Skip to content

Latest commit

 

History

History
5025 lines (3517 loc) · 217 KB

CHANGELOG.md

File metadata and controls

5025 lines (3517 loc) · 217 KB

Changelog

Notable changes to this project are documented in this file. The format is based on Keep a Changelog and this project adheres to Semantic Versioning.

0.15.15

New features:

  • Add --exclude-file to more commands (#4530 by @JordanMartinez)

    This CLI arg was added to the compile command, but not to other commands where such a usage would be relevant (e.g. docs, repl, graph, and ide).

  • Enable passing source input globs via --source-globs-file path/to/file (#4530 by @JordanMartinez)

    --source-globs-file support has been added to the following commands: compile, docs, graph, ide, and publish.

    Due to a shell character limitation on Windows where a large list of source globs cannot be passed (e.g. purs compile ... glob1000/src/**/*.purs), source globs can be stored in a file according to the format below and the file is passed in instead via purs compile ---source-globs-file path/to/file.

    # Lines starting with '#' are comments.
    # Blank lines are ignored.
    # Otherwise, every line is a glob.
    
    .spago/foo-1.2.3/src/**/*.purs
    .spago/bar-2.3.3/src/**/*.purs
    my-package/src/**/*.purs
    my-package/tests/**/*.purs
    

    --source-globs-file is an optional argument. Mixing it with the normal source globs is fine. Assuming .spago/source-globs contains src/**/*.purs, each command below will use the same input globs:

    purs compile src/**/*.purs
    purs compile --source-globs .spago/source-globs
    purs compile --source-globs .spago/source-globs src/**/*.purs 

    In the command...

    purs compile inputGlob1 inputGlob2 --source-globs-file fileWithMoreGlobs --exclude-files excludeGlob1
    

    the files passed to the compiler are: all the files found by inputGlob1, inputGlob2, and all the globs listed in fileWithMoreGlobs minus the files found by excludeGlob1.

0.15.14

Bugfixes:

  • Fix a compilation memory regression for very large files (#4521 by @mjrussell)

    When compiling a a very large file (>12K lines) the CSE pass could balloon memory and result in increased compilation times.

    This fix uses a strict Map instead of a lazy Map to avoid building up unnecessary thunks during the optimization pass.

  • Fix two space leaks while compiling many modules (#4517 by @MonoidMusician)

    The first would interleave compilation of too many modules at once, which would increase memory usage, especially for single threaded builds with +RTS -N1 -RTS. Now the number of concurrent modules is limited to the number of threads available to the GHC runtime system.

    The second would hold on to memory from modules that compiled with warnings until the end of the build when the warnings were printed and the memory freed. This is now fixed with additional NFData instances.

0.15.13

New features:

  • Replace UnusableDeclaration with updated NoInstanceFound (#4513 by @JordanMartinez)

    Previously, the following type class would be invalid because there was no way for the compiler to infer which type class instance to select because the type variable in the class head a was not mentioned in bar's type signature:

    class Foo a where
      bar :: Int

    The recently-added visible type applications (VTAs) can now be used to guide the compiler in such cases:

    class Foo a where bar :: Int
    instance Foo String where bar = 0
    someInt = bar @String -- use the `String` instance

    Without VTAs, the compiler will still produce an InstanceNotFound error, but this error has been updated to note which type variables in the class head can only be disambiguated via visible type applications. Given the following code

    class Single tyVarDoesNotAppearInBody where 
      useSingle :: Int
    
    single :: Int
    single = useSingle

    The error reported for useSingle will be:

    No type class instance was found for
    
      Main.Single t0
    
    The instance head contains unknown type variables.
    
    
    Note: The following type class members found in the expression require visible type applications 
    to be unambiguous (e.g. tyClassMember @Int).
      Main.useSingle
        tyNotAppearInBody
    

    For a multiparameter typeclass with functional dependencies...

    class MultiFdBidi a b | a -> b, b -> a where
      useMultiFdBidi :: Int
    
    multiFdBidi :: Int
    multiFdBidi = useMultiFdBidi

    ...the "Note" part is updated to read

    Note: The following type class members found in the expression require visible type applications 
    to be unambiguous (e.g. tyClassMember @Int).
      Main.useMultiFdBidi
        One of the following sets of type variables:
          a
          b
    

Bugfixes:

  • Fix parsing bug where @var was allowed in type class head (#4523 by @JordanMartinez)

0.15.12

New features:

  • Move the closed record update optimization (#4489 by @rhendric)

    For consumers of CoreFn like alternate backends, the optimization of replacing a closed record update with an object literal has now been moved to the point of desugaring CoreFn into JS. The ObjectUpdate expression constructor now contains a Maybe field holding a list of record labels to be copied as-is, for backends that want to perform this optimization also.

  • Allow instances that require Fail to be empty (#4490 by @rhendric)

    A class instance declaration that has Prim.TypeError.Fail as a constraint will never be used. In light of this, such instances are now allowed to have empty bodies even if the class has members.

    (Such instances are still allowed to declare all of their members, and it is still an error to specify some but not all members.)

Bugfixes:

  • Stop emitting warnings for wildcards in Visible Type Applications (#4492 by @JordanMartinez)

    Previously, the below usage of a wildcard (i.e. _) would incorrectly cause the compiler to emit a warning.

    f :: forall @a. a -> a
    f = identity
    
    x :: { x :: Int }
    x = f @{ x :: _ } { x: 42 }
  • Infer types using VTA inside a record (#4501 by @JordanMartinez)

    Previously, use would fail to compile because the v type variable would not be inferred to String. Now the below code compiles:

    reflect :: forall @t v . Reflectable t v => v
    reflect = reflectType (Proxy @t)
    
    use :: String
    use = show { asdf: reflect @"asdf" }

Internal:

  • Use gh for release artifacts (#4493 by @rhendric, #4509 by @JordanMartinez)

  • Stop triggering CI on non-code-related changes (e.g. Readme) (#4502 by @JordanMartinez)

0.15.11

Please use 0.15.12 instead of this release. There was an issue with the Linux build. This release notes were moved into 0.15.12's release notes.

0.15.10

New features:

  • Implement visible type applications

    The compiler now supports visible type applications, allowing the user to instantiate one or more "visible" type variables to a specific type.

    A "visible" type variable is a type variable in a forall binder that appears prefixed with an @, like the following example:

    id :: forall @a. a -> a  -- or with kinds: `forall (@a :: Type). a -> a`
    id a = a

    We can then use type application syntax to instantiate this binding to a specific type:

    idInt :: Int -> Int
    idInt = id @Int
    
    example :: Int
    example = id @Int 0

    Type variables appearing in class or data are automatically visible, meaning that they do not require annotations:

    data Maybe a = Just a | Nothing
    
    nothingInt :: Maybe Int
    nothingInt = Nothing @Int
    
    class Identity a where
      identity :: a -> a
    
    instance Identity Int where
      identity a = a
    
    identityInt = identity @Int
    
    -- This throws a `NoInstanceFound` error.
    identityNumber = identity @Number

    Lastly, visible type variables can also be skipped with a wildcard (i.e. _)

    data Either a b = Left a | Right b
    
    example = Left @_ @Number 0

    Note that performing a type application with a type that has no visible type variables throws an error:

    module Main where
    
    id :: forall a. a -> a
    id a = a
    
    idInt = id @Int
    
    {-
    Error found:
    in module Main
    at Main.purs:6:9 - 6:16 (line 6, column 9 - line 6, column 16)
    
      An expression of polymorphic type
      with the invisible type variable a:
                      
        forall a. a -> a
                      
      cannot be applied to:
         
        Int
         
    
    while inferring the type of id
    in value declaration idInt
    
    See https://github.com/purescript/documentation/blob/master/errors/CannotApplyExpressionOfTypeOnType.md for more information,
    or to contribute content related to this error.
    -}

    Similarly, monomorphic types also cannot be used for type applications:

    module Main where
    
    idInt :: Int -> Int
    idInt a = a
    
    example = idInt @Int
    
    {-
    Error found:
    in module Main
    at Main.purs:6:11 - 6:21 (line 6, column 11 - line 6, column 21)
    
      An expression of monomorphic type:
                
        Int -> Int
                
      cannot be applied to:
         
        Int
         
    
    while inferring the type of idInt
    in value declaration example
    
    See https://github.com/purescript/documentation/blob/master/errors/CannotApplyExpressionOfTypeOnType.md for more information,
    or to contribute content related to this error.
    -}
  • Exclude files from compiler input (#4480 by @i-am-the-slime)

    The compiler now supports excluding files from the globs given to it as input. This means there's now a new option for purs compile, namely --exclude-files (or the short version -x):

    > purs compile --help
    Usage: purs compile [FILE] [-x|--exclude-files ARG] [-o|--output ARG] ...
    
      Compile PureScript source files
    
    Available options:
      -h,--help                Show this help text
      FILE                     The input .purs file(s).
      -x,--exclude-files ARG   Glob of .purs files to exclude from the supplied
                              files.
      ...

    This allows you to keep related files closer together (that is, colocate them).

    Consider a setup like the following:

    src/
        Main.purs
        View/
            LoginPage.purs
            LoginPageTest.purs
            LoginPageStories.purs

    In order to exclude the files in the example above you can now invoke purs like this and it will only compile LoginPage.purs:

    purs compile "src/**/*.purs" --exclude-files "src/**/*Stories.purs" -x "src/**/*Test.purs"

    With spago, the equivalent command is:

    spago build --purs-args '-x "src/**/*Test.purs" -x "src/**/*Stories.purs"'

0.15.9

New features:

  • Add release artifacts for Linux and macOS running on the ARM64 architecture. (#4455 by @f-f)

Bugfixes:

  • Fix prerelease version number on macOS (#4461 by @rhendric)

  • Consider fixity declarations during linting (#4462 by @ozkutuk)

  • Defer monomorphization for data constructors (#4376 by @purefunctor)

    In 0.15.4 and earlier, the compiler monomorphizes type constructors early, yielding the following type:

    > :t Nothing
    forall (a1 :: Type). Maybe a1
    
    > :t { a : Nothing }
    forall (a1 :: Type).
      { a :: Maybe a1
      }

    With this change, the monomorphization introduced in #835 is deferred to only when it's needed, such as when constructors are used as values inside of records.

    > :t Nothing
    forall a. Maybe a
    
    > :t { a : Nothing }
    forall (a1 :: Type).
      { a :: Maybe a1
      }

    Also as a consequence, record updates should not throw ConstrainedTypeUnified in cases such as:

    v1 :: { a :: Maybe Unit }
    v1 = { a : Just Unit }
    
    v2 :: { a :: Maybe Unit }
    v2 = let v3 = v1 { a = mempty } in v3
  • Update installer to version 0.3.5 to support ARM builds (#4468 and #4469 by @rhendric)

  • Fix exhaustiveness checking to account for case guards (#4467 by @purefunctor)

Internal:

  • Refactor module imports to make identifiers' origins obvious (#4451 by @JordanMartinez)

  • Require comments not to cause Haddock warnings (#4456 by @rhendric)

0.15.8

New features:

  • Generated documentation now supports dark mode (#4438 by @sometimes-i-send-pull-requests)

    PureScript documentation has a new dark theme available. It will automatically be used based on your browser or system's color scheme preferences.

Bugfixes:

  • Fix instance deriving regression (#4432 by @rhendric)

  • Outputs what label the type-error occurred on when types don't match (#4411 by @FredTheDino)

  • Account for typed holes when checking value declarations (#4437 by @purefunctor)

    The compiler now takes into account typed holes when ordering value declarations for type checking, allowing more top-level values to be suggested instead of being limited by reverse lexicographical ordering.

    Given:

    module Main where
    
    newtype K = K Int
    
    aRinku :: Int -> K
    aRinku = K
    
    bMaho :: K
    bMaho = ?help 0
    
    cMuni :: Int -> K
    cMuni = K
    
    dRei :: Int -> K
    dRei _ = bMaho

    Before:

      Hole 'help' has the inferred type
              
        Int -> K
              
      You could substitute the hole with one of these values:
                             
        Main.cMuni  :: Int -> K
        Main.K      :: Int -> K
    

    After:

      Hole 'help' has the inferred type
              
        Int -> K
              
      You could substitute the hole with one of these values:
                              
        Main.aRinku  :: Int -> K
        Main.cMuni   :: Int -> K
        Main.K       :: Int -> K
    

Other improvements:

  • Bump Stackage snapshot to lts-20.9 and GHC to 9.2.5 (#4422, #4428, and #4433 by @purefunctor, @JordanMartinez, and @andys8)

Internal:

  • Update license/changelog scrips to latest Stack resolver (#4445 by @JordanMartinez)

0.15.7

New features:

  • Allow IDE module rebuilds eschewing the filesystem (#4399 by @i-am-the-slime)

    This allows IDE clients to typecheck the module the user is currently typing in without modifying the output. This allows for faster feedback cycles in editors and avoids producing a broken /output before the user actually saves the file.

  • Add purs ide dependency/imports filter (#4412 by @nwolverson)

    This allows IDE tooling to filter type searches according to the imports of a given module, restricting to identifiers in scope.

  • Shorten custom user-defined error message's prefix (#4418 by @i-am-the-slime)

    Improves clarity and gets to the relevant information faster.

  • The compiler can now derive instances for more types and type classes (#4420 by @rhendric)

    New type classes that the compiler can derive:

    • Bifunctor
    • Bifoldable
    • Bitraversable
    • Contravariant
    • Profunctor

    Moreover, the compiler can also use these classes when deriving Functor, Foldable, and Traversable, enabling more instances to be derived whereas before such instances would need to be written manually.

Bugfixes:

  • Update installer to 0.3.3 to fix a few installation issues (#4425 by @JordanMartinez)

Other improvements:

  • Improve DuplicateDeclarationsInLet error so that it mentions what variable names were duplicated, reporting several in separate errors as necessary. (#4405 by @MonoidMusician)

  • Fix various typos in documentation and source comments. (#4415 by @Deltaspace0)

  • Bump Stackage snapshot to 2022-11-12 and GHC to 9.2.4 (#4422 by @purefunctor)

Internal:

  • Organize the compiler's internal constants files (#4406 by @rhendric)

  • Enable more GHC warnings (#4429 by @rhendric)

0.15.6

Bugfixes:

  • Make FromJSON instance for Qualified backwards compatible (#4403 by @ptrfrncsmrph)

    Prior to #4293, Qualified was encoded to JSON such that

    >>> encode $ Qualified Nothing "foo"
    [null,"foo"]
    >>> encode $ Qualified (Just $ ModuleName "A") "bar"
    ["A","bar"]

    The type of Qualified has changed so that null no longer appears in JSON output, but for sake of backwards-compatibility with JSON that was produced prior to those changes (pre-v0.15.2), we need to accept null, which will be interpreted as Qualified ByNullSourcePos.

  • Fix extraneous qualifiers added to references to floated expressions (#4401 by @rhendric)

0.15.5

New features:

  • Increases the max number of typed holes displayed from 5 up to 30 (#4341 by @JordanMartinez)

  • Add a compiler optimization for ST functions with up to 10 arity, similar to Effect optimizations. (#4386 by @mikesol)

  • Enable the compiler to derive Foldable and Traversable instances (#4392 by @rhendric)

    These instances follow the same rules as derived Functor instances. For details, see the PureScript language reference.

Bugfixes:

  • Qualify references to expressions floated to the top level of a module by the compiler (#4364 by @rhendric)

  • Fix replicated type hole suggestions due to malformed source spans (#4374 by @PureFunctor)

    In PureScript 0.15.4, the following code will produce multiple entries in the type hole suggestions. This is due to malformed source spans that are generated when desugaring value declarations into case expressions.

    module Main where
    
    data F = X | Y
    
    f :: forall a. F -> a -> a
    f X b = ?help
    f Y b = ?help
  • Improve error spans for class and instance declarations (#4383 and #4391 by @PureFunctor and @rhendric)

    This improves the error spans for class and instance declarations. Instead of highlighting the entire class or instance declaration when UnknownName is thrown, the compiler now highlights the class name and its arguments.

    Before:

    [1/2 UnknownName]
    
      5  class G a <= F a
         ^^^^^^^^^^^^^^^^
    
      Unknown type class G
    
    [2/2 UnknownName]
    
      7  instance G a => F a
         ^^^^^^^^^^^^^^^^^^^
    
      Unknown type class G

    After:

    [1/2 UnknownName]
    
      5  class G a <= F a
               ^^^
    
      Unknown type class G
    
    [2/2 UnknownName]
    
      7  instance G a => F a
                  ^^^
    
      Unknown type class G
  • Fix a bug where the compiler did not consider interactions of all functional dependencies in classes. (#4195 by @MonoidMusician) In particular, combinations of multiple parameters determining other parameter(s) were not handled properly, affecting overlapping instance checks and the selection of which parameters are fully determined.

Other improvements:

  • Bump actions environment to macOS-11 (#4372 by @PureFunctor)

Internal:

  • Enable OverloadedRecordDot extension throughout codebase (#4355 by @JordanMartinez)

  • Ensure order of args remain unchanged in freeTypeVariables (#4369 by @JordanMartinez)

  • Bump HLint to version 3.5 and address most of the new hints (#4391 by @rhendric)

  • Remove purescript-cst from Makefile (#4389 by @ptrfrncsmrph)

  • Bump depend NPM purescript-installer to ^0.3.1 (#4353 by @imcotton)

  • Remove base-compat as a dependency (#4384 by @PureFunctor)

0.15.4

Bugfixes:

  • Fix name clash in guard clauses introduced in #4293 (#4385 by @PureFunctor)

    As a consequence, a problem with the compiler not being able to see imported names if they're shadowed by a guard binder is also solved.

    import Data.Foldable (fold)
    import Data.Maybe (Maybe(..))
    import Data.Monoid.Additive (Additive(..))
    
    test :: Maybe Int -> Int
    test = case _ of
      m | Just fold <- m -> fold
        -- Previously would complain about `fold` being undefined
        | otherwise -> case fold [] of Additive x -> x

Internal:

  • Add Guard handler for the everywhereWithContextOnValuesM traversal. (#4385 by @PureFunctor)

0.15.3

New features:

  • Float compiler-synthesized function applications (#3915 by @rhendric)

    This is a limited implementation of common subexpression elimination for expressions created by the compiler in the process of creating and using typeclass dictionaries. Users can expect code that heavily uses typeclasses to produce JavaScript that is shorter, simpler, and faster.

    Common subexpression elimination is not applied to any expressions explicitly written by users. If you want those floated to a higher scope, you have to do so manually.

  • Add support for optional shebang lines (#4214 by @colinwahl and @JordanMartinez)

    One or more shebang line are only allowed as the first lines of a file

    #! a shebang line
    #! another shebang line
    -- | module doc comment
    -- other comment
    module MyModule where
    
    #! Using a shebang here will fail to parse
    foo :: String
    foo = ""

Bugfixes:

  • Stop requiring bower.json devDependencies when publishing (#4332 by @JordanMartinez)

  • Stop emitting source spans with negative line/column numbers (#4343 by @j-nava and @JordanMartinez)

Internal:

  • Accommodate internally-generated identifiers that start with digits (#4334 by @rhendric)

  • Enable -Wincomplete-uni-patterns and -Wincomplete-record-updates by default (#4336 by @hdgarrood)

    Update purescript.cabal so that the PureScript compiler is built with the flags -Wincomplete-uni-patterns and -Wincomplete-record-updates enabled by default.

  • Setup infrastructure for testing source maps (#4335 by @JordanMartinez)

  • Removed a couple of unused SimpleErrorMessage constructors (#4344 by @hdgarrood)

  • Compare json files through aeson in tests (#4354 by @PureFunctor)

    This fixes the tests for the graph and source map outputs, as the ordering is inconsistent between stack test and cabal test.

  • Add version bounds to the test suite's build-depends. (#4354 by @PureFunctor)

  • Update GHC to 9.2.3 (#4351 by @hdgarrood and @JordanMartinez)

  • Add qualification for locally-bound names (#4293 by @PureFunctor)

    This change makes it so that Qualified names can now be qualified by either a ModuleName for module-level declarations or the starting SourcePos for bindings introduced locally. This makes disambiguation between references to local bindings much easier in AST-driven analysis.

0.15.2

New features:

  • Check for partially applied synonyms in kinds, ctors (#4169 by @rhendric)

    This check doesn't prevent any programs from compiling; it just makes sure that a more specific PartiallyAppliedSynonym error is raised instead of a KindsDoNotUnify error, which could be interpreted as implying that a partially applied synonym has a valid kind and would be supported elsewhere if that kind is expected.

  • Support deriving instances for type synonyms (#4315 by @rhendric)

Bugfixes:

  • Do not emit warnings about type wildcards used in binders (patterns). (#4309 by @fsoikin)

    Type wildcards in the following examples no longer trigger a warning:

    f :: Int
    f = 42 # \(x :: _) -> x
    
    g :: Maybe Int
    g = do
      x :: _ <- getX
      pure $ x + 5
    
  • Fix issue with unnamed instances using type operators (#4311 by @rhendric)

  • Fix incorrect Prim.Int (class Compare) docs: Int & Ordering, not Symbol (#4313 by @JordanMartinez)

  • Fix bad interaction between module renaming and inliner (#4322 by @rhendric)

    This bug was triggered when modules that the compiler handles specially are shadowed by local constructors. For example, a constructor named Prim could have caused references to Prim_1["undefined"] to be produced in the compiled code, leading to a reference error at run time. Less severely, a constructor named Control_Bind would have caused the compiler not to inline known monadic functions, leading to slower and less readable compiled code.

  • Update Prim docs for Boolean, Int, String/Symbol, Number, Record, and Row (#4317 by @JordanMartinez)

  • Fix crash caused by polykinded instances (#4325 by @rhendric)

    A polykinded instance is a class instance where one or more of the type parameters has an indeterminate kind. For example, the kind of a in

    instance SomeClass (Proxy a) where ...

    is indeterminate unless it's somehow used in a constraint or functional dependency of the instance in a way that determines it.

    The above instance would not have caused the crash; instead, instances needed to be of the form

    instance SomeClass (f a) where ...

    in order to cause it.

  • Fix bad interaction between newtype deriving and type synonyms (#4315 by @rhendric)

    See #3453.

  • Fix bad interaction between instance deriving and type synonyms (#4315 by @rhendric)

    See #4105.

  • Fix spurious kind unification error triggered by newtype deriving, type synonyms, and polykinds (#4315 by @rhendric)

    See #4200.

Internal:

  • Deploy builds continuously to GitHub and npm (#4306 and #4324 by @rhendric)

    (Builds triggered by changes that shouldn't affect the published package are not deployed.)

  • Fix incomplete type traversals (#4155 by @rhendric)

    This corrects oversights in some compiler internals that are not known to be the cause of any user-facing issues.

  • Drop dependency on microlens libraries (#4327 by @rhendric)

0.15.1

Release skipped; use 0.15.2.

0.15.0

Breaking changes:

  • Switch from Common JS to ES modules (#4232 by @sigma-andex)

    Previously, Purescript used Common JS for FFI declarations.

    Before, FFI was declared like this...

    const mymodule = require('mymodule')
    
    exports.myvar = mymodule.myvar

    ...and will be changed to this...

    import * as M from 'mymodule';
    export const myvar = M.myvar

    ...or using the short version...

    export { myvar } from 'mymodule';
  • FFI is annotated with /* #__PURE__ */ so that bundlers can perform DCE

  • The current LTS Node.js version 12 is now the required minimum version

  • Improve apartness checking (#4149 by @rhendric)

    See details in https://github.com/purescript/documentation/blob/master/language/Type-Classes.md#instance-chains

  • Disable type class constraints in FFI (#4240 by @JordanMartinez)

    Previously, one could write FFI like the following:

    foreign import foo :: forall a. Show a => a -> String

    Type class dictionaries are "magically" handled by the compiler. By including them in the above FFI, one can depend on their representation. Since the representation can change without notice, this may silently break code.

