October 2017 - ghc-tickets

Re: [GHC] #7169: Warning for incomplete record field label used as function
by GHC 26 Oct '17

26 Oct '17

#7169: Warning for incomplete record field label used as function -------------------------------------+------------------------------------- Reporter: goldfire | Owner: nakaji_dayo Type: feature request | Status: closed Priority: normal | Milestone: 8.4.1 Component: Compiler | Version: 7.7 Resolution: fixed | Keywords: Warnings, | newcomer Operating System: Unknown/Multiple | Architecture: Type of failure: Incorrect | Unknown/Multiple warning at compile-time | Test Case: Blocked By: | Blocking: Related Tickets: | Differential Rev(s): Phab:D4083 Wiki Page: | -------------------------------------+------------------------------------- Comment (by bgamari): Yes, but I don't see an alternative here; we unfortunately only have so many options. There is already precedent within GHC to treat underscore- prefixed identifiers specially. I would much rather continue relying on that same convention than introduce yet more special cases. Ultimately the right solution here would likely be something like #602, but seeing as that's not on the horizon, I think we'll just have to make due with what we have. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/7169#comment:30> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 0

[GHC] #14377: false result in the addition
by GHC 26 Oct '17

26 Oct '17

#14377: false result in the addition -------------------------------------+------------------------------------- Reporter: vanto | Owner: (none) Type: bug | Status: new Priority: normal | Milestone: Component: Compiler | Version: 8.2.1 Keywords: | Operating System: Unknown/Multiple Architecture: | Type of failure: Incorrect result Unknown/Multiple | at runtime Test Case: | Blocked By: Blocking: | Related Tickets: Differential Rev(s): | Wiki Page: -------------------------------------+------------------------------------- {{{ Prelude> 1.1+2.2 3.3000000000000003 Prelude> 1.11+2.22 3.33 Prelude> 1.111+2.222 3.333 Prelude> 1.1111+2.2222 3.3333 Prelude> 1.11111+2.22222 3.33333 Prelude> 1.111111+2.222222 3.3333329999999997 Prelude> 1.1111111+2.2222222 3.3333333 Prelude> 1.11111111+2.22222222 3.33333333 Prelude> 1.111111111+2.222222222 3.333333333 Prelude> 1.1111111111+2.2222222222 3.3333333333 Prelude> 1.11111111111+2.22222222222 3.33333333333 Prelude> 1.111111111111+2.222222222222 3.333333333333 Prelude> 1.1111111111111+2.2222222222222 3.3333333333333 Prelude> 1.11111111111111+2.22222222222222 3.3333333333333304 Prelude> 1.111111111111111+2.222222222222222 3.333333333333333 }}} Given the other additions, three results can be improved. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/14377> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 18

[GHC] #14389: Improved results in GHCi during basic calculations
by GHC 26 Oct '17

26 Oct '17

#14389: Improved results in GHCi during basic calculations -------------------------------------+------------------------------------- Reporter: vanto | Owner: (none) Type: feature | Status: new request | Priority: normal | Milestone: Component: Compiler | Version: 8.2.1 Keywords: | Operating System: Unknown/Multiple Architecture: | Type of failure: Other Unknown/Multiple | Test Case: | Blocked By: Blocking: | Related Tickets: Differential Rev(s): | Wiki Page: -------------------------------------+------------------------------------- We can use GHCi as a calculator using {{{+}}}, {{{-}}}, {{{*}}}, {{{/}}}, {{{sin}}}, {{{cos}}}, and others. sometimes the result does not meet our expectations. for example if I calculate 3 + 2, the result is 5 and in line with our expectations. \\ {{{ Prelude> 3+2 5 }}} but sometimes it is not.\\ for the two examples below, please, can we have in GHCi a result other than the one given?\\ {{{ Prelude> exp(log(1000)-log(10)) 99.99999999999996 }}} can we have a result equal to {{{100}}}? \\ {{{ Prelude> 12*(1/3-1/4)-1 -2.220446049250313e-16 }}} can we have a result equal to {{{0}}}? \\ -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/14389> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 7

