RyanGlScott (Ryan Scott)
User

Projects

User does not belong to any projects.

Today

  • Clear sailing ahead.

Tomorrow

  • Clear sailing ahead.

Friday

  • Clear sailing ahead.

Badges

User Since
Jan 8 2015, 11:07 PM (110 w, 5 d)
Availability
Available

Recent Activity

Today

RyanGlScott added a comment to D3019: Fixes a spaceleak in `maximumBy` and `minimumBy` (#10830)..

Sorry for responding so late on this. I'll also give my 👍 on this to provide one more vote of confidence from a CLC member.

Wed, Feb 22, 8:48 AM
RyanGlScott added a comment to D3152: Bump process version in GHC [skip ci].

Well that was fast... process-1.6 is now released, so this is technically out of date.

Wed, Feb 22, 8:27 AM

Yesterday

RyanGlScott added a comment to D3153: Bump win32 version in GHC.

This has been superseded by http://git.haskell.org/ghc.git/commit/240b43e4969595418bb3c0d1519111be59572672, no?

Tue, Feb 21, 9:33 AM
RyanGlScott committed rGHC611f998fd545: Replace some pushTcLevelM's with pushTcLevelM_ (authored by RyanGlScott).
Replace some pushTcLevelM's with pushTcLevelM_
Tue, Feb 21, 8:54 AM
RyanGlScott committed rGHC4080a63c510f: Minor spelling, grammar, and formatting fixes (authored by RyanGlScott).
Minor spelling, grammar, and formatting fixes
Tue, Feb 21, 8:54 AM

Fri, Feb 10

RyanGlScott committed rGHC639e702b6129: Refactor DeriveAnyClass's instance context inference (authored by RyanGlScott).
Refactor DeriveAnyClass's instance context inference
Fri, Feb 10, 3:24 PM
RyanGlScott closed D2961: Refactor DeriveAnyClass's instance context inference by committing rGHC639e702b6129: Refactor DeriveAnyClass's instance context inference.
Fri, Feb 10, 3:24 PM
RyanGlScott committed rGHC283a346586e5: Prevent Template Haskell splices from throwing a spurious TypeInType error (authored by RyanGlScott).
Prevent Template Haskell splices from throwing a spurious TypeInType error
Fri, Feb 10, 9:35 AM
RyanGlScott closed D3022: Prevent Template Haskell splices from throwing a spurious TypeInType error by committing rGHC283a346586e5: Prevent Template Haskell splices from throwing a spurious TypeInType error.
Fri, Feb 10, 9:35 AM
RyanGlScott updated the diff for D2961: Refactor DeriveAnyClass's instance context inference.
  • Accept improved error messages
Fri, Feb 10, 9:19 AM

Thu, Feb 9

RyanGlScott updated the diff for D3022: Prevent Template Haskell splices from throwing a spurious TypeInType error.
  • Address Richard's comment
Thu, Feb 9, 7:42 PM
RyanGlScott accepted D3043: TH-spliced class instances are pretty-printed incorrectly post-#3384.

Template Haskell pretty-printing is fixed, so this looks good to me.

Thu, Feb 9, 7:35 PM
RyanGlScott updated the diff for D2961: Refactor DeriveAnyClass's instance context inference.

Rebase on top of master

Thu, Feb 9, 7:32 PM
RyanGlScott added a comment to D2961: Refactor DeriveAnyClass's instance context inference.

Thanks, @simonpj! We took a winding path to get here, but I think this finally does what it advertised. I've updated the comments and Note accordingly.

Thu, Feb 9, 7:30 PM
RyanGlScott updated the diff for D2961: Refactor DeriveAnyClass's instance context inference.
  • It works!
Thu, Feb 9, 7:20 PM
RyanGlScott added a comment to D2961: Refactor DeriveAnyClass's instance context inference.

Unfortunately, removing checkNoErrs doesn't change anything. I tried with and without the insolubleWC check. For posterity's sake, I've uploaded a Diff without checkNoErrs.

Thu, Feb 9, 10:02 AM
RyanGlScott updated the diff for D2961: Refactor DeriveAnyClass's instance context inference.
  • Don't use checkNoErrs
Thu, Feb 9, 10:02 AM
RyanGlScott added a comment to D2961: Refactor DeriveAnyClass's instance context inference.

Well I'm happy that we can delete a bunch of code, but sadly, doing so does not fix the testsuite failures (in fact, even more tests fail now).

Thu, Feb 9, 9:50 AM
RyanGlScott updated the diff for D2961: Refactor DeriveAnyClass's instance context inference.
  • Simon's suggestions
Thu, Feb 9, 9:48 AM

Wed, Feb 8

RyanGlScott added a comment to D2961: Refactor DeriveAnyClass's instance context inference.

