https://linear.app/expo/issue/ENG-13525/update-eas-cli-for-new-envvar-features)
Rework the EnvironmentVariableUpdate function - user is able to alter name, and environments.
# How
* Added `--variable-name` and `--variable-environment` parameters to identify the variable to update
* Using parameters from `--name` and `--environment` to update the variable using the new `updateEnvironmentVariable` method.
All commands that were using name parameter to identify a single variable will use variable-name and variable-environment instead. This change will enable user to distinguish between name and environment that is identifying variable that is acted upon, and name and environment that is a parameter for command.
For example, when the user wants to update variable TEST that has environment production and change its name and change its environment, they can use:
eas env:update --variable-name TEST --variable-environment production --name TEST-update --environment production --environment preview
# Test Plan
Added tests
<!-- If this PR requires a changelog entry, add it by commenting the PR with the command `/changelog-entry [breaking-change|new-feature|bug-fix|chore] [message]`. -->
<!-- You can skip the changelog check by labeling the PR with "no changelog". -->
# Why
[ENG-13525: Update EAS-CLI for new EnvVar features](https://linear.app/expo/issue/ENG-13525/update-eas-cli-for-new-envvar-features)
Improve the selector for the variable to delete, rename parameter `name` to `variable-name`.
All commands that were using name parameter to identify a single variable will use variable-name and variable-environment instead. This change will enable user to distinguish between name and environment that is identifying variable that is acted upon, and name and environment that is a parameter for command.
For example, when the user wants to update variable TEST that has environment production and change its name and change its environment, they can use:
eas env:update --variable-name TEST --variable-environment production --name TEST-update --environment production --environment preview

# Test Plan
Added tests
<!-- If this PR requires a changelog entry, add it by commenting the PR with the command `/changelog-entry [breaking-change|new-feature|bug-fix|chore] [message]`. -->
<!-- You can skip the changelog check by labeling the PR with "no changelog". -->
# Why
[ENG-13525: Update EAS-CLI for new EnvVar features](https://linear.app/expo/issue/ENG-13525/update-eas-cli-for-new-envvar-features)
Polished the interface:
- user is prompted for environment only when the name is ambiguous.
- changed `name` parameter to `variable-name`.
All commands that were using `name` parameter to identify a single variable will use `variable-name` and `variable-environment` instead. This change will enable user to distinguish between `name` and `environment` that is identifying variable that is acted upon, and `name` and `environment` that is a parameter for command.
For example, when the user wants to update variable `TEST` that has environment `production` and change its name and change its environment, they can use:
`eas env:update --variable-name TEST --variable-environment production --name TEST-update --environment production --environment preview`
# Test Plan
Added tests
<!-- If this PR requires a changelog entry, add it by commenting the PR with the command `/changelog-entry [breaking-change|new-feature|bug-fix|chore] [message]`. -->
<!-- You can skip the changelog check by labeling the PR with "no changelog". -->
# Why
[ENG-13525: Update EAS-CLI for new EnvVar features](https://linear.app/expo/issue/ENG-13525/update-eas-cli-for-new-envvar-features)
Update `eas env:create` to accept multiple `--environment` parameters.
# How
* Use `EASMultiEnvironmentFlag` in `EnvironmentVariableCreate`
* Update variable display with `environment` and `updatedAt` fields
# Test Plan
Added tests
<!-- If this PR requires a changelog entry, add it by commenting the PR with the command `/changelog-entry [breaking-change|new-feature|bug-fix|chore] [message]`. -->
<!-- You can skip the changelog check by labeling the PR with "no changelog". -->
# Why
[ENG-13525: Update EAS-CLI for new EnvVar features](https://linear.app/expo/issue/ENG-13525/update-eas-cli-for-new-envvar-features)
EnvVar will support having multiple environments, CLI should support this.
# How
* Add an optional parameter `multiple` to `promptVariableEnvironmentAsync`
* Update commands using `promptVariableEnvironmentAsync`
* Add a new flag `EASMultiEnvironmentFlag`
# Test Plan
Tested manually, tests will be added in the following PRs
<!-- If this PR requires a changelog entry, add it by commenting the PR with the command `/changelog-entry [breaking-change|new-feature|bug-fix|chore] [message]`. -->
<!-- You can skip the changelog check by labeling the PR with "no changelog". -->
# Why
[ENG-13525 Update EAS-CLI for new EnvVar features](https://linear.app/expo/issue/ENG-13525/update-eas-cli-for-new-envvar-features)
Bring up to date `create` and `update` EnvVar mutations.
# How
* Use `updateEnvironmentVariable` operation instead of `createEnvironmentVariable` with an 'overwrite' option.
* Add `type` parameter
* Add `environment` parameter for shared variables
# Test Plan
Tested manually, tests will be added in the following PRs
<!-- If this PR requires a changelog entry, add it by commenting the PR with the command `/changelog-entry [breaking-change|new-feature|bug-fix|chore] [message]`. -->
<!-- You can skip the changelog check by labeling the PR with "no changelog". -->
# Why
[ENG-13525 Update EAS-CLI for new EnvVar features
](https://linear.app/expo/issue/ENG-13525/update-eas-cli-for-new-envvar-features)
Update EnvVar queries - allow quering using environments.
# How
* Added optional `environment` parameter to queries
* Fix typing of the `--environment` parameter
# Test Plan
Tested manually
* Update API functions for new upload endpoint and deploy result
* Fix listWorkerFilesAsync file type
* Add call to finalize deployment
* Update CHANGELOG
* Add retries to /deploy/finalize call
* Apply lints
<!-- If this PR requires a changelog entry, add it by commenting the PR with the command `/changelog-entry [breaking-change|new-feature|bug-fix|chore] [message]`. -->
<!-- You can skip the changelog check by labeling the PR with "no changelog". -->
# Why
https://exponent-internal.slack.com/archives/C017N0N99RA/p1726226230206999
Right now the only way to exit the `eas credentials` command is to kill the process by doing ctrl+c. Let's add an `exit` option to the interactive menu.
# How
Add exit option to menu.
