Wonky sexp detection

[plexus]

Seems the code to detect the start/end of the current sexp is still a bit wonky. See the video:

https://youtu.be/NIHWV3FH808

Also seeing a lot of these in the browser console

Error: Unmatched delimiter: %s("]") [at line 191, column 1]

[plexus]

Here's the code I had in the buffer

https://gist.github.com/c4fdf414aa83a90271a10e5e3b943e71

[plexus]

Seems I was missing a closing paren somewhere. This could arguably be handled more gracefully :)

[mhuebert]

You've certainly found a weakness: bracket highlighting, eval, and all other commands rely on parsed Clojure code. when the code in the panel does not represent a valid form, the reader chokes. This is happening in the 'mhuebert/magic-tree' library, where we dig quite deeply into the 'reading' process rather than using the standard clojure reader. So we have some flexibility. It would be nice to have: 1. Accurate error messages, as soon as this occurs 2. Perhaps modify our reader to be more forgiving, and still create an AST under 'broken' conditions when possible

…

On Tue, Jun 6, 2017 at 11:27 AM Arne Brasseur ***@***.***> wrote: Seems I was missing a closing paren somewhere. This could arguably be handled more gracefully :) — You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub <#19 (comment)>, or mute the thread <https://github.com/notifications/unsubscribe-auth/AAKFZ-YMdEt4k9eJw15Wti9MKit2oPMTks5sBRuOgaJpZM4NxEda> .

[mhuebert]

Closing this just to move the discussion to #20.

Also, a bit of further explanation of the video you posted (thanks btw, that's a really clear way to show the problem!) - currently whenever we are unable to generate an AST, we re-use the AST from the last valid state of the document. This can lead to really weird effects where seemingly random characters are bolded: these are the 'ghosts of parentheses past' ;-).