Sorry, I meant arc patch D2961.

Wed, Feb 8, 10:19 AM
RyanGlScott added a comment to D2961: Refactor DeriveAnyClass's instance context inference.

Four error messages; that's strange. Is there a branch I can build. Or some way to repro?

Wed, Feb 8, 10:18 AM
RyanGlScott added a comment to D2961: Refactor DeriveAnyClass's instance context inference.

@simonpj, I feel I'm stuck again. I've tried two different things:

Wed, Feb 8, 9:05 AM
RyanGlScott updated the diff for D2961: Refactor DeriveAnyClass's instance context inference.
  • in re simonpj
Wed, Feb 8, 9:05 AM

Tue, Feb 7

RyanGlScott planned changes to D2961: Refactor DeriveAnyClass's instance context inference.
Tue, Feb 7, 10:39 AM
RyanGlScott updated the diff for D2961: Refactor DeriveAnyClass's instance context inference.
  • Latest attempt
Tue, Feb 7, 10:07 AM

Mon, Feb 6

RyanGlScott accepted D3072: Derive <$.

Looks great @dfeuer! One minor comment inline.

Mon, Feb 6, 4:28 PM
RyanGlScott committed rGHCa28a55211d6f: Remove unnecessary use of -DGENERICS flag (authored by RyanGlScott).
Remove unnecessary use of -DGENERICS flag
Mon, Feb 6, 1:34 PM

Sun, Feb 5

RyanGlScott accepted D3031: Add liftA2 to Applicative class.

Assuming it validates, LGTM.

Sun, Feb 5, 5:04 PM
RyanGlScott added a comment to D3043: TH-spliced class instances are pretty-printed incorrectly post-#3384.
In D3043#90517, @alanz wrote:

I have updated the patch to do away with pprParendExpr and pprParendLExpr, to ensure that any hsSyn AST generation does actually insert parens in the right place. Doing the same for pprParendPat is also trivial.

Before tackling pprParendHsType I am looking for some feedback as to whether this makes sense, in particular whether there are any debug features that require a heavily parenthesised output that will be broken by this.

Sun, Feb 5, 3:12 PM
RyanGlScott requested changes to D3072: Derive <$.

Thanks for working on this, @dfeuer!

Sun, Feb 5, 9:19 AM
RyanGlScott requested changes to D3031: Add liftA2 to Applicative class.

Looks generally good. What about @rwbarton's comment which asks to add a performance test?

Sun, Feb 5, 9:09 AM

Sat, Feb 4

RyanGlScott added a comment to D3072: Derive <$.
In D3072#90199, @dfeuer wrote:

@RyanGlScott, I think the basic issue is this. Suppose we have

data T ... a =  C (f a)

Then we want x <$ C q = C (x <$ q).

But if we have

data U ... a = C (f (g a))

then we want

x <$ C q = C (fmap (x <$) q)

But we only have one "case" in FFoldType for both.

Sat, Feb 4, 10:02 AM

Fri, Feb 3

RyanGlScott added inline comments to D3072: Derive <$.
Fri, Feb 3, 9:19 PM
RyanGlScott added inline comments to D3072: Derive <$.
Fri, Feb 3, 9:11 PM

Thu, Feb 2

RyanGlScott added a comment to D2961: Refactor DeriveAnyClass's instance context inference.

@simonpj, I've updated mkMinimalBySCs to remove duplicate constraints, so now we don't need to call the simplifier twice.

Thu, Feb 2, 6:44 PM
RyanGlScott updated the diff for D2961: Refactor DeriveAnyClass's instance context inference.
  • Don't call the simplifer twice needlessly
Thu, Feb 2, 6:36 PM
RyanGlScott updated the diff for D2961: Refactor DeriveAnyClass's instance context inference.

Oops, use right base commit

Thu, Feb 2, 2:04 PM
RyanGlScott updated the diff for D2961: Refactor DeriveAnyClass's instance context inference.

Make mkMinimalBySCs remove duplicate constraints

Thu, Feb 2, 2:04 PM

Wed, Feb 1

RyanGlScott added inline comments to D2961: Refactor DeriveAnyClass's instance context inference.
Wed, Feb 1, 8:36 PM
RyanGlScott accepted D3059: base: Derive Enum, Bounded for VecCount, VecElem.

Ah! Then by all means, derive all the instances you find convenient.

Wed, Feb 1, 7:59 PM
RyanGlScott added a comment to D3059: base: Derive Enum, Bounded for VecCount, VecElem.

