Issue parsing nested HTML elements

First of all thanks for you work on the library. We’ve recently adopted it and encountered an issue when trying to parse and render some markdown mixed with HTML (which is there to provide backwards compatibility with another system, hence can’t be removed).

In fact when trying to parse three nested <div>s:

<div>
  <div>
    <div></div>
  </div>
</div>

markdown-to-jsx will output this:

<div>
  <div></div>
  <div></div>
</div>
<p>&lt;/div&gt;</p>

It can be reproduced on the demo site, and it seems to happen when triple-nesting any HTML element. The issue looks similar to https://github.com/probablyup/markdown-to-jsx/issues/168 but I’m not sure if it’s exactly the same problem.

[probablyup]

True, but none of those are small enough to fit the size requirement of this library. It will never be perfect, but it’s good enough for most use cases.

On Sat, Oct 12, 2019 at 9:36 AM James Thistlewood [e-mail] wrote:

Ok, so I’ve taken a long look at this. The problem basically stems from the fact that regex is being used to parse HTML. HTML is not a regular language, and you cannot use regex to parse it https://stackoverflow.com/questions/1732348/regex-match-open-tags-except-xhtml-self-contained-tags/1732454#1732454. You can make it seem like it works, but it will never properly work, hence bugs like this.

I’d recommend using a real XML/HTML parser instead.

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/probablyup/markdown-to-jsx/issues/255?email_source=notifications&email_token=AAELFVVKI2KKQSKBY53SPEDQOHHF7A5CNFSM4HXJOSXKYY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGOEBB7TBY#issuecomment-541325703, or unsubscribe https://github.com/notifications/unsubscribe-auth/AAELFVRHOFUK5WMS3DGWC4LQOHHF7ANCNFSM4HXJOSXA .

[jthistle]

Ok, so I’ve taken a long look at this. The problem basically stems from the fact that regex is being used to parse HTML. HTML is not a regular language, and you cannot use regex to parse it. You can make it seem like it works, but it will never properly work, hence bugs like this.

I’d recommend using a real XML/HTML parser instead.