# Test Plan
Test manually
<!-- If this PR requires a changelog entry, add it by commenting the PR with the command `/changelog-entry [breaking-change|new-feature|bug-fix|chore] [message]`. -->
<!-- You can skip the changelog check by labeling the PR with "no changelog". -->
# Why
On Friday's meeting, we agreed that having both a flag and a field in the build profile for setting the environment might be too much. Additionally, we already have a `--profile` flag that has a `-e` alias which makes it even more confusing. We decided to delete the flag to avoid confusion and make it as straightforward as possible.
# How
Delete the `--environment` flag for build-related commands, rely only on the `environment` field in eas.json.
# Test Plan
Tests
<!-- If this PR requires a changelog entry, add it by commenting the PR with the command `/changelog-entry [breaking-change|new-feature|bug-fix|chore] [message]`. -->
<!-- You can skip the changelog check by labeling the PR with "no changelog". -->
# Why
Add the `eas env:exec` command. This command allows you to run any arbitrary bash command with env vars from a given environment.
# How
Add `eas env:exec` command
# Test Plan
Test manually

* refactor(eas-cli): split prompt from dev domain assignment and support it non-interactive
* docs: add changelog entry
* docs: fix changelog entry
* fix(eas-cli): use proper name in "already taken" prompt
* chore: resolve linting issue
# Why
Almost a year ago we've migrated from using a single file for submission logs to multi-file logs, accommodating need to display logs from both the worker and the scheduler that is starting the worker VM.
Companion to https://github.com/expo/universe/pull/16873.
# How
Replaced `logsUrl` with `logFiles`, adjusted code.
# Test Plan
Logs printed
<img width="1055" alt="Zrzut ekranu 2024-09-24 o 23 06 22" src="https://github.com/user-attachments/assets/6c9a45c0-6760-4629-82e3-a79d5ef99ebe">
<!-- If this PR requires a changelog entry, add it by commenting the PR with the command `/changelog-entry [breaking-change|new-feature|bug-fix|chore] [message]`. -->
<!-- You can skip the changelog check by labeling the PR with "no changelog". -->
# Why
I want to add support for submissions as part of workflows, where, when we start the job, we don't have access to the repository. So, we need to resolve submission config _somewhere_, ideally where we are isolated, can easily clone project, install dependencies and resolve dynamic config.
As we do with `eas build:internal`, I expect it will make sense to do `eas submit:internal` which will do what `eas submit` does usually — resolve all the credentials, using local filesystem or Expo servers and then fetch the _actual_ keys from Expo servers.
# How
First, I changed structure of `SubmitCommand`/`Submitter` a bit — previously, `SubmitCommand` would create the `Submitter` and call `submitAsync` _inside_ `runAsync()`. After my changes `SubmitCommand` will create _and return_ the `Submitter` and the call site will need to call `submitAsync`.
By doing it this way I was able to add a `getSubmissionInputAsync` to `BaseSubmitter` which returns all-resolved submission input, as-it-would-have-been-used-in-submission-mutation. This ensures the same behavior for `eas submit` and `eas submit:internal`.
Once we have that, `eas submit:internal` is relatively straightforward:
1. create `SubmitCommand`, get `Submitter`.
2. `getSubmissionInputAsync`.
3. fetch necessary credentials by IDs from `www` (https://github.com/expo/universe/pull/16793).
4. validate that we have it all (https://github.com/expo/eas-build/pull/443) and print the JSON.
# Test Plan
I was able to run `easd submit:internal` locally and got the JSON output.
<!-- If this PR requires a changelog entry, add it by commenting the PR with the command `/changelog-entry [breaking-change|new-feature|bug-fix|chore] [message]`. -->
<!-- You can skip the changelog check by labeling the PR with "no changelog". -->
# Why
In a submit job we're going to have an IPA/APK to submit. We don't want to resolve any bundle identifiers from the project, we can use one from that IPA/APK.
# How
Let the override env variable override bundle identifier resolution if the workflow is managed too.
# Test Plan
I used it to submit an IPA with workflows.
* refactor(eas-cli): enhance the `--environment` help description
* docs(eas-cli): add changelog entry
* docs(eas-cli): add missing periods to `eas worker:alias --help`
* refactor(eas-cli): enhance the `--environment` help description
* fix(eas-cli): hide project export date when not older than 1m
* fix(eas-cli): add back no export error
* docs(eas-cli): add changelog entry
* chore(eas-cli): add `--platform=web` to error command
* docs(eas-cli): fix changelog entry
* refactor(eas-cli): always default to static uploads and add modified time
* docs(eas-cli): add changelog entry
* chore(eas-cli): resolve linting issue
* fix(eas-cli): error when there are no deployments
* fix(eas-cli): remove unused `exp` var
* fix(eas-cli): fix json output again
* feature(eas-cli): add `worker:alias --prod` flag to promote existing deployments
This also includes some clean ups and log utility sharing
* feature(eas-cli): support `worker:promote --prod --id`
* refactor(eas-cli): rename `/serverless/deployments` -> `/hosting/deployments`
* docs(eas-cli): add changelog entry
* chore(eas-cli): resolve linting issues
* fix(eas-cli): disable ora when possibly asking for a dev domain
* chore(eas-cli): resolve typing issue
* feature(eas-cli): add `worker --export-dir` flag to select export folder
* docs(eas-cli): add changelog entry
* fix(eas-cli): use `path.join` instead of `path.resolve`
`path.resolve` throws when the path does not exist, `path.join` uses our own validation errors.
* docs(eas-cli): fix changelog entry
* refactor(eas-cli): reuse same working in `eas worker:alis` from `eas worker`
* refactor(eas-cli): reuse same result table from `eas worker`
* docs(eas-cli): update changelog entry
* refactor(eas-cli): use `Alias URL` instead of `Aliased URL`
* refactor(eas-cli): add support for both `--prod` and `--production` flags in `eas worker`
* chore: add changelog entry
* refactor(eas-cli): also tweak `--help` output
* refactor(eas-cli): move `--prod` before `--alias`/`--id` in usage
* refactor(eas-cli): clean up ora spinners and add table output
* refactor(eas-cli): simplify project type detection
* refactor(eas-cli): simplify alias and production spinners
* docs(eas-cli): update changelog entry
* chore: resolve linting issues
* refactor(eas-cli): use `Alias URL` instead of `Aliased URL`
* fix(eas-cli): mark `production` as bold in ora spinners
* fix(eas-cli): avoid adding `expo.extra` data from config plugins in `eas init`
Note that in the newer `@expo/config` package, modifying now also merges existing properties. Once thats released, we can use that instead of manually merging this data.
