kibana/packages/kbn-pm
Tiago Costa 20bdf7d4a9
Migration to Babel7 and @babel/preset-typescript (#33093)
* chore(NA): first changes on every package.json order to support new babel 7. chore(NA): build for kbn-pm with babel 7.

* chore(NA): patch babel register to load typescrit

* chore(NA): first working version with babel 7 replacing typescript compiler.

* fix(NA): common preset declaration in order to make it work with babel-loader.

* chore(na): organizing babel preset env package json.

* chore(NA): mocha tests enabled.

* fix(NA): typo on importing

* test(NA): majority of x-pack tests ported to use babel-jest

* fix(NA): report info button test with babel-jest.

* fix(NA): polling service tests.

* test(na): fix server plugins plugin tests.

* test(NA): batch of test fixs for jest tests under babel-jest hoisting.

* chore(NA): add babel plugin to hoist mock prefixed vars on jest tests.

* chore(NA): update yarn.lock file.

* chore(NA): tests passing.

* chore(NA): remove wrong dep

* chore(NA): fix tsconfig

* chore(NA): skip babel for ts-jest.

* chore(NA): selectively apply the plugin to strip off namespace from ts files.

* chore(NA): remove not needed changes from ts tests

* chore(NA): removed ts-jest dependency. chore(NA): migrate ts tests on x-pack to use babel-jest with the new pattern.

* chore(NA): migrate kibana default distribution typescript tests to run with babel-jest and the new test mock pattern.

* chore(NA): merge and solve conflicts with master.

* chore(NA): fix problems reported by eslint

* chore(NA): fix license ovveride for babel-plugin-mock-imports

* chore(NA): update jest integration tests for kbn pm

* chore(NA): update babel jest integration tests for kbn pm.

* test(NA): update jest integration snapshot for kbn pm.

* chore(NA): apply changes according to the pull request reviews.

* chore(NA): apply changes according to the pull request reviews.

* refact(NA): migrate jest tests to the new pattern.

* fix(NA): babel 7 polyfill in the tests bundle.

* chore(NA): restore needed step in order to compile x-pack with typescript.

* chore(NA): change build to compile typescript with babel for the oss code. chore(NA): change transpile typescript task to only transpile types for x-pack. refact(NA): common preset for babel 7

* Revert "chore(NA): change build to compile typescript with babel for the oss code. chore(NA): change transpile typescript task to only transpile types for x-pack. refact(NA): common preset for babel 7"

This reverts commit 2707d538f5.

* fix(NA): import paths for tabConfigConst

* chore(NA): fix transpiling error on browser tests

* chore(NA): simplify kbn babel preset package.

* chore(NA): migrate build to use babel transpiler for typescript excluding xpack.

* fix(NA): introduced error on test quick task.

* fix(NA): fix preset for client side code on build.

* fix(NA): build with babel

* fix(NA): negated patterns in the end.

* fix(NA): kbn_tp_sample_panel_action creation.

* fix(NA): babel typescript transform plugin workaround when exporting interface name.

* refact(NA): remove not needed type cast to any on jest test.

* docs(NA): add developement documentation about jest mocks test pattern.

* chore(NA): missing unmerged path.

* chore(NA): fix jest tests for template.

* [CCR] Client integration tests (table lists) (#33525)

* Force user to re-authenticate if token refresh fails with `400` status code. (#33774)

* Improve performance of the Logstash Pipeline Viewer (#33793)

Resolves #27513.

_This PR is a combination of #31293 (the code changes) + #33570 (test updates). These two PRs were individually reviewed and merged into a feature branch. This combo PR here simply sets up the merge from the feature branch to `master`._

Summary of changes, taken from #31293:

## Before this PR
The Logstash Pipeline Viewer UI would make a single Kibana API call to fetch all the information necessary to render the Logstash pipeline. This included information necessary to render the detail drawer that opens up when a user clicks on an individual vertex in the pipeline.