    In v0.14.x, a warning was emitted if these were used. Now it will fail to compile. Rather, one should write something like the following where the members of the type class are passed explicitly to the FFI function as arguments:

    foo :: forall a. Show a => a -> String
    foo val = fooImpl show val
    
    foreign import fooImpl :: forall a. (a -> String) -> a -> String
  • Removes deprecated syntax for rows (i.e. #) and kinds (i.e. kind-keyword) (#4239 by @JordanMartinez)

  • Apply precedence rules to operator sections (#4033 by @rhendric)

    Previously, (_ * 4 + 1) would desugar to \x -> x * (4 + 1), even though * has higher precedence than +. Conversely, (3 * 2 + _) would not compile, even though * has higher precedence than +. These bugs have now been fixed; (_ * 4 + 1) is an error, and (3 * 2 + _) desugars to \x -> 3 * 2 + x.

    If you have code that relied on the old behavior, add an extra pair of parentheses around the expression in the section.

  • If FFI parsing succeeds & CommonJS is detected, fail; otherwise, do not error or warn (#4250 by @sigma-andex)

    Previously, the compiler would emit an error if it failed to parse the FFI JavaScript file. Since the underlying JavaScript parser (i.e. language-javascript) fails to parse even valid JavaScript files, we cannot consider every failed parse to mean invalid JS files. Fixing the parser would require a lot of effort, so we are planning to remove it instead in v0.16.x.

    If the parse succeeds and a CommonJS module is detected, a compiler error is now emitted. If the parse fails, we no longer emit a compiler error. While we could emit a warning, such a warning will quickly become annoying for FFI files that trigger the buggy paths of language-javascript. Moreover, we presume that all will be migrating their code to ES modules now that CommonJS is being deprecated in the larger JavaScript ecosystem.

  • Warn on ad-hoc non-single-line case expression syntax (#4241 by @JordanMartinez)

    The following code will now produce a compiler warning. These were originally supported to ease the migration to the new CST parser.

    -- before: `arg` isn't indented "past" the `Foo arg` binder
    case foo of Foo arg ->
      arg
    -- after
    case foo of Foo arg ->
                  foo

    Dropping the above syntax make case expressions more similar to how let bindings work:

    let ok = 1
    let
      ok = 1
    let ok =
          1
    let notOk =
      1
  • Drop support for browser backend for repl (i.e. purs repl --port 1234) (#4255 by @JordanMartinez)

    Running this command will print a link that directs users to use Try PureScript instead.

  • Remove purs bundle (#4255 by @JordanMartinez)

    Users of purs bundle should switch to a standalone bundler such as esbuild, webpack or parcel.

  • Lazy initialization for recursive bindings (#4283 by @rhendric)

    This is unlikely to break a working program, but the upshot for users is that it's now possible to get a run-time error when dereferencing an identifier in a recursive binding group before it has been initialized, instead of silently getting an undefined value and having that maybe or maybe not lead to an error somewhere else.

    This change can cause code that relies on tail-call optimization to no longer compile with that optimization. If you find that code that previously compiled to a TCO loop no longer does but does include $lazy initializers, please report the issue.

    Alternate backend maintainers: for you, this change represents a clarification of a responsibility shared by all backends. The identifiers bound in a recursive binding group need to behave as if those identifiers have call-by-need semantics during the initialization of the entire binding group. (Initializing the binding group entails ensuring every initializer has been executed, so after the binding group is initialized, these identifiers can be considered call-by-value again.)

    If an identifier is needed during its own call-by-need initialization, the backend must ensure that an explicit run-time error is raised appropriate for your target platform. This error may be raised at compile time instead if the backend can determine that such a cycle is inevitable. Returning your target language's equivalent of JavaScript's undefined, as purs did in earlier releases in some cases, is not permitted.

    If your target language natively has call-by-need semantics, you probably don't have to do anything. If your target language is call-by-value and you are using PureScript as a library, you can use the function Language.PureScript.CoreFn.Laziness.applyLazinessTransform to your CoreFn input to satisfy this responsibility; if you do, you will need to do the following:

    • Translate InternalIdent RuntimeLazyFactory and InternalIdent (Lazy _) identifiers to appropriate strings for your backend

    • Ensure that any output file that needs it has a reference to a function named InternalIdent RuntimeLazyFactory, with type forall a. Fn3 String String (Unit -> a) (Int -> a), and with the same semantics as the following JavaScript (though you should customize the error raised to be appropriate for your target language):

      function (name, moduleName, init) {
          var state = 0;
          var val;
          return function (lineNumber) {
              if (state === 2) return val;
              if (state === 1) throw new ReferenceError(name + " was needed before it finished initializing (module " + moduleName + ", line " + lineNumber + ")", moduleName, lineNumber);
              state = 1;
              val = init();
              state = 2;
              return val;
          };
      };

    If neither of the previous cases apply to you, you can meet this responsibility most easily simply by ensuring that all recursive bindings are lazy. You may instead choose to implement some light analysis to skip generating lazy bindings in some cases, such as if every initializer in the binding group is an Abs. You also may choose to reimplement applyLazinessTransform, or even develop a more sophisticated laziness transform for your backend. It is of course your responsibility to ensure that the result of whatever analysis you do is equivalent to the expected semantics.

New features:

  • Implement the Reflectable type class (#4207 by @PureFunctor)

    The Reflectable type class is a common interface for reflecting type-level values down to the term-level. Its instances are automatically solved by the compiler, and it allows Symbol, Int, Boolean, and Ordering kinded types to be reflected to their term-level representations.

  • Implement native type-level integers (#4207 and #4267 by @PureFunctor and @JordanMartinez)

    Added support for type-level integers and compiler-solved operations such as Add, Mul, Compare, and ToString. Type-level integers use the Int type as their kind.

  • Print compilation progress on the command line (#4258 by @PureFunctor)

    This feature makes it so purs compile and purs docs now show compilation progress on the command line. Example output:

    [ 1 of 59] Compiling Type.Proxy
    [ 2 of 59] Compiling Type.Data.RowList
    ...
    [58 of 59] Compiling Effect.Class.Console
    [59 of 59] Compiling Test.Main
  • Restore names of quantified variables during generalization (#4257 by @PureFunctor)

    This makes the compiler aware of the names of quantified variables instantiated into unification variables, such that when the latter is generalized, semantic information is restored. For example:

    addNumberSuffix :: forall a b c d. a -> b -> c -> d -> a
    addNumberSuffix a _ _ _ = a
    
    addNumberSuffix' = addNumberSuffix 0

    Previously, inferring top-level declarations without type signatures would use t suffixed with an integer for type variables.

    forall t6 t7 t8. t6 -> t7 -> t8 -> Int

    Now, the inferred type would refer back to their original names.

    forall b6 c7 d8. b6 -> c7 -> d8 -> Int
  • Support Glibc versions >= 2.24 (#4228 by @sd-yip)

    Previously, purs required a Glibc version greater than or equal to 2.27. This requirement is relaxed to support a Glibc version down to 2.24.

Bugfixes:

  • Fix warning suppression for wildcard types (#4269 by @rhendric)

    This bug was triggered by defining recursive partial functions or recursive bindings that contained wildcards in inner type annotations. Recursive partial function declarations now no longer cause spurious wildcard warnings to be emitted, and actual user-written wildcards now accurately emit warnings if and only if they don't appear within a binding (recursive or otherwise) with a complete (wildcard-free) type signature.

  • Remove compiler-generated identifiers from type search results (#4260 by @PureFunctor)

Other improvements:

  • Improve "Unknown value bind" and "Unknown value discard" errors (#4272 by @mhmdanas)

    The previous error implies that do-notation compiles down to only bind or to only discard (depending on whether the symbol not found was bind or discard respectively), which is somewhat misleading, especially in the latter case. Now, the error states correctly that do-notation compiles down to both functions.

Internal:

  • Document the HSPEC_ACCEPT flag for generating golden files (#4243 by @JordanMartinez)

  • Fail test if PureScript file(s) don't have a Main module (#4243 by @JordanMartinez)

  • Update CI to use windows-2019 since windows-2016 is deprecated (#4248 by @JordanMartinez)

  • Move lib/purescript-cst into src/ (#4290 by @JordanMartinez)

  • Update tests and their bower deps to 0.15.0-compatible versions (#4300 by @JordanMartinez)

0.14.7

New features:

  • Make Prim.TypeError's Quote work on all kinds, not just kind Type. (#4142 by @xgrommx)

  • Display role annotations in HTML docs (#4121 by @JordanMartinez)

    Previously, the HTML docs would not indicate which types could be safely coerced and which could not:

    -- cannot be coerced
    data Foo1 a = Foo1 a
    type role Foo1 nominal
    
    -- can be coerced
    data Foo2 a = Foo2
    type role Foo2 phantom
    
    -- can be coerced in some contexts
    data Foo3 a = Foo3 a
    type role Foo3 representational

    The HTML docs now display the role annotations either explicitly declared by the developer or those inferred by the compiler.

    Since role annotations are an advanced feature and since most type parameters' roles are the representational role, the phantom and nominal role annotations are displayed in documentation whereas the representational role is not, similar to "uninteresting" kind signatures.

    Lastly, FFI declarations like below...

    foreign import data Foo :: (Type -> Type) -> Type
    type role Foo nominal

    ...will be rendered as though they are data declarations:

    data Foo :: (Type -> Type) -> Type
    data Foo t0
    type role Foo nominal

    One can distinguish FFI declarations with roles separately from normal data declarations that have roles based on the name of the type parameters. Since FFI declarations' type parameters are implicit and thus unnamed, the compiler will generate their name: t0, t1, ..., tN where N is a zero-based index of the type parameter.

    Note: the resulting documentation will display the roles, but the roles will not be selectable when selecting the type in case one wants to copy-paste the type into source code.

  • Rewrite Partial optimization to be cleaner (#4208 by @rhendric)

    This feature shrinks the generated JS code for declarations that use empty type classes, such as Partial, but is otherwise not expected to have user-visible consequences.

  • Add support for publishing via the purs.json manifest format (#4233 by @thomashoneyman)

    This feature expands compiler support for publishing packages with different manifest formats. Previously, packages had to have a bower.json manifest; now, packages can choose to have a purs.json manifest instead.

    This feature provides only partial support for packages published to the PureScript registry using the purs.json manifest format. Registry packages are allowed to be hosted anywhere (not just GitHub), and do not need to be Git repositories at all. However, purs publish and its primary consumer, Pursuit, both require packages to be available on GitHub and for their version to be a SemVer-compliant Git tag. Therefore, this feature only supports registry packages that are compatible with these restrictions.

Bugfixes:

  • Add missing source spans to data constructors when generating docs (#4202 by @PureFunctor)

  • Check role declarations arity during type checking (#4157 by @kl0tl)

  • Optimize newtype applications with the ($) operator (#4205 by @PureFunctor)

  • Properly deserialize unused identifiers in the CoreFn (#4221 by @sjpgarcia)

    This mostly affects downstream consumers of the CoreFn as discussed in #4201. This makes it so CoreFn deserialization properly reads $__unused into UnusedIdent instead of an Ident. This is particularly useful for downstream consumers of the CoreFn such as alternative backends that don't allow arguments to be omitted from functions.

  • Fix type operators in declaration param kinds (#4220 by @rhendric)

    This fixes an internal error triggered by using a type operator in the kind of a type parameter of a data declaration, type synonym declaration, or class declaration.

  • Scope type vars when type checking typed values (#4216 by @rhendric)

    When the compiler is checking an expression that is annotated with a type against another expected type, and the annotation introduces a type variable, the compiler needs to introduce that type variable to the scope of any types used inside the expression.

    One noteworthy case of this pattern is member signatures inside instances. This fix allows type variables introduced in member signatures to be used in the member declaration itself.

Internal:

  • Bump PureScript to building with GHC-8.10.7, as well as from LTS-17 to LTS-18. (#4199 by @cdepillabout)

  • Prevent hangs on internal errors (#4126 by @rhendric)

  • The explicit disabling of Nix has been removed from stack.yaml. (#4198 by @cdepillabout)

    For developers on NixOS, this means that you should be able to build PureScript by running stack build instead of stack build --nix. For other developers, this shouldn't affect you.

  • Build the entire latest package set in CI (#4217 by @rhendric)

    See #4128.

  • Create test machinery for optimizations (#4205 by @PureFunctor)

    This adds machinery for testing code generation for optimizations.

    Partially extracted from #3915 to add tests for #4205.

0.14.6

Do not use this release. purescript-cst's version wasn't bumped when this release was made. So, tools like trypurescript cannot depend on it. See 0.14.7 for the same thing.

0.14.5

Bugfixes:

  • Fix a case where kind inference inferred the wrong kind for type synonyms (#4184 by @jy14898)

  • Properly rename module imports in case of conflicts with declarations (#4188 by @PureFunctor)

Internal:

  • Fix command and clarify a few other requirements in release guide (#4177 by @JordanMartinez)

  • Add Functor instance for Language.PureScript.CoreFn.Module. (#4186 by @cdepillabout)

v0.14.4

Bugfixes:

  • Solve Prim.Row.Union left right all constraint for left when all and right are already closed rows, (#3720 by @MonoidMusician) reflecting the existing functional dependency all right -> left

  • Account for redundant parens when excluding uninteresting kind sigs from docs (#4137 by @JordanMartinez)

  • Add a hint for errors in foreign data type declarations (#4161 by @kl0tl)

  • Do not remove bindings referenced in function declarations when bundling (#4044 by @kl0tl)

  • Improve row type error messages (#4159 by @rhendric)

    • Remove a redundant hint that repeats the types in the error
    • Correctly diff rows containing duplicate items
    • Erase kind applications from rows in errors (by default)
  • Fix bad interaction between superclasses and type synonyms (#4164 by @rhendric)

    See #4101.

  • Fix regression in row unification (#4168 by @rhendric)

  • Fix backtick operator rule (#4172 by @JordanMartinez)

Other improvements:

  • Add developer guide to readme (#3900 by @milesfrain)

Internal:

  • Move unreleased changelog entries to CHANGELOG.d (#4132 by @rhendric)

    See CHANGELOG.d/README.md for details.

  • Clarify in RELEASE_GUIDE what to do when broken releases are made (#4147 by @JordanMartinez)

  • Miscellaneous updates/clarifications to the release guide (#4131 by @JordanMartinez)

  • Run Weeder in CI and make it happy (#4148 by @rhendric)

  • Add golden tests for self cycles in type class declarations, kind declarations and foreign data type declarations (#4162 by @kl0tl)

  • Represent class dictionaries as newtypes (#4125 by @rhendric)

v0.14.3

New features:

  • Display kind signatures and their comments in documentation (#4100 and #4119 by JordanMartinez)

    The compiler now displays kind signatures for data, newtype, type synonym, and type class declarations in generated documentation. The compiler now also includes documentation-comments (i.e. those which start with a | character) both above and below the associated kind signature declaration (if any) in generated documentation, whereas previously documentation-comments above a kind signature declaration were ignored.

    Both explicitly declared and inferred kinds are included in documentation. The compiler omits including a kind signature in generated documentation only when the kind is considered "uninteresting". An uninteresting kind is defined as one where all of the declaration's type parameters have kind Type.

Bugfixes:

  • Ensure unnamed instances appear in documentation (#4109 by @JordanMartinez)

  • Allow fixity, kind, role declarations in REPL (#4046, @rhendric)

  • Pin OS versions used in CI (#4107, @f-f)

  • Fix UnusedName warnings for multiple non-recursive let bindings (#4114 by @nwolverson)

  • Remove generated names from errors about instances (#4118 by @rhendric)

Internal:

  • Fix for Haddock (#4072 by @ncaq and @JordanMartinez, #4139 by @JordanMartinez)

  • Update RELEASE_GUIDE.md with more details (#4104 by @JordanMartinez)

  • Use GenIdent for anonymous instances (#4096, @rhendric)

  • Desugar type class instance names in type class desugaring pass (#4099 by @JordanMartinez)

v0.14.2

New features:

  • Make type class instance names optional (#4085, @JordanMartinez)

    Previously, one would be required to define a unique name for a type class instance. For example

    -- instance naming convention:
    -- classNameType1Type2Type3
    instance fooIntString :: Foo Int String

    Now, the name and :: separator characters are optional. The above instance could be rewritten like so:

    instance Foo Int String

    Note that generated instance names can change without warning as a result of changes elsewhere in your code, so do not rely upon these names in any FFI code.

Bugfixes:

  • Unused identifier warnings now report smaller and more relevant source spans (#4088, @nwolverson)

    Also fix incorrect warnings in cases involving a let-pattern binding shadowing an existing identifier.

Internal:

  • Drop libtinfo dependency (#3696, @hdgarrood)

    Changes the build configuration so that by default, compiler binaries will not have a dynamic library dependency on libncurses/libtinfo. This should alleviate one of the most common pains in getting the compiler successfully installed, especially on Linux. The cost is a slight degradation in the REPL experience when editing long lines, but this can be avoided by building the compiler with the libtinfo dependency by setting the terminfo flag of the haskeline library to true.

  • Migrate CI from Travis to GitHub Actions (#4077, @rhendric)

  • Remove tasty from test suite and just use hspec (#4056, @hdgarrood)

  • Avoid compiling tests with diagnostics twice in test suite (#4079, @hdgarrood)

  • Do less work in test initialization (#4080, @rhendric)

  • Follow more HLint suggestions (#4090, @rhendric)

  • Export rebuildModule' to speed up Try PureScript! slightly (#4095 by @JordanMartinez)

  • Merge purescript-ast into purescript-cst (#4094 by @JordanMartinez)

v0.14.1

New features:

  • Support TCO for functions with tail-recursive inner functions (#3958, @rhendric)

    Adds support for optimizing functions that contain local functions which call the outer function in tail position, as long as those functions themselves are only called from tail position, either in the outer function or in other such functions.

    This enables hand-written mutually-tail-recursive function groups to be optimized, but more critically, it also means that case guards which desugar to use local functions don't break TCO.

  • Add warnings for unused names and values (#3819, @nwolverson)

    The compiler now emits warnings when it encounters unused names in binders and unused value declarations. A declaration is considered to be unused if it is not exported and is also not reachable by any of the exported declarations. The compiler will not currently produce unused warnings about other kinds of declarations such as data and type class declarations, but we intend to produce warnings for these in the future as well.

Bugfixes:

  • Make close punctuation printable in errors (#3982, @rhendric)

  • Desugar type operators in top-level kind signatures (#4027, @natefaubion)

  • Use type annotation hint only when needed (#4025, @rhendric)

  • Fix pretty printing of "hiding" imports (#4058, @natefaubion)

  • Instantiate polymorphic kinds when unwrapping newtypes while solving Coercible constraints (#4040, @kl0tl)

  • Fix row unification with shared unknown in tails (#4048, @rhendric)

  • Fix kinded declaration reordering in desugaring (#4047, @rhendric)

  • Fix wildly off kind unification positions (#4050, @natefaubion)

  • Fix incorrect incremental builds with different --codegen options (#3911, #3914, @hdgarrood)

    This bug meant that after invoking the compiler with different --codegen options, it was easy to end up with (for example) an outdated docs.json or corefn.json file in your output directory which would be incorrectly considered up-to-date by the compiler.

Other improvements:

  • Add white outline stroke to logo in README (#4003, @ptrfrncsmrph)

    The previous logo.png was not legible against a dark background (#4001).

  • Show the constraints that were being solved when encountering a type error (@nwolverson, #4004)

  • Removed all shift/reduce conflicts in parser (#4063, @JordanMartinez).

    Happy defaults to using "shift" rather than "reduce" in shift/reduce conflicts. This change merely makes explicit what is already happening implicitly.

Internal:

  • Upgrade tests Bower dependencies (#4041, @kl0tl)
  • Remove unused Data.Foldable.foldr import (#4042, @kl0tl)

v0.14.0

Polykinds

Polymorphic kinds, based on the Kind Inference for Datatypes paper (#3779, #3831, #3929, #4007, @natefaubion, @kl0tl)

Just as types classify terms, kinds classify types. But while we have polymorphic types, kinds were previously monomorphic.

This meant that we were not able to abstract over kinds, leading for instance to a proliferation of proxy types:

data Proxy (a :: Type) = Proxy
data SProxy (a :: Symbol) = SProxy
data RProxy (row :: # Type) = RProxy
data RLProxy (row :: RowList) = RLProxy

Now we can have a single proxy type, whose parameter has a polymorphic kind.

Type :: Type

The old Kind data type and namespace is gone. Kinds and types are the same and exist in the same namespace.

Previously one could do:

foreign import kind Boolean
foreign import data True :: Boolean
foreign import data False :: Boolean

Where the kind Boolean and type Boolean were two different things. This is no longer the case. The Prim kind Boolean is now removed, and you can just use Prim type Boolean in the same way. This is a breaking change.

The compiler still supports the old foreign import kind syntax but it warns that it's deprecated.

foreign import kind Foo

Foreign kind imports are deprecated and will be removed in a future release. Use empty 'data' instead.

It is treated internally as:

data Foo

Note that foreign import data declarations are not deprecated. They are still necessary to define types with kinds other than Type since constructors are not lifted as in GHC with DataKinds.

Likewise, kind imports and exports are deprecated and treated the same as a type import or export.

Kind imports are deprecated and will be removed in a future release. Omit the 'kind' keyword instead.

The special unary # syntax for row kinds is still supported, but deprecated and will warn. There is now Prim.Row :: Type -> Type which can be used like a normal type constructor.

Unary '#' syntax for row kinds is deprecated and will be removed in a future release. Use the 'Row' kind instead.

All of these deprecations have suggested fixes in the JSON output, so tools like purescript-suggest (or your IDE plugin) can automatically apply them.

Kind Signatures

With PolyKinds, all type-level declarations are generalized.

data Proxy a = Proxy

Previously, this had the Type-defaulted kind Type -> Type. Now this will be generalized to forall k. k -> Type. Such signature can be written with a kind signature declarations, similar to standalone kind signatures in GHC.

data Proxy :: forall k. k -> Type
data Proxy a = Proxy

In GHC, all signatures use the type prefix, but we reuse the same keyword as the subsequent declaration because we already have foreign import data (rather than foreign import type) and because it makes things align nicer. Signatures have the same rule as value-level signatures, so they must always be followed by the "real" declaration.

It's better to be explicit about polymorphism by writing signatures. Since we don't really quantify over free type variables, it's also necessary in the case that two poly-kinded arguments must have the same kind. The compiler will warn about missing kind signatures when polymorphic kinds are inferred.

Classes can have signatures too, but they must end with the new Constraint kind instead of Type. For example, here's the new definition of Prim.Row.Cons:

class Cons :: forall k. Symbol -> k -> Row k -> Row k -> Constraint
class Cons label a tail row | label a tail -> row, label row -> a tail

Safe zero-cost coercions

Coercible constraints, based on the Safe Zero-cost Coercions for Haskell paper (#3351, #3810, #3896, #3873, #3860, #3905, #3893, #3909, #3931, #3906, #3881, #3878, #3937, #3930, #3955, #3927, #3999, #4000, @lunaris, @rhendric, @kl0tl, @hdgarrood)

Prim.Coerce.Coercible is a new compiler-solved class, used to relate types with the same runtime representation. One can use Safe.Coerce.coerce (from the new safe-coerce library) instead of Unsafe.Coerce.unsafeCoerce to safely turn a a into a b when Coercible a b holds.

