blockquotes violate principle of uniformity

[vassudanagunta]

principle of uniformity: if a chunk of text has a certain meaning, it will continue to have the same meaning when put into a container block (such as a list item or blockquote).

The following:

List with sublist:

- a
  - b

Exact same lines put in a blockquote container block:

>- a
>  - b

is rendered by CommonMark 0.29 as

Please compare the above with the example given in the spec just below the definition of the principle of uniformity

A plurality of markdown implementations, including the original Gruber, correctly follow the principle in this regard (I'm counting all the CommonMark implementations as one).

Another example of the issue:

     indented 5 spaces
    indented 4 spaces
     indented 5 spaces

The above lines, in a blockquote:

>     indented 5 spaces
>    indented 4 spaces
>     indented 5 spaces

is rendered as:

Though in this case, all the implementations except MultiMarkdown get it wrong. I would say this latter case is esoteric and unimportant. The first case, the handling of nested lists as demonstrated in the first example, should determine the proper course of action.

[vassudanagunta]

This issue has common cause (double entendre intended) with #421 and #460, both of which are currently attached to the 0.30 milestone.

I think the many proposals considered under #466 should also be evaluated against at least the first example above.

[Crissov]

The spec should really disallow blockquote markers without a single space following it,

>- a
>  - b

<p>&gt;- a
&gt;  - b</p>

but since that would break backwards compatibility, they should tu least not be mend together in a single blockquote, cf. different bullet list markers.

Then, this Commonmark input

>- a
>  - b

would be identical to

>- a

>  - b

and both would yield the same (normalized) HTML output:

<blockquote><ul><li>a</li></ul></blockquote>
<blockquote><ul> <li>b</li></ul></blockquote>

[vassudanagunta]

That interpretation, @Crissov, to me is very counterintuitive and runs counter to Markdown's design principles.

[jgm]

Yeah, this is a problem for sure. As noted in my comment on #460, it's a tricky one.

[Crissov]

@vassudanagunta Blockquote markers are the only line prefixes that do not require a space following them, since hash-sign headings have been fixed.

PS: Blockquotes - whitespace after > thread in the CM Discourse