* docs(eas-cli): add changelog entry
* feature(eas-cli): add `worker:alias` command to assign aliases
* feature(eas-cli): add worker selection when assigning alias
* refactor(eas-cli): clean up comments and prompt for alias
* refactor(eas-cli): clean up command and flag descriptions
* chore: resolve linting issues
* docs: add changelog entry
* Add `@expo/env`
* Source env variables from ".env" via `@expo/env` to deployment
* Add CHANGELOG
* Increase requested range for `@expo/env`
* Apply lints
* Added --non-interactive and --force support when --id is not passed to eas init command
* Removed id dependency from init command flags
* Bump undici from 5.19.1 to 5.26.3
Bumps [undici](https://github.com/nodejs/undici) from 5.19.1 to 5.26.3.
- [Release notes](https://github.com/nodejs/undici/releases)
- [Commits](https://github.com/nodejs/undici/compare/v5.19.1...v5.26.3)
---
updated-dependencies:
- dependency-name: undici
dependency-type: indirect
...
Signed-off-by: dependabot[bot] <[email protected]>
* Revert "Bump undici from 5.19.1 to 5.26.3"
* Apply suggestions from code review
Co-authored-by: Szymon Dziedzic <[email protected]>
* Fixed initializeWithoutExplicitIDAsync not being used
* Require force flag to use default account when no account name is specified in non-interactive mode
* Changed error message when user has access to multiple accounts in non-interactive mode
* Added changelog entry
* Update CHANGELOG.md
Co-authored-by: Szymon Dziedzic <[email protected]>
---------
Signed-off-by: dependabot[bot] <[email protected]>
Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
Co-authored-by: Ryan Button <[email protected]>
Co-authored-by: Szymon Dziedzic <[email protected]>
* [fix]: correctly retrieve entitlements for extensions using their own target name
Fixed an issue where extensions were incorrectly retrieving the main app's entitlements instead of their own. Updated the code to use `dependency.name` for fetching the correct entitlements.
* update CHANGELOG.md
---------
Co-authored-by: Harris Han <[email protected]>
<!-- If this PR requires a changelog entry, add it by commenting the PR with the command `/changelog-entry [breaking-change|new-feature|bug-fix|chore] [message]`. -->
<!-- You can skip the changelog check by labeling the PR with "no changelog". -->
# Why
companion to https://github.com/expo/third-party/pull/104
# How
This is my first PR adding new capabilities so please double-check if I did everything correctly, I based it on 6389f52
I got new entitlements from https://developer.apple.com/documentation/bundleresources/entitlements?changes=latest_major and added all available capabilities corresponding to these.
# Test Plan
Run prod build manually, add new entitlements for new capabilities to app config, and verify that they were enabled on the provisioning profile
<!-- If this PR requires a changelog entry, add it by commenting the PR with the command `/changelog-entry [breaking-change|new-feature|bug-fix|chore] [message]`. -->
<!-- You can skip the changelog check by labeling the PR with "no changelog". -->
# Why
https://exponent-internal.slack.com/archives/C02123T524U/p1724909908574979
It seems like SDK 51 config-related packages update broke builds for older SDKs
# How
I believe we didn't add any new config-related features in SDK 51, so resolving the config using the package version for SDK 50 should be ok for now and shouldn't cause any trouble.
Revert #2498
# Test Plan
Tests
<!-- If this PR requires a changelog entry, add it by commenting the PR with the command `/changelog-entry [breaking-change|new-feature|bug-fix|chore] [message]`. -->
<!-- You can skip the changelog check by labeling the PR with "no changelog". -->
# Why
https://github.com/expo/eas-cli/issues/2519
I was able to reproduce the issue and it is indeed true. It seems like we mistakenly added a static project config definition to the `eas build` command context in https://github.com/expo/eas-cli/commit/56510f092527e458aa9dca19feac0b5d425dac56#diff-cd2e00f4d286a4714f8736e14c635f4c3b98c402367cd71a83e4cd743034bd79. The output from it wasn't really used anywhere, but adding it caused the app configs to be evaluated without `buildProfile.env` in the first pass when trying the resolve project ID and slug at the begging of command execution and our slug/project ID assertions to fail. It only affected people with dynamic project IDs and slug fields.
# How
Remove unnecessary static context from context definitions, therefore avoid resolving project ID and slug from app config without env vars as a first thing when the `eas build` command is executed.
# Test Plan
Tested manually
* Add fingerprint sources to build metadata
* Upload fingerprint to google cloud
* Rename fingerprintGCSLocation -> fingerprintGCSBucketKey
* Add fingerprint source in an object
* Update dependencies
* Add comment about fingerprintSources
* Add a changelog entry
* Add support for local build plugin
* Add safety check that a temp directory exists
* [eas-cli] make sure we filter out any comments in `build.gradle` file before passing to to `gradle-to-js` parser
* use trim
* add test
* add changelog
* chore: update node versions to test with to 18, 20, 22
* update CHANGELOG.md
* chore(eas-cli)!: bump minimum supported Node version to 18
This makes the PR a breaking change.
* docs(eas-cli): add breaking change(log) entry
* chore: update `engines.node` to 18 everywhere
<!-- If this PR requires a changelog entry, add it by commenting the PR with the command `/changelog-entry [breaking-change|new-feature|bug-fix|chore] [message]`. -->
<!-- You can skip the changelog check by labeling the PR with "no changelog". -->
# Why
We want to save the relationship between a job run and an update.
# How
See https://github.com/expo/universe/pull/15329.
# Test Plan
Tested with staging.
<!-- If this PR requires a changelog entry, add it by commenting the PR with the command `/changelog-entry [breaking-change|new-feature|bug-fix|chore] [message]`. -->
<!-- You can skip the changelog check by labeling the PR with "no changelog". -->
# Why
I noticed logs look off in CI when running `eas update --auto`.
<img width="1000" alt="Zrzut ekranu 2024-05-10 o 17 50 10" src="https://github.com/expo/eas-cli/assets/1151041/ed50d014-62f1-4342-bde2-0673f49d52b4">
# How
I pinpointed it to Ora being restarted on every log.
# Test Plan
Ran `CI=1 easd update --auto` and got ⬇️ which I think is much more readable.