Roles

Types parameters now have roles, which depend on how they affect the runtime representation of their type. There are three roles, from most to least restrictive:

  • nominal parameters can only be coerced to themselves.

  • representational parameters can only be coerced to each other when a Coercible constraint holds.

  • phantom parameters can be coerced to anything.

Role annotations

The compiler infers nominal roles for foreign data types, which is safe but can be too constraining sometimes. For example this prevents the coercion of Effect Age to Effect Int, even though they have the same runtime representation.

The roles of foreign data types can thus be loosened with explicit role annotations, similar to the RoleAnnotations GHC extension.

Here's the annotation we added to Effect:

type role Effect representational

Conversely, we might want to strengthen the roles of parameters with invariants invisible to the type system. Maps are the canonical example of this: the shape of their underlying tree rely on the Ord instance of their keys, but the Ord instance of a newtype may behave differently than the one of the wrapped type so it would be unsafe to allow coercions between Map k1 a and Map k2 a, even when Coercible k1 k2 holds.

In order to forbid such unsafe coercion we added a nominal annotation to the first parameter of Map:

type role Map nominal representational

Annotated roles are compared against the roles inferred by the compiler so it is not possible to compromise safety by ascribing too permissive roles, except for foreign types.

Other changes

Breaking

  • Add compiler support for Coercible based Newtype (#3975, @fsoikin)

We added a Coercible superclass to Data.Newtype.Newtype in order to implement unwrap, wrap and most newtype combinators with coerce (see purescript/purescript-newtype#22). This is only a breaking change for non derived instances because the Newtype class has no members anymore and can now only be implemented for representationally equal types (those satisfying the new superclass constraint).

For example the instance for newtype Additive a = Additive a no longer implements unwrap and wrap:

+instance newtypeAdditive :: Newtype (Additive a) a
-instance newtypeAdditive :: Newtype (Additive a) a where
-  wrap = Additive
-  unwrap (Additive a) = a

Derived instances don't require any modifications.

  • Reform handling of quote characters in raw strings (#3961, @rhendric)

Quotes behaved rather unexpectedly in various edge cases inside raw strings. This clears things up by enforcing the following specification:

'"""' '"'{0,2} ([^"]+ '"'{1,2})* [^"]* '"""'

Meaning that raw strings can contain up to two successive quotes, any number of times, but three successive quotes are not allowed inside.

  • Unsupport bare negative literals as equational binders (#3956, @rhendric)

It used to be possible to match on negative literals, such as -1, but this prevented parsing matches on constructors aliased to -. The compiler will reject matches on bare negative literals, but they can still be matched by wrapping them in parentheses.

  • Forbid partial data constructors exports (#3872, @kl0tl)

Exporting only some of the constructors of a type meant that changes internal to a module, such as adding or removing an unexported constructor, could cause unexhaustive pattern match errors in downstream code. Partial explicit export lists will have to be completed with the missing constructors or replaced by implicit export lists.

  • Print compile errors to stdout, progress messages to stderr (#3839, @JordanMartinez)

Compiler errors and warnings arising from your code are now printed to stdout rather than stderr, and progress messages such as "Compiling Data.Array" are now printed to stderr rather than stdout. Warnings and errors arising from incorrect use of the CLI, such as specifying input files which don't exist or specifying globs which don't match any files, are still printed to stderr (as they were before). This change is useful when using the --json-errors flag, since you can now pipe JSON errors into other programs without having to perform awkward gymnastics such as 2>&1.

Fixes

  • Only include direct dependencies in the output for purs graph instead of their transitive closure (#3993, @colinwahl)

  • Fix the reversal of the qualifier of qualified operators (#3971, @rhendric)

Qualified operators, for instance Data.Array.(!), were interpreted with a reversed qualifier, like Array.Data.(!).

  • Check all recursive paths in data binding groups (#3936, @natefaubion)

The compiler was not catching recursive type synonyms when some recursive paths were guarded by data types or newtypes.

  • Desugar type operator aliases inside parens (#3935, @natefaubion)

The compiler did not accept type operators inside parens in prefix position, except (->).

  • Pin language-javascript to a specific version (#3904, @hdgarrood)

Allowing the compiler to be built against various versions of language-javascript meant that multiple builds of the same version of the compiler could accept different syntaxes for JavaScript foreign modules, depending on how they were built.

Improvements

  • Improves protocol errors from the IDE server (#3998, @kritzcreek)

The IDE server now respond with more descriptive error messages when failing to parse a command. This should make it easier to contribute fixes to the various clients.

  • Extend IDE ImportCompletion with declarationType (#3997, @i-am-the-slime)

By exposing the declaration type (value, type, typeclass, etc.) downstream tooling can annotate imports with this info so users know what they are about to import. The info can also be mapped to a namespace filter to allow importing identifiers that appear more than once in a source file which throws an exception without such a filter.

  • Improve error message when negate isn't imported (#3952, @mhmdanas)

This shows a specific message when using negative literals but Data.Ring.negate is out of scope, similar to the messages shown when using do notation if Control.Bind.bind and Control.Bind.discard are out of scope.

  • Add source spans to PartiallyAppliedSynonym errors (#3951, @rhendric)

PartiallyAppliedSynonym errors were usually rethrown with the appropriate source span, but not when deriving instances. This annotates those errors with the source span of the partially applied synonyms themselves, which is more robust and accurate than rethrowing the error with an approximate source span.

  • Allow type synonyms in instances heads and superclass constraints (#3539, #3966, #3965, @garyb, @kl0tl)

This allows declarations such as

type Env = { port :: Int }
newtype App a = App (ReaderT Env Aff a)
derive newtype instance monadAskApp :: MonadAsk Env App

or

class (Monad m, MonadAsk Env m) <= MonadAskEnv m
  • Improve incremental rebuild times for modules with large dependencies (#3899, @milesfrain)

Other

  • Warn against exported types with hidden constructors but Generic or Newtype instances (#3907, @kl0tl)

Types with hidden constructors are supposed to be opaque outside of their module of definition but Generic and Newtype instances allow to construct them with Data.Generic.Rep.to or Data.Newtype.wrap and examine their content with Data.Generic.Rep.from or Data.Newtype.unwrap, thus making void any invariant those types may witness.

  • Have module re-exports appear in generated code (#3883, @citizengabe)

This is the first step towards smarter incremental rebuilds, which could skip rebuilding downstream modules when the interface of a module did not change (see #3724).

  • Add a printer for CST modules (#3887, @kritzcreek)

  • Deprecate constraints in foreign imports (#3829, @kl0tl)

Constrained foreign imports leak instance dictionaries, hindering the compiler ability to optimize their representation. Manipulating dictionaries in foreign code should be avoided and foreign imports should accept the class members they need as additional arguments instead of being constrained.

  • Deprecate primes (the ' character) in identifiers exported from foreign modules (#3792, @kl0tl)

We are going to output ES modules instead of CommonJS in the next breaking release but named exports of ES modules, unlike CommonJS exports, have to be valid JavaScript identifiers and so cannot contain primes.

Docs

  • Generate a changelog from the GitHub releases and add a pull request template (#3989, @JordanMartinez)

  • Detail license related error messages and fix incorrect SPDX sample licenses (#3970, @fsoikin)

  • Remove a spurious doc comment on the CoreFn Module type (#3552, @jmackie)

  • Add a link to the releases page (#3920, @milesfrain)

  • Update CONTRIBUTING.md (#3924, @hdgarrood)

  • Add troubleshooting steps for libtinfo and EACCES errors (#3903, @milesfrain)

  • Update an outdated link to the book (#3916, @sumew)

Internal

  • Simplify the Ord instances of some AST types (#3902, @milesfrain)

  • Update the desugaring pipeline to work on individual modules (#3944, @kl0tl)

  • Remove the unmaintained and ignored core libraries tests (#3861, @kl0tl)

  • Configure Travis to run hlint (#3816, #3864, @joneshf, @hdgarrood)

  • Remove support for the legacy Bower resolutions format in purs publish (#3847, @kl0tl)

  • Add GitHub issue templates for bugs and proposals (#3853, @joneshf)

  • Add support for Happy >=1.19.10 (#3837, @arrowd)

  • Use the same default extensions in all packages (#3823, #3908, @natefaubion, @i-am-the-slime)

  • Relax purescript-ast dependency on microlens-platform to microlens (#3817, @joneshf)

  • Extract the AST and CST types, and related functions, into their own purescript-ast and purescript-cst packages for ease of consumption by external tooling (#3793, #3821, #3826, @joneshf, @natefaubion)

  • Fix various typos in documentation, comments and bindings names (#3795, @mhmdanas)

  • Add golden tests for errors and warnings (#3774, #3811, #3808, #3846, @dariooddenino, @rhendric, @kl0tl)

  • More descriptive protocol errors from the ide server (@kritzcreek)

v0.13.8 - 2020-05-23

Bug Fixes

  • Update incremental build cache information properly on IDE rebuilds (#3789, @kritzcreek)

    Fixes a bug where triggering a rebuild via the IDE would not update the output/cache-db.json file, which in certain situations could lead to unnecessary rebuilds, as well as modules not being rebuilt when they should have been.

  • Don't include compiler-internal declarations in IDE completions (#3850, @kritzcreek)

    IDE completions would previously include pseudo-declarations such as RowToList$Dict which only exist internally, due to how type class desugaring inside the compiler works. These declarations are now suppressed.

  • Fix corefn JSON version parsing (#3877, @paulyoung)

    Fixes a bug where the parser for the functional core (or "corefn") JSON format would ignore all but the first component of the compiler version stored in the JSON. This does not affect the compiler directly, but will be useful for other tooling which depends on the corefn JSON parser provided by the compiler library.

Improvements

  • Add purs graph subcommand for graphing module dependencies (#3781, @jmackie, @f-f)

    This adds a new graph subcommand which allows tools to consume information about which modules depend on which other modules. The format is as follows:

    { "Prelude":
        { "path": "src/Prelude.purs"
        , "depends": ["Data.Semiring", "Data.Ring", ...]
        },
      "Data.Ring":
        { "path": "src/Data/Ring.purs"
        , "depends": []
        },
      ...
    }
    

    Each property in the returned object has exactly two properties; path, which is a string containing the file path relative to the directory where the command was run, and depends, which is an array of the names of all directly imported modules.

  • purs ide is better at reloading changes (#3799, @kritzcreek)

    The IDE would previously sometimes miss changes that were made outside of the editor, like building with new dependencies or recompiling larger parts of the project on the console.

    The IDE will now notice when this happened on the next command issued to it and refresh its state before processing that command. This might cause the first command after an external change to take a long time to execute, but should increase reliability in general.

  • Switch to a binary encoding for externs files (#3841, @kritzcreek)

    This change should result in significant performance improvements in both IDE load times and incremental builds where lots of modules are already built.

  • Represent module names as a single Text value internally (#3843, @kritzcreek)

    Boosts compiler performance by representing module names as a single Text value, rather than a list of Text values as it was previously.

  • Extract documentation for type classes in purs ide (#3856, @kritzcreek)

    This changes makes documentation comments on type classes visible to the IDE.

Other

  • Declare explicit upper bounds on Cabal and haskeline rather than relying on stack's pvp-bounds (#3777, @coot)

v0.13.7 - 2020-05-23

release withdrawn due to CI mishap

v0.13.6 - 2020-01-17

Bug Fixes

  • Reset IDE state before performing a full reload. (#3766, @kritzcreek)

    This prevents a space leak in the IDE.

  • Added source spans to ado desugaring. (#3758, @dariooddenino)

    Previously errors in ado desugaring might have had no line information.

  • Generate correct arity failure case for some guarded matches. (#3763, @nwolverson)

    Specifically when a multi-way case contains a pattern guard or multiple guard expressions, the desugared case expression could contain a guard with a different arity to the matched expressions, resulting in an error.

Improvements

  • Improved ambiguous variable check for functional dependencies. (#3721, @MonoidMusician)

    Previously the compiler might warn about ambiguous variables that aren't actually ambiguous due to functional dependencies. This check now fully takes functional dependencies into consideration.

  • Optimize import desugaring for full builds (#3768, @colinwahl)

    The compiler was performing redundant work when resolving dependencies for modules resulting in poor asymptotics. This work is now shared across modules yielding a 30-40% improvement in build times for full builds.

  • Use PureScript escapes in string pretty-printing (#3751, @hdgarrood)

    Previously the compiler might print invalid escape sequences when pretty-printing code for error messages. It now prints correctly escaped code based on PureScript's lexical grammar.

  • Optimize away binds to wildcards in do-notation (#3220, @matthewleon, @hdgarrood)

    This avoids generating variable assignments if no variables are actually bound in do-notation. Previously the compiler would emit a unique variable name that went unused.

  • Output docs.json files for Prim modules (#3769, @f-f)

    This change allows downstream tools such as spago to obtain documentation data for Prim modules. Please note, however, that the API for the docs.json files is unstable and may change without warning.

Other

  • Fix various typos in source comments (#3760, @bwignall)

v0.13.5 - 2019-11-13

This is a small bugfix release to address some issues which were introduced in 0.13.4.

Bug fixes

  • Fix "too many open files" during compiling (#3743, @hdgarrood)

    The compiler would not promptly close files after opening them, which could easily lead to reaching the open file limit, causing the compiler to crash.

  • Fix incorrect unused import warnings when kinds are re-exported (#3744, @hdgarrood)

    Fixes a bug in which unused import warnings were generated for kinds which were re-exported (and therefore should have been considered "used").

Other

  • Fix Haddock markup error preventing Haddock docs being generated (#3745, @cdepillabout)
  • Add upper bound on Protolude to prevent 0.2.4 from being selected (#3752, @hdgarrood)

v0.13.4 - 2019-10-20

Enhancements

  • Use content hashes when determining whether a file needs rebuilding (#3708, @hdgarrood)

    We now calculate and store content hashes of input files during compilation. If a file's modification time has changed since the last compile, we compare the hash to the previous hash; if the hash is unchanged, this allows us to skip rebuilding this file, speeding up the build.

  • Include import declaration qualifiers in unused import warnings (#3685, @matthew-hilty)

    Previously, warnings didn't distinguish between import declarations from the same module. Code like the following

    import A.B (x) -- `x` is used.
    import A.B (y) as C -- `y` is not used.

    would induce a warning like The import of module A.B is redundant even though only the qualified import declaration C is actually redundant. The warning now would be The import of module A.B (qualified as C) is redundant.

  • Include kind imports when determining unused import warnings (#3685, @matthew-hilty)

    Previously, kind imports were ignored. The linter wouldn't emit any warnings for code like the following.

    import A.B (kind K) -- `kind K` is not used.

    And the linter, disregarding kind K, would emit an UnusedImport instead of an UnusedExplicitImport for code like the following.

    import A.B (x, kind K) -- `x` is not used, but `kind K` is.
  • Better reporting of I/O errors (#3730, @hdgarrood)

    If an unexpected I/O error occurs during compiling, we now include details in the error message. For example, when trying to write compilation results onto a device which has run out of space, we previously would have received a "CannotWriteFile" error with no further information. Now, we receive the underlying error message too:

    I/O error while trying to write JSON file: ./output/cache-db.json
    
      ./output/cache-db.json: hClose: resource exhausted (No space left on device)
    

Bug fixes

  • Improve type class resolution in the presence of constrained higher-order functions (#3558, @matthew-hilty)

    This is perhaps best illustrated with an example.

    newtype LBox row a = LBox (∀ r. (∀ lbl _1. Row.Cons lbl a _1 row  IsSymbol lbl  SProxy lbl  r)  r)
    
    unLBox   row a r. ( lbl _1. Row.Cons lbl a _1 row  IsSymbol lbl  SProxy lbl  r)  LBox row a  r
    unLBox g (LBox f) = f g
    
    read   row a. Record row  LBox row a  a
    read rec = unLBox \lbl → Record.get lbl rec

    The read function would previously fail with the error

    No type class instance was found for
    
        Prim.Row.Cons lbl4
                      a5
                      t2
                      row6
    

    although that dictionary should have been available in the function passed to unLBox. Now, it type checks successfully.

  • Fix cache invalidation false negatives by storing timestamps (#3705, @hdgarrood)

    Previously, an input file would be considered 'modified', and thus requiring rebuilding on a subsequent compile, if its modification time specifies a point in time after any of the modification times of the corresponding output files. This has turned out to be insufficient; files can often change in a way that this algorithm misses, because the input file might still have a timestamp older than the output files. Often this can happen by switching between git branches or by updating a dependency.

    This problem can manifest as compiler errors which don't appear to make sense or correspond to what is inside a source file, and which (until now) would need to be fixed by a clean rebuild (e.g. rm -r output).

    We now make a note of the modification time when we read an input file, and we consider that input file to have changed on a subsequent compile if the modification time is different to what it was before.

    The hope with this fix is that it should never be necessary to remove an output directory to get a build to run successfully. If you do run into this problem again, it is a bug: please report it.

  • Fix exports incorrectly being identified as unused in purs bundle (#3727, @rhendric)

    References to properties on the exports object would previously not be picked up by purs bundle as uses of those properties, which could lead to them being incorrectly removed. For example:

    'use strict';
    
    exports.foo = 1;
    exports.bar = exports.foo;

    would remove the exports.foo = 1; statement, breaking the assignment to exports.bar, if foo were not used elsewhere. This statement is now no longer removed.

  • Show entire rows in type errors in the presence of the --verbose-errors flag (#3722, @Woody88)

    The row diffing feature, which elides common labels in rows occurring in type errors, did not previously respect the --verbose-errors flag, giving the same output regardless of whether it was set or not. Now, if the flag has been supplied, we always show the entire row.

Other

  • Add Makefile command to run license generator (#3718, @hdgarrood)

  • Update language-javascript to 0.7.0.0 (@rhendric, @hdgarrood)

    This enables a number of newer JavaScript syntactic constructs to be used in FFI files. Please see the language-javascript release notes for details.

  • Fix for object shorthand syntax in FFI files (#3742, @hdgarrood)

v0.13.3 - 2019-08-18

Enhancements

  • Eliminate empty type class dictionaries in generated code (#2768, @LiamGoodacre)

    Empty type class dictionaries — dictionaries which do not contain any type class member implementations at runtime — are often used to provide evidence at compile-time to justify that a particular operation will not fail; for example, Prim.Row.Cons can be used to justify that we can expect a record to contain a particular field with a particular type. Unfortunately, constructing empty dictionaries can be costly, especially in more complex scenarios such as type-level programming. This release implements a new optimization which avoids the need to build empty dictionaries at runtime by instead inserting undefined into the generated code. This optimization can both reduce code size and improve performance in certain contexts.

  • Render doc-comments for data constructors and type class members in HTML documentation (#3507, @marcosh)

    Documentation comments for data constructors and type class members are now picked up by purs docs, and will soon start appearing in Pursuit too. For example:

    -- | Doc-comments like this one were always rendered in Pursuit
    data Maybe a =
      -- | Now this one (for the Just constructor) will be rendered too
      = Just a
      -- | And this one (for Nothing)
      | Nothing
    
    -- | Doc-comments like this one were always rendered in Pursuit
    class Eq a where
      -- | Now this one (for the `eq` method) will be rendered too
      eq :: a -> a -> Boolean
  • Show diffs of rows in errors and hints (#3392, @dariooddenino)

    In type mismatches between rows, we now elide common labels so that the problem is easier to identify. For example, consider the following code, which has a type error due to the types of the b fields in the two records not matching:

    foo =
      { a: 1, b: "hi", c: 3, d: 4, e: 5 }
    bar =
      { a: 1, b: 2, c: 3, d: 4, e: 5 }
    baz =
      [ foo, bar ]

    Previously, the type error would include the entirety of each record type:

    Could not match type
    
      String
    
    with type
    
      Int
    
    while trying to match type ( a :: Int
                               , b :: String
                               , c :: Int
                               , d :: Int
                               , e :: Int
                               )
    with type ( a :: Int
              , b :: Int
              , c :: Int
              , d :: Int
              , e :: Int
              )
    

    This can become quite difficult to read in the case of large record types. Now, we get this:

    Could not match type
    
      String
    
    with type
    
      Int
    
    while trying to match type
                               ( b :: String
                               ...
                               )
    
    with type
                ( b :: Int
                ...
                )
    

Bug fixes

  • Remove more dead code in purs bundle (#3551, @rhendric)

    The dead code elimination in purs bundle now no longer incorrectly considers declarations to be used in the presence of local variables which happen to share their names, and is therefore able to remove these declarations when they are unused.

  • Fix parsing of comma-separated guards in let statements (#3713, @natefaubion)

    The 0.13 parser would previously choke on guards separated by commas in let statements within do/ado blocks, such as

    test = ado
      let
        foo
          | bar
          , baz =
            42
          | otherwise = 100
      in
        foo

    This has now been fixed.

Other

  • Add placeholder purs.bin to fix npm installs (#3695, @hdgarrood)
  • Refactor and simplify BuildPlan a little (#3699, @hdgarrood)
  • Update link to partial type class guide in error message hints (#3717, @alextes)

v0.13.2 - 2019-07-05

Enhancements

  • Add --debug flag to purs bundle command (#3666, @rhendric)

    This flag causes an optimized-for-humans JSON representation of the modules being bundled to be dumped to stderr, prior to dead code elimination.

  • Ignore duplicate file inputs to CLI commands (#3653, @dyerw)

    If, after expanding globs, a particular file path appears more than once, the compiler will now ignore the extra occurrences, instead of emitting a DuplicateModule error.

Bug fixes

  • Fix printing of tokens with string escapes (#3665, @hdgarrood)
  • Fix multiple "let"s in ado before the final "in" (#3675, @natefaubion)
  • Throw a parse error (not internal error) when using quoted labels as puns (#3690, @natefaubion)

Other

  • Parser: Remove partial type signatures for parameterized productions (#3667, @natefaubion)
  • Make git consider *.out files as binary for the golden tests (#3656, @kritzcreek)
  • Fix build failures on older GHCs by tightening base lower bound (#3659, @hdgarrood)
  • Pin happy version to address build failures when building with Cabal (#3660, @hdgarrood)
  • Add upper bounds when producing source distributions (#3661, @hdgarrood)
  • Update test dependency on typelevel-prelude (#3649, @hdgarrood)
  • Update author and maintainer sections of cabal file (#3663, @hdgarrood)
  • Update to GHC 8.6.5, Stackage LTS 13.26 (#3688, @hdgarrood)
  • Various CI maintenance (#3687, @hdgarrood)
  • Move the "purescript" npm package into the compiler repo (#3691, @hdgarrood)

v0.13.1 - 2019-07-04

Notice: This release has been unpublished due to an error in the package tarball.

v0.13.0 - 2019-05-30

Grammar/Parser Changes

0.13 is a very exciting release for me (@natefaubion). For the past few months I've been working on a complete rewrite of the existing parser. The old parser has served us very well, but it has grown very organically over the years which means it's developed some unsightly limbs! Throughout the process I've tried to iron out a lot of dark corner cases in the language grammar, and I hope this release will set us on a firm foundation so we can start to specify what "PureScript the Language" actually is. This release is definitely breaking, but I think you'll find the changes are modest. I also hope that this release will open up a lot of opportunities for syntactic tooling, both using the existing parser or even using alternative parsers (which are now possible).

Breaking

There are a number of breaking changes, but I think you'll find that most code will continue to parse fine. We've tested the parser against the existing ecosystem and several large production applications at Awake, Lumi, and SlamData. The migration burden was either non-existent or only involved a few changes.

