glasgow_exts.rst 599 KB
Newer Older
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33
.. index::
   single: language, GHC extensions

As with all known Haskell systems, GHC implements some extensions to the
standard Haskell language. They can all be enabled or disabled by command line
flags or language pragmas. By default GHC understands the most recent Haskell
version it supports, plus a handful of extensions.

Some of the Glasgow extensions serve to give you access to the
underlying facilities with which we implement Haskell. Thus, you can get
at the Raw Iron, if you are willing to write some non-portable code at a
more primitive level. You need not be “stuck” on performance because of
the implementation costs of Haskell's "high-level" features—you can
always code "under" them. In an extreme case, you can write all your
time-critical code in C, and then just glue it together with Haskell!

Before you get too carried away working at the lowest level (e.g.,
sloshing ``MutableByteArray#``\ s around your program), you may wish to
check if there are libraries that provide a "Haskellised veneer" over
the features you want. The separate
`libraries documentation <../libraries/index.html>`__ describes all the
libraries that come with GHC.

.. _options-language:

Language options
================

.. index::
   single: language; option
   single: options; language
   single: extensions; options controlling

34
The language extensions control what variation of the language are
35 36 37 38 39 40 41 42 43 44 45 46
permitted.

Language options can be controlled in two ways:

-  Every language option can switched on by a command-line flag
   "``-X...``" (e.g. ``-XTemplateHaskell``), and switched off by the
   flag "``-XNo...``"; (e.g. ``-XNoTemplateHaskell``).

-  Language options recognised by Cabal can also be enabled using the
   ``LANGUAGE`` pragma, thus ``{-# LANGUAGE TemplateHaskell #-}`` (see
   :ref:`language-pragma`).

47 48 49 50
GHC supports these language options:

.. extension-print::
    :type: table
51

52 53
Although not recommended, the deprecated :ghc-flag:`-fglasgow-exts` flag enables
a large swath of the extensions supported by GHC at once.
54

55
.. ghc-flag:: -fglasgow-exts
56 57 58 59
    :shortdesc: Deprecated. Enable most language extensions;
        see :ref:`options-language` for exactly which ones.
    :type: dynamic
    :reverse: -fno-glasgow-exts
60
    :category: misc
61

62 63
    The flag ``-fglasgow-exts`` is equivalent to enabling the following extensions:

64
    .. include:: what_glasgow_exts_does.rst
65 66 67 68

    Enabling these options is the *only* effect of ``-fglasgow-exts``. We are trying
    to move away from this portmanteau flag, and towards enabling features
    individually.
69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84

.. _primitives:

Unboxed types and primitive operations
======================================

GHC is built on a raft of primitive data types and operations;
"primitive" in the sense that they cannot be defined in Haskell itself.
While you really can use this stuff to write fast code, we generally
find it a lot less painful, and more satisfying in the long run, to use
higher-level language features and libraries. With any luck, the code
you write will be optimised to the efficient unboxed version in any
case. And if it isn't, we'd like to know about it.

All these primitive data types and operations are exported by the
library ``GHC.Prim``, for which there is
85
:ghc-prim-ref:`detailed online documentation <GHC.Prim.>`. (This
86 87 88 89 90
documentation is generated from the file ``compiler/prelude/primops.txt.pp``.)

If you want to mention any of the primitive data types or operations in
your program, you must first import ``GHC.Prim`` to bring them into
scope. Many of them have names ending in ``#``, and to mention such names
91
you need the :extension:`MagicHash` extension.
92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116

The primops make extensive use of `unboxed types <#glasgow-unboxed>`__
and `unboxed tuples <#unboxed-tuples>`__, which we briefly summarise
here.

.. _glasgow-unboxed:

Unboxed types
-------------

Most types in GHC are boxed, which means that values of that type are
represented by a pointer to a heap object. The representation of a
Haskell ``Int``, for example, is a two-word heap object. An unboxed
type, however, is represented by the value itself, no pointers or heap
allocation are involved.

Unboxed types correspond to the “raw machine” types you would use in C:
``Int#`` (long int), ``Double#`` (double), ``Addr#`` (void \*), etc. The
*primitive operations* (PrimOps) on these types are what you might
expect; e.g., ``(+#)`` is addition on ``Int#``\ s, and is the
machine-addition that we all know and love—usually one instruction.

Primitive (unboxed) types cannot be defined in Haskell, and are
therefore built into the language and compiler. Primitive types are
always unlifted; that is, a value of a primitive type cannot be bottom.
117 118 119
(Note: a "boxed" type means that a value is represented by a pointer to a heap
object; a "lifted" type means that terms of that type may be bottom. See
the next paragraph for an example.)
120 121 122 123 124 125 126 127
We use the convention (but it is only a convention) that primitive
types, values, and operations have a ``#`` suffix (see
:ref:`magic-hash`). For some primitive types we have special syntax for
literals, also described in the `same section <#magic-hash>`__.

