## How does this work?
Any time a lint rule is added or changed, you can run `yarn lint:fix` to handle all the auto-fixable situations.
But not all lints are auto-fixable -- for those, lint-to-the-future has tooling to automatically ignore present violations.
An alias has been added for lint-to-the-future to ignore new violations, `yarn lttf:ignore`.
The command will add lint-ignore declarations throughout all the files with present violations, which should then be committed.
An excerpt from lint-to-the-future's [README](https://github.com/mansona/lint-to-the-future#lint-to-the-future-dashboard):
> The point of Lint to the Future is to allow you to progressively update your codebase using new lint rules without overwhelming you with the task. You can easily ignore lint rules using project-based ignores in your config files but that doesn't prevent you from making the same errors in new files.
> We chose to do the ignores on a file basis as it is a perfect balance and it means that the tracking/graphing aspects of Lint to the Future provide you with achievable goals, especially in large codebases.
## How do I view progress?
lint-to-the-future provides graphs of violations-over-time per lint rule in a dashboard format, so we can track how well we're doing at cleaning up the violations.
To view the dashboard locally, run `yarn lint-progress` and visit `http://localhost:8084` (or whatever the port it chose, as it will choose a new port if 8084 is preoccupied)
Also there is a `list` command which shows a JSON object of:
```ts
{
[date: string]: { // yyyy-mm-dd
[pluginName: string]: {
[fileName: string]: string[]; // list of files with violations
}
}
}
```
```bash
yarn lint-to-the-future list --stdout
```
## What about lint-todo?
Lint todo is another system available for both eslint and ember-template-lint that _forces_ folks to "leave things better than they found them" by being transparent / line-specific ignoring of violations.
It was decided that for _this_ project, it made more sense, and would be less disruptive to new contributors to have the ignore declarations explicitly defined in each file (whereas in lint-todo, they are hidden).
To effectively use lint-todo, a whole team needs to agree to the workflow, and in open source, we want "just anyway" to be able to contribute, and throwing surprises at them can deter contributions.
We provide a `.ruby-version.sample` file that we use for warning developers
about the minimum recommended Ruby version to run Discourse.
d24dfe8f96/config/application.rb (L15-L20)
But if people copy the sample to a `.ruby-version` file it would be added
next time they commit.
This adds the `.ruby-version` file to `.gitignore` so it doesn't get commited
by mistake and developers can test Discourse on other versions of Ruby if
they want to.
- `GeoList2-ASN.mmdb` is one of the MaxMind databases we use for geolocation
- `.bundle/` is created when you run something like `bundle config set --local ...`
(inspired by martin-brennan's recent adventure with `tags` exclusion)
**tl;dr: some of it is obsolete, some is dev-env specific and should live in *your* ~/.gitignore.**
---
To enable global gitignore (`~/.gitignore`) run:
```bash
git config --global core.excludesFile '~/.gitignore'
```
Then create that file and add your VIM/VSCode/Sublime/Emacs/Eclipse/RubyMine/JetBrains/macOS/Arch/Windows 95-specific stuff.
---
Reasons for removal:
* `bin` - never really ignored, all the files are checked in
* `.DS_Store`, `._.DS_Store` - OS specific
* `.sass-cache/*`, `/.bundle`, `/bundle/*`, `/cache`, `/logfile`, `!/plugins/discourse-nginx-performance-report`, `MiniProfiler/Ruby/rack-mini-profiler-2.0.1a.gem`, `config/fog_credentials.yml`, `/public/stylesheet-cache/*`, `script/download_db`, `script/refresh_db`, `config/version.rb` - no longer used?