  • The only whitespace now allowed in code is ASCII space and line endings. Since you must use indentation to format PureScript code (unlike Haskell), we felt it was best to be more restrictive in what you can write instead of allowing potentially confusing behavior (implicit tab-width, zero-width spaces, etc). You can still use unicode whitespace within string literals.
  • The only escapes accepted in string literals are \n\r\t\'\"\\, \x[0-9a-fA-F]{1,6} (unicode hex escape), and \[\r\n ]+\ (gap escapes). We had inherited a vast zoo of escape codes from the Parsec Haskell Language parser. We decided to minimize what we support, and only add things back if there is significant demand.
  • Octal and binary literals have been removed (hex remains).
  • \ is no longer a valid operator. It conflicts with lambda syntax.
  • @ is no longer a valid operator. It conflicts with named binder syntax.
  • forall is no longer a valid identifier for expressions. We wanted a consistent rule for type identifiers and expression identifiers.
  • Precedence of constructors with arguments in binders (a@Foo b must be a@(Foo b)).
  • Precedence of kind annotations (a :: Type -> Type b :: Type must now be (a :: Type -> Type) (b :: Type)).
  • Precedence of type annotations (:: has lowest precedence, rather than sitting between operators and function application).
  • Various edge cases with indentation/layout. Again, most code should work fine, but there were some cases where the old parser let you write code that violated the offside rule.

Fixes

  • Many fixes around parse error locations. The new parser should yield much more precise error locations, especially for large expressions (like in HTML DSLs).
  • Reported source spans no longer include whitespace and comments.
  • Reported source span for the last token in a file is now correct.

Enhancements

  • where is still only sugar for let (it does not introduce bindings over guards), but it is now usable in case branches in the same manner as declarations.
  • _ is now allowed in numeric literals, and is an ignored character (ie. 1_000_000 == 1000000).
  • Raw string literals (triple quotes) can now contain trailing quotes (ie. """hello "world"""" == "hello \"world\"").
  • Kind annotations are now allowed in forall contexts (#3576 @colinwahl).
  • The new parser is much faster and can avoid parsing module bodies when initially sorting modules. We also do more work in parallel during the initialization phase of purs compile. This means that time to start compiling is faster, and incremental builds are faster. In my testing, a noop call to purs compile on the Awake codebase went from ~10s to ~3s.

Other Changes

Breaking

  • Fix sharing in function composition inlining (#3439 @natefaubion). This is really a bugfix, but it has the potential to break code. Previously, you could write recursive point-free compositions that the compiler inadvertently eta-expanded into working code by eliminating sharing. We've changed the optimization to respect strict evaluation semantics, which can cause existing code to stack overflow. This generally arises in instance definitions. Unfortunately, we don't have a way to disallow the problematic code at this time.
  • Fail compilation when a module imports itself (#3586 @hdgarrood).
  • Disallow re-exporting class and type with the same name (#3648 @joneshf).

Enhancements

  • Better illegal whitespace errors (#3627 @hdgarrood).
  • Only display class members that are not exported from the module when throwing a TransitiveExportError for a class (#3612 @colinwahl).
  • Tweaks to type pretty printing (#3610 @garyb).
  • Unify matching constraints (#3620 @garyb).
  • Improve error message on ModuleNotFound error for Prim modules (#3637 @ealmansi).

Docs

  • Make markdown format behave like html. Remove --docgen opt. Separate directories for html and markdown docs (#3641 @ealmansi).
  • Make html the default output format (#3643 @ealmansi).
  • Write ctags and etags to filesystem instead of stdout (#3644 @ealmansi).
  • Add --output option for purs docs (#3647 @hdgarrood).
  • Use externs files when producing docs (#3645 @hdgarrood). docs is now a codegen target for purs compile where documentation is persisted as a docs.json file in the output directory.

Internal

  • Remove failable patterns and NoMonadFailDesugaring extension (#3610 @adnelson).
  • Add tests for grammar fixes addressed by CST (#3629 #3631 @hdgarrood).
  • Keep Parser.y ASCII to avoid locale issues with happy (#3640 @jmackie).
  • Improve display of internal errors (#3634 @hdgarrood).

v0.12.5 - 2019-04-15

This small release fixes three issues which were introduced in 0.12.4.

Filter out module declarations when suggesting imports (#3591)

When determining candidates for imports, ignore modules. This allows you to easily import types which come from modules of the same name, like Effect. (@kRITZCREEK)

Running purs ide server crashes on macOS (#3594)

Running purs ide server on macOS would immediately crash with the error purs: Network.Socket.listen: unsupported operation (Operation not supported on socket); this has now been fixed. (@f-f)

Take qualification into consideration when determining type class cycles (#3595)

When checking for cycles in type classes, the compiler is now able to distinguish classes which have come from different modules, meaning that e.g. class SomeOtherModule.Foo <= Foo is no longer incorrectly reported as a class having itself as a superclass. (@hdgarrood)

v0.12.4 - 2019-04-07

Enhancements

[purs ide] Treat module declarations like any other (#3541)

This means we can now complete module names with the completion API as well as being able to query for module level documentation and goto-defintion for module names.

The list loadedModules command has also been deprecated, since you can now use the completion command with a filter for modules instead. (@kRITZCREEK)

Truncate types in errors (#3401)

Large types in error messages are now truncated. For example:

module Main where

data Id a = Id a

foo :: Id (Id (Id (Id (Id Int))))
foo = "hi"

now produces

  Could not match type

    String

  with type

    Id (Id (Id (... ...)))

The previous behaviour of printing the types in full may be recovered by passing the --verbose-errors flag to the compiler. (@hdgarrood)

Don't generate unused imports in JavaScript output (#2177)

In both CommonJS compiler output and JavaScript purs bundle output, we no longer emit JS imports for modules whose use sites have all been optimized out. This reduces the number of warnings produced by other JavaScript bundlers or compressors such as "Side effects in initialization of unused variable Control_Category". (@rhendric)

Simplify purs publish resolutions format (#3565)

The format for resolutions files passed via the CLI to purs publish has been simplified. A new-style resolutions file should look something like this:

{
  "purescript-prelude": {
     "version": "4.0.0",
     "path": "bower_components/purescript-prelude"
  },
  "purescript-lists": {
     "version": "6.0.0",
     "path": "bower_components/purescript-lists"
  },
  ...
}

The version field is used for generating links between packages on Pursuit, and the path field is used to obtain the source files while generating documentation: all files matching the glob "src/**/*.purs" relative to the path directory will be picked up.

The version field is optional, but omitting it will mean that no links will be generated for any declarations from that package on Pursuit. The "path" field is required.

The old format is still accepted, but it has been deprecated, and purs publish will now produce a warning when consuming it.

This change allows us to work around a bug in Bower which prevented packages with larger dependency trees (such as Halogen) from being uploaded to Pursuit (purescript-contrib/pulp#351). (@hdgarrood)

Improve error messages for cycles in type class declarations (#3223)

A cycle in type class declarations, such as

class C a <= D a
class D a <= C a

now produces a more informative error, which no longer confusingly refers to type synonyms, and which displays all of the classes involved in the cycle. (@Saulukass)

Bug fixes

  • Naming a constructor PS no longer causes JS runtime errors when using purs bundle (#3505, @mhcurylo)
  • purs publish now warns instead of failing if not all dependencies have a resolved version, e.g. if some have been installed via a branch or commit reference instead of a version range (#3061, @hdgarrood)
  • Fix handling of directive prologues like "use strict" in purs bundle (#3581, @rhendric)

Other

  • Raise upper bound on aeson in package.yaml (#3537, @jacereda)
  • Add Nix test dependencies to stack.yaml (#3525, @jmackie)
  • [purs ide] Represent filters as a data type rather than functions (#3547, @kRITZCREEK)
  • Carry data constructor field names in the AST (#3566, @garyb)
  • Convert prim docs tests to use tasty (#3568, @hdgarrood)
  • Bump bower version used in tests (#3570, @garyb)
  • Add tests for purs bundle (#3533, @mhcurylo)
  • Update to GHC 8.6.4 (#3560, @kRITZCREEK)
  • Rerun some of the compiler tests to test with purs bundle (#3579, @rhendric)

v0.12.3 - 2019-02-24

Enhancements

  • Add better positions for UnknownName errors for types/kinds (#3515, @colinwahl)

    Previously an UnknownName error (arising from e.g. referring to a non-existent type, or a type which you forgot to import) would have a span covering the whole type annotation. Now, the error span only covers the relevant part of the type.

  • Boost performance of purs docs by simplifying re-export handling (#3534, @hdgarrood)

Bug fixes

  • Fix applicative do notation breaking API documentation generation with purs docs (#3414, @hdgarrood)
  • Fix the REPL browser backend (#3387, @dariooddenino)

Other

  • Make the license generator a proper stack script (@kRITZCREEK)
  • Include the module from which something was imported for re-exports in externs files (@hdgarrood)
  • Add AppVeyor build status to README.md (@hdgarrood)

v0.12.2 - 2019-01-13

New features

  • Named type wildcards (#3500, @natefaubion)

    It's now possible to use ?hole style syntax in type signatures where you want the compiler to tell you the missing type. This was previously possible by using _ in a type signature, but now _ can be used without raising a warning, as long as it does not appear in a top level declaration.

Enhancements

  • Improve error message for missing node.js in the repl (#3456, @justinwoo)
  • Add Boolean kind to Prim.Boolean (#3389, @justinwoo)
  • Link to documentation repo as docs for non-Prim built-in types/kinds (#3460, @JordanMartinez)
  • PSCi: Support multiple command types in paste-mode (#3471, @LiamGoodacre)
  • Add row:column printing for source positions in error messages (#3473, @justinwoo)
  • Add :print directive for customizable repl printing (#3478, @hdgarrood)
  • Implement qualified do (#3373, @pkamenarsky)
  • Add better source positions to kind errors (#3495, @natefaubion)

Fixes

  • Remove references to previous kinds * and ! (#3458, @LiamGoodacre)
  • Fix linting of unused type variables (#3464, @LiamGoodacre)
  • Avoid dropping super class dicts for the same class (#3461, @LiamGoodacre)
  • Fix issue where Partial can foil TCO optimizations (#3218, @matthewleon)
  • Fix quoting of record labels in error messages (#3480, @hdgarrood)
  • Prevent invalid JS being generated from awkward record labels (#3486, @hdgarrood)
  • Fix unnecessary quoting of reserved names when used as labels (#3487, @hdgarrood)
  • Fix source spans for binding groups (#3462, @LiamGoodacre)
  • Fix kind error for recursive data type (#3511, @natefaubion)

Other (internals)

  • Add annotations to Type and Kind (#3484, @natefaubion)
  • Use handwritten JSON instances for Type/Kind (#3496, @natefaubion)
  • Remove pretty print constructors from Type (#3498, @natefaubion)
  • Add JSON compatibility tests (#3497, @hdgarrood)
  • Remove the concept of the 'current module' in Docs (#3506, @hdgarrood)

v0.12.1 - 2018-11-12

Enhancements

  • Print types of missing typeclass members (#3398, @fehrenbach)
  • Added Prim.TypeError.QuoteLabel for pretty printing labels in custom type errors (#3436, @dariooddenino)
  • purs ide accepts codegen targets for the rebuild command (#3449, @kRITZCREEK)

Fixes

  • Fixes errors spans for CannotFindDerivingType (#3425, @kRITZCREEK)
  • Fixes a traversal bug where ObjectNestedUpdate was surviving desugaring (#3388, @natefaubion)
  • Fixes type operators reexports (#3410, @natefaubion)
  • Fixes ST magic-do and inlining (#3444, @natefaubion)
  • Fixes missing span information when using do-syntax without importing bind or discard (#3418, @natefaubion)
  • Fixes missing span information when shadowing an open import with a module definition (#3417, @natefaubion)
  • Fixes stale :browse environment after :reload (#3001, @rndnoise)

Other

  • Fix test-support dependency versions and update psci browse test (#3374, @LiamGoodacre)
  • Changes to build with GHC 8.4.3 (#3372, @kRITZCREEK)
  • Set --haddock flag based on BUILD_TYPE (#3409, @justinwoo)
  • Use microlens-platform instead of lens (#3400, @joneshf)
  • Avoid Data.ByteString.Lazy.toStrict (#3433, @coot)
  • Add ffiCodegen to MakeActions (#3434, @coot)
  • Add nix config to stack.yaml (#3435, @f-f)

v0.12.0 - 2018-05-21

Breaking changes

  • Added applicative-do notation; ado is now a keyword. An full explanation of the behaviour and usage of ado is available in a comment on the issue. (#2889, @rightfold)
  • Removed wrapper scripts for the old binary names (psc, psci, etc.) (#2993, @hdgarrood)
  • Removed compiler support for deriving purescript-generics. purescript-generics-rep is still supported. (#3007, @paf31)
  • Instances with just one method now require the method to be indented (bug fix, but potentially breaking) (#2947, @quesebifurcan)
  • Overlapping instances are now an error rather than a warning, but can be resolved with the new instance chain groups feature (#2315, @LiamGoodacre)
  • Reworked the CoreFn json representation. This change enables use of the Zephyr tree shaking tool for PureScript. (#3049, #3342, @coot)
  • It is no longer possible to export a type class that has superclasses that are not also exported (bug fix, but potentially breaking) (#3132, @parsonsmatt)
  • Eq and Ord deriving will now rely on Eq1 and Ord1 constraints as necessary where sometimes previously Eq (f _) would be required. Eq1 and Ord1 instances can also be derived. (#3207, @garyb)
  • Some Prim type classes have been renamed/moved, so will require explicit importing (#3176, @parsonsmatt):
    • RowCons is now Prim.Row.Cons
    • Union is now Prim.Row.Union
    • Fail is now Prim.TypeError.Fail
    • Warn is now Prim.TypeError.Warn
  • Users can no longer specify modules under the Prim namespace (#3291, @parsonsmatt)
  • TypeConcat and TypeString have been replaced because they were in kind Symbol but weren't literals. The Prim.TypeError.Doc kind and related constructors (Text, Quote, Beside, Above) have been added in their place. The Fail and Warn type classes now accept a Doc instead of a Symbol. (#3134, @LiamGoodacre)
  • In simple cases instance overlaps are now checked at declaration time rather than being deferred until an attempt is made to use them. (#3129, @LiamGoodacre)
  • Chaining non-associative or mixed associativity operators of the same precedence is no longer allowed (#3315, @garyb)
  • The --dump-corefn and --source-maps arguments to purs compile have been removed. There is now a --codegen argument that allows the specific codegen targets to be specified - for example, --codegen corefn will not produce JS files, --codgen js,corefn will produce both. If the sourcemaps target is used js will be implied, so there's no difference between --codegen js,sourcemaps and --codegen sourcemaps). If no targets are specified the default is js. (#3196, @garyb, @gabejohnson)
  • Exported types that use foreign kinds now require the foreign kinds to be exported too (bug fix, but potentially breaking) (#3331, @garyb)
  • The pursuit commands were removed from purs ide due to lack of use and editor tooling implementing the features instead (#3355, @kRITZCREEK)

Enhancements

  • Added Cons compiler-solved type class for Symbol (#3054, @kcsongor)
  • The Append compiler-solved type class for Symbol can now be run in reverse (#3025, @paf31)
  • Find Usages for values and constructors in purs ide (#3206, @kRITZCREEK)
  • purs ide treats hiding imports the same as open imports when sorting (#3069, @kRITZCREEK)
  • Added inlining for fully saturated usages of runEffFn/mkEffFn (#3026, @nwolverson)
  • Improved explanation of UnusableDeclaration error (#3088, #3304, @i-am-tom)
  • Improved rendering of comments in generated JavaScript by removing additional newlines (#3096, @brandonhamilton)
  • Instance chain support. (#2315, @LiamGoodacre)

    We can now express an explicit ordering on instances that would previously have been overlapping. For example we could now write an IsEqual type class to compute if two types are equal or apart:

    class IsEqual (l :: Type) (r :: Type) (o :: Boolean) | l r -> o
    instance isEqualRefl :: IsEqual x x True
    else instance isEqualContra :: IsEqual l r False
    

    Note the else keyword that links the two instances together. The isEqualContra will only be up for selection once the compiler knows it couldn't possible select isEqualRefl - i.e that l and r are definitely not equal.

  • Improved orphan instance error to include locations where the instance would be valid (#3106, @i-am-tom)
  • Added an explicit error for better explanation of duplicate type class or instance declarations (#3093, @LiamGoodacre)
  • purs ide now provide documentation comments (#2349, @nwolverson)
  • Clarified meaning of duplicate labels in a Record row (#3143, @paf31)
  • Explicit import suggestions consistently use (..) for constructors now (#3142, @nwolverson)
  • Improved tab completion in purs repl (#3227, @rndnoise)
  • Large compiler perfomance improvement in some cases by skipping source spans in Eq, Ord for binders (#3265, @bitemyapp)
  • Added support for error/warning messages to carry multiple source spans (#3255, @garyb)
  • Improved tab completion in purs repl when parens and brackets are involved (#3236, @rndnoise)
  • Improved completion in purs repl after :kind and :type (#3237, @rndnoise)
  • Added the "magic do" optimisation for the new simplified Effect type (Control.Monad.Eff is still supported) (#3289, @kRITZCREEK, #3301, @garyb)
  • Improvide build startup times when resuming a build with incremental results (#3270, @kRITZCREEK)
  • Added compiler-solved Prim.Row.Nub type class (#3293, @natefaubion)
  • Improved docs for Prim.Row.Cons and Prim.Row.Union (#3292, @vladciobanu)
  • Functor can now be derived when quantifiers are used in constructors (#3232, @i-am-tom)
  • purs repl will now complete types after :: (#3239, @rndnoise)
  • Added compiler-solved Prim.Row.Lacks type class (#3305, @natefaubion)
  • Added current output path to missing output error message from purs ide (#3311, @rgrinberg)
  • Improved parser error messages for .purs-repl (#3248, @rndnoise)
  • require in generated JavaScript now includes full index.js file paths (#2621, @chexxor)
  • Added more compiler-solved type classes and supporting types and kinds to Prim:
    • Prim.Ordering module with kind Ordering, type LT, type EQ, type GT
    • Prim.RowList module with class RowToList, kind RowList, type Nil, type Cons
    • Prim.Symbol module with class Compare, class Append, class Cons (#3312, @LiamGoodacre, @kRITZCREEK)
  • Generated code for closed records now explicitly reconstructs the record rather than looping (#1493, @fehrenbach, blog post with more details)
  • Enhanced purs --help message to include hint about using --help with commands (#3344, @hdgarrood)
  • IncorrectConstructorArity error message now includes a hint of how many arguments are expected for the constructor (#3353, @joneshf)
  • purs ide now uses absolute locations for file paths for better experience in some editors (#3363, @kRITZCREEK)

Bug fixes

  • Fixed a bug with names cause by Prim always being imported unqualified (#2197, @LightAndLight)
  • Fixed overlapping instances error message to reflect its new status as an error (#3084, @drets)
  • Added source position to TypeClassDeclaration errors (#3109, @b123400)
  • Fixed entailment issues with skolems and matches in the typechecker (#3121, @LiamGoodacre)
  • Fixed multiple parentheses around a type causing a crash (#3085, @MonoidMusician)
  • Fixed purs ide inserting conflicting imports for types (#3131, @nwolverson)
  • Fixed constraints being inferred differently for lambda expressions compared with equational declarations (#3125, @LiamGoodacre)
  • Updated glob handling to prevent excessive memory usage (#3055, @hdgarrood)
  • Added position information to warnings in type declarations (#3174, @b123400)
  • Fixed documentation generated for Pursuit rendering functional dependency variables as identifier links (#3180, @houli)
  • Naming a function argument __unused no longer breaks codegen (#3187, @matthewleon)
  • Added position information to ShadowedName warning (#3213, @garyb)
  • Added position information to UnusedTypeVar warning (#3214, @garyb)
  • Added position information to MissingClassMember, ExtraneousClassMember, ExpectedWildcard errors (#3216, @garyb)
  • Added position information to ExportConflict errors (#3217, @garyb)
  • Fixed ctags and etags generation when explicit exports are involved (#3204, @matthewleon)
  • Added position information to ScopeShadowing warning (#3219, @garyb)
  • Added position information for various FFI related errors and warnings (#3276, @garyb)
  • Added all available positions to CycleInModule and DuplicateModule errors (#3273, @garyb)
  • Added position information for IntOutOfRange errors (#3277, @garyb, @kRITZCREEK)
  • Warnings are now raised when a module re-exports a qualified module with implicit import (#2726, @garyb)
  • purs repl now shows results for :browse Prim (#2672, @rndnoise)
  • Added position information to ErrorParsingFFIModule (#3307, @nwolverson)
  • Added position information for ScopeConflict cause by exports (#3318, @garyb)
  • Added position information to errors that occur in binding groups and data binding groups (#3275, @garyb)
  • Fixed a scoping issue when resolving operators (#2803, @kRITZCREEK, @LightAndLight)
  • Type synonyms are now desugared earlier when newtype deriving (#3325, @LiamGoodacre)
  • Fixed subgoals of compiler-solved type classes being ignored (#3333, @LiamGoodacre)
  • Added position information to type operator associativity errors (#3337, @garyb)
  • Updated description of purs docs command (#3343, @hdgarrood)
  • Fixed purs docs issue with re-exporting from Prim submodules (#3347, @hdgarrood)
  • Enabled purs ide imports for Prim submodules (#3352, @kRITZCREEK)
  • Fixed purs bundle failing to bundle in the 0.12-rc1 (#3359, @garyb)
  • Enabled :browse for Prim submodules in purs repl (#3364, @kRITZCREEK)

Other

  • Updated installation information to include details about prebuild binaries (#3167, @MiracleBlue)
  • Test suite now prints output when failing cases are encountered (#3181, @parsonsmatt)
  • Updated test suite to use tasty (#2848, @kRITZCREEK)
  • Improved performance of repl test suite (#3234, @rndnoise)
  • Refactored let pattern desugaring to be less brittle (#3268, @kRITZCREEK)
  • Added makefile with common tasks for contributors (#3266, @bitemyapp)
  • Added ghcid and testing commands to makefile (#3290, @parsonsmatt)
  • Removed old unused MultipleFFIModules error (#3308, @nwolverson)
  • mod and div for Int are no longer inlined as their definition has changed in a way that makes their implementation more complicated - purescript/purescript-prelude#161 (#3309, @garyb)
  • The test suite now checks warnings and errors have position information (#3211, @garyb)
  • The AST was updated to be able to differentiate between let and where clauses (#3317, @joneshf)
  • Support for an optimization pass on CoreFn was added (#3319, @matthewleon)
  • Clarified note in the purs ide docs about the behaviour of --editor-mode (#3350, @chexxor)
  • Updated bundle/install docs for 0.12 (#3357, @hdgarrood)
  • Removed old readme for psc-bundle (a leftover from before the unified purs binary) (#3356, @Cmdv)

v0.12.0-rc1 - 2018-04-29

Breaking changes

  • Added applicative-do notation; ado is now a keyword. An full explanation of the behaviour and usage of ado is available in a comment on the issue. (#2889, @rightfold)
  • Removed wrapper scripts for the old binary names (psc, psci, etc.) (#2993, @hdgarrood)
  • Removed compiler support for deriving purescript-generics. purescript-generics-rep is still supported. (#3007, @paf31)
  • Instances with just one method now require the method to be indented (bug fix, but potentially breaking) (#2947, @quesebifurcan)
  • Overlapping instances are now an error rather than a warning, but can be resolved with the new instance chain groups feature (#2315, @LiamGoodacre)
  • Reworked the CoreFn json representation (#3049, @coot)
  • It is no longer possible to export a type class that has superclasses that are not also exported (bug fix, but potentially breaking) (#3132, @parsonsmatt)
  • Eq and Ord deriving will now rely on Eq1 and Ord1 constraints as necessary where sometimes previously Eq (f _) would be required. Eq1 and Ord1 instances can also be derived. (#3207, @garyb)
  • Some Prim type classes have been renamed/moved, so will require explicit importing (#3176, @parsonsmatt):
    • RowCons is now Prim.Row.Cons
    • Union is now Prim.Row.Union
    • Fail is now Prim.TypeError.Fail
    • Warn is now Prim.TypeError.Warn
  • Users can no longer specify modules under the Prim namespace (#3291, @parsonsmatt)
  • TypeConcat and TypeString have been replaced because they were in kind Symbol but weren't literals. The Prim.TypeErrer.Doc kind and related constructors (Text, Quote, Beside, Above) have been added in their place. The Fail and Warn type classes now accept a Doc instead of a Symbol. (#3134, @LiamGoodacre)
  • In simple cases instance overlaps are now checked at declaration time rather than being deferred until an attempt is made to use them. (#3129, @LiamGoodacre)
  • Chaining non-associative or mixed associativity operators of the same precedence is no longer allowed (#3315, @garyb)
  • The --dump-corefn and --source-maps arguments to purs compile have been removed. There is now a --codegen argument that allows the specific codegen targets to be specified - for example, --codegen corefn will not produce JS files, --codgen js,corefn will produce both. If the sourcemaps target is used js will be implied, so there's no difference between --codegen js,sourcemaps and --codegen sourcemaps). If no targets are specified the default is js. (#3196, @garyb, @gabejohnson)
  • Exported types that use foreign kinds now require the foreign kinds to be exported too (bug fix, but potentially breaking) (#3331, @garyb)

Enhancements

  • Added Cons compiler-solved type class for Symbol (#3054, @kcsongor)
  • The Append compiler-solved type class for Symbol can now be run in reverse (#3025, @paf31)
  • Find Usages for values and constructors in purs ide (#3206, @kRITZCREEK)
  • purs ide treats hiding imports the same as open imports when sorting (#3069, @kRITZCREEK)
  • Added inlining for fully saturated usages of runEffFn/mkEffFn (#3026, @nwolverson)
  • Improved explanation of UnusableDeclaration error (#3088, #3304, @i-am-tom)
  • Improved rendering of comments in generated JavaScript by removing additional newlines (#3096, @brandonhamilton)
  • Instance chain support. (#2315, @LiamGoodacre)

    We can now express an explicit ordering on instances that would previously have been overlapping. For example we could now write an IsEqual type class to compute if two types are equal or apart:

    class IsEqual (l :: Type) (r :: Type) (o :: Boolean) | l r -> o
    instance isEqualRefl :: IsEqual x x True
    else instance isEqualContra :: IsEqual l r False
    

    Note the else keyword that links the two instances together. The isEqualContra will only be up for selection once the compiler knows it couldn't possible select isEqualRefl - i.e that l and r are definitely not equal.