```
CI=1 easd update --auto
- Exporting...
[expo-cli] Starting Metro Bundler
[expo-cli] Android Bundling complete 3492ms
[expo-cli] iOS Bundling complete 3493ms
[expo-cli] iOS Building Hermes bytecode for the bundle
[expo-cli] Android Building Hermes bytecode for the bundle
[expo-cli]
[expo-cli] Bundle Size
[expo-cli] ┌ index.ios.hbc 858 kB
[expo-cli] ├ index.android.hbc 864 kB
[expo-cli] ├ index.ios.hbc.map 3.45 MB
[expo-cli] └ index.android.hbc.map 3.48 MB
[expo-cli]
[expo-cli] 💡 JavaScript bundle sizes affect startup time. Learn more: https://expo.fyi/javascript-bundle-sizes
[expo-cli] Finished saving JS Bundles
[expo-cli] Dumping asset map
[expo-cli] Dumping source maps
[expo-cli] Preparing additional debugging files
[expo-cli] Export was successful. Your exported files can be found in dist
✔ Exported bundle(s)
- Uploading...
✔ Uploaded 2 app bundles
✔ Uploading assets skipped - no new assets found
ℹ 1 iOS asset, 1 Android asset (maximum: 20000 total per update). Learn more about asset limits: https://expo.fyi/eas-update-asset-limits
- Publishing...
✔ Published!
```
<!-- If this PR requires a changelog entry, add it by commenting the PR with the command `/changelog-entry [breaking-change|new-feature|bug-fix|chore] [message]`. -->
<!-- You can skip the changelog check by labeling the PR with "no changelog". -->
# Why
Implementing @tchayen's suggestion.
# How
Changed wording
# Test Plan
None.
* [eas-cli] add eas onboarding flow
* mark command as hidden and prompt user before starting any action
* add gql queries
* regenerate gql schema
* fix type issues
* mark isCLIDone as true
* apply suggested changes
* improvements after cr
* more improvements
* [eas-cli] don't allow to specify custom global version of deprecated `expo-cli`
* update CHANGELOG.md
* update dependencies
* print a warning vs throwing an error
* mark as deprecated
* print warning nicer
* fix changelog entry
* fix changelog
* [eas-cli] don't prompt to set up push notifications if `promptToConfigurePushNotifications` is `undefined` and `expo-notifications` is not present
* update eas.schema.json
* update CHANGELOG.md
* [eas-cli] improve error message displayed when EAS CLI version doesn't satisfy minimal version required specified in `eas.json`
* fix typo
* update CHANGELOG.md
* [eas-cli] don't prompt user to set `android.package` and `ios.bundleIdentifier` value when runing in non-interactive mode
* update CHANGELOG.md
* rename test
* [ci] bump `@expo/eas-json` version in universe after the release
* Update .github/workflows/release.yml
* test
* edit eas.json
* Revert "edit eas.json"
This reverts commit 3979a1b1fb.
* Revert "test"
This reverts commit 69db39f548.
* [eas-cli] mark `default` image tag as deprecated
* add new image tags to `eas.schema.json`
* update CHANGELOG.md
* update CHANGELOG.md
* deprecate JDK 8 and NDK r19c images
* update CHANGELOG.md
* use better warn message for Xcode < 14 images and link to our changelog post
* update warn message
* don't deprecate Androdi images in this PR
* fix changelog
* use correct latest image
* [eas-cli] create new `init-build` command under the `credentials` topic
* get hacky version of cli command working
* add all option to flag
* extend the Manage<Platform> actions to reduce code duplication
* prompt user for platform if it doesn't exist
* prompt the user to choose a profile
* add a changelog entry
* call SetUpBuildCredentials for android
* [eas-cli] bump `@expo/apple-utils` to switch between `developer.apple.com` and `developer-mdn.apple.com` domains when one of them is not working
* update CHANGELOG.md
* bump version
* [eas-cli] send a path to custom build config to EAS Build server with posix separator for Windows
* add process.platform to if statement
* update CHANGELOG.md
* apply suggestions from cr
* remove unnecessary check
* apply suggested change
# Why
We want users to feel like they're adding something we support.
# How
Added `config` property to build profile level and platform-specific profile properties levels.
Also prettified the file according to my JSON Tools VS Code extension. Let me know if it should be deprettified.
# Test Plan
None.
* [fcmv1] Manage FCM V1 Google Service Account Key in CLI
# Why
NOTE: DO NOT LAND THIS UNTIL GRAPHQL CHANGES LAND IN WWW
We're adding support for FCM V1 credentials, as Google is shutting down
the FCM Legacy API for sending Android notifications in June.
# How
Add new prompts and refactor existing prompts to match this schematic:
# Test Plan
Verified all functionality e2e:
* update CHANGELOG.md
<!-- If this PR requires a changelog entry, add it by commenting the PR with the command `/changelog-entry [breaking-change|new-feature|bug-fix|chore] [message]`. -->
<!-- You can skip the changelog check by labeling the PR with "no changelog". -->
# Why
It is currently not possible to use `if: ${ always() }` in Custom Build workflows due to outdated `@expo/steps` dependency.
# How
Added a GitHub workflow, similar to the one we have in `turtle-v2`, which will automatically upgrade all dependencies from the `eas-build` repo.
# Test Plan
https://github.com/expo/eas-cli/pull/2220
* [ci] add update eas cli used on production for builds triggered with GitHub trigger once new version is published
* remove unnecessary docs from releasing
<!-- If this PR requires a changelog entry, add it by commenting the PR with the command `/changelog-entry [breaking-change|new-feature|bug-fix|chore] [message]`. -->
<!-- You can skip the changelog check by labeling the PR with "no changelog". -->
# Why
Simplification, unification.
# How
Removed bare-specific template.
# Test Plan
None.
* test against node 20 and drop support for node 14
* update CHANGELOG.md
* update @types/node
* bump yarn version
* Revert "update @types/node"
This reverts commit 027e26b002.
* update @types/node
* try to debug
* Revert "try to debug"
This reverts commit 0d557d02f8.
* test sth
* update lock
* Revert "update lock"
This reverts commit 184b412655.
* Revert "test sth"
This reverts commit 77606cf5be.
* diff once again
* fix issues
* Revert "diff once again"
This reverts commit 526ab57453.
