search

jeffreykegler / yahc

Yet Another Hoon Compiler
MIT License
0 stars 1 forks source link

LUTE status #25

Closed jeffreykegler closed 5 years ago

jeffreykegler commented 5 years ago

Here is the lint output for lute's, which will be the basis of comments to follow.

lute.lint.txt

jeffreykegler commented 5 years ago

There is a kind of pattern to the whitespace, but it looks awfully like it might be special to this case. I wonder if I should just leave lute's totally "free form" -- no lint warnings, except perhaps for problematic interior comments?

@ohAitch -- what do you think?

ohAitch commented 5 years ago

Lute is the weird tall-form [] right? Not enforcing whitespace seems reasonable yup, I think its main use is going to be this kind of "we're not using normal irregular :* because we wanted to align things"; and multiline nock literals, which will want to be indented according to the nock operators which show up as "freeform" extra indents e.g.

[ %6  [%3 %0 1]
    [%1 0]
  [%0 2]
]

Besides interior comments, "the [ and ] are either on the same line or same column, and in the latter case contents are indented past that parent column" might be good?

jeffreykegler commented 5 years ago

Will do. Since the multiline form has no examples in the arvo/ corpus, I may continue to treat it as NYI, with a comment in the code as to what you wanted done for when the case comes up.

I will mark this as my task and then, assuming I have no further questions, mark it "ready to close"

jeffreykegler commented 5 years ago

I have implemented lutes, and updated [the whitespace doc](https://github.com/jeffreykegler/yahc/blob/master/misc/whitespace.md. accordingly. One-line lutes are free-form, multi-line lutes (of which arvo/ contains no examples) will cause a "not yet implemented" warning.

I will mark this "ready for closing", and if nothing comes up, close it shortly.

jeffreykegler commented 5 years ago

Closed as per the above.