Primitive values are often represented by a simple bit-pattern, such as
``Int#``, ``Float#``, ``Double#``. But this is not necessarily the case:
a primitive value might be represented by a pointer to a heap-allocated
128 129
object. Examples include ``Array#``, the type of primitive arrays. Thus,
``Array#`` is an unlifted, boxed type. A
130 131 132 133
primitive array is heap-allocated because it is too big a value to fit
in a register, and would be too expensive to copy around; in a sense, it
is accidental that it is represented by a pointer. If a pointer
represents a primitive value, then it really does point to that value:
134
no unevaluated thunks, no indirections. Nothing can be at the other end
135 136 137 138
of the pointer than the primitive value. A numerically-intensive program
using unboxed types can go a *lot* faster than its “standard”
counterpart—we saw a threefold speedup on one example.

139 140
Unboxed type kinds
------------------
141

142
Because unboxed types are represented without the use of pointers, we
143 144
cannot store them in use a polymorphic datatype at an unboxed type.
For example, the ``Just`` node
145 146 147 148 149 150
of ``Just 42#`` would have to be different from the ``Just`` node of
``Just 42``; the former stores an integer directly, while the latter
stores a pointer. GHC currently does not support this variety of ``Just``
nodes (nor for any other datatype). Accordingly, the *kind* of an unboxed
type is different from the kind of a boxed type.

151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169
The Haskell Report describes that ``*`` (spelled ``Type`` and imported from
``Data.Kind`` in the GHC dialect of Haskell) is the kind of ordinary datatypes,
such as ``Int``. Furthermore, type constructors can have kinds with arrows; for
example, ``Maybe`` has kind ``Type -> Type``. Unboxed types have a kind that
specifies their runtime representation. For example, the type ``Int#`` has kind
``TYPE 'IntRep`` and ``Double#`` has kind ``TYPE 'DoubleRep``. These kinds say
that the runtime representation of an ``Int#`` is a machine integer, and the
runtime representation of a ``Double#`` is a machine double-precision floating
point. In contrast, the kind ``Type`` is actually just a synonym for ``TYPE
'LiftedRep``. More details of the ``TYPE`` mechanisms appear in the `section
on runtime representation polymorphism <#runtime-rep>`__.

Given that ``Int#``'s kind is not ``Type``, it then it follows that ``Maybe
Int#`` is disallowed. Similarly, because type variables tend to be of kind
``Type`` (for example, in ``(.) :: (b -> c) -> (a -> b) -> a -> c``, all the
type variables have kind ``Type``), polymorphism tends not to work over
primitive types. Stepping back, this makes some sense, because a polymorphic
function needs to manipulate the pointers to its data, and most primitive types
are unboxed.
170 171

There are some restrictions on the use of primitive types:
172 173 174 175 176 177 178 179 180

-  You cannot define a newtype whose representation type (the argument
   type of the data constructor) is an unboxed type. Thus, this is
   illegal:

   ::

         newtype A = MkA Int#

181
   However, this restriction can be relaxed by enabling
182
   :extension:`UnliftedNewtypes`.  The `section on unlifted newtypes
183 184
   <#unlifted-newtypes>`__ details the behavior of such types.

185 186 187 188 189 190 191
-  You cannot bind a variable with an unboxed type in a *top-level*
   binding.

-  You cannot bind a variable with an unboxed type in a *recursive*
   binding.

-  You may bind unboxed variables in a (non-recursive, non-top-level)
Richard Eisenberg's avatar
Richard Eisenberg committed
192 193 194
   pattern binding, but you must make any such pattern-match strict.
   (Failing to do so emits a warning :ghc-flag:`-Wunbanged-strict-patterns`.)
   For example, rather than:
195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216

   ::

         data Foo = Foo Int Int#

         f x = let (Foo a b, w) = ..rhs.. in ..body..

   you must write:

   ::

         data Foo = Foo Int Int#

         f x = let !(Foo a b, w) = ..rhs.. in ..body..

   since ``b`` has type ``Int#``.

.. _unboxed-tuples:

Unboxed tuples
--------------

217 218
.. extension:: UnboxedTuples
    :shortdesc: Enable the use of unboxed tuple syntax.
219 220

    :since: 6.8.1
221

222 223

Unboxed tuples aren't really exported by ``GHC.Exts``; they are a
224
syntactic extension (:extension:`UnboxedTuples`). An
225
unboxed tuple looks like this: ::
226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274

    (# e_1, ..., e_n #)

where ``e_1..e_n`` are expressions of any type (primitive or
non-primitive). The type of an unboxed tuple looks the same.

Note that when unboxed tuples are enabled, ``(#`` is a single lexeme, so
for example when using operators like ``#`` and ``#-`` you need to write
``( # )`` and ``( #- )`` rather than ``(#)`` and ``(#-)``.