* Revert "[ENG-8956][eas-build] Improve EAS CLI tsconfig.json for local development (#1933)"
This reverts commit e963558ef6.
* add start-allow-unused option
* update CONTRIBUTING.md
* fix type error
<!-- If this PR requires a changelog entry, add it by commenting the PR with the command `/changelog-entry [breaking-change|new-feature|bug-fix|chore] [message]`. -->
<!-- You can skip the changelog check by labeling the PR with "no changelog". -->
# Why
x-ref https://github.com/expo/eas-build/pull/302
# How
Upgraded `@expo/eas-build-job` and fixed the only TS error.
# Test Plan
None.
<!-- If this PR requires a changelog entry, add it by commenting the PR with the command `/changelog-entry [breaking-change|new-feature|bug-fix|chore] [message]`. -->
<!-- You can skip the changelog check by labeling the PR with "no changelog". -->
# Why
I've noticed tests ran for an old commit even though I pushed a new one.
# How
Added `concurrency` where I thought it was applicable.
# Test Plan
None.
<!-- If this PR requires a changelog entry, add it by commenting the PR with the command `/changelog-entry [breaking-change|new-feature|bug-fix|chore] [message]`. -->
<!-- You can skip the changelog check by labeling the PR with "no changelog". -->
# Why
It's nice to be able to download both application archive and "build artifact".
# How
Added `buildArtifactsUrl` to GraphQL query. It contains the thing I'm looking for.
# Test Plan
Ran `easd build:view 7bab0c66-d917-4873-9241-c5fa0b01dec1 --json` and confirmed it has the new field.
* [eas-cli] Add context fields for vcs clients
* [eas-cli] Replace vcs client helper with new context field
* [eas-cli] Pass instance of vcs client in helpers instead of re-calling vcs helper
* update CHANGELOG.md
* [FIX] Remove semver validation from node version read from .nvmrc
Node version can be specified as an alias (such as `iojs`, `lts/hydrogen` etc), so validating it as semver may cause errors for some users.
* update CHANGELOG.md
* fix(eas-cli): implement variadic `--platform` flag when creating EAS updates
* fix(eas-cli): finish todo and set minimum `@expo/[email protected]` version for multiple `--platform` flags during export
* update CHANGELOG.md
* fix(eas-cli): be verbose about `--platform=<flag>` in bundle script
* [eas-cli] handle server side defined error when `intel-medium` resource class is not available
* don't suggest using `intel-medium` in `eas.schema.json`
* update CHANGELOG.md
* update CHANGELOG.md
* [eas-cli] allow for `rollout` option when configuring submission
* update CHANGELOG.md
* add entry to `eas.schema.json`
* improve eas.schema.json
* make rollout default to 1 if not set
* Revert "make rollout default to 1 if not set"
This reverts commit 2ab689b02c.
* generate schema against staging
* update fragment
Changed noUnusedLocals and noUnusedParameters to false for easier local development and experimenting.
These checks do not run during local development. Instead, `yarn build` catches these cases on github.
Why
---
Fail fast with a clear message if the app config contradicts itself or the developer's behavior.
How
---
Throw if useClassicUpdates is set in the app config and the developer is running an EAS Update-related command or if the app config uses an EAS Update URL. If the developer is using a static app config (app.json), offer to modify the config file for them.
Test Plan
---
Added a unit test for ensureEASUpdateIsConfiguredAsync. Manually tested as wlel.
* added alias -m for --message
* align with the other messages
* updated README
* moving change in update/republish to a seperate PR
* removed README comment
* update CHANGELOG.md
---------
Co-authored-by: Steven Pu <[email protected]>
Co-authored-by: Steven Pu <[email protected]>
* added alias in republish
* reverted change in README
* update CHANGELOG.md
---------
Co-authored-by: Steven Pu <[email protected]>
Co-authored-by: Steven Pu <[email protected]>
* [eas-cli] validate that `expo-updates` are in project dependencies when somebody tries to run a build with channel set
* apply suggested changes
* remove unnecessary change
* fix warning
* improve validation logic
* improve insatlled message
* warn when expo-updates are in dev dependencies for update command
* impreve warn messsages
* improve warning messages
* one more warn message improvement
* update CHANGELOG.md
* [eas-cli][ENG-8763] Select default android keystore
- Add option to change default keystore from CLI (previously available only from dashboard)
- Streamline setting up new keystore
Instead of overwriting existing keystore, new one is created every time. Also, option to immediately set it as default is added, since it is also easier to change it from CLI.
* update CHANGELOG.md
<!-- If this PR requires a changelog entry, add it by commenting the PR with the command `/changelog-entry [breaking-change|new-feature|bug-fix|chore] [message]`. -->
<!-- You can skip the changelog check by labeling the PR with "no changelog". -->
# Why
Cleanup after https://github.com/expo/eas-build/pull/233.
# How
Upgraded `@expo/steps` package. It is the only `eas-build` package in here.
# Test Plan
It's a type-only change so CI success will be enough.
* [eas-cli] make large resource class only available for paid users
* update eas.schema.json
* Revert "update eas.schema.json"
This reverts commit 319d998b76a4c6c87d137174d6cf9f90d64b3886.
* apply suggested changes
* update CHANGELOG.md
* [eas-cli][bugfix] Fix ios version-test
Tests were passing, but were not testing the right thing, essentially always defaulting to 'managed' workflow. `projectDir` should be set to the same value as EAS_PROJECT_ROOT.
* [eas-cli][ENG-8571] Validate CFBundleShortVersionString
* update CHANGELOG.md
* [eas-cli][eas-json] add `large` resource class for iOS as allowed value in `eas.json`
* add deprecation warning for `m-large`
* add test
* update CHANGELOG.md
* [eas-cli] don't resolve default resource class for iOS based on metadata
* don't set m1-medium as default in `eas build:configure`
* pass sdkVersion and reactNativeVersion in buildParams
* convert it to sync function
* update gql schema
* update CHANGELOG.md
* [eas-cli] allow users to select app to run if multiple apps are found in the build tarball
* rename variable
* improve error message
* apply suggested changes
* add dot
* update CHANGELOG.md
* [eas-cli] throw error if somebody tries to run iOS with `large` resource class selected through deprecated `--resource-class` flag
* apply suggested changes
* improve error message
* update CHANGELOG.md
* [eas-cli] Update apple utilities to avoid the maintenance error
* [eas-cli] Add changelog entry
* [eas-cli] Dont change the eas-cli version
* [eas-cli] Fix changelog entry
* [eas-cli] fix it again
* Unify error display method
Overriden the `Command`'s `catch` method to catch all the errors thrown by the commands, log the error message there the same way as it was previously handled in `src/build/build.ts` and re-throw generic `Error` instance with predefined base error message.