I've no personal objections. Although I'm curious what the use case is—aren't these supposed to only ever appear at the type level?

Wed, Feb 1, 7:47 PM
RyanGlScott added a comment to D3058: Make interface loading for COMPLETE pragmas lazy.

Huh, I knew about the giant fixM block in the typechecker to tie an enormous recursive knot, but I never knew about forkM until now. Tricky stuff.

Wed, Feb 1, 5:00 PM

Tue, Jan 31

RyanGlScott added a comment to D2961: Refactor DeriveAnyClass's instance context inference.

What is the status of this for GHC 8.2?

Tue, Jan 31, 5:12 PM

Mon, Jan 30

RyanGlScott updated the diff for D2961: Refactor DeriveAnyClass's instance context inference.

Rebase on top of master

Mon, Jan 30, 8:16 PM
RyanGlScott added a comment to D3043: TH-spliced class instances are pretty-printed incorrectly post-#3384.
In D3043#89268, @alanz wrote:
In particular, converted expressions still aren't wrapped up in `HsPar`s, and they're only pretty-printed correctly because the `Outputable` instance for `HsExpr` behaves differently from those of `HsType` and `Pat`.

For HsPat, Convert makes use of mkParPat which in turn calls hsPatNeedsParens.
For HsExpr, Convert makes use of mkLHsPar which in trun calls hsExprNeedsParens.

Each of these do much the same as wrap_apps does for HsType.

Mon, Jan 30, 6:20 PM
RyanGlScott added a comment to D2983: Check that a default type signature aligns with the non-default signature.

Thanks @bgamari. If @simonpj doesn't speak up, I'll plan on landing this in the next couple of days.

Mon, Jan 30, 10:39 AM
RyanGlScott committed rGHCf60287c478a5: Fix mismatched tick in GHC.Generics documentation (authored by RyanGlScott).
Fix mismatched tick in GHC.Generics documentation
Mon, Jan 30, 8:35 AM

Sun, Jan 29

RyanGlScott requested changes to D3043: TH-spliced class instances are pretty-printed incorrectly post-#3384.

Thanks @alanz, this is looking much better.

Sun, Jan 29, 4:36 PM

Sat, Jan 28

RyanGlScott committed rGHCbc42e2b03a87: Convert pprTrace in isPredTy to a WARN (authored by RyanGlScott).
Convert pprTrace in isPredTy to a WARN
Sat, Jan 28, 4:44 PM
RyanGlScott closed D3033: Convert pprTrace in isPredTy to a WARN by committing rGHCbc42e2b03a87: Convert pprTrace in isPredTy to a WARN.
Sat, Jan 28, 4:44 PM
RyanGlScott updated the diff for D3033: Convert pprTrace in isPredTy to a WARN.
  • Fix Haddocks
Sat, Jan 28, 9:49 AM

Fri, Jan 27

RyanGlScott accepted D3025: Fixes bug #11046.

Thanks, Iavor.

Fri, Jan 27, 5:11 PM
RyanGlScott requested changes to D3029: Expand list of always loaded Windows shared libs.

Also, a reminder to make a mention of this in the release notes, since I'll inevitably forget about all of these little differences in the linker between 8.0 and 8.2.

Fri, Jan 27, 2:53 PM
RyanGlScott added a comment to D3022: Prevent Template Haskell splices from throwing a spurious TypeInType error.

Urgh, there is a downside to this patch, though: this program (which legitimately requires -XTypeInType):

Fri, Jan 27, 10:51 AM
RyanGlScott retitled D3033: Convert pprTrace in isPredTy to a WARN from to Convert pprTrace in isPredTy to a WARN.
Fri, Jan 27, 9:11 AM

Thu, Jan 26

RyanGlScott requested changes to D3025: Fixes bug #11046.

Actually, one tiny (but important) thing: can you mention this in the 8.2 release notes?

Thu, Jan 26, 9:17 PM
RyanGlScott requested changes to D3031: Add liftA2 to Applicative class.

Oops, I didn't catch that you said you'd open a Trac issue in https://phabricator.haskell.org/D3031#88747. Splendid.

Thu, Jan 26, 8:56 PM
RyanGlScott added a comment to D3031: Add liftA2 to Applicative class.

Indeed, you should at the very least:

Thu, Jan 26, 8:44 PM
RyanGlScott accepted D3025: Fixes bug #11046.

Seems reasonable to me.

Thu, Jan 26, 6:18 PM
RyanGlScott added a comment to D3030: Fix nofib benchmarks on Windows..
In D3030#88723, @Phyx wrote:

It requires msys, just because the shell detection is done using msys and because of the unix tools used. tr, sed, diff and perl. and of course make.

