@davidlj95/website-v2

@davidlj95/website

This project was generated with Angular CLI version 16.2.0.

To start

First, run the prebuild command, to generate files that are required for the app to work

yarn run prebuild

Development server

Run ng serve for a dev server. Navigate to http://localhost:4200/. The application will automatically reload if you change any of the source files.

Code scaffolding

Run ng generate component component-name to generate a new component. You can also use ng generate directive|pipe|service|class|guard|interface|enum|module.

Build

Run ng build to build the project. The build artifacts will be stored in the dist/ directory.

Running unit tests

Run ng test to execute the unit tests via Karma.

Running end-to-end tests

Run ng e2e to execute the end-to-end tests via a platform of your choice. To use this command, you need to first add a package that implements end-to-end testing capabilities.

Commit message guidelines

Commit messages follow the conventional commits guidelines. This allows for automating the semantic versioning release process using semantic release

Commit message lint

To enforce that, commitlint is used. It reads the commit messages on a PR and ensures they follow the conventional commits convention.

You can use the following run script to ensure the last commit follows the guidelines:

yarn run commitlint-last

Release

Semantic Release is used to automate the release process. Based on commit messages, a new major, minor or patch release is generated for every push to main branch. For every release, a Git tag will be created, a GitHub release will be created and a new version of the app will be published to npm public registry. All of this is managed automatically by Semantic Release and CI/CD pipelines.

Config has been tweaked so all commits appear in the release notes. To do so, types of commits included are all of those defined in Conventional Changelog type defaults. But none is hidden. That list comprehends

all commit types that commitlint

will lint.

Getting release info

In order to embed release information in the app, a script was created to export that kind of information using Semantic Release's JavaScript API.

To generate the release info, run

yarn run generate-release-file

It will generate a release.json file in the root of the repository containing all info about releases that Semantic Release provides.

⚠️ Next release can be fake

The API only outputs information when a new release has to be generated. As a workaround, a fake patch release is generated in case no release was needed to get access to all that info.

To distinguish a real run from a faked run, a fake property exists in that JSON with value set to true. The only faked properties are about next release (nextRelease and releases at the moment of writing this). Info about last release remains correct even when faking the release to generate the output (lastRelease at the moment of writing this)

⚠️ Next release can be a preview

If running the script from a branch that's not configured to trigger releases, the default API behaviour is to output no data. As a workaround, we simulate being on the main branch and see what would happen in that scenario when including commits from the current branch.

To distinguish a real run from a faked run, a preview property exists in that JSON with value set to true. This is because it's a simulation of this branch being merged into main branch. But you could want to merge that against another (the script could be smarter and simulate that too, but not yet :P).

⚠️ Next release can be fake and preview If both of previous situations happen at same time (commits of branch are not enough to trigger a release and branch is not one configured for releases)

Git hooks

Not a fan of, but it's useful to enforce conventional commits. Indeed, it's only one, a pre-commit hook. And checks your commit message complies with the conventional commits format and configuration.

To use it, run

yarn run git-hooks

It will install Git hooks via husky. Now everytime you commit, you'll be sure commit message guidelines are enforced.

This is very useful if you can push to main directly. Because there's no way to reject commits that don't follow the commit message guidelines. And once you push to main, you cannot amend that commit as per branch protection rules.

Autogenerated files from templates

Some metadata about the app (like title, author, theme color...) was repeated across files. So used LiquidJS to create template files (.liquid extension). Then, created a small script to render templates into generated files using metadata defined in metadata.ts file (and some hardcoded constants in the generation script)

If you update some metadata in that file, run the script to update files generated from templates. Otherwise, they won't be in sync.

Automatically updating generated files is not straightforward. Doing something before/after an Angular app build runs is not an option right now. Adding a bunch of pre npm run scripts left the package.json file quite cluttered. Also, won't work if issuing ng commands. So decided to leave the task to update generated files to the developer. At some point a CI check could be introduced to ensure generated files are up-to-date. Generated files are left in the repo because of that reason (and also so you can clone & run without issues). That metadata won't change often anyway.

Quirks

Symlinked static assets don't work when serving locally

When using Angular development server (either regular or SSR one), symbolic links seem not to work for static assets. So for instance /profile.jpg which is symlinked to /assets/img/og.jpg won't be served. And the server will redirect you towards main page (/). It works when building though, so it will appear in the built version.

Rendering font subsets

Some fonts included are a subset of a big font file. Before doing anything, please run

yarn run font-subsets

To generate them. Otherwise, those fonts won't be found and you may get some errors. Also, remember to run it if changing the glyphs included in each font. Or to add the new glyphs if you want to use more glyphs of a font.

CI/CD

GitHub Actions are used

Commands ran by CI/CD

In order to ease running CI/CD commands locally, the .ci directory contains a Makefile intended to contain all commands that will be run in CI/CD pipelines. So you can test those locally easily

Just:

cd .ci && make test # for instance

And see how a command run in the CI/CD behaves locally. Notice your machine's state may differ from the CI/CD machine one.

Running GitHub Actions workflows locally

You can use act to run GitHub Action workflows in your development machine. If using macOS, install it using brew:

brew install act

Then, use one of the many run- targets in the CI/CD Makefile to simulate a CI/CD workflow run. For instance:

cd .ci && make run-main

Simulates the workflow triggered by a push to main branch.

If you get a message related to Apple M-series chip:

WARN  ⚠ You are using Apple M-series chip and you have not specified container architecture, you might encounter issues while running act. If so, try running it with '--container-architecture linux/amd64'. ⚠                      7s

You can add --container-architecture linux/amd64 (notice a space when line begins) to your ~/.actrc file. Create it if it doesn't exist.

ℹ️ Images are set to js-* ones via the .actrc file as medium default ones don't include yarn :sadparrot:

Release

Further help

To get more help on the Angular CLI use ng help or go check out the Angular CLI Overview and Command Reference page.