entering opening parens ends up with an extra space at the end:
then from this state:
attempting tab completion yields a FailedToPerform (doesn't advance caret as before)
then from this state:
pressing down key yields incorrect caret position:
then from this state:
entering "|" somewhat unexpectedly puts the arrow on the next line:
and pressing backspace leaves the whole rule as a ghost:
if I start entering the pattern, the arrow then pops up to the current line:
then from this state:
if I enter space and then "=" the following hole stays put, but if i just enter "=" with no preceeding space, the hole jumps up to the current line:
and then completing the arrow by pressing ">" causes the subsequent rule indentation to change:
and then from here:
inserting a wrapping parens inserts the closing on the next line, causing the "in" to move somewhat distractingly:
and then from here
inserting an opening parens results in the closing ghost ending up outside? which i assume is a bug but am not totally sure:
anyway, most of these aren't egregious but things are going to add up... i feel that about ~20-40% of token boundary actions causes something slightly unexpected to happen.
entering opening parens ends up with an extra space at the end:
attempting tab completion yields a FailedToPerform (doesn't advance caret as before)
pressing down key yields incorrect caret position:
entering "|" somewhat unexpectedly puts the arrow on the next line:
and pressing backspace leaves the whole rule as a ghost:
if I start entering the pattern, the arrow then pops up to the current line:
if I enter space and then "=" the following hole stays put, but if i just enter "=" with no preceeding space, the hole jumps up to the current line:
inserting a wrapping parens inserts the closing on the next line, causing the "in" to move somewhat distractingly:
inserting an opening parens results in the closing ghost ending up outside? which i assume is a bug but am not totally sure:
anyway, most of these aren't egregious but things are going to add up... i feel that about ~20-40% of token boundary actions causes something slightly unexpected to happen.