Naturally, this single API call fetched _a lot_ of data, not just from the Kibana server but also, in turn, from Elasticsearch as well. The "pro" of this approach was that the user would see instantaneous results if they clicked on a vertex in a pipeline and opened the detail drawer for that vertex. The "cons" were the amount of computation Elasticsearch had to perform and the amount of data being transferred over the wire between Elasticsearch and the Kibana server as well as between the Kibana server and the browser.

## With this PR
This PR makes the Kibana API call to fetch data necessary for **initially** rendering the pipeline — that is, with the detail drawer closed — much lighter. When the user clicks on a vertex in a pipeline, a second API call is then made to fetch data necessary for the detail drawer.

## Gains, by the numbers

Based on a simple, 1-input, 1-filter, and 1-output pipeline.

* Before this PR, the Elasticsearch `logstash_stats` API responses (multiple calls were made using the `composite` aggregation over the `date_histogram` aggregation) generated a total of 1228 aggregation buckets (before any `filter_path`s were applied but across all `composite` "pages"). With this PR, the single `logstash_stats` API response (note that this is just for the initial rendering of the pipeline, with the detail drawer closed) generated 12 buckets (also before any `filter_path`s were applied). That's a **99.02% reduction** in number of buckets.

* Before this PR, the Elasticsearch `logstash_stats` API responses added up to 70319 bytes. With this PR, the single `logstash_stats` API response for the same pipeline is 746 bytes. That's a **98.93% reduction** in size.

* Before this PR, the Elasticsearch `logstash_state` API response was 7718 bytes. With this PR, the API response for the same pipeline is 2328 bytes. That's a **69.83% reduction** in size.

* Before this PR the Kibana API response was 51777 bytes. With this PR, the API response for the same pipeline is 2567 bytes (again, note that this is just for the initial rendering of the pipeline, with the detail drawer closed). That's a **95.04% reduction** in size.

* [Maps] split settings into layer and source panels (#33788)

* [Maps] split settings into layer and source panels

* fix SCSS import

* [env] exit if starting as root (#21563)

* [env] exit if starting as root

* fix windows

* s/--allow-root

* Typescript sample panel action (#33602)

* Typescript sample panel action

* Update EUI version to match main cabana version

* update yarn.lock

* add back typings include

* use correct relative path

* Home page "recent links" should communicate saved object type #21896 (#33694)

* adds object type for screen order
* adds object type for pointer hovering
* Update src/legacy/ui/public/chrome/directives/header_global_nav/components/header.tsx

Co-Authored-By: rockfield <philipp.b@ya.ru>
2019-03-26 20:44:03 +00:00
..
dist Migration to Babel7 and @babel/preset-typescript (#33093) 2019-03-26 20:44:03 +00:00
src Migration to Babel7 and @babel/preset-typescript (#33093) 2019-03-26 20:44:03 +00:00
.babelrc Migration to Babel7 and @babel/preset-typescript (#33093) 2019-03-26 20:44:03 +00:00
cli.js Apache 2.0 license headers (#19383) 2018-05-28 20:06:30 -07:00
package.json Migration to Babel7 and @babel/preset-typescript (#33093) 2019-03-26 20:44:03 +00:00
README.md Prefer third-party plugin development in ./plugins instead of ../kibana-extra (#31748) 2019-03-07 17:04:29 -06:00
tsconfig.json Migration to Babel7 and @babel/preset-typescript (#33093) 2019-03-26 20:44:03 +00:00
webpack.config.js Migration to Babel7 and @babel/preset-typescript (#33093) 2019-03-26 20:44:03 +00:00

@kbn/pm — The Kibana project management tool

@kbn/pm is a project management tool inspired by Lerna, which enables sharing code between Kibana and Kibana plugins.

To run @kbn/pm, go to Kibana root and run yarn kbn.

Why @kbn/pm?