* `/log/*.log` - covered by /log
* `bootsnap-load-path-cache`, `bootsnap-compile-cache` - bootsnap now keeps its ~~trash~~ temporary files in `/tmp`
* `/.project`, `/.buildpath`, `/.byebug_history`, `/.idea`, `discourse.sublime-workspace`, `*~`, `*.swp`, `*.swo`, `*.swm`, `config/multisite1.yml`, `.rvmrc`, `.ruby-version`, `.ruby-gemset`, `.rbenv`, `bundler_stubs/*`, `*.db`, `*.iml`, `*.swn`, `/package-lock.json`, `.vscode`, `.envrc`, `tags` - dev-env specific, see the note at the top of the commit
In development we regularly restart/reload Rails, which wipes out the schema cache. This then has to be regenerated using DDL queries on the database.
Instead, we can make use of the `rake db:schema:cache:dump` command. This will dump the schema cache to a YAML file, and then load it when needed. This is significantly faster than rebuilding the cache from DDL queries every time.
This PR adds a new relative-time component, that is an input box with a SK dropdown of minutes, hours, days, and months which outputs the duration selected in minutes. This new component is used in the time shortcuts list (used by bookmarks and topic timers) as a new Relative Time shortcut.
Also in this PR, I have made the "Auto-Close After Last Post" timer into a top level timer type in the UI, and removed the "based on last post" custom time shortcut.
We want to allow developers to customize their own vscode environment.
Including launch.json/tasks.json files in the repository makes this very
difficult.
These were originally added for GitHub codespaces. Once codespaces is
more widely available, we can look into automatically copying the
`.vscode-sample` directory to `.vscode` when the codespace boots.
This has a basic config to use the new Codespaces feature in GitHub.
If you have access to the feature all you need to do is:
1. Visit https://github.com/discourse/discourse
2. Click on Code > Open with Codespaces > New codespace
3. Wait it to prepare the environment
4. Click on File > Run > Run without Debug or press
<kbd>CTRL</kbd>+<kbd>F5</kbd>
5. When prompted click on the green pop-up on the lower left to open the
running Discourse app in a new tab
This plugin is only useful for developers, however, making it core allows us to centralize any component modification in one commit.
This integration also adds a new site_setting: `styleguide_admin_only` which allows to enable a styleguide on a live site while restricting visibility to admins only.
By default, styleguide is disabled.
On some installations, there may be a leftover symlink which uses the
old plugin name:
public/plugins/discourse-internet-explorer ->
-> plugins/discourse-internet-explorer/public
And don't load javascript assets if plugin is disabled.
* precompile auto generated plugin js assets
* SPEC: remove spec test functions
* remove plugin js from test_helper
Co-Authored-By: Régis Hanol <regis@hanol.fr>
* DEV: using equality is slightly easier to read than inequality
Co-Authored-By: Régis Hanol <regis@hanol.fr>
* DEV: use `select` method instead of `find_all` for readability
Co-Authored-By: Régis Hanol <regis@hanol.fr>
This PR is intended to work in concert with the discourse_docker changes
in https://github.com/discourse/discourse_docker/pull/292. I have used
those changes to build a local "discourse_dev" image, and then
_**these**_ changes to easily spin up and work on Discourse and plugin
functionality. It's working well for me, but of course YMMV.
Add `--init` and `--patch` options to bin/docker/boot_dev to make it
easier to spin up a container running Discourse out of a development
directory.
Add `bin/docker/README.md` to explain how to go about using the docker
command-line tools.
Tweak the Docker command-line tools to use the `-u` option to specify
user rather than changing to the user "inside" the container via `chpst`
(This way, we don't have to explicitly specify the HOME environment
variable, either.)
Add `bin/docker/shell` command to make it easy to jump inside the
running container.
NullVoxPopuli
Jan Cernik
Régis Hanol
Roman Rizzi
Penar Musaraj
David Taylor
Jarek Radosz
Vinoth Kannan
Martin Brennan
Rafael dos Santos Silva
Joffrey JAFFEUX
Robin Ward
Bianca Nenciu
Kyle E. Mitchell
Dan Ungureanu
Blake Erickson
Gerhard Schlager
Arpit Jalan
Vinoth Kannan
Tim Lange
Sam
Andrew Schleifer
Guo Xiang Tan
Jared Reisinger
Jeff Atwood
Neil Lalonde