Adding support for subscript, highlighted and underlined text

[bismitpanda]

• One '~' produces a subscript instead of strikethrough (when enabled)
• "==" highlights the text inside it
• "++" underlines the text inside

[kivikakk]

Hi there! Thanks so much for this high-quality submission!

There's one major issue, which is that this changes the behaviour of the strikethrough extension, even when subscript isn't enabled. This causes an incompatibility with GitHub Flavored Markdown.

I'd be happy to accept the proposed change if this was addressed -- I'd be okay with subscript being enabled changing strikethrough's behaviour accordingly, but not with it disabled (i.e. GFM defaults). Right now there's an "interesting" bug that results where enabling only strikethrough (not subscript) and then using one ~ results in emphasis being rendered (!); see https://github.com/kivikakk/comrak/actions/runs/4902153007/jobs/8760508796?pr=302#step:4:659.

I'll leave some more comments inline.

[kivikakk]

I suspect (though haven't checked) this will cause spec failure with this example. The spec requires three or more tildes to not generate a strikethrough; since we've already subtracted use_delims (2 for a strikethrough of length 2 or more) from opener_num_chars, if we had 3 or more, opener_num_chars > 0 would be true and so we'd return None here.

[bismitpanda]

If subscript and strikethrough, both are enabled, it will be a subscript as well as a strikethrough\ like in cases of italics and bolds. Considering this, I removed the condition.

[bismitpanda]

Adding and additional && to check if subscript is enabled. Like

if (self.options.extension.strikethrough
    && opener_char == b'~'
    && (opener_num_chars != closer_num_chars || opener_num_chars > 0))
    && !self.options.extension.subscript
{
    return None;
}

[bismitpanda]

Can I add a new feature of inserted text i.e. basically underline using ++? Got inspiration from https://markdown-it.github.io/

[bismitpanda]

Can I get some info regarding the skip_chars array in Subject struct in src/parser/inline.rs@41:4?

[bismitpanda]

Any other changes/improvements required? As I see it is still open.

[kivikakk]

I'm currently sick and unable to review the PR. I will when I can.

[bismitpanda]

any updates on this?

[kivikakk]

No updates. With my limited energy I've been prioritising bug fixes and upstream compliance.