This workspace contains the libraries that can be used to develop an SPA in combination with Bloomreach.

The SDKs are located under the packages folder and example SPAs are located in the examples folder for each of the supported frameworks. See the READMEs of the respective packages for more detailed information.

The SDKs:

• SPA SDK
• React SDK
• Vue SDK
• Vue 3 SDK
• Angular SDK

The examples:

• React CSR example
• Next.js example
• Vue CSR example
• Vue 3 CSR example
• Nuxt.js example
• Angular CSR/Universal example

The latest SPA SDK versions support both the Paas and Saas backends from Bloomreach. For more information please read the latest developer guides on the documentation website.

The supported frameworks are currently:

• React and Next.js
• Vue (2 and 3) and Nuxt.js (2)
• Angular and Angular Universal

The versions of these frameworks that the SDKs currently is verified to work with are indicated in the release notes of each package and the minimum required versions are set in the package.json files of the respective packages.

The integration between an SPA and Bloomreach Content is achieved by using a Framework SDK which itself has the SPA SDK as a dependency. Depending on the use case one might import interfaces or functions from the SPA SDK directly.

The SPA SDK itself is written in typescript and is framework independent, it holds the core code that sets up a connection to the Page Model API of the Bloomreach Content instance. When initializing it will use the provided configuration to do a call to the PMA and transform the response from a PageModel to a Page object that provides methods to easily query and work with the model. The SPA SDK also automatically detects whether it is in preview mode and sets up the connection with the Bloomreach Experience Manager Preview if that is the case. The Framework SDKs use this Page object to derive what needs to be rendered on the page.

In short the SPA SDK contains:

• Page Model API Client
• Page Model Javascript implementation
• URL Generator
• Integration with Bloomreach Experience Manager Preview

and the Framework SDK contains:

• HTTP Client
• Framework specific components that render the required DOM

The repository contains multiple self-contained npm packages that also depend on other packages in the repository. To easily manage the dependencies between them and execute commands taking into account topological order we use lerna-lite to manage its packages and builds, however we do not use the bootstrap mechanism of lerna as it caused duplicate framework instances when symlinking the dependencies. Instead we use Yalc to (optionally) link the dependencies.

This lerna workspace configuration is located in lerna.json.

Installation of the dependencies of all the workspaces is done by running npm ci in the root of the workspace.

Running npm run build will not only build all packages but also link them together such that you can develop packages locally without having to publish 'prerelease' versions.

Running npm run dev in the root of the monorepo will watch each SDK package and build and yalc push it when changes happen. Combining this with running the dev server for one of the example apps allows for live development with any SDK.

You can also use the lerna cli to run commands if you require topological ordering such as npx lerna run build --scope @bloomreach/example-react --include-dependencies which wil build example-react as well as its dependency structure.

The root workspace commands are straightforward and are listed in the package.json scripts property.

There is the main branch which is protected and requires MR's to recieve changes, it should only contain those changes/commits that are actually in a released version of the SPA SDKs.

There is also the development branch which should be used as the baseline for any other working branches.

Generally speaking for any kind of development one would:

• Branch the development branch to a new branch e.g. mybranch
• Do work on mybranch until finished
• Create an MR of mybranch to be merged into development
• Create a test release if it has to be published to npm for test purposes but use prerelease strategy and use dist-tag that mentions jira ticket or general effort for easier recognition by running npm run release -- prerelease --dist-tag [your Jira issue / effort name]
• Get reviews & approval
• After the pipelines are green: merge the MR to development

• You need access to the npm @bloomreach organization and rights to publish packages
• You need access to the heroku dashboard and rights to deploy apps
• You need access to the SPA SDK Jira project to manage releases

Pre release actions:

• Check if issues mentioned in the commits since the last release have the fixVersion assigned as the version you are about to release
• Check if all issues in the upcoming release are closed, any open issues must be moved to a different release, or the release must be postponed
• Check out the development branch and make sure your local branch its up to date and the pipeline for development is green
• Make sure that running npm run docs can be done succesfully

Release actions:

• Run npm run bump in the workspace root to update the versions in all package files
• Commit these version changes with git commit -am "[your Jira issue] Bumping versions to [new version]"
• Create a new annotated tag on the development branch using git tag -a spa-sdk-[new-version]
• Push to origin with git push --follow-tags
• Make MR to the main branch and merge it after approve

Post release actions:

• Check that release pipeline is green
• Check that all example apps have been deployed to Heroku
• Set the Jira release version as released
• Create release notes on github
• Check that Documentation Portal has been deployed correctly to github pages
• Check that TypeDoc has been deployed correctly to github pages