Fork me on GitHub
#parinfer
<
2017-06-16
>
shaunlebron15:06:38

“oakinfer” was a failed experiment. non-indentation based inference was a rabbit hole of edge cases, even under conservative constraints

shaunlebron15:06:03

v3 will have onBadInnerParen: 'erase' | 'suspend'. It will default to suspend, and highlight the offending paren. rationale: https://github.com/shaunlebron/parinfer/issues/131#issuecomment-309053770

shaunlebron15:06:23

i’ll add it to the demo to help plugin authors decide which option they want

cfleming22:06:35

@shaunlebron Thanks for all the great documentation of your thought process on this (and parinfer in general)

cfleming22:06:58

It’s very easy to skip this because it’s a lot of work, but it’s amazingly useful

shaunlebron22:06:56

@cfleming thanks for the encouragement!

shaunlebron22:06:26

it’s definitely a lot of work. it helps me keep context which I’m always afraid of losing

cfleming22:06:37

No problem, I wish I had the discipline to document my own work like this, since even in a solo-developer project it’s very useful

cfleming22:06:43

Cursive has a bunch of very different components, I tend to work on one and then come back to it months later - as you say, it helps context and remember the justification for decisions.

shaunlebron22:06:34

you probably have a better memory than I do then!

cfleming22:06:30

Not at all - I spend a lot of time re-reading my code 😞

cfleming22:06:07

Really, long term, documenting is almost certainly time-efficient, it just doesn’t feel like it when you’re doing it.

shaunlebron22:06:37

that’s a really good way to put that

shaunlebron22:06:08

i find it’s like climbing and taking the time to setup an anchor to clip into

shaunlebron22:06:07

i got tired of not finishing projects some years ago, and writing helped me overcome that