What would an additional syntax to “put comments in the source” text accomplish?
I thought that according to the 0.22 CommonMark spec an “HTML Comment” (which is also an “XML Comment”, and is a restricted form of a ISO 8879:1986 SGML comment declaration) would be
[…] parsed as a raw HTML tag and will be rendered in HTML without escaping.
What would be the need that introducing a “native” CommonMark comment syntax should satisfy? And how would the content of such a comment be represented in the output, if at all?
If the only goal is to discard the “comment in the source”, and not transfer it into the output document, I would prefer
-
an option that a CommonMark processor discards all “HTML comments”, and/or
-
a “special form” of “HTML comments” that will be discarded in any case by a CommonMark processor, like the one cited by @jgm above.
Because by definition a comment declaration aka “HTML comment” aka “XML comment” is not part of the (SGML/XML/HTML/XHMTL) document content, I see no compelling reason why such comments should be passed on into the CommonMark processor’s output anyway (by default at least).
In case a comment is supposed to “mean” something for a processor down the line: that’s what processing instructions are for, and they too are mentioned in the CommonMark spec as being passed through (IMO rightly so).