Unboxed tuples are used for functions that need to return multiple
values, but they avoid the heap allocation normally associated with
using fully-fledged tuples. When an unboxed tuple is returned, the
components are put directly into registers or on the stack; the unboxed
tuple itself does not have a composite representation. Many of the
primitive operations listed in ``primops.txt.pp`` return unboxed tuples.
In particular, the ``IO`` and ``ST`` monads use unboxed tuples to avoid
unnecessary allocation during sequences of operations.

There are some restrictions on the use of unboxed tuples:

-  The typical use of unboxed tuples is simply to return multiple
   values, binding those multiple results with a ``case`` expression,
   thus:

   ::

         f x y = (# x+1, y-1 #)
         g x = case f x x of { (# a, b #) -> a + b }

   You can have an unboxed tuple in a pattern binding, thus

   ::

         f x = let (# p,q #) = h x in ..body..

   If the types of ``p`` and ``q`` are not unboxed, the resulting
   binding is lazy like any other Haskell pattern binding. The above
   example desugars like this:

   ::

         f x = let t = case h x of { (# p,q #) -> (p,q) }
                   p = fst t
                   q = snd t
               in ..body..

   Indeed, the bindings can even be recursive.

275 276 277 278 279
.. _unboxed-sums:

Unboxed sums
------------

280 281
.. extension:: UnboxedSums
    :shortdesc: Enable unboxed sums.
282 283

    :since: 8.2.1
284 285 286 287 288 289 290 291

    Enable the use of unboxed sum syntax.

`-XUnboxedSums` enables new syntax for anonymous, unboxed sum types. The syntax
for an unboxed sum type with N alternatives is ::

    (# t_1 | t_2 | ... | t_N #)

292 293
where ``t_1`` ... ``t_N`` are types (which can be unlifted, including unboxed
tuples and sums).
294 295 296 297 298

Unboxed tuples can be used for multi-arity alternatives. For example: ::

    (# (# Int, String #) | Bool #)

299 300
The term level syntax is similar. Leading and preceding bars (`|`) indicate which
alternative it is. Here are two terms of the type shown above: ::
301 302 303 304 305

    (# (# 1, "foo" #) | #) -- first alternative

    (# | True #) -- second alternative

306
The pattern syntax reflects the term syntax: ::
307 308 309 310 311 312 313 314

    case x of
      (# (# i, str #) | #) -> ...
      (# | bool #) -> ...

Unboxed sums are "unboxed" in the sense that, instead of allocating sums in the
heap and representing values as pointers, unboxed sums are represented as their
components, just like unboxed tuples. These "components" depend on alternatives
315 316 317 318 319 320 321
of a sum type. Like unboxed tuples, unboxed sums are lazy in their lifted
components.

The code generator tries to generate as compact layout as possible for each
unboxed sum. In the best case, size of an unboxed sum is size of its biggest
alternative plus one word (for a tag). The algorithm for generating the memory
layout for a sum type works like this:
322 323 324 325 326

- All types are classified as one of these classes: 32bit word, 64bit word,
  32bit float, 64bit float, pointer.

- For each alternative of the sum type, a layout that consists of these fields
327 328 329
  is generated. For example, if an alternative has ``Int``, ``Float#`` and
  ``String`` fields, the layout will have an 32bit word, 32bit float and
  pointer fields.
330 331

- Layout fields are then overlapped so that the final layout will be as compact
332
  as possible. For example, suppose we have the unboxed sum: ::
333

334 335
    (# (# Word32#, String, Float# #)
    |  (# Float#, Float#, Maybe Int #) #)
336

337
  The final layout will be something like ::
338 339 340

    Int32, Float32, Float32, Word32, Pointer

341 342 343 344 345 346
  The first ``Int32`` is for the tag. There are two ``Float32`` fields because
  floating point types can't overlap with other types, because of limitations of
  the code generator that we're hoping to overcome in the future. The second
  alternative needs two ``Float32`` fields: The ``Word32`` field is for the
  ``Word32#`` in the first alternative. The ``Pointer`` field is shared between
  ``String`` and ``Maybe Int`` values of the alternatives.
347

348 349
  As another example, this is the layout for the unboxed version of ``Maybe a``
  type, ``(# (# #) | a #)``: ::
350 351 352

    Int32, Pointer

353 354 355 356 357 358 359 360 361 362 363 364
  The ``Pointer`` field is not used when tag says that it's ``Nothing``.
  Otherwise ``Pointer`` points to the value in ``Just``. As mentioned
  above, this type is lazy in its lifted field. Therefore, the type ::

    data Maybe' a = Maybe' (# (# #) | a #)

  is *precisely* isomorphic to the type ``Maybe a``, although its memory
  representation is different.