Thu, Jan 26, 5:56 PM
RyanGlScott added a comment to D3028: Load dependent dlls..

Ah, I understand now. Given your description, I think this discrepancy might be acceptable. But we should advertise this in docs/users_guide/bugs.rst just in case (in addition to the release notes).

Thu, Jan 26, 5:55 PM
RyanGlScott added a comment to D3030: Fix nofib benchmarks on Windows..

To be clear, does this require MSYS? Or does it work Cygwin with too?

Thu, Jan 26, 5:50 PM
RyanGlScott added a comment to D3028: Load dependent dlls..

I'm a bit lost. Can you give an example of a program which, using a version of GHC with this patch, would fail to compile but would work with GHCi?

Thu, Jan 26, 5:43 PM
RyanGlScott added a comment to D3029: Expand list of always loaded Windows shared libs.

Or maybe this should be punted till a proper implementation can be made for 8.4.1?

Thu, Jan 26, 5:39 PM
RyanGlScott added a comment to D3027: Slighly clean up symbol loading error..

Any thoughts on this or would it not be useful?

Thu, Jan 26, 5:34 PM
RyanGlScott committed rGHC3eebd1f5fd56: Generalizes the type of asProxyTypeOf (#12805) (authored by dalaing).
Generalizes the type of asProxyTypeOf (#12805)
Thu, Jan 26, 11:48 AM
RyanGlScott committed rGHCad3d2dfa19a1: Don't unnecessarily qualify TH-converted instances with empty contexts (authored by RyanGlScott).
Don't unnecessarily qualify TH-converted instances with empty contexts
Thu, Jan 26, 11:48 AM
RyanGlScott closed D3018: Don't unnecessarily qualify TH-converted instances with empty contexts by committing rGHCad3d2dfa19a1: Don't unnecessarily qualify TH-converted instances with empty contexts.
Thu, Jan 26, 11:48 AM
RyanGlScott closed D3017: Generalizes the type of asProxyTypeOf (#12805) by committing rGHC3eebd1f5fd56: Generalizes the type of asProxyTypeOf (#12805) (authored by dalaing).
Thu, Jan 26, 11:48 AM
RyanGlScott committed rNOFIBce4b36b56727: Add .arcconfig file to allow patches by phabricator (authored by mpickering).
Add .arcconfig file to allow patches by phabricator
Thu, Jan 26, 11:36 AM
RyanGlScott closed D2945: Add .arcconfig file to allow patches by phabricator by committing rNOFIBce4b36b56727: Add .arcconfig file to allow patches by phabricator (authored by mpickering).
Thu, Jan 26, 11:36 AM
RyanGlScott abandoned D2945: Add .arcconfig file to allow patches by phabricator.

I landed this patch (by hand) in http://git.haskell.org/nofib.git/commit/ce4b36b56727f1bdfc318fc0f7f4c4b192781537.

Thu, Jan 26, 11:33 AM

Wed, Jan 25

RyanGlScott committed rGHC88a89b761ae3: Nix typo and redundant where-clauses (authored by RyanGlScott).
Nix typo and redundant where-clauses
Wed, Jan 25, 10:53 PM
RyanGlScott committed rGHCe4ab8ba72af2: Add pragCompleteDName to templateHaskellNames (authored by RyanGlScott).
Add pragCompleteDName to templateHaskellNames
Wed, Jan 25, 10:36 PM
RyanGlScott retitled D3022: Prevent Template Haskell splices from throwing a spurious TypeInType error from to Prevent Template Haskell splices from throwing a spurious TypeInType error.
Wed, Jan 25, 9:31 PM
RyanGlScott added a comment to D2983: Check that a default type signature aligns with the non-default signature.

I like your deeplyInstantiate2 much more than tcSplitNestedSigmaTys. Use the former not the latter.

Wed, Jan 25, 6:54 PM
RyanGlScott accepted D3021: Update Win32 submodule to fix Windows build.
Wed, Jan 25, 8:57 AM

Tue, Jan 24

RyanGlScott updated subscribers of D3018: Don't unnecessarily qualify TH-converted instances with empty contexts.
Tue, Jan 24, 11:35 PM
RyanGlScott retitled D3018: Don't unnecessarily qualify TH-converted instances with empty contexts from to Don't unnecessarily qualify TH-converted instances with empty contexts.
Tue, Jan 24, 11:30 PM
RyanGlScott accepted D3017: Generalizes the type of asProxyTypeOf (#12805).

