Mitchell Dalvi Rosen comments

Results 292 comments of


                                            Mitchell Dalvi Rosen

round trip error with long abilities lists

the `Semi True` is coming from [this line](https://github.com/unisonweb/unison/blob/8a27961e9aa4fb1eea2d231a604cb722ee30dc6b/unison-syntax/src/Unison/Syntax/Lexer.hs#L227)

round trip error with long abilities lists

Just an idea: I see we don't emit virtual semis inside parens. Do we also want to avoid emitting virtual semis inside curly brackets like in this example? @pchiusano maybe...

round trip error with long abilities lists

Yeah @aryairani, it does

delete.branch should allow multiple arguments

I think this is because `delete.branch` isn't variadic, if `delete.branch 1-2` is expanding to `delete.branch allow_debug_totext error-details`. A `delete.branch 2` does work.

delete.branch should allow multiple arguments

No reason, I think it could/should be made variadic, too

diff.namespace produces malformed numbered args and `view` doesn't work for full hashes

@pchiusano I think `view` does work with full term/decl hashes, but not namespace hashes. Is that what used to work?

diff.namespace produces malformed numbered args and `view` doesn't work for full hashes

Oh! I see: `diff.namespace` is outputting numbered args like this: - `:` and `view` doesn't like those.

diff.namespace produces malformed numbered args and `view` doesn't work for full hashes

Oh? Am I looking at the wrong output above? Where's the ``? Also, I can't reproduce the behavior of `view` requiring the hash you give it to exist in the...

refactor: show more info about connection errors in output

What should we do with this one @aryairani?

Using quotation marks for arguments in repl, that allows for using sp…

Sorry for the delay; it's kind of hard for me to think through all the implications, but I'm leaning towards "this change makes sense". It does seem to me that...