To avoid duplication, old `Log.error()` calls have been removed from `src/build/build.ts`.
Also added tests to see if the `catch()` method works as intended.
* Adjust base error messages
Base error messages have been adjusted after @dsokal suggested that the error may be caused by something different from a request and displaying message about a failed request is misleading
* update CHANGELOG.md
* Add test for unclean error message
Checks if the `[GraphQL] ` prefix gets removed when logging the error message
* Update handling GraphQL errors
Post-review update to how graphQL errors are handled
* Print different message for graphQL errors
When a graphQL error is caught a different base error message is printed
* Print original error to debug
When a graphQL error is caught and the message is cleaned, an original message is printed to debug (or just the same message otherwise)
* Add custom error classes
Created some custom error classes for known server side defined errors
* Log the error, not the message
Changed the debug log to log the error object, not the error message
* Fix
Fixed testing leftover
* Change phrasing
Edited the phrasing in default messages related to EAS Build
* Update if condition
Changed the if condition after review comment
* Create error base class
Created a new base class for the errors expected in build.ts, so they can be handled separately in `catch()`
* Move EasCommand properties
Moved the class properties above the class methods
* Remove helper method
Removed the helper method after review comment
* Change const type
Updated the type of SERVER_SIDE_DEFINED_ERRORS constant
* Change default base error message
Set the default base error message to `<command_name> command failed.`
* Remove comments
Removed leftover comments
* Change optional property
Changed empty string to optional property
* Post-review tweaks
Self-explanatory
* Fix failing test
One of tests started failing after changing the coalescing operators (|| -> ??). This fixes that test
See: https://linear.app/expo/issue/ENG-6945/improve-eas-cli-error-logging
* Remove error class
Removed `GenericGraphQLError` class after @wkozyra95 comment, since atm there's no need to handle it differently than a generic `Error`
See: https://linear.app/expo/issue/ENG-6945/improve-eas-cli-error-logging
* Move `EasCommandError` base class
Moved `EasCommandError` base class after @wkozyra95 comment. The class is not specific to `Build` command and so it has been moved to `commandUtils` directory
See: https://linear.app/expo/issue/ENG-6945/improve-eas-cli-error-logging
* Make message mandatory
Made `message` parameter in `EasCommandError`s mandatory, to remove need for providing a fallback string and making the server the source of truth for the errors messages
See: https://linear.app/expo/issue/ENG-6945/improve-eas-cli-error-logging
* Update packages/eas-cli/src/commandUtils/EasCommand.ts
Co-authored-by: Wojciech Kozyra <[email protected]>
* Move `baseGraphQLErrorMessage` property
Made the property into a constant in a global scope, since atm it is not changed in any of the subclasses
See: https://linear.app/expo/issue/ENG-6945/improve-eas-cli-error-logging
* Suppress ESLint warning
ESLint erroneously complains about the constructor of `EasCommandError` to be useless, which is not true is the constructor of `Error` parent class allows for optional `message` parameter and the constructor of `EasCommandError` does not
See: https://linear.app/expo/issue/ENG-6945/improve-eas-cli-error-logging
* Add error handling tests
Added a bunch of new tests related to the `handleBuildRequestError` function
See: https://linear.app/expo/issue/ENG-6945/improve-eas-cli-error-logging
* Directly export `handleBuildRequestError`
Dropped the `testExports` convention after team discussion
See: https://linear.app/expo/issue/ENG-6945/improve-eas-cli-error-logging
---------
Co-authored-by: Wojciech Kozyra <[email protected]>
* [eas-cli] link to build phase in error message
* add better fallback
* use fallback for all links
* use terminalLink.isSupported
* improvements after code review
* use old fallback
* update CHANGELOG.md
* [eas-cli] display build message data when picking build in `submit`, `build:run` and `build:resign` commands
* display only 200 first characters of message if it exists
* print only defined values
* add '...' to message if trimmed
* use array instead of object
* update CHANGELOG.md
* [eas-cli] fix issue with `eas.json` envs not available when resolving local config
* fix type definitions to prevent it from happening again
* update CHANGELOG.md
* [eas-cli] set `m1-medium` resource class for builds which don't specify resource class for SDK 48 and RN 0.71
* refactor resolveBuildResourceClassAsync function
* apply suggested changes
* drop warning for ios large resource classes
* call getDynamicProjectConfigAsync in create context function
* update CHANGELOG.md
<!-- Thanks for contributing to _EAS CLI_! Before you submit your pull request, please make sure to check the following boxes by putting an x in the [ ] (don't: [x ], [ x], do: [x]) -->
# Checklist
- [ ] I've added an entry to [CHANGELOG.md](https://github.com/expo/eas-cli/blob/main/CHANGELOG.md#main) if necessary. You can comment this pull request with `/changelog-entry [breaking-change|new-feature|bug-fix|chore] [message]` and CHANGELOG.md will be updated automatically.
<!-- If this PR requires a changelog entry, add it by commenting the PR with the command `/changelog-entry [breaking-change|new-feature|bug-fix|chore] [message]`. -->
<!-- You can skip the changelog check by labeling the PR with "no changelog". -->
message:${{ contains(github.event.pull_request.labels.*.name, 'no changelog') && '⏩ The changelog entry check has been skipped since the "no changelog" label is present.' || '✅ Thank you for adding the changelog entry!' }}
message-failure:|
❌ It looks like a changelog entry is missing for this PR. Add it manually to CHANGELOG.md.
⏩ If this PR doesn't require a changelog entry, such as if it's an internal change that doesn't affect the user experience, you can add the "no changelog" label to the PR.
# packages are ordered according to output of `lerna list --toposort`
# https://github.com/expo/turtle-v2/pull/568
for PACKAGE in "@expo/downloader" "@expo/eas-build-job" "@expo/logger" "@expo/template-file" "@expo/steps" "@expo/turtle-spawn" "@expo/build-tools"; do
Set up an alias for the EAS CLI so you can try it in projects all around your computer. The project is compiled on the fly so you don't need to run a build (watch) command.
