Closed wooorm closed 8 years ago
I've updated mast to remark, and renamed the project/repo.
As for the detection algorithm, I'll have to spend some time on this :) So I'll repurpose this issue. Thanks for the input.
Trying to make sense of your comments....
So basically, in your PPS you're saying that mdast-util-to-nlcst
as is now cannot really be used towards this lint rule, right?
I went ahead and merged #15 which gives pretty good behavior. What do you think of that PR?
Since that PR shows pretty satisfactory behavior (at least for my uses), I'm going to close this issue, but please feel free to provide any feedback :) As always, thanks for all the great work @wooorm
I no longer know what I was referring to in my P.P.S. Weird.
Anyway, thanks @chcokr for upgrading to remark!
P.S. Also: I’ve done quite some work on disambiguating regular dots from sentence stops with
retext
, maybe investigate usingmdast-util-to-nlcst
here in order to disambiguate between the two.In NLCST, check for the children of each
SentenceNode
(starting from back to front), and if you find aPunctuationNode
, check whether it contains (withnlcst-to-string
) just terminal-punctuation (?
,!
, and.
). If that’s the case, the sentence ends in a terminal marker, which should be followed by aWhiteSpaceNode
containing anewline
.P.P.S. I just though of the fact that with the current set-up, it’s impossible to add a break after a
.
(markdown needs two spaces, then a newline, for that).