  In the degenerate case where all the alternatives have zero width, such
  as the ``Bool``-like ``(# (# #) | (# #) #)``, the unboxed sum layout only
  has an ``Int32`` tag field (i.e., the whole thing is represented by an integer).
365

366 367 368 369 370 371 372 373 374 375 376 377
.. _unlifted-newtypes:

Unlifted Newtypes
-----------------

.. extension:: UnliftedNewtypes
    :shortdesc: Enable unlifted newtypes.

    :since: 8.10.1

    Enable the use of newtypes over types with non-lifted runtime representations.

378 379 380
GHC implements an :extension:`UnliftedNewtypes` extension as specified in
`this GHC proposal <https://github.com/ghc-proposals/ghc-proposals/blob/master/proposals/0013-unlifted-newtypes.rst>`_.
:extension:`UnliftedNewtypes` relaxes the restrictions around what types can appear inside
381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414
of a `newtype`. For example, the type ::

    newtype A = MkA Int#

is accepted when this extension is enabled. This creates a type
``A :: TYPE 'IntRep`` and a data constructor ``MkA :: Int# -> A``.
Although the kind of ``A`` is inferred by GHC, there is nothing visually
distictive about this type that indicated that is it not of kind ``Type``
like newtypes typically are. `GADTSyntax <#gadt-style>`__ can be used to
provide a kind signature for additional clarity ::

    newtype A :: TYPE 'IntRep where
      MkA :: Int# -> A

The ``Coercible`` machinery works with unlifted newtypes just like it does with
lifted types. In either of the equivalent formulations of ``A`` given above,
users would additionally have access to a coercion between ``A`` and ``Int#``.

As a consequence of the
`levity-polymorphic binder restriction <#levity-polymorphic-restrictions>`__,
levity-polymorphic fields are disallowed in data constructors
of data types declared using ``data``. However, since ``newtype`` data
constructor application is implemented as a coercion instead of as function
application, this restriction does not apply to the field inside a ``newtype``
data constructor. Thus, the type checker accepts ::

    newtype Identity# :: forall (r :: RuntimeRep). TYPE r -> TYPE r where
      MkIdentity# :: forall (r :: RuntimeRep) (a :: TYPE r). a -> Identity# a

And with `UnboxedSums <#unboxed-sums>`__ enabled ::

    newtype Maybe# :: forall (r :: RuntimeRep). TYPE r -> TYPE (SumRep '[r, TupleRep '[]]) where
      MkMaybe# :: forall (r :: RuntimeRep) (a :: TYPE r). (# a | (# #) #) -> Maybe# a

415 416 417 418 419
This extension also relaxes some of the restrictions around data family
instances. In particular, :extension:`UnliftedNewtypes` permits a
``newtype instance`` to be given a return kind of ``TYPE r``, not just
``Type``. For example, the following ``newtype instance`` declarations would be
permitted: ::
420

421 422 423 424 425 426 427 428 429 430
     class Foo a where
       data FooKey a :: TYPE 'IntRep
     class Bar (r :: RuntimeRep) where
       data BarType r :: TYPE r

     instance Foo Bool where
       newtype FooKey Bool = FooKeyBoolC Int#
     instance Bar 'WordRep where
       newtype BarType 'WordRep = BarTypeWordRepC Word#

431 432 433 434 435 436
It is worth noting that :extension:`UnliftedNewtypes` is *not* required to give
the data families themselves return kinds involving ``TYPE``, such as the
``FooKey`` and ``BarType`` examples above. The extension is
only required for ``newtype instance`` declarations, such as ``FooKeyBoolC``
and ``BarTypeWorkRepC`` above.

437 438 439 440
This extension impacts the determination of whether or not a newtype has
a Complete User-Specified Kind Signature (CUSK). The exact impact is specified
`the section on CUSKs <#complete-kind-signatures>`__.

441 442 443 444 445 446 447 448 449 450
.. _syntax-extns:

Syntactic extensions
====================

.. _unicode-syntax:

Unicode syntax
--------------

451 452
.. extension:: UnicodeSyntax
    :shortdesc: Enable unicode syntax.
453 454

    :since: 6.8.1
455 456 457 458

    Enable the use of Unicode characters in place of their equivalent ASCII
    sequences.

459
The language extension :extension:`UnicodeSyntax` enables
460 461 462
Unicode characters to be used to stand for certain ASCII character
sequences. The following alternatives are provided:

463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494
+--------------+---------------+-------------+-----------------------------------------+
| ASCII        | Unicode       | Code point  | Name                                    |
|              | alternative   |             |                                         |
+==============+===============+=============+=========================================+
| ``::``       | ∷             | 0x2237      | PROPORTION                              |
+--------------+---------------+-------------+-----------------------------------------+
| ``=>``       | ⇒             | 0x21D2      | RIGHTWARDS DOUBLE ARROW                 |
+--------------+---------------+-------------+-----------------------------------------+
| ``->``       | →             | 0x2192      | RIGHTWARDS ARROW                        |
+--------------+---------------+-------------+-----------------------------------------+
| ``<-``       | ←             | 0x2190      | LEFTWARDS ARROW                         |
+--------------+---------------+-------------+-----------------------------------------+
| ``>-``       | ⤚             | 0x291a      | RIGHTWARDS ARROW-TAIL                   |
+--------------+---------------+-------------+-----------------------------------------+
| ``-<``       | ⤙             | 0x2919      | LEFTWARDS ARROW-TAIL                    |
+--------------+---------------+-------------+-----------------------------------------+
| ``>>-``      | ⤜             | 0x291C      | RIGHTWARDS DOUBLE ARROW-TAIL            |
+--------------+---------------+-------------+-----------------------------------------+
| ``-<<``      | ⤛             | 0x291B      | LEFTWARDS DOUBLE ARROW-TAIL             |
+--------------+---------------+-------------+-----------------------------------------+
| ``*``        | ★             | 0x2605      | BLACK STAR                              |
+--------------+---------------+-------------+-----------------------------------------+
| ``forall``   | ∀             | 0x2200      | FOR ALL                                 |
+--------------+---------------+-------------+-----------------------------------------+
| ``(|``       | ⦇             | 0x2987      | Z NOTATION LEFT IMAGE BRACKET           |
+--------------+---------------+-------------+-----------------------------------------+
| ``|)``       | ⦈             | 0x2988      | Z NOTATION RIGHT IMAGE BRACKET          |
+--------------+---------------+-------------+-----------------------------------------+
| ``[|``       | ⟦             | 0x27E6      | MATHEMATICAL LEFT WHITE SQUARE BRACKET  |
+--------------+---------------+-------------+-----------------------------------------+
| ``|]``       | ⟧             | 0x27E7      | MATHEMATICAL RIGHT WHITE SQUARE BRACKET |
+--------------+---------------+-------------+-----------------------------------------+
495 496 497 498 499 500

.. _magic-hash:

The magic hash
--------------

501 502
.. extension:: MagicHash
    :shortdesc: Allow ``#`` as a postfix modifier on identifiers.
503 504

    :since: 6.8.1
505

506
    Enables the use of the hash character (``#``) as an identifier suffix.
507

508
The language extension :extension:`MagicHash` allows ``#`` as a postfix modifier
509
to identifiers. Thus, ``x#`` is a valid variable, and ``T#`` is a valid type
510 511 512 513 514
constructor or data constructor.

The hash sign does not change semantics at all. We tend to use variable
names ending in "#" for unboxed values or types (e.g. ``Int#``), but
there is no requirement to do so; they are just plain ordinary
515
variables. Nor does the :extension:`MagicHash` extension bring anything into
516
scope. For example, to bring ``Int#`` into scope you must import
517
``GHC.Prim`` (see :ref:`primitives`); the :extension:`MagicHash` extension then
518 519 520 521 522
allows you to *refer* to the ``Int#`` that is now in scope. Note that
with this option, the meaning of ``x#y = 0`` is changed: it defines a
function ``x#`` taking a single argument ``y``; to define the operator
``#``, put a space: ``x # y = 0``.

523
The :extension:`MagicHash` also enables some new forms of literals (see
524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545
:ref:`glasgow-unboxed`):

-  ``'x'#`` has type ``Char#``

-  ``"foo"#`` has type ``Addr#``

-  ``3#`` has type ``Int#``. In general, any Haskell integer lexeme
   followed by a ``#`` is an ``Int#`` literal, e.g. ``-0x3A#`` as well as
   ``32#``.

-  ``3##`` has type ``Word#``. In general, any non-negative Haskell
   integer lexeme followed by ``##`` is a ``Word#``.

-  ``3.2#`` has type ``Float#``.

-  ``3.2##`` has type ``Double#``

.. _negative-literals:

Negative literals
-----------------

546 547
.. extension:: NegativeLiterals
    :shortdesc: Enable support for negative literals.
548

549 550
    :since: 7.8.1

551 552
    Enable the use of un-parenthesized negative numeric literals.

553 554
The literal ``-123`` is, according to Haskell98 and Haskell 2010,
desugared as ``negate (fromInteger 123)``. The language extension
555
:extension:`NegativeLiterals` means that it is instead desugared as
556 557 558 559 560 561 562 563 564 565 566 567
``fromInteger (-123)``.

This can make a difference when the positive and negative range of a
numeric data type don't match up. For example, in 8-bit arithmetic -128
is representable, but +128 is not. So ``negate (fromInteger 128)`` will
elicit an unexpected integer-literal-overflow message.

.. _num-decimals:

Fractional looking integer literals
-----------------------------------

568
.. extension:: NumDecimals
569
    :shortdesc: Enable support for 'fractional' integer literals.
570

571 572
    :since: 7.8.1

573 574
    Allow the use of floating-point literal syntax for integral types.

575 576 577
Haskell 2010 and Haskell 98 define floating literals with the syntax
``1.2e6``. These literals have the type ``Fractional a => a``.

