Closed teetrinkers closed 8 years ago
I'm used to eslint
. Any argument in favor of JSHint?
Not really. JSHint is just what I use on my other projects. I haven't used eslint yet.
Would it be ok to add configs for both? I think in Atom there is no way to select the linter per project, so jshint runs on matterfront even if there is no .jshintrc.
I'm used to eslint. Any argument in favor of JSHint? -@lloeki
I'm going to argue against it. I am currently integrating eslint
, and prefer it to JShint. It's far better than JSHint in many ways, especially in a project that has multiple run/execution environments (such as an electron based application).
For example, I noticed that in webview.js, $ is used. Is jquery included in Electron?
I'm currently reviewing any deps in webview and moving them into the work being done for the CLI/build stuff. They'll be listed as dependencies in the package.json
for the next minor update of the matterfront npm module.
I think that this should have been an issue and not a PR. There are no changes to package.json
either that would reflect the inclusion and execution of either linter, which means the developer would have to depend on an editor to have a plugin to do that.
I prefer it to be set up explicitly, so that linting can be part of the build & test pipeline, (which is also an item in scope of my current branch)
I think in Atom there is no way to select the linter per project, so jshint runs on matterfront even if there is no .jshintrc.
That's why I prefer things so critical to the build & test pipeline to not be dependent upon an editor. An authoring environment should be a personal choice of preference, and I'd prefer to avoid any assumptions or project dependencies that lock our contributors in to an environment they may not be comfortable with.
The config settings are up for discussion of course. For example, I noticed that in webview.js,
$
is used. Is jquery included in Electron? If so, we should enable thejquery
environment in .jshintrc to silence those warnings.