Line blocks

[jgm]

Pandoc offers a line block format which is useful for addresses and such things as poetry. In this format, newlines are hard breaks, and all spaces are significant, even leading spaces.

| The limerick packs laughs anatomical
| Into space that is quite economical.
|   But the good ones I've seen
|   So seldom are clean
| And the clean ones so seldom are comical.

Note that the same effect can be achieved with backslash-newline and backslash-space, but it arguably looks less natural.

The limerick packs laughs anatomical\
Into space that is quite economical\
\ \ But the good ones I've seen\
\ \ So seldom are clean\
And the clean ones so seldom are comical.

In addition, the pipes may create confusion with pipe tables. So, not sure this feature is worth it.

[uvtc]

I didn't realize you could use backslash-space to make consecutive spaces that would not be condensed down to a single space. (I've previously used something like \hspace{0.5cm} for that (when targetting pdf)).

[matklad]

I think this can be generalized to a block, which applies only inline formatting, ignores block formatting and keeps whitespace. That is, verbatim + inlines.

This potentially could be useful for https://github.com/jgm/djot/discussions/49.

[Omikhleia]

Note that the same effect can be achieved with backslash-newline and backslash-space, but it arguably looks less natural.

Just my two cents regarding the semantics: what does that backlash-space really means here actually?

In regular justified content, it's usually a non-breaking space, so typically a U+00A0... But in proper typography those are stretchable and shrinkable as any interword-space, i.e. have a variable width. See Unicode TR14 §3.

In regular poetry, I'd say that the initial "indent" of a verse is (usually) a fixed-width space. It could be, for instance an em-quad (U+2001)... Whatever, it's usually not adjustable.

To work around this, a rendering engine could say, well, that \ may be seen as stretchable except when it's initial in a paragraph or after a forced line break, but that seems a bit contrived, perhaps?[^1]

I am not saying that Djot shall necessarily have Pandoc-like Line Blocks --- just that the suggested "same effect" using existing syntax might not be fully equivalent, and might impose a mixed interpretation of backslash-space depending on context.

[^1]: For the record, that's how I interpret the initial space in Pandoc-like Line Blocks in markdown.sile in the context of poetry typesetting.