feature/cong-fsharp-basic-types #31
Conversation
|
Thanks for this - looks good. The CI for this repo has apparently rotted quite badly, so I'll fix that up first. I'll also drop netframework support and target only netstandard, so we'll be able to get rid of the NETFRAMEWORK clause. |
I had a go here: #30 |
|
Thanks - I'm pretty opinionated about CI (sorry) so I'll copy-paste a bunch of that into a fresh branch. In particular I'll add steps ensuring the NuGet pack succeeds, for example, and I'll enable a single step we can use as a gate (as in e.g. https://github.com/G-Research/ApiSurface/blob/ffdd180be34a51fb8294f11750e4bfe775562c9d/.github/workflows/dotnet.yaml#L127-L134). |
I based it on another GResearch repo FWIW, but no worries 👍 |
|
We are large, we contain multitudes, I'm afraid. (If it was an F# repo, though, I'd like to remain consistent; which repo was it?) I'm also now tracking down the configured owner of TypeEquality, because it's definitely misconfigured; I don't have write access right now. There's going to be a delay on this, I'm afraid! |
Sorry, I can't recall. My branch doesn't push any Nuget artefacts, so probably needs to be redone anyways 👍 |
TypeEquality/Teq.fsi
Outdated
| /// returns the type equality for the corresponding Map type. | ||
| val map<'k1, 'v1, 'k2, 'v2 when 'k1 : comparison and 'k2 : comparison> : Teq<'k1, 'k2> -> Teq<'v1, 'v2> -> Teq<Map<'k1, 'v1>, Map<'k2, 'v2>> | ||
|
|
||
| #if !NETFRAMEWORK |
There was a problem hiding this comment.
Sigh, and all those changes through this FSI file too, I'm afraid!
Co-authored-by: Patrick Stevens <3138005+Smaug123@users.noreply.github.com>
Co-authored-by: Patrick Stevens <3138005+Smaug123@users.noreply.github.com>
Co-authored-by: Patrick Stevens <3138005+Smaug123@users.noreply.github.com>
Co-authored-by: Patrick Stevens <3138005+Smaug123@users.noreply.github.com>
Co-authored-by: Patrick Stevens <3138005+Smaug123@users.noreply.github.com>
Co-authored-by: Patrick Stevens <3138005+Smaug123@users.noreply.github.com>
|
Sorry, I merged |
…njlr/TypeEquality into feature/cong-fsharp-basic-types
TypeEquality/Teq.fs
Outdated
| believeMe prf | ||
|
|
||
| /// Given a type equality between two seq types, returns the type equality on the corresponding element types. | ||
| let seqOf<'a, 'b> (prf : Teq<'a seq, 'b seq>) : Teq<'a, 'b> = |
There was a problem hiding this comment.
Actually I think this one is unsound. Teq.seqOf-shaped things in general only work when the types are sealed. This function lets you construct a Teq<'a list, 'a ResizeArray>. The others all look fine to me.
TypeEquality/Teq.fs
Outdated
| (fst fstPrf) | ||
| (snd sndPrf) | ||
|
|
||
| /// Given a type equality between two types, returns the type equality on the corresponding Async types. |
There was a problem hiding this comment.
I know you're just being consistent with the rest of this file, but there's already an FSI file, so these doc comments aren't going to be surfaced anywhere. Probably worth just omitting them. (I only say this because I have incoming comments about docstrings, and it's not worth making the same markups twice.)
Co-authored-by: Patrick Stevens <3138005+Smaug123@users.noreply.github.com>
Co-authored-by: Patrick Stevens <3138005+Smaug123@users.noreply.github.com>
Co-authored-by: Patrick Stevens <3138005+Smaug123@users.noreply.github.com>
Co-authored-by: Patrick Stevens <3138005+Smaug123@users.noreply.github.com>
Co-authored-by: Patrick Stevens <3138005+Smaug123@users.noreply.github.com>
…njlr/TypeEquality into feature/cong-fsharp-basic-types
|
Thanks! Merge at will. |
Who can merge? 😅 |
|
Sorry - I thought you would get the button! |
2 participants
Extends
Congwith FSharp basic types