DEV Community

Cover image for Lerna reborn - What's new in v6?
Nx profile image Juri Strumpflohner
Juri Strumpflohner for Nx

Posted on • Edited on • Originally published at blog.nrwl.io

Lerna reborn - What's new in v6?

#monorepo #javascript #webdev

Lerna v6 is out!! Here's everything you need to know about the new Lerna experience!

Table of Contents

Lerna continues to evolve

If you already know this, feel free to skip ahead. But surprisingly many still haven't heard that Lerna is back, far from obsolete or deprecated and getting brand new features. We from Nrwl are the creators of Nx and given our long history in the monorepo space, we offered to take over stewardship of Lerna when it was declared "dead" in April 2022.

Since we took over, in May 2022, it has been an absolute rollercoaster. We launched a brand new website, updated the content of the docs, and made Lerna 10x faster. And now, Lerna v6 is out!

Fast Lerna with caching by default

Up until Lerna v4, either the p-map or p-queue npm packages have been used to delegate the task scheduling. With v5.1 we introduced nx as an additional mechanism to schedule tasks. The advantage? Nx has caching built-in, which also gives Lerna caching support, making it lightning fast. A recent benchmark test resulted in Lerna being 2.5x faster than Lage and around 4x faster than Turbo (as of Oct 2022; test it out by yourself).

So far you had to enable "Nx support" by setting the useNx flag in lerna.json:

// lerna.json
{
    ...
    "useNx": true
}
Enter fullscreen mode Exit fullscreen mode

We've been testing this opt-in for the last couple of months and got tons of amazing feedback from companies and open source projects. As a result, with v6 all Lerna workspaces have the useNx set to true by default even if you don't have it in your Lerna config file. If you don't want to use it, you can disable it by setting the flag to false.

To experience fast caching, ensure you have a nx.json file at the root of your Lerna workspace where you define the cacheable operations. Check out the docs for more details. Here's an example configuration file:

{
  "tasksRunnerOptions": {
    "default": {
      "runner": "nx/tasks-runners/default",
      "options": {
        "cacheableOperations": ["build", "test"]
      }
    }
  }
}
Enter fullscreen mode Exit fullscreen mode

Note that you can also run..

npx lerna add-caching
Enter fullscreen mode Exit fullscreen mode

..to automatically generate a nx.json configuration file based on your existing Lerna workspace.

Remote caching with Lerna

By using Nx as the task scheduler for Lerna it inherits all the capabilities Nx comes with. That not only just includes local caching, but also the possibility of having remote caching and distributed task execution.

Remote caching allows you to distribute your local cache with your co-workers and your CI system. This is done via Nx Cloud. But distributed caching is just one aspect. Nx Cloud also comes with a "run view" that visualizes your CI run with easy grouping and filtering capabilities, but in particular, it comes with the ability to distribute your tasks dynamically across multiple machines. All by optimizing for the best parallelization and machine utilization.

Nx Cloud Features Overview

All you need to set this up is to run..

npx nx connect-to-nx-cloud
Enter fullscreen mode Exit fullscreen mode

..in your Lerna workspace, which will guide you through a couple of questions and set you up with an Nx Cloud.

Read more on the docs.

Defining a task pipeline

When running tasks in a monorepo environment, you want to maximize the parallelization, but still account for potential dependencies among tasks. For example, assume you have a Remix application that depends on some shared-ui library. You want to ensure that shared-ui is built before either building or serving the Remix application.

With Lerna v6 you can do so in the nx.json file by defining the targetDefaults:

// nx.json
{
  ...
  "targetDefaults": {
    "build": {
      "dependsOn": ["^build"]
    },
    "dev": {
      "dependsOn": ["^build"]
    }
  }
}
Enter fullscreen mode Exit fullscreen mode

In this case, whenever you run either build or dev, Lerna would first run the build task on all the dependent packages.

Read more on our docs.

Lerna add-caching command

If you don’t have caching or your task pipeline set up just yet, no worries. We wanted to make it as easy as possible by providing a dedicated command:

npx lerna add-caching
Enter fullscreen mode Exit fullscreen mode

This will scan your workspace, find all your package.json scripts and then guide you through the configuration of both, your cacheable operations as well as your task pipeline.