LGTM!

Tue, Jan 24, 10:34 PM
RyanGlScott added a comment to D2038: TysPrim: Generalize kind of (->).

@bgamari, I fear you might have lost some commits during rebasing?

Tue, Jan 24, 6:21 PM
RyanGlScott requested changes to D3017: Generalizes the type of asProxyTypeOf (#12805).

Looks great @dalaing, thanks. Can you also add a note to libraries/base/changelog.md about this change?

Tue, Jan 24, 5:28 PM
RyanGlScott added a comment to D2999: base: Improve Prelude.read error message.

@erikd's the author, not me ;)

Tue, Jan 24, 4:08 PM
RyanGlScott committed rGHC9fd87ef8a16f: Don't put foralls in front of TH-spliced GADT constructors that don't need them (authored by RyanGlScott).
Don't put foralls in front of TH-spliced GADT constructors that don't need them
Tue, Jan 24, 9:51 AM
RyanGlScott closed D3002: Don't put foralls in front of TH-spliced GADT constructors that don't need them by committing rGHC9fd87ef8a16f: Don't put foralls in front of TH-spliced GADT constructors that don't need them.
Tue, Jan 24, 9:51 AM
RyanGlScott accepted D3013: Skip path_with_commas when dyn unavailable.

That fixes the issue for me! I now realize that I was using a build with DYNAMIC_GHC_PROGRAMS=NO, which was triggering the bug.

Tue, Jan 24, 8:32 AM

Mon, Jan 23

RyanGlScott added a comment to D2983: Check that a default type signature aligns with the non-default signature.

@simonpj, I've added a blurb at the end of Note [Default method type signatures must align] describing why the methods' contexts are allowed to differ.

Mon, Jan 23, 8:14 PM
RyanGlScott updated the diff for D2983: Check that a default type signature aligns with the non-default signature.
  • in re simonpj
Mon, Jan 23, 8:14 PM
RyanGlScott added a comment to rGHC729a5e452db5: Don't quantify implicit type variables when quoting type signatures in TH.

Hopefully https://phabricator.haskell.org/D3002 resolves this issue.

Mon, Jan 23, 11:01 AM
RyanGlScott retitled D3002: Don't put foralls in front of TH-spliced GADT constructors that don't need them from to Don't put foralls in front of TH-spliced GADT constructors that don't need them.
Mon, Jan 23, 11:00 AM
RyanGlScott raised a concern with rGHC729a5e452db5: Don't quantify implicit type variables when quoting type signatures in TH.

Urk, it turns out that th-desugar has a test case that this commit broke:

Mon, Jan 23, 9:41 AM

Jan 23 2017

RyanGlScott added inline comments to D2983: Check that a default type signature aligns with the non-default signature.
Jan 23 2017, 9:09 AM
RyanGlScott updated the diff for D2983: Check that a default type signature aligns with the non-default signature.
  • mpickering's comments
Jan 23 2017, 9:09 AM
RyanGlScott committed rGHC729a5e452db5: Don't quantify implicit type variables when quoting type signatures in TH (authored by RyanGlScott).
Don't quantify implicit type variables when quoting type signatures in TH
Jan 23 2017, 8:17 AM
RyanGlScott closed D2974: Don't quantify implicit type variables when quoting type signatures in TH by committing rGHC729a5e452db5: Don't quantify implicit type variables when quoting type signatures in TH.
Jan 23 2017, 8:17 AM

Jan 22 2017

RyanGlScott added a comment to D2974: Don't quantify implicit type variables when quoting type signatures in TH.

Good point @bgamari, I've added a Note (and references to it) in DsMeta.

Jan 22 2017, 12:47 PM
RyanGlScott updated the diff for D2974: Don't quantify implicit type variables when quoting type signatures in TH.
  • Add a Note to DsMeta
Jan 22 2017, 12:47 PM
RyanGlScott committed rGHC560bc289fc6a: Revert "Remove unnecessary isTyVar tests in TcType" (authored by RyanGlScott).
Revert "Remove unnecessary isTyVar tests in TcType"
Jan 22 2017, 12:42 PM
RyanGlScott added a reverting commit for rGHCa0899b2f66a4: Remove unnecessary isTyVar tests in TcType: rGHC560bc289fc6a: Revert "Remove unnecessary isTyVar tests in TcType".
Jan 22 2017, 12:42 PM
RyanGlScott closed D2931: Revert "Remove unnecessary isTyVar tests in TcType" by committing rGHC560bc289fc6a: Revert "Remove unnecessary isTyVar tests in TcType".
Jan 22 2017, 12:42 PM