Long-term we want to get rid of Webpack from production (basically, it's causing a lot of problems, using a lot of memory and adding a lot of complexity). Ideally we want each plugin to build its own separate production bundles for both server and UI. To get there all Kibana plugins (including x-pack) need to be able to build their production bundles separately from Kibana, which means they need to be able to depend on code from Kibana without import-ing random files directly from the Kibana source code.

From a plugin perspective there are two different types of Kibana dependencies: runtime and static dependencies. Runtime dependencies are things that are instantiated at runtime and that are injected into the plugin, for example config and elasticsearch clients. Static dependencies are those dependencies that we want to import. eslint-config-kibana is one example of this, and it's actually needed because eslint requires it to be a separate package. But we also have dependencies like datemath, flot, eui and others that we control, but where we want to import them in plugins instead of injecting them (because injecting them would be painful to work with). (Btw, these examples aren't necessarily a part of the Kibana repo today, they are just meant as examples of code that we might at some point want to include in the repo while having them be importable in Kibana plugins like any other npm package)

Another reason we need static dependencies is that we're starting to introduce TypeScript into Kibana, and to work nicely with TypeScript across plugins we need to be able to statically import dependencies. We have for example built an observable library for Kibana in TypeScript and we need to expose both the functionality and the TypeScript types to plugins (so other plugins built with TypeScript can depend on the types for the lib).

However, even though we have multiple packages we don't necessarily want to npm publish them. The ideal solution for us is being able to work on code locally in the Kibana repo and have a nice workflow that doesn't require publishing, but where we still get the value of having "packages" that are available to plugins, without these plugins having to import files directly from the Kibana folder.

Basically, we just want to be able to share "static code" (aka being able to import) between Kibana and Kibana plugins. To get there we need tooling.

@kbn/pm is a tool that helps us manage these static dependencies, and it enables us to share these packages between Kibana and Kibana plugins. It also enables these packages to have their own dependencies and their own build scripts, while still having a nice developer experience.

How it works

Internal usage

For packages that are referenced within the Kibana repo itself (for example, using the @kbn/i18n package from an x-pack plugin), we are leveraging Yarn's workspaces feature. This allows yarn to optimize node_modules within the entire repo to avoid duplicate modules by hoisting common packages as high in the dependency tree as possible.

To reference a package from within the Kibana repo, simply use the current version number from that package's package.json file. Then, running yarn kbn bootstrap will symlink that package into your dependency tree. That means you can make changes to @kbn/i18n and immediately have them available in Kibana itself. No npm publish needed anymore — Kibana will always rely directly on the code that's in the local packages.

External Plugins

For external plugins, referencing packages in Kibana relies on link: style dependencies in Yarn. With link: dependencies you specify the relative location to a package instead of a version when adding it to package.json. For example:

"@kbn/i18n": "link:packages/kbn-i18n"

Now when you run yarn it will set up a symlink to this folder instead of downloading code from the npm registry. This allows external plugins to always use the versions of the package that is bundled with the Kibana version they are running inside of.

"@kbn/i18n": "link:../../kibana/packages/kbn-date-math"

This works because we moved to a strict location of Kibana plugins, ./plugins/{pluginName} inside of Kibana, or ../kibana-extra/{pluginName} relative to Kibana. This is one of the reasons we wanted to move towards a setup that looks like this:

elastic
└── kibana
    └── plugins
        ├── kibana-canvas
        └── x-pack-kibana

Relying on link: style dependencies means we no longer need to npm publish our Kibana specific packages. It also means that plugin authors no longer need to worry about the versions of the Kibana packages, as they will always use the packages from their local Kibana.

The kbn use-cases

Bootstrapping

Now, instead of installing all the dependencies with just running yarn you use the @kbn/pm tool, which can install dependencies (and set up symlinks) in all the packages using one command (aka "bootstrap" the setup).