Here’s a quick walkthrough video:

You are obviously always free to create the nx.jsonby hand.

PNPM support for Lerna

In the past, Lerna didn't properly support PNPM. We fixed this in v6. Now whenever you use Lerna in combination with PNPM, we make sure to detect packages based on the pnpm-workspace.yaml, to enforce useWorkspaces: true , we update the pnpm-lock.yaml
accordingly when using lerna version and we also added proper support for the workspace: protocol that PNPM uses.

You can now finally use one of the fastest package managers in combination with a new fast Lerna experience. Also, make sure to check out our docs for all the details.

Dynamic terminal output

When running tasks in parallel across a large number of projects, it can become quite difficult to follow along in the terminal with what got built and where tasks failed. That's why the new Lerna version comes with a dynamic terminal output that only shows only what is most relevant at a given moment.

Lerna dynamic terminal screenshot powered by Nx

Note that you would still see all of the output as usual on CI.

VSCode extension for Lerna workspaces

Lerna now has a dedicated VSCode extension to help you navigate your monorepo. This allows you to run commands directly from the context menu (by right-clicking on a project):

Screenshot of the Nx Console integration into VSCode right-click menu

Or visualize a project and its relationships with other projects in the workspace.

Screenshot showing the Nx Console graph visualization in a Lerna workspace

You will also get intelligent autocompletion in configuration files. Here's an example of Nx console providing context-based information when editing the nx.json task dependencies.

Screenshot of Nx Console autocompletion in a nx.json file

Lerna Repair

Lerna v6 comes with a built-in lerna repair command. Running this command will automatically fix your Lerna configuration. For instance, in Lerna v6, there's no need to have useNx: true in your lerna.json since that will be the new default going forward. Running lerna repair fixes this.

Screenshot of lerna repair console log output

This allows you always to have the most up-to-date Lerna setup and it will become even more powerful as we keep adding migrations in the future.

Lerna and Prettier

Prettier is part of the standard toolchain of every developer nowadays. In Lerna v6 we added a feature to detect whether Prettier is set up in the workspace. If so, we automatically apply it to all files that get updated by running the lerna version command. No more follow-up commits just to fix the file formatting!

Migrating to Lerna v6

Migrating from Lerna v5 to v6 is non-breaking. We increased the major because we changed some defaults and wanted to be cautious about that and communicate it properly.

Similarly, if you're still on v4 and want to migrate to v6 it should be pretty straightforward and not be breaking in most cases.

Just update the Lerna package version to the latest and then run..

npx lerna add-caching
Enter fullscreen mode Exit fullscreen mode

..to enable and configure caching for your workspace.

Are you stuck or need help? There is also a community-organized Slack channel where also Nx and Lerna core members hang out. Join https://go.nrwl.io/join-slack and enter the #lerna channel.

Lerna is using Nx now. Can I keep using my Lerna commands?

Absolutely! One of the key advantages of the new integration of Lerna with Nx is that you can keep using your existing Lerna commands without migrating them to a new syntax. They will now just be a lot faster.

You can read more about that on our docs.

Are you maintaining an OSS repository using Lerna?

If you are an OSS maintainer and you use a Lerna workspace, let us know!
Ping the Lerna team on Twitter or ping me directly. We'd love to have a look and help with the migration, look at the repository and make sure it is configured in the best optimal way in terms of monorepo setup and features like caching.

That said, as an open-source maintainer you also get unlimited free computation caching with Nx Cloud. So we'd love to set you up with that.

Learn more

Also, if you liked this, click the ❤️ and make sure to follow Juri and Lerna on Twitter for more!

#nx

Top comments (0)

Read next

siisee11 profile image

Langgraph Human In The Loop with socket

Jaeyoun Nam -

truongpx396 profile image

🔐 Solidity Limitations, Solutions, Best Practices and Gas Optimization 🚀

Truong Phung -

dmitry_bas profile image

Using SCP with ssh2 in Node.js Applications without SFTP

Dmitry Bas -

whiskeygerman profile image

Web canvas with generative neural networks with elements of visual programming.

German Whiskey -