An ESLint plugin that allows you to lint the content of <script> tags in arbitrary text.

Unlike eslint-plugin-html, which parses HTML to find <script>s, this plugin uses a regular expression.

The advantage of using a regular expression, and the reason this plugin exists, is that by avoiding a parser you can find and lint <script>s in arbitrary text — text that might not parse nicely as HTML, such as Markdown files with HTML fenced code blocks that contain <script>s. Like this (imagine the '''s are backticks):

Here is some *markdown*.

'''html
<div>
<script>
  // Here is JS you'd like to lint!
  var a = "foo";
</script>
'''

(If you want to lint JS fenced code blocks, use eslint-plugin-markdown).

The disadvantage of using a regular expression, of course, is that it's not as robust as a real parser. If your <script> is within a comment, for example, it will still be linted: the regular expression doesn't know about the comment. You might like that, or you might not.

Use this module like any other ESLint plugin.

Unfortunately, we do need to provide a finite list of file extensions to check. Here's the current list:

• .html
• .md
• .jsp
• .tag
• .hbs
• .ejs
• .php
• .vue
• .erb

There's no reason not to add more: so if you'd like another, please PR!

Expected indentation (if you use the indent rule) is determined by the first line of code. This can get hairy if you do something sloppy like:

<script>var a = 'foo';
  var b = 'bar';
</script>

So I wouldn't do that.

Ideally this module could allow you to arbitrarily modify the expected tags regular expression, from <script>...</script> to something like {% highlight js %}...{% endhighlight %} or [javascript-block]...[end javascript-block], etc. However, to do so would require adding options, and it doesn't seem that ESLint's built options into the processor API. If you would like this feature and want to work on adding options, your help would be welcome.