578
The language extension :extension:`NumDecimals` allows you to also use the
579 580 581 582 583 584 585 586
floating literal syntax for instances of ``Integral``, and have values
like ``(1.2e6 :: Num a => a)``

.. _binary-literals:

Binary integer literals
-----------------------

587 588
.. extension:: BinaryLiterals
    :shortdesc: Enable support for binary literals.
589

590 591
    :since: 7.10.1

592 593
    Allow the use of binary notation in integer literals.

594 595 596 597
Haskell 2010 and Haskell 98 allows for integer literals to be given in
decimal, octal (prefixed by ``0o`` or ``0O``), or hexadecimal notation
(prefixed by ``0x`` or ``0X``).

598
The language extension :extension:`BinaryLiterals` adds support for expressing
599 600
integer literals in binary notation with the prefix ``0b`` or ``0B``. For
instance, the binary integer literal ``0b11001001`` will be desugared into
601
``fromInteger 201`` when :extension:`BinaryLiterals` is enabled.
602

603 604 605 606 607
.. _hex-float-literals:

Hexadecimal floating point literals
-----------------------------------

608
.. extension:: HexFloatLiterals
Douglas Wilson's avatar
Douglas Wilson committed
609
    :shortdesc: Enable support for :ref:`hexadecimal floating point literals <hex-float-literals>`.
610 611 612 613 614 615 616 617 618 619 620 621

    :since: 8.4.1

    Allow writing floating point literals using hexadecimal notation.

The hexadecimal notation for floating point literals is useful when you
need to specify floating point constants precisely, as the literal notation
corresponds closely to the underlying bit-encoding of the number.

In this notation floating point numbers are written using hexadecimal digits,
and so the digits are interpreted using base 16, rather then the usual 10.
This means that digits left of the decimal point correspond to positive
622
powers of 16, while the ones to the right correspond to negative ones.
623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643

You may also write an explicit exponent, which is similar to the exponent
in decimal notation with the following differences:
- the exponent begins with ``p`` instead of ``e``
- the exponent is written in base ``10`` (**not** 16)
- the base of the exponent is ``2`` (**not** 16).

In terms of the underlying bit encoding, each hexadecimal digit corresponds
to 4 bits, and you may think of the exponent as "moving" the floating point
by one bit left (negative) or right (positive).  Here are some examples:

-  ``0x0.1``     is the same as ``1/16``
-  ``0x0.01``    is the same as ``1/256``
-  ``0xF.FF``    is the same as ``15 + 15/16 + 15/256``
-  ``0x0.1p4``   is the same as ``1``
-  ``0x0.1p-4``  is the same as ``1/256``
-  ``0x0.1p12``  is the same as ``256``




644 645 646 647 648
.. _numeric-underscores:

Numeric underscores
-------------------

649
.. extension:: NumericUnderscores
650 651 652 653 654 655 656 657 658
    :shortdesc: Enable support for :ref:`numeric underscores <numeric-underscores>`.

    :since: 8.6.1

    Allow the use of underscores in numeric literals.

GHC allows for numeric literals to be given in decimal, octal, hexadecimal,
binary, or float notation.

659
The language extension :extension:`NumericUnderscores` adds support for expressing
660 661
underscores in numeric literals.
For instance, the numeric literal ``1_000_000`` will be parsed into
662
``1000000`` when :extension:`NumericUnderscores` is enabled.
663
That is, underscores in numeric literals are ignored when
664
:extension:`NumericUnderscores` is enabled.
665 666
See also :ghc-ticket:`14473`.

667 668 669
For example:

.. code-block:: none
670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700

    -- decimal
    million    = 1_000_000
    billion    = 1_000_000_000
    lightspeed = 299_792_458
    version    = 8_04_1
    date       = 2017_12_31

    -- hexadecimal
    red_mask = 0xff_00_00
    size1G   = 0x3fff_ffff

    -- binary
    bit8th   = 0b01_0000_0000
    packbits = 0b1_11_01_0000_0_111
    bigbits  = 0b1100_1011__1110_1111__0101_0011

    -- float
    pi       = 3.141_592_653_589_793
    faraday  = 96_485.332_89
    avogadro = 6.022_140_857e+23

    -- function
    isUnderMillion = (< 1_000_000)

    clip64M x
        | x > 0x3ff_ffff = 0x3ff_ffff
        | otherwise = x

    test8bit x = (0b01_0000_0000 .&. x) /= 0

701 702 703
About validity:

