fix(v2): remove Markdown syntax from excerpt

[lex111]

Motivation

Resolve #2607

Have you read the Contributing Guidelines on pull requests?

Yes

Test Plan

See preview and tests.

Related PRs

(If this PR adds or changes functionality, please take some time to update the docs at https://github.com/facebook/docusaurus, and link to your PR here.)

[docusaurus-bot]

Deploy preview for docusaurus-2 ready!

Built with commit 3ec542d

https://deploy-preview-2701--docusaurus-2.netlify.app

[yangshun]

Nice! Have you considered using https://github.com/remarkjs/strip-markdown instead? It's based on remark while remove-markdown uses regex.

[lex111]

Ok, I used Remark for these purposes, but I noticed that there was a slight decrease in performance compared to the previous solution that uses regular expressions. I commented lines related to the previous solution so that you can see for yourself which option is better.

UPD: Moreover, we do not need to use a third-party "remove-markdown" package as ready-made solution that removes MD via regular expressions, since there are many replace calls that are not needed in our case.
We can take a limited number of these regexps and thus improve performance. WDYT?

[yangshun]

I think we should allow the full syntax to also handle cases like this:

[lex111]

I did not understand which of the solutions we use in the end?

[lex111]

cc @yangshun

[yangshun]

Which do you recommend?

[lex111]

Personally, I like a solution using regular expressions. We can opt out of using the npm strip-markdown package, and instead choose the regexps we really need to solve current issue.

[yangshun]

Ok let's do your way then. Introducing a perf regression for this isn't really worth it.

[lex111]

Done ✔️

Time building on my laptop:
Before 54.65s
After 55.22s.

Probably not bad, although I think we need a more accurate way to comparing than just starting the building website with a cleared cache.