  • Improved orphan instance error to include locations where the instance would be valid (#3106, @i-am-tom)
  • Added an explicit error for better explanation of duplicate type class or instance declarations (#3093, @LiamGoodacre)
  • purs ide now provide documentation comments (#2349, @nwolverson)
  • Clarified meaning of duplicate labels in a Record row (#3143, @paf31)
  • Explicit import suggestions consistently use (..) for constructors now (#3142, @nwolverson)
  • Improved tab completion in purs repl (#3227, @rndnoise)
  • Large compiler perfomance improvement in some cases by skipping source spans in Eq, Ord for binders (#3265, @bitemyapp)
  • Added support for error/warning messages to carry multiple source spans (#3255, @garyb)
  • Improved tab completion in purs repl when parens and brackets are involved (#3236, @rndnoise)
  • Improved completion in purs repl after :kind and :type (#3237, @rndnoise)
  • Added the "magic do" optimisation for the new simplified Effect type (Control.Monad.Eff is still supported) (#3289, @kRITZCREEK, #3301, @garyb)
  • Improvide build startup times when resuming a build with incremental results (#3270, @kRITZCREEK)
  • Added compiler-solved Prim.Row.Nub type class (#3293, @natefaubion)
  • Improved docs for Prim.Row.Cons and Prim.Row.Union (#3292, @vladciobanu)
  • Functor can now be derived when quantifiers are used in constructors (#3232, @i-am-tom)
  • purs repl will now complete types after :: (#3239, @rndnoise)
  • Added compiler-solved Prim.Row.Lacks type class (#3305, @natefaubion)
  • Added current output path to missing output error message from purs ide (#3311, @rgrinberg)
  • Improved parser error messages for .purs-repl (#3248, @rndnoise)
  • require in generated JavaScript now includes full index.js file paths (#2621, @chexxor)
  • Added more compiler-solved type classes and supporting types and kinds to Prim:
    • Prim.Ordering module with kind Ordering, type LT, type EQ, type GT
    • Prim.RowList module with class RowToList, kind RowList, type Nil, type Cons
    • Prim.Symbol module with class Compare, class Append, class Cons (#3312, @LiamGoodacre, @kRITZCREEK)
  • Generated code for closed records now explicitly reconstructs the record rather than looping (#1493, @fehrenbach, blog post with more details)

Bug fixes

  • Fixed a bug with names cause by Prim always being imported unqualified (#2197, @LightAndLight)
  • Fixed overlapping instances error message to reflect its new status as an error (#3084, @drets)
  • Added source position to TypeClassDeclaration errors (#3109, @b123400)
  • Fixed entailment issues with skolems and matches in the typechecker (#3121, @LiamGoodacre)
  • Fixed multiple parentheses around a type causing a crash (#3085, @MonoidMusician)
  • Fixed purs ide inserting conflicting imports for types (#3131, @nwolverson)
  • Fixed constraints being inferred differently for lambda expressions compared with equational declarations (#3125, @LiamGoodacre)
  • Updated glob handling to prevent excessive memory usage (#3055, @hdgarrood)
  • Added position information to warnings in type declarations (#3174, @b123400)
  • Fixed documentation generated for Pursuit rendering functional dependency variables as identifier links (#3180, @houli)
  • Naming a function argument __unused no longer breaks codegen (#3187, @matthewleon)
  • Added position information to ShadowedName warning (#3213, @garyb)
  • Added position information to UnusedTypeVar warning (#3214, @garyb)
  • Added position information to MissingClassMember, ExtraneousClassMember, ExpectedWildcard errors (#3216, @garyb)
  • Added position information to ExportConflict errors (#3217, @garyb)
  • Fixed ctags and etags generation when explicit exports are involved (#3204, @matthewleon)
  • Added position information to ScopeShadowing warning (#3219, @garyb)
  • Added position information for various FFI related errors and warnings (#3276, @garyb)
  • Added all available positions to CycleInModule and DuplicateModule errors (#3273, @garyb)
  • Added position information for IntOutOfRange errors (#3277, @garyb, @kRITZCREEK)
  • Warnings are now raised when a module re-exports a qualified module with implicit import (#2726, @garyb)
  • purs repl now shows results for :browse Prim (#2672, @rndnoise)
  • Added position information to ErrorParsingFFIModule (#3307, @nwolverson)
  • Added position information for ScopeConflict cause by exports (#3318, @garyb)
  • Added position information to errors that occur in binding groups and data binding groups (#3275, @garyb)
  • Fixed a scoping issue when resolving operators (#2803, @kRITZCREEK, @LightAndLight)
  • Type synonyms are now desugared earlier when newtype deriving (#3325, @LiamGoodacre)
  • Fixed subgoals of compiler-solved type classes being ignored (#3333, @LiamGoodacre)
  • Added position information to type operator associativity errors (#3337, @garyb)

Other

  • Updated installation information to include details about prebuild binaries (#3167, @MiracleBlue)
  • Test suite now prints output when failing cases are encountered (#3181, @parsonsmatt)
  • Updated test suite to use tasty (#2848, @kRITZCREEK)
  • Improved performance of repl test suite (#3234, @rndnoise)
  • Refactored let pattern desugaring to be less brittle (#3268, @kRITZCREEK)
  • Added makefile with common tasks for contributors (#3266, @bitemyapp)
  • Added ghcid and testing commands to makefile (#3290, @parsonsmatt)
  • Removed old unused MultipleFFIModules error (#3308, @nwolverson)
  • mod and div for Int are no longer inlined as their definition has changed in a way that makes their implementation more complicated - purescript/purescript-prelude#161 (#3309, @garyb)
  • The test suite now checks warnings and errors have position information (#3211, @garyb)
  • The AST was updated to be able to differentiate between let and where clauses (#3317, @joneshf)
  • Support for an optimization pass on CoreFn was added (#3319, @matthewleon)

v0.11.7 - 2017-11-15

Enhancements

  • Add position to type class declaration errors (@b123400)
  • Add valid location list to orphan instance errors (@i-am-tom)
  • Expand error message for UnusableDeclaration (#3088, @i-am-tom)
  • Inline Unsafe.Coerce.unsafeCoerce (@coot)

Bug Fixes

  • Correctly quote uppercased field labels in errors (@Thimoteus)
  • purs ide inserts conflicting imports for types (#3131, @nwolverson)
  • Instantiate abstraction body during inference to fix a type checking bug (@LiamGoodacre)
  • Fix a bug related to the desugaring of nested parentheses (@MonoidMusician)
  • Fix a loop in the kind checker (@paf31)
  • Fix a bug in type operator precedence parsing (@paf31)
  • Eliminate some redundant whitespace in the generated JS output (@matthewleon)
  • Only add newline before initial group of comment lines during code generation (@brandonhamilton)
  • Treat kinds as used in import warnings (@nwolverson)

purs ide

  • Add an "editor mode" (@kRITZCREEK)

    When the editor-mode flag is specified at startup the server will not start afile watcher process any more. Instead it only reloads after successful rebuild commands. This is a lot less fragile than relying on the file system APIs, but will mean that a manual load needs to be triggered after builds that didn't go through purs ide.

  • purs ide now groups hiding imports with implicit ones (@kRITZCREEK)

  • Return documentation comments in purs ide completions (@nwolverson)

  • Add an actualFile parameter to the rebuild command (@kRITZCREEK)

  • Add qualified explicit import (@nwolverson)

  • Fixed case-splitting on local non-exported datatypes (@LightAndLight)

  • Make the filters parameter in the type command optional (@b123400)

purs docs

  • Embed CSS for HTML docs (@hdgarrood)
  • Fix source links for re-exports (@felixSchl)
  • Use order given in export list in generated docs (@hdgarrood)
  • Prevent browser from treating the title and source link as one word (@Rufflewind)
  • Fix fragment links to type constructors in HTML (@hdgarrood)

purs repl

  • Add :complete directive to purs repl to support completion in more editors (@actionshrimp)

Other

  • Add docs for duplicate labels in record types (@paf31)
  • Adds a document for the design of purs ide. (@kRITZCREEK)
  • Update PROTOCOL.md docs for purs ide (@BjornMelgaard)
  • Upgrade to GHC version 8.2 (@kRITZCREEK)
  • Allow blaze-html-0.9 (@felixonmars)
  • Bump Glob dependency (@mjhoy)
  • Use Hspec in TestDocs (@hdgarrood)
  • Fix AppVeyor deployment (#2774) (@hdgarrood)
  • Various type safety improvements to the AST (@kRITZCREEK)
  • Remove some references to old executables (@hdgarrood)
  • Update the installation documentation (@hdgarrood)
  • Update test dependencies (@hdgarrood)
  • Only build master and versioned tags in AppVeyor (@hdgarrood)

v0.11.6 - 2017-07-10

New Features

RowToList support

(@LiamGoodacre)

There is a new type class in typelevel-prelude called RowToList, which turns a row of types into a type-level list. This allows us to work with closed rows in more ways at the type level. The compiler will now solve these constraints automatically for closed rows of types.

Enhancements

  • Allow things to be hidden from Prim (@garyb)
  • Re-evaluate REPL globs on :reload (@hdgarrood)
  • Include comments in child declarations in HTML docs (@hdgarrood)

IDE Enhancements

  • Collect data constructors (@kRITZCREEK)
  • Adds declarations for Prim (@kRITZCREEK)
  • Repopulates the rebuild cache when populating volatile state (@kRITZCREEK)
  • Add declaration type filter (#2924) (@sectore)
  • Improve reexport bundling (@kRITZCREEK)
  • Resolve synonyms and kinds (@kRITZCREEK)

Bug Fixes

  • Replace synonyms in instance constraints (@LiamGoodacre)
  • Encode PSCI's server content as UTF-8 string (@dgendill)
  • Fix child declaration ordering in docs (@hdgarrood)
  • Improve instance ordering in HTML docs (@hdgarrood)
  • Fix links to type operators in HTML docs (@hdgarrood)

Other

  • Add source span annotations to Declaration (@garyb)
  • Add source span annotations to DeclarationRef (@garyb)
  • Remove purescript.cabal and add to .gitignore (@garyb)
  • Raise upper bound on aeson in package.yaml (@garyb)
  • Only build master and semver tags in Travis (@hdgarrood)

v0.11.5 - 2017-06-05

Compiler

Enhancements

Type signatures in instances

(@cdepillabout)

Type class instances can now include type signatures for class members, as documentation:

data MyType = MyType String

instance showMyType :: Show MyType where
  show :: MyType -> String
  show (MyType s) = "(MyType " <> show s <> ")"

Bug Fixes

  • Encode HTML content as UTF8 when using purs repl with --port (@dgendill)
  • Disallow some invalid newtype-derived instances (@paf31)
  • Disallow forall within constraints (#2874, @sectore)
  • Convert \r\n into \n after reading files (@kRITZCREEK)
  • Fix PSCi tests (@kRITZCREEK)
  • Better variable naming hygiene in TCO. (#2868, @houli)
  • Simplify TCO generated code (@matthewleon)
  • Remove newlines from printed custom type errors (@matthewleon)
  • Fix some purs command line help message issues (@Cmdv)
  • Apply relative paths during pretty printing of errors (@kRITZCREEK)
  • Desugar let properly when generating docs (@paf31)
  • Fix kind signature for RowCons type class in documentation (@tslawler)
  • Fix an issue with error messages for TypesDoNotUnify involving duplicate labels (#2820, @thoradam)

Other

  • Update package.yaml (@sol)
  • Parse support modules from actual test support purs (@noraesae)
  • Update build command to run tests (@sectore)
  • Bumps lower bound for directory (@kRITZCREEK)
  • Switch core-tests to psc-package (#2830, @matthewleon)
  • Small fix for the copyright dates (@seanwestfall)
  • Update CONTRIBUTING.md for "new contributor" label (@thoradam)

purs ide

Features

  • Add a new namespace filter (#2792, @sectore, @stefanholzmueller)

A new filter, which restricts query results to the value, type and/or kind namespaces, which allows improvements to the completion and import commands.

  • Adds a command to add qualified imports (@kRITZCREEK)

This empowers editor plugins to add imports for qualified identifiers, for example in the Emacs plugin.

  • New import formatting (@kRITZCREEK)
  • Group reexports in completions (@kRITZCREEK)

Editors can now choose to let purs ide group reexports for the same value, to reduce noise when completing values like Data.Functor.map which are reexported a lot and show up that many times in the completion list.

Enhancements

  • Parse modules in parallel (@kRITZCREEK)

This can yield significant speedups in the initial load times. For example a full load of slamdata/slamdata improves from 11 to 6 seconds

  • Introduce completion options (@kRITZCREEK)

Bug Fixes

  • Resolve synonyms and kinds (@kRITZCREEK)
  • Work around laziness when measuring command performance (@kRITZCREEK)
  • Simplify state type (@kRITZCREEK)
  • Extract namespace ADT (@kRITZCREEK)
  • Decodes source files as UTF8 when parsing out the imports (@kRITZCREEK)
  • Fix the import command for kinds (@kRITZCREEK)
  • Reads files in text mode for adding imports (@kRITZCREEK)
  • Add -h/--help to ide subcommands (@simonyangme)

v0.11.4 - 2017-04-17

Enhancements

  • purs executable will now display help text by default (@matthewleon)
  • Adding -h/--help to ide subcommands (@simonyangme)
  • Some simplifications to the tail call optimization (@matthewleon)

Bug Fixes

  • Remove newline from printed custom type errors (@matthewleon)
  • Fix pretty printing of rows in error messages (#2820, @thoradam)
  • Allow user to propagate Warn constraints (@paf31)
  • Match type level strings in docs renderer (#2772, @hdgarrood)
  • Fix encoding bug in purs ide list import command (@kRITZCREEK)
  • purs ide now reads files in text mode for adding imports (@kRITZCREEK)

Other

  • Bump aeson lower bound to 1.0 (@hdgarrood)
  • Add a bunch of NFData instances (@hdgarrood)
  • Turn off coveralls upload for now (@paf31)
  • purs command line help message fixes (@Cmdv)
  • Switch core-tests to psc-package (#2830, @matthewleon)
  • Update CONTRIBUTING.md notes (@thoradam)

v0.11.3 - 2017-04-08

Bug Fixes

  • Fix the exhaustivity check for pattern guards (@alexbiehl)

Other

  • Require directory >=1.2.3.0 for XDG support (@bergmark)
  • @noraesae has refactored some PSCi code to improve the test suite.
  • Use hpack to generate the .cabal file (@kRITZCREEK)
  • Use XDG Base Directory Specification for psci_history (@legrostdg)

v0.11.2 - 2017-04-02

New Features

Polymorphic Labels

(@paf31)

A new RowCons constraint has been added to Prim. RowCons is a 4-way relation between

  1. Symbols
  2. Types
  3. Input rows
  4. Output rows

which appends a new label (1) with the specified type (2) onto the front of the input row (3), to generate a new output row (4). The constraint can also be run backwards to subtract a label from an output row.

This allows us to quantify types over labels appearing at the front of a row type, by quantifying over the corresponding symbol/type pair. This gives us a limited form of polymorphic labels which enables things like writing a single lens for any record accessor.

Enhancements

  • Use XDG Base Directory Specification for the location of the psci_history file (@legrostdg)
  • Collect more information for classes and synonyms in purs ide (@kRITZCREEK)

Bug Fixes

  • Desugar pattern guards after type checking, to avoid an issue with the exhaustivity checker (@alexbiehl)

Other

  • A new PSCi evaluation test suite was added (@noraesae)
  • Use hpack to generate the .cabal file (@kRITZCREEK)

v0.11.1 - 2017-03-28

Bug Fixes

Compiler

  • Enable TCO for variable intros and assignments #2779 (@paf31)
  • Fixed special case in codegen for guards #2787 (@paf31)

Docs generation

  • Wrap decl title in span for better double-click selection #2786 (@rightfold)
  • List instance info under correct sections, fix #2780 (@paf31)

v0.11.0 - 2017-03-25

This release includes several breaking changes, in preparation for the 1.0 release, as well as many enhancements and bug fixes.

Most users will probably want to wait until all aspects of the release have been finalized. Progress on libraries and tools is being tracked here.

Many thanks to the contributors who helped with this release!

Breaking Changes

(@garyb, @paf31)

=> now acts like a binary type operator

It was previously possible to specify many constraints in the same context by separating them with commas inside parentheses on the left of the =>:

runFreeT ::  m f. (Functor f, Monad m) => ...

This is no longer allowed. Instead, => now acts like a binary operator, with a constraint on the left and a type on the right. Multiple constraints must be introduced using currying, as with regular function arguments:

runFreeT ::  m f. Functor f => Monad m => ...

This is in preparation for adding constraint kinds, at which point => will become an actual binary type operator, defined in Prim.

* and ! kinds have been removed

The kind symbols * (for the kind of types) and ! (for the kind of effects) have been removed from the parser. Instead of *, use Type, which is defined in Prim. Instead of !, use Effect, which can now be imported from Control.Monad.Eff.

The # symbol, which is used to construct a row kind, is still supported. We cannot move this kind into Prim (because it is polykinded, and we do not support kind polymorphism).

One single consolidated executable

The various psc-* executables have been replaced with a single executable called purs. The various subcommands are documented on the --help page:

bundle     Bundle compiled PureScript modules for the browser
compile    Compile PureScript source files
docs       Generate Markdown documentation from PureScript source files
hierarchy  Generate a GraphViz directed graph of PureScript type classes
ide        Start or query an IDE server process
publish    Generates documentation packages for upload to Pursuit
repl       Enter the interactive mode (PSCi)

Wrapper scripts will be provided in the binary distribution.

psc-package was removed

psc-package has been removed from the main compiler distribution. It will still be maintained along with the package sets repo, but will not be bundled with the compiler.

A binary distribution which is compatible with this release is available.

Implicitly discarded values in do blocks now raise errors

Code which discards the result of a computation in a do block:

duplicate :: Array a -> Array a
duplicate xs = do
  x <- xs
  [true, false] -- the result here is discarded
  pure x

will now raise an error. The compiler allows values of certain types to be discarded, based on the Discard class in Control.Bind. The only type which can be discarded is Unit, but the feature was implemented using a type class to enable support for alternative preludes.

No more dependency on the Bower executable

In addition to removing psc-package from the compiler distribution, we have also removed any explicit dependency on the Bower executable. The compiler will not assume use of any particular package manager, but will aim to provide generic support for package managers generally, via command line options and hooks.

purs publish will continue to use the Bower JSON formats. The bower.json format is now referred to as the "manifest file", while the output of bower list --json, which is used by purs publish internally, is referred to as the "resolutions file".

Enhancements

Pattern Guards

(@alexbiehl)

In addition to regular guards:

foo x | condition x = ...

the compiler now supports pattern guards, which let the user simultaneously test a value against a pattern, and bind names to values.

For example, we can apply a function fn to an argument x, succeeding only if fn returns Just y for some y, binding y at the same time:

bar x | Just y <- fn x = ... -- x and y are both in scope here

Pattern guards can be very useful for expressing certain types of control flow when using algebraic data types.

HTML Documentation

(@hdgarrood)

The --format html option has been added to purs docs. The HTML format uses the Pursuit template, and is very useful for rendering documentation for offline use.

Here is an example of the generated HTML.

Duplicate Labels

(@paf31)

Row types now support duplicate labels, which can be useful when using the Eff monad. For example, we could not previously use the catchException function if the resulting action also required the EXCEPTION effect, since otherwise the type of the inner action would contain a duplicate label.

Rows are now unordered collections (of labels and types) with duplicates. However, the collection of types for a specific label within a row is ordered. Conceptually, a row can be thought of as a type-level Map Label (NonEmptyList Type).

A type constructor (such as Record) which takes a row of types as an argument should define what its meaning is on each row. The meaning of a value of type Record r is a JavaScript object where the type of the value associated with each label is given by the head element of the non-empty list of types for that label.

Row Constraints

(@doolse, @paf31)

A new constraint called Union has been added to Prim. Union is a three-way relation between rows of types, and the compiler will solve it automatically when it is possible to do so.

Union is a left-biased union of rows which takes into account duplicate labels. If the same label appears in rows l and r, and Union l r u holds, then the label will appear twice in u.

Union makes it possible to give a type to the function which merges two records:

merge :: forall r1 r2 r3. Union r1 r2 r3 => Record r1 -> Record r2 -> Record r3

Note that this is a left-biased merge - if the two input record contain a common label, the type of the label in the result will be taken from the left input.

Patterns in let expressions

(@noraesae)

Let expressions and where clauses can now use binders on the left hand side of a declaration:

map f xs =
  let { head, tail } = uncons xs
  in [f head] <> map f tail

Unlike in Haskell, declarations with these patterns cannot appear in dependency cycles, and bound names can only be used in declarations after the one in which they are brought into scope.