## Getting Started
> The only exception is when you just cloned the repository or there have been changes in `packages/eas-json`. In that case, you'll have to run `yarn build` in the root.
If this is your first time running `eas-cli` locally, make sure you install and build the packages. In the project root, run:
```sh
yarn install
yarn build
```
From there, set up an alias for the EAS CLI so you can try it in projects all around your computer. The project is compiled on the fly so you don't need to run a build (watch) command.
> The only exception is if there have been changes in `packages/eas-json`. In that case, you'll have to run `yarn build` in the root.
Open your `.zshrc` or other config file and add:
```
```sh
aliaseasd="/PATH/TO/eas-cli/bin/run"
```
@@ -14,10 +23,29 @@ Then use it with `easd` like `easd build`.
If you're making changes to `packages/eas-json` or prefer/need to work with production code, start the build command in watch mode with:
```
```sh
yarn start
```
If it is easier for you to work with the code locally with compiler options set to:
```json
...
"compilerOptions":{
...
"noUnusedLocals":false,
"noUnusedParameters":false,
...
}
...
```
you can consider using:
```sh
yarn start-allow-unused
```
## Format
- Be sure to update the [`CHANGELOG.md`](./CHANGELOG.md) with changes for every PR. There is a changelog bot that can generate the proper entry for you. The instructions on how to use it are in the PR description placeholder.
1.Run `yarn release` in the repository root folder. The next version is chosen automatically based on the changelog entries. If you want to use different version, pass the version string as a positional argument to the command, e.g. `yarn release 1.2.3`.
1.Invoke the GitHub workflow ["Manually trigger a release"](https://github.com/expo/eas-cli/actions/workflows/trigger-release.yml). The next version is chosen automatically based on the changelog entries. If you want to use different version, pass the version string as an input to the workflow, in semver format, e.g. "1.2.3"
2. That's it! GitHub Actions is going to take care of the rest. Watch the #eas-cli Slack channel for a successful release notification.
This file provides guidance for working with the **eas-cli** package.
## Overview
The main CLI tool for Expo Application Services (EAS). Published as `eas-cli` on npm.
- **95+ commands** organized by domain
- Built on **oclif** (Open CLI Framework)
- Entry point: `bin/run`
## Development Commands
### Running Locally
```bash
# From repository root:
yarn eas <command>
# Or use the binary directly:
packages/eas-cli/bin/run <command>
# Recommended alias for development:
aliaseasd="$(pwd)/packages/eas-cli/bin/run"
easd build --help
```
### Build & Development
```bash
yarn build # TypeScript compilation (src → build)
yarn watch # Watch mode
yarn watch-allow-unused # Watch mode with relaxed TS rules
yarn typecheck # Type check without building
yarn typecheck-for-build # Type check for production build
```
### Testing
```bash
yarn test# Run all tests
yarn test <path-to-file> # Run specific test file
```
### GraphQL Code Generation
```bash
yarn generate-graphql-code # Generate TypeScript types from GraphQL schema
yarn verify-graphql-code # Verify GraphQL code is up to date (CI check)
```
The GraphQL schema is fetched from `https://staging-api.expo.dev/graphql` and generates:
-`src/graphql/generated.ts` - 499KB+ of TypeScript types
- Types for all queries, mutations, and fragments
### oclif Commands
```bash
yarn oclif readme # Auto-generate README.md from command metadata
```
## Command Architecture
### Base Command Class
All commands extend `EasCommand` (in `src/commandUtils/EasCommand.ts`).
### Context System
Commands declare required context via `ContextOptions`. This is a powerful system that provides commands with pre-initialized resources based on their needs.
"description":"When parents visit the Kids category on the App Store, they expect the apps they find will protect their children’s data, provide only age-appropriate content, and require a parental gate in order to link out of the app, request permissions, or present purchasing opportunities. It’s critical that no personally identifiable information or device information be transmitted to third parties, and that advertisements are human-reviewed for age appropriateness in order to be displayed.\n@see https://developer.apple.com/news/?id=091202019a",
"markdownDescription":"When parents visit the Kids category on the App Store, they expect the apps they find will protect their children’s data, provide only age-appropriate content, and require a parental gate in order to link out of the app, request permissions, or present purchasing opportunities. It’s critical that no personally identifiable information or device information be transmitted to third parties, and that advertisements are human-reviewed for age appropriateness in order to be displayed.\n\n[Learn more](https://developer.apple.com/news/?id=091202019a)"
"description":"When parents visit the Kids category on the App Store, they expect the apps they find will protect their children’s data, provide only age-appropriate content, and require a parental gate in order to link out of the app, request permissions, or present purchasing opportunities. It’s critical that no personally identifiable information or device information be transmitted to third parties, and that advertisements are human-reviewed for age appropriateness in order to be displayed.\n@see https://developer-mdn.apple.com/news/?id=091202019a",
"markdownDescription":"When parents visit the Kids category on the App Store, they expect the apps they find will protect their children’s data, provide only age-appropriate content, and require a parental gate in order to link out of the app, request permissions, or present purchasing opportunities. It’s critical that no personally identifiable information or device information be transmitted to third parties, and that advertisements are human-reviewed for age appropriateness in order to be displayed.\n\n[Learn more](https://developer-mdn.apple.com/news/?id=091202019a)"
},
"seventeenPlus":{
"type":"boolean",
"deprecated":true,
"deprecationMessage":"This property is deprecated, use `ageRatingOverride` instead.",
"description":"If your app rates 12+ or lower, and you believe its content may not be suitable for children under 17, you can manually set the age rating to 17+.\n@see https://help.apple.com/app-store-connect/#/dev599d50efb",
"markdownDescription":"If your app rates 12+ or lower, and you believe its content may not be suitable for children under 17, you can manually set the age rating to 17+.\n\n[Learn more](https://help.apple.com/app-store-connect/#/dev599d50efb)"
"description":"If your app rates 12+ or lower, and you believe its content may not be suitable for children under 17, you can manually set the age rating to 17+.\n\nIf your app will be distributed through alternative app marketplaces or websites in the European Union and you believe your app is [Unrated](https://developer.apple.com/help/app-store-connect/reference/age-ratings), set this to `UNRATED`.\n\n@see https://help.apple.com/app-store-connect/#/dev599d50efb",
"markdownDescription":"If your app rates 12+ or lower, and you believe its content may not be suitable for children under 17, you can manually set the age rating to 17+.\n\nIf your app will be distributed through alternative app marketplaces or websites in the European Union and you believe your app is [Unrated](https://developer.apple.com/help/app-store-connect/reference/age-ratings), set this to `UNRATED`.\n\n[Learn more](https://help.apple.com/app-store-connect/#/dev599d50efb)"
"description":"If your app rates 12+ or lower, and you believe its content may not be suitable for children under 15 or 19, you can manually set the age rating to 15+ or 19+.\n\nIf your app will be distributed through alternative app marketplaces or websites in the European Union and you believe your app is [Unrated](https://developer.apple.com/help/app-store-connect/reference/age-ratings), set this to `UNRATED`.\n\n@see https://help.apple.com/app-store-connect/#/dev599d50efb",
"markdownDescription":"If your app rates 12+ or lower, and you believe its content may not be suitable for children under 15 or 19, you can manually set the age rating to 15+ or 19+.\n\nIf your app will be distributed through alternative app marketplaces or websites in the European Union and you believe your app is [Unrated](https://developer.apple.com/help/app-store-connect/reference/age-ratings), set this to `UNRATED`.\n\n[Learn more](https://help.apple.com/app-store-connect/#/dev599d50efb)"
},
"lootBox":{
"type":"boolean",
"description":"Does your app contain purchasable loot boxes? These virtual containers provide players with in-game items, including functional cards or cosmetic modifications, based on random chance."