.. code-block:: none
704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731

    x0 = 1_000_000   -- valid
    x1 = 1__000000   -- valid
    x2 = 1000000_    -- invalid
    x3 = _1000000    -- invalid

    e0 = 0.0001      -- valid
    e1 = 0.000_1     -- valid
    e2 = 0_.0001     -- invalid
    e3 = _0.0001     -- invalid
    e4 = 0._0001     -- invalid
    e5 = 0.0001_     -- invalid

    f0 = 1e+23       -- valid
    f1 = 1_e+23      -- valid
    f2 = 1__e+23     -- valid
    f3 = 1e_+23      -- invalid

    g0 = 1e+23       -- valid
    g1 = 1e+_23      -- invalid
    g2 = 1e+23_      -- invalid

    h0 = 0xffff      -- valid
    h1 = 0xff_ff     -- valid
    h2 = 0x_ffff     -- valid
    h3 = 0x__ffff    -- valid
    h4 = _0xffff     -- invalid

732 733 734 735 736
.. _pattern-guards:

Pattern guards
--------------

737 738 739
.. extension:: NoPatternGuards
    :shortdesc: Disable pattern guards.
        Implied by :extension:`Haskell98`.
740

741
    :implied by: :extension:`Haskell98`
742 743
    :since: 6.8.1

744
Disable `pattern guards
745
<http://www.haskell.org/onlinereport/haskell2010/haskellch3.html#x8-460003.13>`__.
746

747 748 749 750 751
.. _view-patterns:

View patterns
-------------

752 753
.. extension:: ViewPatterns
    :shortdesc: Enable view patterns.
754 755

    :since: 6.10.1
756 757 758

    Allow use of view pattern syntax.

759
View patterns are enabled by the language extension :extension:`ViewPatterns`. More
760
information and examples of view patterns can be found on the
761
:ghc-wiki:`Wiki page <view-patterns>`.
762 763 764 765 766

View patterns are somewhat like pattern guards that can be nested inside
of other patterns. They are a convenient way of pattern-matching against
values of abstract types. For example, in a programming language
implementation, we might represent the syntax of the types of the
767
language as follows: ::
768 769 770 771 772 773 774 775 776 777 778 779

    type Typ

    data TypView = Unit
                 | Arrow Typ Typ

    view :: Typ -> TypView

    -- additional operations for constructing Typ's ...

The representation of Typ is held abstract, permitting implementations
to use a fancy representation (e.g., hash-consing to manage sharing).
Rik Steenkamp's avatar
Rik Steenkamp committed
780
Without view patterns, using this signature is a little inconvenient: ::
781 782 783 784 785 786 787 788 789 790 791

    size :: Typ -> Integer
    size t = case view t of
      Unit -> 1
      Arrow t1 t2 -> size t1 + size t2

It is necessary to iterate the case, rather than using an equational
function definition. And the situation is even worse when the matching
against ``t`` is buried deep inside another pattern.

View patterns permit calling the view function inside the pattern and
792
matching against the result: ::
793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813

    size (view -> Unit) = 1
    size (view -> Arrow t1 t2) = size t1 + size t2

That is, we add a new form of pattern, written ⟨expression⟩ ``->``
⟨pattern⟩ that means "apply the expression to whatever we're trying to
match against, and then match the result of that application against the
pattern". The expression can be any Haskell expression of function type,
and view patterns can be used wherever patterns are used.

The semantics of a pattern ``(`` ⟨exp⟩ ``->`` ⟨pat⟩ ``)`` are as
follows:

-  Scoping:
   The variables bound by the view pattern are the variables bound by
   ⟨pat⟩.

   Any variables in ⟨exp⟩ are bound occurrences, but variables bound "to
   the left" in a pattern are in scope. This feature permits, for
   example, one argument to a function to be used in the view of another
   argument. For example, the function ``clunky`` from
814
   :ref:`pattern-guards` can be written using view patterns as follows: ::
815 816 817 818 819 820 821

       clunky env (lookup env -> Just val1) (lookup env -> Just val2) = val1 + val2
       ...other equations for clunky...

   More precisely, the scoping rules are:

   -  In a single pattern, variables bound by patterns to the left of a
822
      view pattern expression are in scope. For example: ::
823 824

          example :: Maybe ((String -> Integer,Integer), String) -> Bool
825
          example (Just ((f,_), f -> 4)) = True
826 827 828

      Additionally, in function definitions, variables bound by matching
      earlier curried arguments may be used in view pattern expressions
829
      in later arguments: ::
830 831 832 833 834 835 836 837 838 839 840

          example :: (String -> Integer) -> String -> Bool
          example f (f -> 4) = True

      That is, the scoping is the same as it would be if the curried
      arguments were collected into a tuple.

   -  In mutually recursive bindings, such as ``let``, ``where``, or the
      top level, view patterns in one declaration may not mention
      variables bound by other declarations. That is, each declaration
      must be self-contained. For example, the following program is not
841
      allowed: ::
842 843 844 845 846 847 848 849 850 851

          let {(x -> y) = e1 ;
               (y -> x) = e2 } in x

   (For some amplification on this design choice see :ghc-ticket:`4061`.