Find record accessors in Type Directed Search

(@kRITZCREEK)

Type-directed search will now include results for record accessors. This can be very useful when working with extensible records with a type-driven programming workflow.

Other Enhancements

  • Add basic usability check and error for ambiguously-typed type class members (@LiamGoodacre)
  • Improved skolem escape check (@paf31)
  • Fix links to declarations in Prim (@hdgarrood)
  • Emit _ instead of false case for if then else to improve optimizations (@rightfold)
  • Add InvalidDerivedInstance error to improve errors for derived instances (@paf31)
  • Make generated code for superclass instances less ugly (@paf31)
  • Support polymorphic types in typed binders (@paf31)
  • Make file paths relative in error messages (@paf31)
  • Improve errors from module sorter (@paf31)
  • Improve error for unused type variables (@paf31)
  • Include source span in externs file for error reporting purposes (@paf31)
  • Improve instance arity errors (@mrkgnao)

purs ide

Features

Improve import parsing

  • purs ide now uses a new import parser, which allows purs ide to handle any import section that the compiler would accept correctly. (@kRITZCREEK)
  • Parse imports with hanging right paren (@matthewleon)
  • Reuses lenient import parsing for the list import command (@kRITZCREEK)

Don't create the output/ directory if it can't be found

(@kRITZCREEK)

purs ide will now no longer leave empty output/ directories behind when it is started in a directory that is not a PureScript project.

Collect type class instances

(@kRITZCREEK)

purs ide collects instances and stores them with their respective type class. There's no way to retrieve these yet, but we will extend the protocol soon.

Bug Fixes

  • No longer strip trailing dots for Pursuit queries (@kRITZCREEK)
  • Fix #2537 (psc-ide shouldn't crash when building a non-existent file) (@kRITZCREEK)
  • Fix #2504 (fix a crash related to prematurely closed handles) (@kRITZCREEK)
  • Speed up rebuilding by x2, by rebuilding with open exports asynchronously (@kRITZCREEK)
  • Return operators in purs ide imports list (@nwolverson)
  • Also detect location information for operators (@kRITZCREEK)

Cleanup

  • Removes unnecessary clause in import pretty printing (@kRITZCREEK)
  • Removes the deprecated --debug option (@kRITZCREEK)
  • Restructure testing to avoid running the server (@kRITZCREEK)

purs repl

  • Add back .purs-repl file support (@paf31)
  • PSCi command changes, add :clear (@noraesae)
  • Declarations no longer require let (@noraesae)
  • Improve CLI error and startup messages (@noraesae)

Bug Fixes

  • Changes to help the tail call optimization fire more consistently (@paf31)
  • Fix everythingWithScope traversal bug #2718 (@paf31)
  • Errors for open rows in derived instances (@paf31)
  • Instantiate types in record literals as necessary (@paf31)
  • Fix Generic deriving with synonyms (@paf31)
  • Rebuild modules if necessary when using --dump-corefn (@paf31)
  • Ensure solved type classes are imported (@LiamGoodacre)
  • Allow for older Git versions in purs publish (@mcoffin)
  • Fix purs publish --dry-run (@hdgarrood)
  • Exported data constructors can now contain quotes (@LiamGoodacre)

Documentation

  • Capitalise *script into *Script (@noraesae)

Performance

  • Optimize keepImp (@paf31)
  • Replace nub with ordNub (@matthewleon)
  • Combine inlining optimizations into a single pass (@paf31)

Other

  • Add HasCallStack to internalError (@alexbiehl)
  • Use Stackage LTS 8.0 (@noraesae)
  • Address Travis timeout issues (@hdgarrood)
  • Improve module structure in PSCi test suite (@noraesae)
  • Fix the PSCi script (@mrkgnao)
  • Include Git commit information in non-release builds (@hdgarrood)
  • Add test case for #2756 (@int-index)
  • Some code cleanup in the module imports phase (@matthewleon)

v0.10.7 - 2017-02-11

This release contains a bug fix for a bug in psc-bundle which was introduced in 0.10.6.

v0.10.6 - 2017-02-07

Enhancements

  • Add support for user defined warnings via the Warn type class (@LiamGoodacre, blog post)
  • Support nested record update (@LiamGoodacre, blog post)
  • Inline unsafePartial (@paf31)
  • Fail early when bind is brought into scope inside do (@paf31)

Bug Fixes

  • Disallow polymorphic types in binders, preventing a crash (@paf31)
  • Rebuild modules if necessary when using --dump-corefn (@paf31)
  • TypeLevelString/TypeConcat should not be quoted (@michaelficarra)
  • Generate JS static member accesses whenever possible (@michaelficarra)
  • Require dependencies to exist during sorting phase (@paf31)
  • Fix inlining for negateInt (@paf31)
  • Fix object key quoting (@hdgarrood)
  • Don't expand synonyms until after kind checking (@paf31)
  • Fix 'Unknown type index' on mismatch between class and instance argument counts (@LiamGoodacre)
  • Style comment types differently (@matthewleon)

psc-ide

  • Return operators in psc-ide imports list (@nwolverson)
  • Collect type class instances (@kRITZCREEK)
  • Log failing to accept or parse an incoming command (@kRITZCREEK)
  • Fix #2537 (@kRITZCREEK)
  • Fix #2504 (@kRITZCREEK)
  • Also detect location information for operators (@kRITZCREEK)
  • Speeds up rebuilding by x2 (@kRITZCREEK)
  • Restructure testing to avoid running the server (@kRITZCREEK)

psc-publish

  • Add modules for rendering HTML documentation (@hdgarrood)
  • Fix psc-publish --dry-run (@hdgarrood)
  • Fix failure to parse git tag date in psc-publish (@hdgarrood)
  • Add git tag time to psc-publish JSON (@hdgarrood)
  • Remove Docs.Bookmarks (@hdgarrood)

Performance

  • Combine inlining optimizations into a single pass (@paf31)
  • Use Map.foldlWithKey' instead of foldl (@hdgarrood)
  • Minor memory usage improvements in Language.PureScript.Docs (@hdgarrood)

Other

  • Generate data constructors without IIFEs (@hdgarrood)
  • Add stack-ghc-8.0.2.yaml (@noraesae)
  • Add HasCallStack to internalError (@alexbiehl)
  • Update psc-package to use turtle 1.3 (@taktoa)
  • Remove JSAccessor; replace with JSIndexer (@michaelficarra)
  • Store more information in RenderedCode (@hdgarrood)

v0.10.5 - 2017-01-06

Enhancements

  • Adds specific error message when failing to import bind (@FrigoEU)

Bug Fixes

  • Detect conflicting data constructor names (@LiamGoodacre)
  • Update pretty printer for Kinds (@hdgarrood)
  • Restore JSON backwards compatibility for PSString (@hdgarrood)
  • Replace type wildcards earlier (@paf31)
  • Restore backwards compatibility for parsing Kinds (@hdgarrood)

Other

  • Update bower-json to 1.0.0.1 (@hdgarrood)

v0.10.4 - 2017-01-02

New Features

Deriving Functor

(@LiamGoodacre, #2515)

The Functor type class can now be derived using the standard derive instance syntax:

newtype F a = F { foo :: Array a, bar :: a }

derive instance functorF :: Functor F

User-Defined Kinds

(@LiamGoodacre, #2486)

Custom kinds can now be defined using the foreign import kind syntax:

foreign import kind SymbolList

Custom kinds can be ascribed to types using foreign import data declarations, as usual:

foreign import data Nil :: SymbolList
foreign import data Cons :: Symbol -> SymbolList -> SymbolList

Note that kind arguments are not supported.

User defined kinds can be imported/exported using the kind prefix, for example:

import Type.SymbolList (kind SymbolList)

Source Maps in psc-bundle

(@nwolverson)

psc-bundle will now generate source maps if the--source-maps flag is used.

Solving CompareSymbol and AppendSymbol

(@LiamGoodacre, #2511)

Support for the new purescript-typelevel-prelude library has been added to the compiler. CompareSymbol and AppendSymbol constraints will now be solved automatically for literal symbols.

New psc-package Features

(@paf31)

Two new commands have been added to psc-package to support library authors and package set curators.

  • The updates command (#2510) is used to update packages in the set.
  • The verify-set command (#2459) is used to verify the health of a package set. This command replicates the work done by the package-sets CI job, and can be used to test modifications to the package set locally before making a pull request.

Enhancements

  • Update orphan instance check to use covering sets when functional dependencies are involved (@LiamGoodacre)
  • Add --node-path option to PSCi to modify the path to the Node executable (#2507, @paf31)
  • Add package information to re-exports (@hdgarrood)
  • Add Prim docs to the library (#2498, @hdgarrood)

Bug Fixes

  • Derive instances when data types use type synonyms (#2516, @paf31)
  • Unwrap KindedType when instance solving (@LiamGoodacre)
  • Update links to wiki (#2476, @LiamGoodacre)
  • Update websocket host to fix PSCi on Windows (#2483, @seungha-kim)
  • Fix psc-ide tests on windows (@kRITZCREEK)
  • Fix some issues with the pretty printer (#2039, @paf31)

Other

  • More robust license generator script (@hdgarrood)
  • Further conversions to Text in the Docs modules (#2502, @hdgarrood)
  • Add upper bound on turtle, fixes #2472, (@hdgarrood)
  • Fix version bounds on language-javascript (@hdgarrood)

v0.10.3 - 2016-12-11

Enhancements

Solving IsSymbol instances

(@LiamGoodacre)

The compiler will now derive Data.Symbol.IsSymbol instances for type-level string literals.

This enables interesting type-level programming features, such as deriving Show instances using Data.Generics.Rep.

Rows in Instance Heads

(@LiamGoodacre)

The compiler now allows rows to appear in type class instance heads, but only in type arguments which are fully determined by some functional dependency.

This allows instances like

MonadState { field :: Type } MyAppMonad

and also Newtype instances for newtypes which contain records.

Speeds up parsing by reading files as Text

(@kRITZCREEK)

The use of String has been replaced by Text in the compiler, resulting in some non-trivial performance improvements.

Functional Dependencies in psc-docs output

(@soupi, #2439)

psc-docs now includes functional dependency information when rendering type classes.

New psc-package Commands

  • The available command (@andyarvanitis) shows all available packages in the current package set
  • The uninstall command (@joneshf) removes a package from the set of active packages and updates the package configuration file.

Type Class Warning (@joneshf)

A warning was added for shadowed type variables in type class declarations.

Bug Fixes

  • psc-package: display full path in 'packages.json does not exist' error messsage (@andyarvanitis)
  • Use writeUTF8File in psc-bundle (@hdgarrood)
  • Use HTTPS to query Pursuit (@paf31)
  • Moved the expansion of astral code points to UTF-16 surrogate pairs from the JS code generator to the parser (@michaelficarra, #2434)
  • Allow astral code points in record literal keys (@michaelficarra, #2438)
  • Add value source positions (@nwolverson)
  • Update error message of ErrorInDataBindingGroup to include participating identifiers (@LiamGoodacre)

psc-ide

  • Polling option for psc-ide-server (@kRITZCREEK)
  • Better logging and diagnostics (@kRITZCREEK)

Other

  • Dump output of psc tests to file (@andyarvanitis, #2453)
  • Fix windows CI (@hdgarrood)
  • Link to new documentation repo (@hdgarrood)
  • Create documentation for psc-package (@paf31)
  • Fix GHC 8.0.2 build (@RyanGlScott)
  • Add psc-package to release bundle (@marsam)
  • Update for latest language-javascript (@tmcgilchrist)
  • Fix exhaustivity warnings (@charleso)
  • Update CONTRIBUTING.md (@osa1)

v0.10.2 - 2016-11-07

Major Changes

Type-directed search (@kRITZCREEK)

This extends the typed holes error messages to include suggested replacements for a typed hole, by using type subsumption to determine which identifiers in scope are appropriate replacements.

A blog post will accompany this feature soon.

psc-package (@paf31)

This is an experimental package manager for PureScript packages. It supports the following commands:

  • init - create a new project using the package set for the current compiler version
  • update - sync the local package collection with the package set
  • install - install a specific package from the current set and add it to the package config
  • build - run psc on any active packages
  • sources - list source globs for active package versions
  • dependencies - list transitive dependencies of the current project

For example:

$ psc-package init
$ psc-package install transformers
$ psc-package build

Eventually, psc-package might replace the use of Bower, but that will require support from tools like Pulp. For now, package authors should continue to publish packages using Bower and Pursuit.

Data.Generic.Rep.Generic Deriving (@paf31)

This is an alternative generic programming implementation based on GHC.Generics. It should allow deriving of more interesting classes, such as Semigroup. See the purescript-generics-rep package for examples.

Enhancements

  • #2323: Sort IDE-generated explicit imports (@bbqbaron)
  • #2374: Add error message for ambiguous type variables in inferred contexts (@bbqbaron)
  • #934 Add paste mode, remove --multi-line option (@paf31)
  • Allow symbols in data constructors (@brandonhamilton)
  • Fix inliner for integer bitwise operators (@brandonhamilton)
  • Use SSL for pursuit queries (@guido4000)

Bug Fixes

  • #2370, allow rows in instance contexts (@paf31)
  • #2379, add error message for unknown classes (@paf31)
  • Better error messages for bad indentation (@paf31)
  • Fix inliner for Data.Array.unsafeIndex (@brandonhamilton)
  • Fix issue with typed holes in inference mode (@paf31)
  • Fix scope traversal for do-notation bind. (@LiamGoodacre)
  • Handle TypeLevelString when checking orphans (@joneshf)
  • Move unsafeIndex to Data.Array (@brandonhamilton)
  • Pretty-print suggested types differently (@paf31)
  • Traversal should pick up bindings in all value declarations. (@LiamGoodacre)
  • Treat type annotations on top-level expressions as if they were type declarations (@paf31)

Other

  • Refactor subsumes function (@paf31)
  • Refactor to use lens (@kRITZCREEK)
  • Small cleanup to Language.PureScript.Interactive.IO (@phiggins)
  • Speeds up parsing by reading files as Text (@kRITZCREEK)
  • Update outdated comments about Prim types (@rightfold)

v0.10.1 - 2016-10-02

Breaking Changes

The new functional dependencies feature fixes type inference in some cases involving multi-parameter type classes. However, due to a bug in the compiler, some of those expressions were previously type checking where they should not have. As a result, it is necessary to add functional dependencies to some classes in order to make previous code type-check in some cases. Known examples are:

  • MonadEff and MonadAff
  • MonadState, MonadReader, and the rest of the MTL-style classes in transformers

New Features

Data.Newtype Deriving

(@garyb)

It is now possible to derive the Newtype class for any data declaration which is a newtype, using the existing deriving instance syntax:

newtype Test = Test String

derive instance newtypeTest :: Newtype Test _

Note that the second type argument should be specified as a wildcard, and will be inferred.

Added type level string functions

(@FrigoEU)

The Prim module now defines the TypeString and TypeConcat type constructors, which can be used to build more descriptive error messages which can depend on types, using the Fail constraint:

instance cannotShowFunctions
    :: Fail ("Function type " <> TypeString (a -> b) <> " cannot be shown.")
    => Show (a -> b) where
  show _ = "unreachable"

infixl 6 type TypeConcat as <>

--dump-corefn

(@rightfold)

The compiler now supports the --dump-corefn option, which causes the functional core to be dumped in output/**/corefn.json. This should be useful for implementing new backends which interpret the functional core.

Newtype Deriving

(@paf31)

It is now possible to derive type class instances for newtypes, by reusing the instance for the underlying type:

newtype X = X String

derive newtype instance showX :: Show X

Note that it is possible to derive instances for multi-parameter type classes, but the newtype must only appear as the last type argument.

Allow anonymous accessor chains (_.a.b)

(@rvion)

Anonymous record accessor syntax has been extended to work with chains of one or more accessors:

getBaz = _.foo.bar.baz

Functional Dependencies (@paf31)

The type class solver now supports functional dependencies. A multi-parameter type class can define dependencies between its type arguments by using the -> operator:

class Stream el s | s -> el where
  cons :: el -> (Unit -> s) -> s
  uncons :: s -> { head :: el, tail :: s }

Here, the s and el type arguments are related by a single functional dependency, which ensures that there is at most one instance for any given type s. Alternatively, the type s determines the type el, i.e. there is an implicit function from types s to types el. This information can be used by the solver to infer types where it was previously not possible.

See the following examples for more information:

Enhancements

  • Return qualifier from explicit/hiding imports (@nwolverson)
  • Verify entry points exist in psc-bundle (@kRITZCREEK)
  • Improved error messages for record subsumption (@FrigoEU)

psc-ide

  • Resolve types/kinds for operators (@kRITZCREEK)
  • Unify Completion Commands (@kRITZCREEK)
  • Parse type annotations from source files (@kRITZCREEK)
  • Update pursuit JSON parsing (@nwolverson)
  • Remove a pursuit workaround (@kRITZCREEK)
  • Add a suggestion to the UnusedDctorImport warning (@FrigoEU)
  • Return JSON errors for cycles in module dependencies (@kRITZCREEK)

Bug Fixes

  • Fix usage detection for operators (@garyb)
  • Fix handling of duplicate module imports in JS codegen (@garyb)
  • Fix a small bug in the type pretty-printer (@paf31)
  • Fix function application judgment (@paf31)
  • Fix inlining for $ and # operators (@garyb)
  • Fix everywhereOnTypesTopDown (@ianbollinger)
  • Fix unification of string literals (@paf31)

Infrastructure

  • Support aeson-1.0 (@phadej)
  • Support http-client-0.5 (@phadej)
  • Safer installation from source in INSTALL.md (@hdgarrood)

Implementation

  • Fix most HLint warnings (@ianbollinger)
  • Fixing imports (@charleso)
  • Export desugarDecl from Sugar.ObjectWildcards (@rvion)
  • Remove legacy ObjectGetter and update doc (@rvion)

v0.9.3 - 2016-08-01

Enhancements

  • Better context information for typed hole errors (@paf31)
  • Improved error messages in the constraint solver. Type class errors now include better contextual information, including smaller source spans. (@paf31)

Bug Fixes

  • Decode externs with correct encoding (@natefaubion)
  • Fix bad codegen for empty string fields (@LiamGoodacre, #2244)
  • Instantiate types in array literals before unification (@paf31, #2252)

Other

  • Upgrade to protolude 0.1.6 (@ilovezfs)
  • Use latest LTS (@paf31, #2241)
  • Add upper bound to http-client (@paf31, #2237)
  • Combine the sdist and coverage builds. Avoid .tix files during deployment. (@paf31)

v0.9.2 - 2016-07-11

Enhancements

Goto Definition

@kRITZCREEK has added the ability to return position information for expressions in psc-ide. This can be used to implement a Goto Definition feature in IDEs which use psc-ide-server as the backend.

Evaluate PSCi expressions in the browser

(@paf31)

PSCi now features an alternative backend, which can run commands in the browser via a websocket. To use this mode, simply pass the --port option on the command line:

$ pulp psci --port 9000

and open your web browser to localhost on that port.

See https://github.com/paf31/psci-experiment for a demonstration.

psc-ide architecture changes

@kRITZCREEK has worked on changing the architecture of psc-ide generally, to load data in multiple phases and asynchronously. This enables new features like Goto Definition above.

Other

  • Allow pipes version 4.2 (@felixonmars)
  • Elaborate re-exports (@garyb)

Bug Fixes

psc-ide

  • Fix unicode encoding of json responses (@kRITZCREEK)
  • Improved handling of reexports (@kRITZCREEK)

Other

  • Update Data.Function constant for prelude 1.0 (@felixSchl)
  • Include position info in ScopeShadowing warning (@garyb)

v0.9.1 - 2016-06-01

PureScript 0.9.1 is a major stable release of the compiler. It removes features which were deprecated in the 0.8.x series, and contains several useful enhancements and bug fixes.

This release will be accompanied by new releases of the core libraries and a compatible version of Pulp, which have been updated to work with this version.

Due to the relatively large number of breaking changes, library authors are advised that they will probably need to update their libraries to maintain compatibility. Users may prefer to continue using version 0.8.5 until their dependencies have been updated.

Breaking Changes

Name resolving

(@garyb)

The way names are resolved has now been updated in a way that may result in some breakages. The short version is: now only names that have been imported into a module can be referenced, and you can only reference things exactly as you imported them.

Some examples:

Import statement Exposed members
import X A, f
import X as Y Y.A Y.f
import X (A) A
import X (A) as Y Y.A
import X hiding (f) A
import Y hiding (f) as Y Y.A

Qualified references like Control.Monad.Eff.Console.log will no longer resolve unless there is a corresponding import Control.Monad.Eff.Console as Control.Monad.Eff.Console. Importing a module unqualified does not allow you to reference it with qualification, so import X does not allow references to X.A unless there is also an import X as X.

Although the new scheme is stricter it should be easier to understand exactly what the effect of any given import statement is. The old resolution rules for qualified names were obscure and unexpected results could arise when locally-qualified module names overlapped with "actual" module names.

Module re-exports have also been tightened up as a result of these rules. Now if module X is only imported as Y, the re-export must list module Y also. If a module is imported without being re-qualified then the original name is used.

Partial Constraints

(@garyb, @paf31)

The compiler will now generate an error for a missing Partial constraints, where it would previously have issued a warning.

Module Restrictions

(@garyb, @paf31)

  • Imports must now appear before other declarations in a module.
  • A source file must now contain exactly one module.

These restrictions will allow us to improve incremental build times in future, since we will only need to parse a small prefix of each file in order to figure out what needs to be rebuilt. Right now, we need to parse every file fully.

Foreign Function Interface Changes

(@paf31)

Foreign modules are now found by filename rather than by searching for a custom JavaScript comment. The foreign module is found by changing the extension of the corresponding PureScript module from .purs to .js.

This change was made to be more consistent with psc-ide, and also to adopt a simple convention which will port well to other backends.

Operator Aliases

(@garyb)

All operators must be defined as aliases from now on. That is, it is no longer valid to define an operator as a name in local scope (e.g. let (#) x y = x y in ...). This change makes it possible to generate better JavaScript code for operators, by desugaring them to the functions they alias.

Other

  • Deprecated class import/export syntax has been removed (@LiamGoodacre). Classes are now imported using the class keyword, and exported similarly:

    import Prelude (class Show, show)
  • Remove support for = in record binders (@paf31).

    Record binders such as

    f { x = 0 } = true

    are no longer supported. Record binders must now use : instead:

    f { x: 0 } = true
  • Prim.Object has been renamed to Prim.Record (#1768, @paf31)

Enhancements

Programmable Type Errors

(@paf31)

Constraints can now contain type-level strings which can be used as custom error messages using the Fail constraint. For example, one can now document the fact that foreign types such as JSDate cannot be made instances of Generic:

instance dateIsNotGeneric
  :: Fail "JSDate is not Generic. Consider using Int with toEpochMilliseconds instead."
  => Generic JSDate where
    fromSpine   = crashWith "fromSpine: unreachable"
    toSpine     = crashWith "toSpine: unreachable"
    toSignature = crashWith "toSignature: unreachable"

Attempting to derive a Generic instance for a type containing JSDate will then result in

A custom type error occurred while solving type class constraints:

    JSDate is not Generic. Consider using Int with toEpochMilliseconds instead.

