-
-
Notifications
You must be signed in to change notification settings - Fork 415
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Replace type variables in a single sweep during reification #1888
Conversation
packages/term/ansi_term.pony
Outdated
@@ -80,7 +80,8 @@ actor ANSITerm | |||
return | |||
end | |||
|
|||
for c in (consume data).values() do | |||
let data' : Array[U8] ref = consume data |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Sorry, a small style consistency point - could you remove the space preceding the colon here, and in the other three places in this commit where it happens?
The C changes look good to me. However, I don't think we should merge this until we have a plan for #1887, as this will break some people's code and we don't have a plan for the permanent fix yet. |
Yeah, that sounds reasonable. |
This has come up again in #1921. The way I see it, with this bug the compiler is currently doing a type substitution which is just wrong, and it's only a coincidence that this substitution allows the standard library to build today. |
@plietar this is tagged as "CHANGELOG-FIXED". When this is merged the CHANGELOG comment would be "Replace type variables in a single sweep during reification" which I don't think is a not of value in a CHANGELOG. Ideally this would have a changelog comment that is meaningful to the average Pony user. I'd be in favor of either a) changing the title of this PR |
Did we ever get @sylvanc's thoughts on #1887? I'm still not very comfortable merging this until we have a plan of action for #1887. I think @sylvanc will have the best big-picture view of the original intent behind some of this stuff, as well as some other important insights to coming up with the right solution. |
We discussed #1887 on a call - we can indeed replace |
...and we should have the #1887 fix first, so that the manual rcap decay used here isn't needed. |
Again, like in #2503 i took the freedom to rebase this on master in order to get this rolling again. It seems i broke something within the stdlib though, fixing. |
@mfelsche any progress on this? it's been a while. |
Unfortunately not. It seems i broke it and i dont know why yet. Bein realistic here, i dont think i will manage it in the next few weeks. :( |
Currently blocked |
New PR: |
Fixes #1875
The first commit is a workaround for #1887. It can be removed when that issue is fixed (depending on how it is fixed). I don't know if we should wait for it to be fixed first.