Closed tummychow closed 6 years ago
Thanks for this!
Sorry, I didn't spot (and can't now find) any mention of tests in #35. I'm not familiar with the tests myself, but as far as I can see, M-x eval-buffer
in dtrt-indent-tests.el
is sufficient. (When I do that with the current release 0.3 I get no output whatsoever, but that appears fine; eval-ing individual tests produces a reassuring nil
.)
Since there have been changes since the last release (and I hadn't noticed I'd forgotten to make another), I shall make another (the tests seem to pass!) right now.
Further PRs after this one certainly welcome.
I've now set up Travis; could you possibly rebase on top of master and set up Travis on your fork too?
Meanwhile, I think I misunderstood what you were going for with your PR: I thought you wanted to remove the syntax table just for bash, not for all languages. I'm not sure that removing it for all languages is a good idea. Could we start with just removing it for bash (I guess having a regex that simply allows every line would be the way to do it?)?
Also, mixing dtrt-indent and SMIE sounds like it has a use, as you say, but I suspect it'll throw up problems, so could it be behind a setting that defaults to off for now, please? Also, in a separate PR.
Restored all the syntax tables except the bash one, and removed the smie config guess change.
Thanks, that's great. As I say, having dtrt-indent in SMIE mode might well be good too, just ought to be toggle-able…and do bring on those other changes you mentioned.
As we discussed on #35. I actually don't know how to run the tests for this repo (haven't written a lot of elisp), so I can't tell if they're passing. Let me know if there are issues.
By the way, I have a couple more commits for your consideration here. I didn't want to piggyback them into this change because they're not really related (also, they're breaking), but let me know if you're willing to accept these in another PR.