[GHC] #14364: Reduce repetition in derived Read instances
by GHC 26 Oct '17

26 Oct '17

#14364: Reduce repetition in derived Read instances -------------------------------------+------------------------------------- Reporter: bgamari | Owner: (none) Type: task | Status: new Priority: normal | Milestone: 8.4.1 Component: Compiler | Version: 8.2.1 Keywords: | Operating System: Unknown/Multiple Architecture: | Type of failure: None/Unknown Unknown/Multiple | Test Case: | Blocked By: Blocking: | Related Tickets: #10980 #7258 Differential Rev(s): | Wiki Page: -------------------------------------+------------------------------------- While looking at #7258 with tdammers, I noticed that 5000 of the 7000 terms that the `W2` module simplifies to are attributed to `$creadPrec`. In fact, much of this is repetition of the form, {{{#!hs (GHC.Read.expectP (Text.Read.Lex.Ident (GHC.CString.unpackCString# "field1"#))) (>> @ Text.ParserCombinators.ReadPrec.ReadPrec Text.ParserCombinators.ReadPrec.$fMonadReadPrec @ () @ DT (GHC.Read.expectP (Text.Read.Lex.Punc (GHC.CString.unpackCString# "="#))) (>>= @ Text.ParserCombinators.ReadPrec.ReadPrec Text.ParserCombinators.ReadPrec.$fMonadReadPrec @ Int @ DT (Text.ParserCombinators.ReadPrec.reset @ Int (GHC.Read.readPrec @ Int GHC.Read.$fReadInt)) (\ (a1_a1oe :: Int) -> >> @ Text.ParserCombinators.ReadPrec.ReadPrec Text.ParserCombinators.ReadPrec.$fMonadReadPrec @ () @ DT (GHC.Read.expectP (Text.Read.Lex.Punc (GHC.CString.unpackCString# ","#))) }}} Let's factor this pattern out into a `readField` helper in `GHC.Read`, {{{#!hs readField :: String -> ReadPrec a -> ReadPrec a readField fieldName readVal = do expectP (Ident fieldName) expectP (Punc "=") readVal {-# NOINLINE readField #-} }}} This will at least knock off a constant factor from the size of what should not be performance-critical code. We could also try folding the terminal "," into this, although then we would need to somehow handle the last field specially. This might not be worth it. Perhaps instead just factor out the comma case as well, {{{#!hs readComma :: ReadPrec () readComma = expectP (Punc ",") {-# NOINLINE readField #-} }}} It's unclear whether this is worth it, however. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/14364> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 9

[GHC] #14387: listToMaybe doesn't participate in foldr/build fusion
by GHC 26 Oct '17

26 Oct '17

#14387: listToMaybe doesn't participate in foldr/build fusion -------------------------------------+------------------------------------- Reporter: duog | Owner: (none) Type: bug | Status: new Priority: normal | Milestone: Component: Core | Version: 8.2.1 Libraries | Keywords: | Operating System: Unknown/Multiple Architecture: | Type of failure: None/Unknown Unknown/Multiple | Test Case: | Blocked By: Blocking: | Related Tickets: Differential Rev(s): | Wiki Page: -------------------------------------+------------------------------------- I noticed that `Data.OldList.findIndex` seems to use more memory than necessary, and that changing the definition of `listToMaybe` to be in terms of `foldr` fixed the situation. Consider the following module: {{{ {-# LANGUAGE MagicHash #-} {-# OPTIONS_GHC -ddump-to-file -ddump-prep -O #-} module FindIndex where import GHC.Base (Int(I#), build) import GHC.Prim -- | The definitions of listToMaybe, findIndices and findIndex are taken from base listToMaybe :: [a] -> Maybe a listToMaybe [] = Nothing listToMaybe (a:_) = Just a findIndices :: (a -> Bool) -> [a] -> [Int] findIndices p ls = build $ \c n -> let go x r k | p x = I# k `c` r (k +# 1#) | otherwise = r (k +# 1#) in foldr go (\_ -> n) ls 0# {-# inline findIndices #-} findIndex :: (a -> Bool) -> [a] -> Maybe Int findIndex p = listToMaybe . findIndices p -- This is the definition of findIndices when USE_REPORT_PRELUDE is defined findIndices' :: (a -> Bool) -> [a] -> [Int] findIndices' p xs = [ i | (x,i) <- zip xs [0..], p x] {-# inline findIndices' #-} listToMaybe' :: [a] -> Maybe a listToMaybe' = foldr (const . Just) Nothing -- | using listToMaybe', we get a join point findIndex2 :: (a -> Bool) -> [a] -> Maybe Int findIndex2 p = listToMaybe' . findIndices p -- | a "manual" implementaiton, we get a join point findIndex3 :: (a -> Bool) -> [a] -> Maybe Int findIndex3 p = go . zip [0..] where go [] = Nothing go ((i, x) : xs) | p x = Just i | otherwise = go xs -- | alternate version of findIndices, stock listToMaybe, no join point findIndex4 :: (a -> Bool) -> [a] -> Maybe Int findIndex4 p = listToMaybe . findIndices' p -- | alternate version of findIndices, foldr listToMaybe, we get a join point findIndex5 :: (a -> Bool) -> [a] -> Maybe Int findIndex5 p = listToMaybe' . findIndices' p }}} Find attached .dump-prep files with ghc-8.2.1 and ghc-head at commit 5c178012f47420b5dfa417be21146ca82959d273. My interpretation of this is: with both ghc-8.2.1 and ghc-head, findIndex{2,4,5} get join points and findIndex{"",3} don't. Having a join point means constant stack space, not having a join point means linear stack space. I don't understand the simplifier well enough to know whether ghc could do better here, but it seems that changing the definition of `listToMaybe` to {{{ listToMaybe :: [a] -> Maybe a listToMaybe = foldr (const . Just) Nothing }}} would be a win. Are there any downsides? -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/14387> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 9

Re: [GHC] #7169: Warning for incomplete record field label used as function
by GHC 25 Oct '17