}
}
},
@@ -565,6 +600,20 @@
"maxLength":4000
}
}
},
"AppleAgeRatingOverride":{
"enum":[
"NONE",
"SEVENTEEN_PLUS",
"UNRATED"
]
},
"AppleKoreaAgeRatingOverride":{
"enum":[
"NONE",
"FIFTEEN_PLUS",
"NINETEEN_PLUS"
]
}
},
"AppleConfig":{
@@ -775,19 +824,5 @@
}
}
}
},
"properties":{
"configVersion":{
"enum":[0],
"description":"The EAS metadata store configuration version"
exports[`BuildVersionGetView reading version aborts when the appVersionSource is not specified and the user chooses to set it to LOCAL 1`] = `"Aborting..."`;
exports[`BuildVersionGetView reading version when appVersionSource is set to local 1`] = `"This project is not configured for using remote version source. Add [1m{"cli": { "appVersionSource": "remote" }}[22m in eas.json or re-run this command without "--non-interactive" flag."`;
test('reading version sets appVersionSource to LOCAL and aborts when the appVersionSource is not specified and is run in non-interactive mode',async()=>{
test('setting version aborts when the appVersionSource is not specified and the user chooses to set it to LOCAL, and they refuse auto configuration',async()=>{
test('setting version when the appVersionSource is not specified and the user chooses to set it to LOCAL, and they allow auto configuration',async()=>{
* @param {function} options.filterPredicate - A predicate to filter the branches that are shown to the user. It takes a branchInfo object as a parameter and returns a boolean.
* @param {string} options.printedType - The type of branch printed to the user. Defaults to 'branch'.
* @param {number} options.pageSize - The number of branches to show per page. Defaults to 100.
* @param {function} options.beforeEachFilterQuery Optional. If a filter predicate was specified, this callback function will be called before each query.
* @args externalQueryParams The query params for the pagination.
* @args totalNodesFetched The total number of nodes fetched so far.
* @args dataset The dataset so far.
* @param {function} options.afterEachFilterQuery Optional. If a filter predicate was specified, this callback function will be called after each query.
* @args externalQueryParams The query params for the pagination.
* @args totalNodesFetched The total number of nodes fetched so far.
* @args dataset The dataset so far.
* @args willFetchAgain If the query will fetch again to get a complete page.
constEXPECTED_EAS_BUILD_DOWN_MESSAGE=`EAS Build is down for maintenance. Try again later. Check ${link(
'https://status.expo.dev/'
)} for updates.`;
constEXPECTED_MAX_BUILD_COUNT_MESSAGE_ANDROID=`You have already reached the maximum number of pending Android builds for your account. Try again later.`;
constEXPECTED_MAX_BUILD_COUNT_MESSAGE_IOS=`You have already reached the maximum number of pending iOS builds for your account. Try again later.`;
constEXPECTED_GENERIC_MESSAGE=
'Build request failed. Make sure you are using the latest eas-cli version. If the problem persists, report the issue.';
`Remote versions are not configured and EAS CLI was not able to read the current version from your project. Use "eas build:version:set" to initialize remote versions.`
@@ -217,7 +230,10 @@ export async function resolveRemoteVersionCodeAsync(
@@ -362,8 +487,8 @@ async function handleSingleBuildProgressAsync(
spinner.succeed('Build finished');
return{refetch: false};
caseBuildStatus.New:
spinner.text=`Build is waiting to enter the queue. Check your concurrency limit at ${link(
formatAccountSubscriptionsUrl(accountName)
spinner.text=`Build concurrency limit reached for your account. Build will enter queue once a concurrency becomes available. Add additional concurrencies at ${link(
formatAccountBillingUrl(accountName)
)}.`;
break;
caseBuildStatus.InQueue:{
@@ -376,18 +501,16 @@ async function handleSingleBuildProgressAsync(
`The following environment variables are defined in both the "${buildProfileName}" build profile "env" configuration and the "${environment.toLowerCase()}" environment on EAS: ${overlappingKeys.join(
', '
)}. The values from the build profile configuration will be used.`
);
}
}
Log.newLine();
return{...serverEnvVars,...buildProfile.env};
}catch(e){
Log.error(`Failed to pull env variables for environment ${environment} from EAS servers`);
Log.error(e);
Log.error(
'This can possibly be a bug in EAS/EAS CLI. Report it here: https://github.com/expo/eas-cli/issues'
Loaded 100 of 798 files, more files were not shown because too many files have changed in this diff.
Show more
Reference in new issue
Block a user
Blocking a user prevents them from interacting with repositories, such as opening or commenting on pull requests or issues. Learn more about blocking a user.