Typed Hole Improvements

(#2070, @paf31)

Typed hole error messages now include the types of any names in scope, to assist with type-driven development:

> :t \x -> maybe 0 ?f x
Error found:
in module $PSCI
at  line 1, column 8 - line 1, column 22

  Hole 'f' has the inferred type

    t0 -> Int

  in the following context:

    it :: Maybe t0 -> Int
    x :: Maybe t0


in value declaration it

where t0 is an unknown type

Editor Support

  • The results of the last rebuild are now cached by psc-ide, which improves completion support for editor plugins. (@kRITZCREEK)
  • A reset command was added to psc-ide (@kRITZCREEK)
  • The compiler will now suggest replacements to address MissingTypeDeclaration and TypeWildCard warnings (@nwolverson)

PSCi Improvements

(@paf31)

  • The design of PSCi has been changed to improve performance. PSCi now precompiles all dependencies and uses the same incremental rebuilding approach as psc-ide. This means that the :load and :foreign commands have been removed, since dependencies are fixed and pre-compiled when PSCi loads.
  • PSCi now supports alternative base libraries such as Neon, by depending on purescript-psci-support for its supporting code.

Colors in Error Messages

Types and values will now be highlighted in error messages, when the terminal supports it (MacOS and Linux for now) (@soupi).

Type Names

Prime characters are now allowed in type names. (@garyb)

Bug Fixes

  • Parser error messages inside type class and instance declarations were improved (#2128, @bmjames)
  • Editor suggestions for imports now use (..) (@garyb)
  • Source-spans to token end position (@nwolverson)
  • Some pretty printing issues related to string literals in records were fixed (@LiamGoodacre)
  • Some presentation bugs in PSCi's :show import were fixed (@LiamGoodacre)
  • Parsec was updated to the latest version to fix an issue with literal parsing (#2115, @hdgarrood)
  • Fixed a bug related to certain typed binders which would cause the compiler to crash (#2055, @paf31)
  • As-patterns now bind less tightly (@paf31)
  • More identifiers can now be parsed in FFI imports (@michaelficarra)
  • Fixed a performance issue which manifested under certain conditions in psc-ide (#2064, @kika)
  • Fixed a test which contained an unreliable comparison (#2093, @andyarvanitis)
  • The precedence of type application was corrected (#2092, @paf31)
  • An indentation bug in the parser was fixed (@DavidLindbom)
  • License errors from psc-publish were improved (@hdgarrood)

Other

  • The test suite now exercises various compiler warnings (@garyb)
  • The test suite performance was improved by using incremental rebuilds (@paf31)
  • The test suite now tests that passing tests contain a main function (@hdgarrood)
  • The test suite now supports tests which use multiple files (@garyb)
  • Portability of the core library test suite was improved (@bmjames)
  • Performance of import elaboration was improved (@garyb)
  • We now use Stack for our CI builds and release builds (#1974, @hdgarrood)
  • We now use NoImplicitPrelude and enable some global extensions (@garyb)
  • Type-safety in the source-level AST was improved (@garyb)
  • Use HSpec for the compiler tests (@garyb)
  • New Prelude names in 0.9 (@garyb)

v0.9.0 - 2016-05-22

This is pre-release software

This release is provided so that library developers can test the new compiler features.

v0.8.5 - 2016-04-21

New Features

  • Fast recompilation for single files in psc-ide-server #1712 (@kRITZCREEK, @paf31)

    The pscid project makes use of this to watch files as you work and raise errors and warnings when they occur with near instant feedback.

  • Operator aliases can now be declared for types #416 (@garyb)

    infixr 6 type Natural as ~>
  • Underscore wildcards can now be used in case and if expressions #1558 (@garyb)

    case _ of
      Something -> ...
    -- underscores can optionally be used in any part of an `if` expression
    cond = if _ then _ else _
    picker = if _ then "x" else "y"
  • Typed holes #1283 (@garyb)

    example :: forall a. Maybe a -> Unit
    example ma = ?umm
    Hole 'umm' has the inferred type
    
      Unit
    
    in value declaration example
    

    You can use any identifier name after the question mark and that will be used to label the hole in the raised error message.

Breaking changes

  • Type annotations may need parentheses in some situations that they previously did not due to the introduction of type operators. For example, x :: a == y will be now parsed as x :: (a == y) instead of (x :: a) == y.

Enhancements

  • Improved error messages for invalid FFI identifiers #2011 (@hdgarrood)
  • psc-publish now allows publishing of packages with a valid SPDX license field in bower.json #1985 (@hdgarrood)
  • Haddock markdown fix #2001 (@trofi)
  • psc-ide now creates the output folder on startup if it is missing #2030 (@kRITZCREEK)

Bug Fixes

  • Fixed an issue with incorrect suggestions when re-exporting modules #1862 (@garyb)
  • Fixed an issue with invalid redundant import warnings #1823 (@garyb)
  • Fixed an issue where DuplicateSelectiveImport would not fire when it should #2004 (@garyb)
  • Fixed the error that occurs when an invalid newtype is created that belongs to a data binding group #1895 (@garyb)
  • Fixed a case where re-exports included unintended exports #1872 (@garyb)
  • Operator aliases can now be declared for qualified data constructors #2015 (@LiamGoodacre)
  • A single hiding import will no longer raise an "unspecified imports" error #2017 (@garyb)
  • Fixed a case where cycles in modules were being detected when they do not occur #2018 (@garyb)
  • Various cases where files were not being read as UTF-8 on Windows were fixed #2027, #2031 (@garyb, @kRITZCREEK)
  • Fixed some issues in pretty printing of records #2043 (@LiamGoodacre)
  • psci now shows qualified imports correctly #2040 (@LiamGoodacre)
  • Parser errors are now returned as JSON during IDE rebuild #2042 (@paf31)

v0.8.4 - 2016-04-06

This is an interim bug fix release before 0.9.0.

Enhancements

  • Check that FFI imports match with implementations (@hdgarrood)

    This is technically a breaking change, since some existing code might fail to compile if it has missing FFI code (purescript-dom is an example), but these libraries should be fixed soon.

  • Import helper commands in psc-ide (@kRITZCREEK)

Bug Fixes

  • Disallow constraint generalization for recursive functions. (#1978, @paf31)
  • Fix #1991, instantiate polymorphic types before unification (@paf31)
  • Use UTF8 when writing to stdout and stderr (@garyb)
  • Fix for rendered constrained types needing parens. (@LiamGoodacre)
  • everythingWithScope improperly traversing binary ops (@LiamGoodacre)

Other

  • Update to use language-javascript 0.6.x (@nwolverson)

v0.8.3 - 2016-03-26

Breaking Changes

  • We have dropped support for GHC 7.8 and older (@hdgarrood)

Enhancements

  • Infer types with class constraints (@paf31)

    For example, this simple code would previously have failed with a confusing NoInstanceFound error:

    add x y = x + y

    The compiler will now infer the most general type, namely forall a. (Semiring a) => a -> a -> a.

    Note that constraints can only be inferred if they only mention type variables; inference of arbitrary types in constraints is not (yet) supported. So, for example, you would still have to write a type signature for a function which had a constraint such as (MonadEff (console :: CONSOLE | eff) m).

  • Default require path to ../ (@nwolverson)

    The previous default behavior was no require path prefix, which was confusing for some workflows. The new default is ../, which is the prefix used in purs-loader. This option will be removed completely in 0.9.

  • Expose hiding import suggestion in JSON (@nwolverson)

  • Error on missing LICENSE file or missing license field in bower.json (@faineance)

Bug Fixes

  • Fix #1916 (@bagl)
  • Fix detection of single open import (@garyb)
  • Fix true not being treated as an infallible guard (@garyb)
  • Fix pretty printer spinning (@garyb)
  • Fix Windows build script (@garyb)
  • Fix #1889, improve performance by avoiding whitespace operations on large strings (@paf31)

psc-ide

  • Fix a crash related to error messages in the case splitting command (@kRITZCREEK)
  • Escape regex characters when using the flex matcher (@kRITZCREEK)
  • Adds --help commands to the psc-ide executables (@kRITZCREEK)
  • Catches EOF exceptions thrown in acceptCommand (@kRITZCREEK)

Other

  • Switched to Trusty distribution for Travis (@garyb)
  • @kRITZCREEK and @faineance worked on refactoring the compiler.
  • The optparse-applicative dependency was updated to >= 0.12.1 (@stevejb71)
  • The bower-json dependency was bumped (@hdgarrood)
  • Better error message for psc-publish tests (@kRITZCREEK)
  • Use generic Literal in the AST (@garyb)

v0.8.2 - 2016-02-29

Breaking Changes

None

Enhancements

  • psc-ide is now distributed with the compiler! (@kRITZCREEK)

    The psc-ide-server and psc-ide-client executables are now maintained and distributed alongside the compiler. This will ensure that the externs file format used by psc-ide-server is kept in sync with changes in the compiler.

  • Source maps (@nwolverson)

    Source maps can be generated using the --source-maps flag. See the example repository for a full demonstration of source maps using Webpack.

  • Operator aliases for data constructors (@garyb)

    Aliases can now be defined for data constructors. For example:

    data List a = Nil | Cons a (List a)
    
    infixr 6 Cons as :

    Here, the : operator can be used as a function to replace the Cons constructor, and also in binders.

  • Eq and Ord deriving (@paf31)

    Eq and Ord instances can now be derived, using the derive instance syntax:

    derive instance eqList  :: (Eq  a) => Eq  (List a)
    derive instance ordList :: (Ord a) => Ord (List a)
  • Types are now inferred in psc-docs and psc-publish (@hdgarrood)

    If type annotations are missing in source files, they will be inferred by psc-docs and psc-publish before documentation generation.

  • Initial version of new syntax for operator sections (#1846, @paf31)

    Operator sections can now be written using underscores. For example:

    decrementAll :: Array Int -> Array Int
    decrementAll = map (_ - 1)

    which is equivalent to:

    decrementAll :: Array Int -> Array Int
    decrementAll = map (\x -> x - 1)

Bug Fixes

  • Allow one open import without warning (@garyb)

    Warnings for open imports were a pain point for some users after the 0.8 release. This change allows a single open import without a warning. This is still safe in the presence of dependency updates, and does not lead to ambiguity for editor plugins searching for declaration sites.

Other

  • @phadej has updated the Stack build to use the latest LTS and nightly builds.
  • @izgzhen has refactored the PSCi code to be more readable.
  • @hdgarrood has refactored the test suite.

v0.8.1 - 2016-02-29

You are recommended to use v0.8.2 instead.

v0.8.0 - 2016-01-31

A massive thanks to everyone involved in this release!

Breaking Changes

None, but there are lots of new warnings related to upcoming breaking changes in 0.9:

  • Operators as aliases will become mandatory, and regular operators (as functions) will now generate warnings.
  • Non-exhaustive functions will get a Partial constraint in 0.9, so the exhaustivity checker will now attempt to generate warnings by looking for Partial constraints in scope.
  • The qualified import syntax has been deprecated.
  • Class imports will use the new class syntax in 0.9 and the alternative syntax is deprecated.

Enhancements

  • Add native Partial constraint (@garyb)
  • Reduce backtracking in parser to hopefully improve quality of parsing error messages (@paf31)
  • Drop requirement to parenthesize single constraints in instance contexts (@garyb)
  • Case expressions can now match multiple values (@natefaubion)
  • Add operator aliases (@garyb)
  • Show identifiers correctly in ctags (@nwolverson)
  • Fix #1523, add --json-errors flag for editor integrations (@paf31)
  • Error and warning corrections are now available to editors via --json-errors (@nwolverson)
  • Check integer values are within range in codegen (@garyb)
  • Support for unicode operators (@paf31)
  • The parser now supports unicode symbols for forall and function arrows (@DavidLindbom)
  • Module Imports
    • Use class keyword for class references in imports (@garyb)
    • Type imports no longer require () (@garyb)
    • Allow import hiding with qualified imports (@garyb)
    • Naming conflicts are now resolved at the use site (@garyb)
  • Error Messages
    • Fix #1662, display extra type info in instance errors (@paf31)
    • Add information about skolem constants to type errors (@paf31)
    • Sort rows in unification errors (@paf31)
  • Warnings
    • Warn on unspecified imports (@garyb)
    • Warn when import X hiding (..) imports nothing (@garyb)
    • Warn on duplicate imports and exports (@garyb)
    • Warn about unused class imports (@garyb)

Bug Fixes

  • Renamer updates, fixes naming bug in some unlikely situations (@garyb)
  • Fix #1645, implement new indentation rules for types to avoid very wide errors (@paf31)
  • Fix "resource exhausted" issue on MacOS (@mgmeier)
  • Fix #1664, check kind before expanding wildcards. (@paf31)
  • Fix up shadowed module names in JS codegen (@garyb)
  • Fix #1185, fix #1369, add everythingWithScope traversal to correct some scoping issues. (@paf31)
  • Fix two cases where errors were missing context (@garyb)
  • Fix #1636, instantiate polytypes fully, even under constraints. (@paf31)
  • Fix missing data constructors in re-exports (@garyb)
  • Fix codegen error with instance for re-exported class (@garyb)
  • Fix #1479, encode .js files as UTF8. (@paf31)
  • Fix a bug related to redundancy checking in cases (#1853, @nicodelpiano)
  • Fix a TCO/composition inlining bug (@garyb, @hdgarrood)
  • Fix renaming for nested constructor binders (#1839, @sharkdp)
  • Fix generic deriving bug with >1 type argument (@hdgarrood)
  • Fix generate fresh binder names unless all names in case are equal (#1825, @paf31)
  • Fix external require expressions when minifying (#1794, @paf31)
  • Rename foreign argument to fix compiling issue (@anttih)
  • Allow use of bottom integer (@garyb)

Other

  • Fix #1700, remove warnings for syntactic features removed in 0.7.0 (@paf31)
  • Fix psc-publish test (@passy)
  • Relax rules for docs comments (#1820, @hdgarrood)
  • Qualified name lookup is now supported in PSCi (#974, @soupi)
  • https://github.com and [email protected] URLs are now allowed by psc-publish (@passy, @codedmart)
  • Docs are now generated for module re-exports (@hdgarrood)
  • Use friendly module name in psc-docs error (@nwolverson)
  • Distinguish between the different ProperNames (@garyb)
  • Warn about unspecified constructors in type imports (@garyb)
  • Fix warning about values missing from virtual modules (@garyb)

v0.7.6.1 - 2015-11-18

Fixes a bug in generic deriving.

See the release notes for 0.7.6.

v0.7.6 - 2015-11-18

Thanks once again to everyone involved in this release!

This release includes some updates to generic deriving which require updating to the latest version of purescript-generics.

Features

  • Field puns, fix #921 (@balajirrao)

    It is now possble to construct objects by using values in scope with the same name as the field labels. For example, the expression { foo, bar } is equivalent to { foo: foo, bar: bar }. Patterns desugar in the same way.

Enhancements

  • Modules are now parsed in parallel (@paf31)
  • Use Types.Proxy.Proxy instead of Data.Generic.Proxy. This fixes #1573 (@tfausak)
  • Update generic deriving for latest purescript-generics changes (@paf31)
  • New import warnings - unused data constructors, unused imports (@nwolverson)
  • psc-publish: only warn on dirty working tree on dry runs (@hdgarrood)
  • Add more information to psci :browse command (@soupi)
  • Add support for --require-path option to psc-bundle (@natefaubion)
  • Improved error reporting in psc-publish (@hdgarrood)
  • Reduce noise in instance declarations in documentation (@hdgarrood)

Bug Fixes

  • New approach to unification, fixing some loops in the type checker (@paf31)
  • Fix #1632, instantiate type variables in anyProxy calls in generic instances (@paf31)
  • Fix warnings for unqualified implicit imports (@nwolverson)
  • Fix #1596, don't show type checker warnings in the event of an error (@paf31)
  • Fix #1602, improvements around code generation of string literals (@michaelficarra)
  • Fix #1090, allow accessors in operator sections (@paf31)
  • Fix #1590, limit depth of pretty-printed expressions (@paf31)
  • Fix #1591, use the 'negate' in scope (@paf31)
  • Fix #1335, track scoped type variables when skolemizing (@paf31)
  • Fix #1175, check types inside where clauses inside instances (@paf31)
  • Some refactoring (@phadej)
  • Fixed some error messages (@zudov)

Deployment

  • Use base-compat to reduce the need for CPP (@phadej)
  • Write license-generator in Haskell (@phadej)
  • Add GHC 7.10.3 to CI build matrix (@phadej)

v0.7.5.3 - 2015-10-29

Bug Fixes

  • #1072, #1130, #1578, #1577, #1582

v0.7.5.2 - 2015-10-27

Fixes a build issue with GHC versions < 7.10. Functionally equivalent to v0.7.5.1.

v0.7.5.1 - 2015-10-27

Bug Fixes

  • Fix #1169, #1315, #1534, #1543, #1548, #1551, #1557, #1570
  • Fix memory leak caused by WriterT (#1297) by @paf31
  • Display hints after main error (#1563) by @paf31
  • Friendlier errors by @paf31
  • Documentation fixes by @nwolverson
  • Haddock fixes by @trofi

v0.7.5 - 2015-10-20

A big thank you to everyone who was involved in this release, from filing issues, through fixing bugs to testing patches.

The main focus areas for this release, as part of the 0.8 milestone, were error messages and performance.

Breaking Changes

None!

Enhancements

  • Pretty printing of types and expressions in errors was improved (@paf31)

  • Externs files are now saved as JSON (@paf31)

  • Support for parallel builds has been added (@paf31) Builds will now use multiple cores by default, but the number of capabilities can be modified by passing the -N option to the GHC runtime:

    psc <input files> +RTS -N8
    
  • Binders can now be given type annotations (@5outh)

    For example:

    example = do
      n :: Int <- get
      put (n + 1)

    This can be useful when disambiguating types.

  • There is a new warning for missing type signatures on top-level declarations (@paf31)

  • There are new warnings for shadowed and unused type variables (@garyb)

  • Contextual information in warnings was improved (@garyb)

  • The qualified keyword is now optional when importing modules qualified (@michaelficarra)

  • @zudov changed the behavior of PSCi on CTRL+C/D to match GHCi and other REPLs.

  • A bug in row unification was fixed (#1310, @paf31)

  • Constrained types can now be defined without a forall keyword. This is useful in some nullary type class and rank-N scenarios. (@paf31)

Bug Fixes

  • @garyb added some additional checks for transitive module exports.
  • Type synonyms are now expanded more eagerly to avoid some error cases in the type checker (@paf31)
  • Better support for multi-byte UTF-8 characters (@senju)
  • A check has been added to the exhaustivity checker to avoid exponential blowup (@paf31)
  • Empty case statements are no longer syntactically valid (@zudov)

Other

  • @aspidites fixed all compiler warnings in the core libraries.
  • @zudov and @phadej have made improvements to the Stack distribution of the compiler, and the Stackage builds.
  • @garyb has added a warning for operators in type classes, since they will be disallowed before 0.8.

v0.7.4.1 - 2015-08-26

This patch release fixes two bugs related to the new instance resolution algorithm and overlapping instances:

  • psci would not work due to overlaps in the PSCI.Support module
  • free would not build due to its dependency on inject

The solution for now is to make overlapping instances into a warning (instead of an error) at the site of their use.

Later we might revisit this decision and allow the user to express classes like Inject which are necessarily overlapping.

v0.7.4.0 - 2015-08-25

Breaking Changes

  • The type class instance search algorithm has changed. The search will now eagerly match an instance for each subgoal based on the instance head, or fail. This makes certain instances in previous versions of purescript-transformers invalid, so users of this release should upgrade to the latest transformers.
  • A module must be imported to be re-exported.

Enhancements

  • RedefinedModule errors now include position info #1024 (@garyb)
  • Multiple imports of the same module are now resolved correctly, allowing for combinations of qualified and unqualified importing #817 #1112 (@garyb)
  • Errors for unresolvable imports and exports have been clarified #1232 (@garyb)
  • A warning is emitted when importing Type(..) when Type is a synonym or has no constructors. #1391 (@garyb)
  • Superclass constraints can now be relied upon when resolving instances #421 (@paf31)
  • A serious performance regression was partially addressed, memory usage should now be drastically reduced #1297 (@garyb)
  • Module re-export handling has been much improved. If a module is partially imported, only the specifically imported members are re-exported. Qualified modules can also be re-exported. #291 #1244 (@garyb)
  • Parser error messages are now formatted in a manner more consistent with other errors #1098 (@epost)
  • Using -ffi to specify JavaScript FFI files is now optional, files with a .js extension will be detected as FFI files automatically when encountered. #1268 (@mjgpy3)

Bug fixes

  • Fixed an error when attempting to derive for Void #1380 (@nicodelpiano)
  • "The impossible happened in desugarDo" should no longer occur #386 (@paf31)

Other

@zudov, @phadej and @erdeszt made more updates and improvements to the CI build.

v0.7.3 - 2015-08-13

Major Features

  • @gbaz has implemented generic deriving. This allows instances for the Generic class in the purescript-generics package to be derived by the compiler.

    A Generic instance can be derived as follows:

    data Example = Foo String | Bar Int | Baz Boolean
    
    derive instance genericExample :: Generic Example

    purescript-generics provides examples of usage, such as gShow, gEq and gCompare, for printing, equality tests and comparison respectively.

    See #1138.

  • @garyb has implemented a test for orphan instances which will now cause the build to fail with an error. See #1247

Enhancements

  • @mjgpy3 has added a warning when an input glob does not match any files.

Bug Fixes

  • The psc: <<loop>> has been fixed. This was due to a bug in the error pretty printer. (@paf31)
  • An issue with unicode characters in string literals was fixed by @michaelficarra.
  • Compiler errors are now pretty printed in psc-publish (@paf31)
  • Modules are no longer linted if they are not being rebuilt (@paf31)
  • FFI bindings are now reloaded when changed, in PSCi (@paf31)

Other

  • @phadej and @zudov have improved our CI process, so that PureScript now compiles against three versions of GHC and two LTS Stackage releases, as well as the nightly stackage releases.
  • @phadej and @lukerandall have worked on supporting PureScript in Stackage.

v0.7.2.1 - 2015-08-12

Functionally equivalent to v0.7.2. This release fixes a version incompatibility with Stackage.

v0.7.2 - 2015-08-03

Bug fixes

  • Fixed haddock for the Language.PureScript.Bundle module #1262 (@wuzzeb)
  • Some erroneous error positions were fixed for kind and missing instance errors #1086 (@garyb)
  • The number of warnings printed for exhaustivity checks was limited to 5 #1281 (@nicodelpiano)
  • Home directory is checked for .psci file after the current working directory #883 (@mjgpy3)
  • Line numbers now show for shadowed name warnings #1165 (@nicodelpiano)
  • Cabal file was fixed for Nix packaging #1302 (@MasseGuillaume)
  • Kind query for types defined in psci now works #1235 (@mjgpy3)
  • Boolean operations are now being inlined again #1312 (@garyb)
  • Int operations are now being inlined again #1330 (@garyb)
  • "Writing..." and "Compiling..." messages are no-longer printed in psci #1276 (@paf31)

Enhancements

  • Exhaustivity checker was extended to report errors about redundant cases #1289 (@nicodelpiano)
  • Improved triggering of suggestion for errors about using (<<<) instead of (.) #1284 (@mjgpy3)
  • Only show the module name rather than the filename for pattern errors #1296 (@nicodelpiano)
  • Error reporting in psc-bundle was improved #1307 (@hdgarrood)
  • psc-publish code is now part of the library module #1304 (@hdgarrood)
  • psc-publish now has --version and --help options #1300 (@garyb)
  • psc-publish now has a --dry-run option for checking whether the module can be published #1308 (@hdgarrood)
  • psc-publish now requires a clean working tree #1306 (@hdgarrood)
  • psc-publish can now find bower on Windows machines #1317 (@hdgarrood)
  • psc-publish now uses OS-specific path delimiters to fix another bug on Windows #1326 (@hdgarrood)
  • Error list heading was made emacs-friendly #1327 (@epost)

v0.7.1 - 2015-07-13

Minor fixes after 0.7.0:

  • @hdgarrood has worked on improvements to psc-publish to support the new Pursuit website.
  • @mjgpy3 has improved warning messages
  • @wuzzeb has improved the pretty printers
  • @hdgarrood has added CI builds for GHC 7.10 and 7.6

Enhancements

  • @nicodelpiano has added exhaustivity checking as a new warning type. Incomplete pattern matches will now generate warnings like this:

    Warning in module Data.Either.Unsafe:
      Warning in value declaration fromRight:
      Warning at src/Data/Either/Unsafe.purs line 14, column 1 - line 15, column 1:
    Pattern could not be determined to cover all cases.
        The definition has the following uncovered cases:
          (Data.Either.Left _)
      See https://github.com/purescript/purescript/wiki/Error-Code-NotExhaustivePattern for more information, or to contribute content related to this error.
    

v0.7.0 - 2015-06-30

Introduction

This release ("MELTDOWN") aims to handle as many planned breaking changes as possible, to ease the upgrade path before 1.0. It is necessary to upgrade almost all PureScript code to compile with this release.