To bootstrap Kibana:

yarn kbn bootstrap

By default, @kbn/pm will bootstrap all packages within Kibana, plus all Kibana plugins located in ./plugins or ../kibana-extra. There are several options for skipping parts of this, e.g. to skip bootstrapping of Kibana plugins:

yarn kbn bootstrap --skip-kibana-plugins

Or just skip few selected packages:

yarn kbn bootstrap --exclude @kbn/pm --exclude @kbn/i18n

For more details, run:

yarn kbn

Bootstrapping also calls the kbn:bootstrap script for every included project. This is intended for packages that need to be built/transpiled to be usable.

Running scripts

Some times you want to run the same script across multiple packages and plugins, e.g. build or test. Instead of jumping into each package and running yarn build you can run:

yarn kbn run build

And if needed, you can skip packages in the same way as for bootstrapping, e.g. with --exclude and --skip-kibana-plugins:

yarn kbn run build --exclude kibana

Watching

During development you can also use kbn to watch for changes. For this to work package should define kbn:watch script in the package.json:

yarn kbn watch

By default kbn watch will sort all packages within Kibana into batches based on their mutual dependencies and run watch script for all packages in the correct order.

As with any other kbn command, you can use --include and --exclude filters to watch only for a selected packages:

yarn kbn watch --include @kbn/pm --include kibana

Building packages for production

The production build process relies on both the Grunt setup at the root of the Kibana project and code in @kbn/pm. The full process is described in tasks/build/packages.js.

Development

This package is run from Kibana root, using yarn kbn. This will run the "pre-built" (aka built and committed to git) version of this tool, which is located in the dist/ folder. This will also use the included version of Yarn instead of using your local install of Yarn.

If you need to build a new version of this package, run yarn build in this folder.

Even though this file is generated we commit it to Kibana, because it's used before dependencies are fetched (as this is the tool actually responsible for fetching dependencies).

Technical decisions

Why our own tool?

While exploring the approach to static dependencies we built PoCs using npm 5 (which symlinks packages using file: dependencies), Yarn workspaces, Yarn (using link: dependencies), and Lerna.

In the end we decided to build our own tool, based on Yarn, and link: dependencies, and workspaces. This gave us the control we wanted, and it fits nicely into our context (e.g. where publishing to npm isn't necessarily something we want to do).

Some notes from this exploration

file: dependencies in npm<5 and in yarn

When you add a dependency like "foo": "file:../../kibana/packages/foo", both npm<5 and yarn copies the files into the node_modules folder. This means you can't easily make changes to the plugin while developing. Therefore this is a no-go.

file: dependencies in npm5

In npm5 file: dependencies changed to symlink instead of copy the files. This means you can have a nicer workflow while developing packages locally. However, we hit several bugs when using this feature, and we often had to re-run npm install in packages. This is likely because we used an early version of the new file: dependencies in npm5.

This is the same feature as file: dependencies in npm5. However, we did not hit any problems with them during our exploration.

Yarn workspaces

Enables specifying multiple "workspaces" (aka packages/projects) in package.json. When running yarn from the root, Yarn will install all the dependencies for these workspaces and hoist the dependencies to the root (to "deduplicate" packages). However:

Workspaces must be children of the workspace root in term of folder hierarchy. You cannot and must not reference a workspace that is located outside of this filesystem hierarchy.

So Yarn workspaces requires a shared root, which (at least currently) doesn't fit Kibana, and it's therefore a no-go for now.

Lerna

Lerna is based on symlinking packages (similarly to the link feature which exists in both npm and Yarn, but it's not directly using that feature). It's a tool built specifically for managing JavaScript projects with multiple packages. However, it's primarily built (i.e. optimized) for monorepo libraries, so it's focused on publishing packages and other use-cases that are not necessarily optimized for our use-cases. It's also not ideal for the setup we currently have, with one app that "owns everything" and the rest being packages for that app.

Why a local version of Yarn?

See the vendor readme.