25 Oct '17

#7169: Warning for incomplete record field label used as function -------------------------------------+------------------------------------- Reporter: goldfire | Owner: nakaji_dayo Type: feature request | Status: closed Priority: normal | Milestone: 8.4.1 Component: Compiler | Version: 7.7 Resolution: fixed | Keywords: Warnings, | newcomer Operating System: Unknown/Multiple | Architecture: Type of failure: Incorrect | Unknown/Multiple warning at compile-time | Test Case: Blocked By: | Blocking: Related Tickets: | Differential Rev(s): Phab:D4083 Wiki Page: | -------------------------------------+------------------------------------- Comment (by AntC): Replying to [comment:27 Ben Gamari <ben@…>]: > {{{ > Warning on declaring a partial record selector. > However, disable warn with field names that start with underscore. > > }}} Eek. I may be too late. Don't the lenses people use underscore prefix on record labels? So that TH can detect that and produce a lens name with the `_` stripped off? Then they're going to miss out on this warning of partiality. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/7169#comment:29> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 0

Re: [GHC] #7169: Warning for incomplete record field label used as function
by GHC 25 Oct '17

25 Oct '17

#7169: Warning for incomplete record field label used as function -------------------------------------+------------------------------------- Reporter: goldfire | Owner: nakaji_dayo Type: feature request | Status: closed Priority: normal | Milestone: 8.4.1 Component: Compiler | Version: 7.7 Resolution: fixed | Keywords: Warnings, | newcomer Operating System: Unknown/Multiple | Architecture: Type of failure: Incorrect | Unknown/Multiple warning at compile-time | Test Case: Blocked By: | Blocking: Related Tickets: | Differential Rev(s): Phab:D4083 Wiki Page: | -------------------------------------+------------------------------------- Changes (by bgamari): * status: new => closed * resolution: => fixed * milestone: => 8.4.1 -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/7169#comment:28> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 0

[GHC] #13986: TypeApplications causes parse errors in @-patterns with certain Unicode characters
by GHC 25 Oct '17

25 Oct '17

#13986: TypeApplications causes parse errors in @-patterns with certain Unicode characters -------------------------------------+------------------------------------- Reporter: Tikhon | Owner: (none) Type: bug | Status: new Priority: normal | Milestone: Component: Compiler | Version: 8.0.1 (Parser) | Keywords: | Operating System: Unknown/Multiple Architecture: | Type of failure: GHC rejects Unknown/Multiple | valid program Test Case: | Blocked By: Blocking: | Related Tickets: Differential Rev(s): | Wiki Page: -------------------------------------+------------------------------------- The following function definition works without TypeApplications but fails to parse with the extension enabled: {{{#!hs {-# LANGUAGE TypeApplications #-} module Mininal where foo x₁@True = 10 }}} It also parses correctly with "x" in place of "x₁". -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/13986> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 5

[GHC] #14214: Users guide lies about default optimization level
by GHC 25 Oct '17

25 Oct '17

#14214: Users guide lies about default optimization level -------------------------------------+------------------------------------- Reporter: bgamari | Owner: (none) Type: bug | Status: new Priority: high | Milestone: 8.4.1 Component: Compiler | Version: 8.2.1 Keywords: | Operating System: Unknown/Multiple Architecture: | Type of failure: Documentation Unknown/Multiple | bug Test Case: | Blocked By: Blocking: | Related Tickets: Differential Rev(s): | Wiki Page: -------------------------------------+------------------------------------- #14208 points out that the users guide claims that the default optimization level is `-O0`. However this apparently isn't true. We should decide whether this should be the default and, if so, identify and fix the differences. -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/14214> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 8

Re: [GHC] #7169: Warning for incomplete record field label used as function
by GHC 25 Oct '17

25 Oct '17

#7169: Warning for incomplete record field label used as function -------------------------------------+------------------------------------- Reporter: goldfire | Owner: nakaji_dayo Type: feature request | Status: new Priority: normal | Milestone: Component: Compiler | Version: 7.7 Resolution: | Keywords: Warnings, | newcomer Operating System: Unknown/Multiple | Architecture: Type of failure: Incorrect | Unknown/Multiple warning at compile-time | Test Case: Blocked By: | Blocking: Related Tickets: | Differential Rev(s): Phab:D4083 Wiki Page: | -------------------------------------+------------------------------------- Comment (by Ben Gamari <ben@…>): In [changeset:"f7f270eb6ba616feda79d370336db7e66f9ab79c/ghc" f7f270e/ghc]: {{{ #!CommitTicketReference repository="ghc" revision="f7f270eb6ba616feda79d370336db7e66f9ab79c" Implement `-Wpartial-fields` warning (#7169) Warning on declaring a partial record selector. However, disable warn with field names that start with underscore. Test Plan: Added 1 test case. Reviewers: austin, bgamari, simonpj Reviewed By: bgamari, simonpj Subscribers: goldfire, simonpj, duog, rwbarton, thomie GHC Trac Issues: #7169 Differential Revision: https://phabricator.haskell.org/D4083 }}} -- Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/7169#comment:27> GHC <http://www.haskell.org/ghc/> The Glasgow Haskell Compiler

1 0