The latest versions of the core libraries have all been updated to compile with this release. Older versions of the core libraries will not work with this release, and the latest versions of libraries will not build with older compiler releases.

Detailed instructions for those who need to migrate their code can be found on the wiki.

As usual, many thanks go to all of the contributors who helped with this release!

Breaking changes

  • The psc executable has been replaced with psc-make, which has been renamed to psc (in an effort to standardize on CommonJS module output). Features which were previously only available in old psc (dead code elimination, bundling code for the browser) are now handled by the new executable psc-bundle, which works with the output of the new psc (for faster, incremental builds).
  • There are now Int and Number literals. To disambiguate the two, integer Number values must now be written with a decimal place (3.0 rather than 3).
  • The Prelude module is no longer imported automatically, and must be imported the same way as any other module.
  • No modules are included with the compiler now, they have been broken out into their own libraries:
  • Debug.Trace has been renamed to Control.Monad.Eff.Console, and trace has been renamed to log.
  • [] syntax for array types has been removed. It is still possible to use [] array literals however.
    • [] should now be written as Array, and [a] as Array a.
  • Cons patterns for arrays have been removed.
  • Declaring operators in classes will now produce a warning. Changes will be coming to operators in PureScript 0.8, and moving to named members in classes with operators as aliases (e.g. (<$>) = map) should make the transition easier in the future.
  • JavaScript for the FFI can no longer be provided inline.
    • Values must instead be provided in a separate .js file, and passed to the compiler with the -ffi flag.
    • Values should be provided in the form exports.foo = ..., similar to a CommonJS module
    • The file should have a comment // module X.Y.Z where X.Y.Z is the name of the module the JS values are for.
    • See here for an example

Enhancements

  • Module exports (@andyarvanitis). Currently, only full module exports are supported, but imported modules can be re-exported using the following syntax: purescript module M1 (module M2) where import M2
  • Documentation improvements (@hdgarrood):
    • psc-docs can now generate multiple output files, allowing documentation to be collected into functional groups.
    • A new tool psc-publish has been added, which generates module documentation in a JSON format required by Pursuit 2 (coming soon)
  • @hdgarrood has worked on improving the quality of code completion inside psci, and generally tidying up and refactoring that code.
  • @puffnfresh has worked on dramatically increasing the performance of incremental builds, with improvements up to 10x compared to the previous release.
  • The new --require-path option allows the syntax of module imports in generated CommonJS modules to be customized (@garyb).
  • @etrepum has added support for building with Stack.
  • PSCi now supports computations in the Eff monad. (@paf31)
  • The compiler now emits warnings in the following cases:
    • Operator name used in type class definition (@garyb)
    • Type wildcard used (@nicodelpiano)
    • Shadowed variable name (@paf31)
  • @balajirrao has improved the appearance of unknown and rigid types appearing in error messages.
  • @puffnfresh has added position information to pattern match errors.
  • @puffnfresh has added some new optimizations (inlining <<< and $)

Bug Fixes

  • psc, psc-docs and psc-bundle now support file globs as command-line arguments, fixing a bug related to the command length on Windows machines (@paf31)
  • @wuzzeb has fixed some issues in the pretty printer.
  • @mjgpy3 has improved error messages related to incorrect pattern matches on data constructors.

Tools

  • Pulp has been updated:
    • The new psc and psc-bundle binaries are supported
    • FFI modules are now identified and compiled based on a convention
    • pulp docs now generates individual Markdown files for each source module
  • gulp-purescript has been updated:
    • The new psc and psc-bundle binaries are supported
    • FFI modules are now supported

Libraries

  • The following libraries have been moved into the core library set:
    • purescript-lists - Strict and lazy linked list data structures
    • purescript-assert - Low level assertion library for tests
    • purescript-parallel - An applicative functor for parallel composition of asynchronous computations.
    • purescript-arrows - Arrow type classes and standard instances.
    • purescript-tailrec - A type class for stack-safe monadic tail recursion.
  • The requirements for libraries in the purescript-contrib organization have been tightened, to try to ensure that libraries stay maintained.

v0.7.0-rc.1 - 2015-06-07

Important note

This release should be used with the latest versions of the core libraries, which are also tagged as -rc.1.

Breaking changes

  • There are now Int and Number literals. To disambiguate the two, integer Number values must now be written with a decimal place (3.0 rather than 3).
  • The Prelude module is no longer imported automatically, and must be imported the same way as any other module.
  • No modules are included with the compiler now, they have been broken out into their own libraries:
  • [] syntax for array types has been removed. It is still possible to use [] array literals however.
    • [] should now be written as Array, and [a] as Array a.
  • Cons patterns for arrays have been removed.
  • Declaring operators in classes will now produce a warning. Changes will be coming to operators in PureScript 0.8, and moving to named members in classes with operators as aliases (e.g. (<$>) = map) should make the transition easier in the future.
  • JavaScript for the FFI can no longer be provided inline.
    • Values must instead be provided in a separate .js file, and passed to the compiler with the -ffi flag.
    • Values should be provided in the form exports.foo = ..., similar to a CommonJS module
    • The file should have a coment // module X.Y.Z where X.Y.Z is the name of the module the JS values are for.
    • See here for an example

Full release notes coming soon

v0.6.9.5 - 2015-04-25

This release contains two patches:

  • Case statements were generating incorrect function name arguments #1008 (@paf31)
  • Comments and verbose error flags were mixed up #991 (@garyb)

v0.6.9.3 - 2015-03-18

Breaking Changes

  • refEq and refIneq are no longer exported from the Prelude.

Bug Fixes

  • Instances can now be defined before the corresponding class declaration (@paf31)
  • A bug related to imports in psci was fixed. (@paf31)
  • A typechecker bug related to type class dictionaries was fixed. (@garyb)
  • A bug related to operator precedence in codegen was fixed. (@garyb)

Enhancements

  • psci now supports long-form directives (@mrhania)
  • Syntax for imports and other declaration types in psci was improved. (@hdgarrood)
  • Markdown comments can now be included at the module level (@joneshf)
  • Error messages are now represented internally as an algebraic data type, and pretty printing has been improved by using the boxes library. Errors now link to the wiki. (@paf31)
  • psc-docs can now generate tags files for Vim and Emacs (@jacereda)
  • psci now supports a --node-opts flag for passing options to the Node executable. (@MichaelXavier)
  • Code gen now preserves names of more function arguments in case statements (@andyarvanitis)
  • There is now a Semigroup instance for Ordering (@pseudonom)

Documentation

  • The Prelude now has Markdown documentation (various contributors - thank you!)
  • The Pursuit website has been updated with new versions of libraries, including Markdown documentation (@hdgarrood)

Libraries

  • The following libraries are now core libraries:
    • purescript-tailrec - A type class for monadic tail recursion
    • purescript-monad-eff - A type class for monads supporting native effects
    • purescript-integers - Integer numeric type
    • purescript-invariant - Invariant functors
    • purescript-parallel - An applicative functor for parallel composition of asynchronous computations

Other

  • There is an experimental C++11 backend for PureScript called pure11.

v0.6.8 - 2015-02-21

Breaking Changes

  • The Num type class has been refined to allow more interesting instances. The Semiring, ModuloSemiring, Ring and DivisionRing classes have been introduced. Most code should continue to compile, since Number was one of only a handful of instances, but library developers will need to break up their Num instances.

Enhancements

  • @garyb has improved the readability of psc-docs output.

Notes

  • All uses of the deprecated ErrorT have been replaced with ExceptT and the transformers and mtl dependencies bumped accordingly.

v0.6.7.1 - 2015-02-14

Bug Fixes

  • A fix for a bug in the type class instance resolution code (#870, @paf31)

v0.6.7 - 2015-02-12

Enhancements

Scoped Type Variables

(#347, @paf31)

This feature allows type variables which are bound by a forall keyword to be used inside type annotations in the body of the function. For example, suppose we want to define a map function on a List type:

data List a = Nil | Cons a (List a)

map :: forall a b. (a -> b) -> List a -> List b
map f = go
  where
  go Nil = Nil
  go (Cons x xs) = Cons (f x) (map f xs)

To give a type to go, we could previously use type wildcards:

go :: List _ -> List _

Now, we can refer to the types a and b inside the type of go, giving a more precise type:

go :: List a -> List b

Rows In Instance Contexts

(@paf31, @apsk)

This feature allows rows to appear on the left of a => in a type signature. For example, given a MonadEff class:

class MonadEff eff m where
  liftEff :: forall a. Eff eff a -> m a

we can now write the following function which works in any Monad supporting Trace actions:

logging :: forall m a eff. (Monad m, MonadEff (trace :: Trace | eff) m) => String -> m a -> m a
logging s action = do
  liftEff $ trace $ "Starting: " <> s
  a <- action
  liftEff $ trace $ "Done: " <> s
  return a

Improved let bindings in psci

(#782, @paf31)

Any declaration can now be used inside a let binding in psci. For example, we can define data types or foreign imports:

> let data Foo = Foo | Bar | Baz

> let foreign import foo :: Foo -> String

The general form of a let statement in psci now contains one or more declarations of any type, and these declarations simply get added to the current module.

As a bonus, polymorphic functions bound using let now work at multiple type instantiations in psci:

> let f x = x

> if f true then f "true" else f "False"
"true"

Markdown Support in psc-docs

(#802, @paf31)

Markdown can now be used for documentation purposes by using pipe characters to align content. For example:

-- | Create a copy of the array without its first element.
-- |
-- | Running time: `O(n)`, where `n` is the length of the array.
-- |
-- | This function is partial. Specifically, `tail []` is undefined.
tail :: forall a. [a] -> [a]

psc-docs will insert this markdown content verbatim into your generated documentation.

Bug Fixes

  • Modules are rebuilt before a command is executed in psci, to avoid situations where compiled code becomes out-of-date (@paf31)
  • @ is a valid operator name again (#815, @paf31)
  • Reserved module names are now properly escaped (@garyb)

v0.6.6 - 2015-02-09

Breaking Changes

  • The syntax of record getters was changed to _.prop (@garyb)

Enhancements

  • The record part of a record updater can now be made into a wildcard, e.g. _ { foo = 1 } (@garyb)

  • Extended infix expressions are now supported, (@paf31) e.g.

    [1, 2, 3] `zipWith (+)` [4, 5, 6]
    

Bug Fixes

  • Newline issues were fixed in executables (@michaelficarra)

v0.6.5 - 2015-02-08

Enhancements

  • Lightweight record constructors are now supported (@garyb):

    person :: Maybe String -> Maybe Number -> Maybe Address -> Maybe Person
    person = { name: _, age: _, location: _ } <$> name <*> age <*> location
  • Field accessor sections are now supported (@garyb):

    getPersonName :: Maybe String
    getPersonName = (.name) <$> getPersonInfo
  • Syntactic sugar has been introduced for object update functions:

    updateName :: Person -> String -> Person
    updateName person = person { name = _ }
  • Operator sections are now supported (@garyb)

Bug Fixes

  • Some command line options were fixed in psc-make (@paulyoung)
  • Some module import errors were fixed (@garyb)
  • A typechecker bug related to row synonyms was fixed (#795, @paf31)

v0.6.4.1 - 2015-02-03

Various small bug fixes.

v0.6.4 - 2015-01-23

  • Fix some precedence issues in the code generator.
  • Tighten the bounds on utf8-string.
  • Fixed a bug in the typechecker.

v0.6.3 - 2015-01-08

Breaking Changes

Bug Fixes

  • Case statement at end of Eff block not being executed. (#759, @paf31)
  • A bug related to dead code elimination was fixed. (@garyb)
  • Wildcards can now appear in row endings. (@RossMeikleham)

Enhancements

  • There is a new "core functional representation", which will enable certain optimizations, and new features such as rewrite rules. (#710, @garyb)
  • Record pattern matches now allow field names to be separated from binders using : instead of =, to match record construction (#760, @leighman)
  • Some improvements needed for the Pursuit tool (@hdgarrood)
  • The lexer was separated from the parser, and now supports explicit comments in the AST. Documentation generated by psc-docs now contains any inline comments which precede the corresponding declaration, and generated code preserves the same comments. (@paf31)
  • PureScript now builds on GHC 7.6.* again. (@dylex)
  • Proper names can now contain underscores. (@dylex)
  • Several auto-completion improvements and fixes in PSCI. (@vkorablin)

Libraries

  • The Prelude now contains a pureST function to run ST computations in a pure context. (@KMahoney)

Tools

  • The Pursuit tool now runs on the community server, and integrates with Bower. Libraries can be added by submitting a pull request. (@hdgarrood)

v0.6.2 - 2014-11-28

Breaking Changes

  • Command line options with multiplicity 1 now require an equals symbol, e.g.

    psc --main=Main --browser-namespace=PS
    

    The Grunt and Gulp plugins already support this format.

Enhancements

  • Use optparse-applicative instead of cmdtheline (@anthoq88)

Libraries

  • Move STArray out of Prelude. (@paf31)

v0.6.1.2 - 2014-11-24

v0.6.1.1 - 2014-11-19

Breaking Changes

  • The pipe symbol is now a reserved operator.
  • The operators in the Bits type class have been renamed.

Enhancements

  • Fix build on GHC 7.6.* (@dylex)
  • Relax indentation requirements (@paf31)

v0.6.1 - 2014-11-18

Breaking Changes

  • The body of a guarded expression must now be indented past the guard. For example, this is valid:
positive n | n > 0 = true
positive _ = false

but this is not:

positive n | n > 0
  = true
positive _ = false

New Features

  • Type wildcards are now supported (#287, @paf31)

Enhancements

  • Allow unquoted keywords as key names in record literals (#606, @michaelficarra)
  • Import instances when referencing qualified values (#667, @garyb)
  • Multiple guard clauses are now supported (#294, @paf31)
  • Type check let declarations immediately in psci (#615, @garyb)

v0.6.0.2 - 2014-11-09

  • Prevent psci and psc-make from rebuilding everything on every build #692

v0.6.0 - 2014-11-06

For more information on PureScript, see the purescript.org website.

Breaking Changes

  • The Alternative type class hierarchy was refactored. See here.
  • --runtime-type-checks has been removed. The recommended approach is to use purescript-foreign. (@garyb)
  • The Unit type is now used in the Prelude and core libraries to represent values containing no data. (@garyb)
  • The Prelude is no longer distributed as a separate file, but is embedded in the compiler executables. (@paf31)
  • docgen is now called psc-docs.

New Features

  • Newtypes are now supported using the newtype keyword. The runtime representation of a newtype is identical to that of the contained type. (@garyb)
  • Multiline string literals are now supported via triple-quote syntax, making FFI declarations much neater. (@phadej)
  • Kind signatures on types and type constructor arguments are now supported. (@paf31)

Enhancements

  • The runFnN and mkFnN families of functions are now inlined by the optimizer, making interop with JavaScript functions of multiple arguments much simpler. (@paf31)
  • Tail call optimization has been improved for functions using case expressions. (@paf31)
  • Saturated calls to data constructors are now optimized. (@garyb)
  • A new Renamer module now renames identifiers which shadow other names in scope, which greatly simplies code generation. (@garyb)
  • psci now provides the following new options:
    • :b to browse a module (@ardumont)
    • :s to show current imports or modules (@ardumont)
    • :k to find the kind of a type constructor (@5outh)
  • The approach to checking whether a name is initialized in the generated JavaScript was simplified (@paf31)
  • The dependency on the PureScript_paths module has been removed, which makes distribution via binaries simpler. (@paf31)
  • Nested if blocks now get optimized. (@garyb)
  • Generated code for type class dictionaries was simplified. (@garyb, @dylex)
  • The code generator now inserts the version of psc into the file as a comment. (@co-dh)
  • () is now valid syntax, referring to the empty row. (@paf31)
  • The type checker will now display multiple errors for type errors in the same binding group. (@paf31)
  • Imports can now specify hidden names using import ... hiding ( ... ) (@andreypopp)

Bug Fixes

  • Binding group errors in type class members are now caught at compile time. (@dylex)
  • Some errors related to type checking rows with duplicate labels were fixed. (@paf31)
  • Some issues with the calculation of binding groups were fixed. (@paf31)
  • Error messages for invalid case declarations are now generated. (@natefaubion)
  • Some issues related to module exports were fixed. (@garyb)
  • psci now checks imports for validity. (@Bogdanp)

Libraries

  • The Alternative type class hierarchy was refactored (@joneshf, @garyb)
  • The exceptions library no longer supports throwing exceptions of any type.
  • The following libraries have been moved to the core PureScript organisation: (@garyb)
    • purescript-transformers
    • purescript-free
    • purescript-const
    • purescript-identity
    • purescript-lazy
    • purescript-distributive
    • purescript-bifunctors
    • purescript-contravariant
    • purescript-profunctors
    • purescript-maps

Documentation

v0.5.7.1 - 2014-10-30

v0.5.7 - 2014-10-29

0.5.6.1 - 2014-10-06

0.5.6 - 2014-10-06

v0.5.6.3 - 2014-10-06

0.5.6.2 - 2014-09-22

v0.5.5 - 2014-09-02

v0.5.4 - 2014-08-04

This incremental release is provided to provide bug fixes and features required to compile the latest core libraries.

v0.5.0 - 2014-04-27

Breaking Changes

  • Support for blocks has been removed. (paf31)

  • Type class instances must now be named (paf31)

    instance showNumber :: Show Number where
      ...
    
  • Prelude modules now follow a naming scheme similar to haskell (e.g. Data.Maybe, Control.Monad) (garyb)

  • Many modules that were previously part of the Prelude have been split into individual libraries, now distributed via Bower (garyb)

  • Multiple modules with the same name are now disallowed rather than merged (garyb)

  • The Prelude module is now imported automatically. Conflicts can be avoided by using qualified imports or an explicit import list. (garyb, paf31)

  • Overlapping instances are no longer allowed. The Prelude and core libraries have been updated accordingly. (paf31)

  • Functor, Applicative, Monad are now part of a class heirarchy that include Apply and Bind. return is now an alias for pure. (joneshf, paf31, garyb)

  • Semigroupoid is now a superclass of Category (garyb)

  • (:) is now part of Prelude (garyb)

  • (!!) has been renamed to Prelude.Unsafe.unsafeIndex and a safe version has been added to Data.Array (garyb)

New Features

  • Multi parameter typeclasses (paf31)
  • Superclasses (puffnfresh, paf31)
  • FlexibleInstances and FlexibleContexts (paf31)
  • Let bindings are now supported. The let keyword can introduce several local (possibly mutually recursive) bindings, along with optional type signatures. (paf31)
  • where clauses are now supported in value declarations, with the same rules as let bindings (garyb)
  • Empty data declarations and empty type classes are now supported (paf31)
  • A new command line option --codegen controls which modules will have Javascript and externs generated (paf31)
  • psc-make now generates CommonJS-compatible modules, which can be used with require() in node. psc still generates modules for use in the browser. (paf31, garyb)

Enhancements

  • Pretty printing for row types was improved (garyb)
  • Module names can now contain . (garyb)
  • New optimizer rules have been added for code in the ST monad, to reproduce the functionality of the blocks feature, which has been removed (paf31)
  • Pattern binders are now usable in lambda expression arguments (paf31)
  • PSCI now has a :t command for checking the type of a value (paf31)
  • Array pretty printing via show has been improved (joneshf)
  • PSCI completions are sorted (joneshf)
  • PSCI now has help commands (joneshf)
  • PSCI history is in XDG config (joneshf)
  • PSCI allows loading of modules from ~ paths (joneshf)
  • PSCI can accept a list of modules to load on start from the command line (paf31)
  • PSCI can now be configured using a .psci file in the current directory. If such a file exists, it should contain a list of commands to run on startup. (paf31)
  • Type class instances are now named, to enable easier interop with Javascript (paf31)
  • Class names no longer need to be qualified in instance declarations (garyb)
  • Module exports can now be specified explicitly (garyb)
  • Let bindings can now define functions with binders (paf31)
  • Case statements and functions which do not pattern match on their arguments now generate smaller code (paf31)
  • Imported type class instances are now exported (paf31)
  • Some error messages were improved (paf31)
  • Qualfied module imports are now supported as import qualified M as X (garyb)
  • The escape check was removed, since it was too restrictive (paf31)
  • The binary operator reordering step was greatly simplified (paf31) The Object type constructor can now be referenced explicitly as Prim.Object (with kind # * -> *) (paf31)
  • Optimizations are now enabled by default and can be disabled with the --no-tco and --no-magic-do flags (garyb)
  • Unary minus and signed numeric literals are now supported again (paf31, garyb)
  • Type errors have been simplified, the full trace can be enabled with --verbose-errors or -v (paf31)
  • Error messages now display source positions (paf31, garyb)
  • The type classes implementation and code generation was greatly simplified (paf31)
  • Object properties and row labels can now be accessed with arbitrary string names by using string literals (paf31)
  • (++) is now an alias for the Semigroup operator (<>) (paf31)
  • Error messages for classes with undefined or missing members have been improved (garyb)
  • The SYB dependency was removed, and traversals rewritten by hand, for a large performance increase (paf31)

Bug Fixes

  • The subsumes relation has been fixed for object types (paf31)
  • sort no longer mutates arrays (joneshf)
  • PSCI now evaluates expressions (joneshf)
  • Overlapping variables in typeclass instances are rejected (paf31)
  • A bug in the optimizer related to inlining was fixed (paf31)
  • A type checker bug related to array literals was fixed (paf31)
  • Externs files (--externs) are now working again (paf31)
  • Precedence of backticked infix functions have been corrected (paf31)
  • A bug which allowed some incorrect type class instances to pass the type checker was fixed (paf31)
  • Type synonyms are no longer restricted to kind * (paf31)
  • Negative number literals have been restored (garyb)
  • If a type defined in a module appears in an exported declaration it must also be exported from the module (garyb)
  • Error messages for unresolvable types or values include the declaration name again (garyb)
  • Characters in string literals are now properly escaped (garyb)
  • A module containing a single orphan type declaration and no other declarations now fails to compile (garyb)
  • An error involving ordering of type class instances was fixed (garyb, paf31)
  • Externs files no longer include fixity declarations for members that were removed as dead code (garyb)
  • A bug which prevented sequence $ [Just 1] from typechecking was fixed (paf31)

Libraries

Plugins

  • The grunt-purescript plugin has been updated to provide support for new command line options.
  • There is a new gulp-purescript plugin available for compiling with Gulp.

Documentation

  • There is a new hierarchy executable which will generate .dot diagrams based on the type class hierarchy of a module. The Prelude docs have been updated to include such a type class diagram. (joneshf)