-  Typing: If ⟨exp⟩ has type ⟨T1⟩ ``->`` ⟨T2⟩ and ⟨pat⟩ matches a ⟨T2⟩,
   then the whole view pattern matches a ⟨T1⟩.

-  Matching: To the equations in Section 3.17.3 of the `Haskell 98
852
   Report <http://www.haskell.org/onlinereport/>`__, add the following: ::
853 854 855 856 857 858 859 860 861 862 863 864 865 866 867 868 869 870 871 872 873 874 875 876 877 878 879 880 881 882 883 884 885 886 887

       case v of { (e -> p) -> e1 ; _ -> e2 }
        =
       case (e v) of { p -> e1 ; _ -> e2 }

   That is, to match a variable ⟨v⟩ against a pattern ``(`` ⟨exp⟩ ``->``
   ⟨pat⟩ ``)``, evaluate ``(`` ⟨exp⟩ ⟨v⟩ ``)`` and match the result
   against ⟨pat⟩.

-  Efficiency: When the same view function is applied in multiple
   branches of a function definition or a case expression (e.g., in
   ``size`` above), GHC makes an attempt to collect these applications
   into a single nested case expression, so that the view function is
   only applied once. Pattern compilation in GHC follows the matrix
   algorithm described in Chapter 4 of `The Implementation of Functional
   Programming
   Languages <http://research.microsoft.com/~simonpj/Papers/slpj-book-1987/>`__.
   When the top rows of the first column of a matrix are all view
   patterns with the "same" expression, these patterns are transformed
   into a single nested case. This includes, for example, adjacent view
   patterns that line up in a tuple, as in

   ::

       f ((view -> A, p1), p2) = e1
       f ((view -> B, p3), p4) = e2

   The current notion of when two view pattern expressions are "the
   same" is very restricted: it is not even full syntactic equality.
   However, it does include variables, literals, applications, and
   tuples; e.g., two instances of ``view ("hi", "there")`` will be
   collected. However, the current implementation does not compare up to
   alpha-equivalence, so two instances of ``(x, view x -> y)`` will not
   be coalesced.

Simon Peyton Jones's avatar
Simon Peyton Jones committed
888
.. _n-k-patterns:
889

Simon Peyton Jones's avatar
Simon Peyton Jones committed
890 891
n+k patterns
------------
892

893
.. extension:: NPlusKPatterns
894
    :shortdesc: Enable support for ``n+k`` patterns.
895
        Implied by :extension:`Haskell98`.
896

897
    :implied by: :extension:`Haskell98`
898
    :since: 6.12.1
Ben Gamari's avatar
Ben Gamari committed
899

Simon Peyton Jones's avatar
Simon Peyton Jones committed
900
    Enable use of ``n+k`` patterns.
901

Simon Peyton Jones's avatar
Simon Peyton Jones committed
902
.. _recursive-do-notation:
903

Simon Peyton Jones's avatar
Simon Peyton Jones committed
904 905
The recursive do-notation
-------------------------
906

907 908
.. extension:: RecursiveDo
    :shortdesc: Enable recursive do (mdo) notation.
909 910

    :since: 6.8.1
911

Simon Peyton Jones's avatar
Simon Peyton Jones committed
912
    Allow the use of recursive ``do`` notation.
913

Simon Peyton Jones's avatar
Simon Peyton Jones committed
914 915 916 917
The do-notation of Haskell 98 does not allow *recursive bindings*, that
is, the variables bound in a do-expression are visible only in the
textually following code block. Compare this to a let-expression, where
bound variables are visible in the entire binding group.
918

Simon Peyton Jones's avatar
Simon Peyton Jones committed
919 920 921 922 923
It turns out that such recursive bindings do indeed make sense for a
variety of monads, but not all. In particular, recursion in this sense
requires a fixed-point operator for the underlying monad, captured by
the ``mfix`` method of the ``MonadFix`` class, defined in
``Control.Monad.Fix`` as follows: ::
924

Simon Peyton Jones's avatar
Simon Peyton Jones committed
925 926
    class Monad m => MonadFix m where
       mfix :: (a -> m a) -> m a
927

Simon Peyton Jones's avatar
Simon Peyton Jones committed
928 929 930 931
Haskell's ``Maybe``, ``[]`` (list), ``ST`` (both strict and lazy
versions), ``IO``, and many other monads have ``MonadFix`` instances. On
the negative side, the continuation monad, with the signature
``(a -> r) -> r``, does not.
932

Simon Peyton Jones's avatar
Simon Peyton Jones committed
933 934
For monads that do belong to the ``MonadFix`` class, GHC provides an
extended version of the do-notation that allows recursive bindings. The
935
:extension:`RecursiveDo` (language pragma: ``RecursiveDo``) provides the
Simon Peyton Jones's avatar
Simon Peyton Jones committed
936 937 938 939 940
necessary syntactic support, introducing the keywords ``mdo`` and
``rec`` for higher and lower levels of the notation respectively. Unlike
bindings in a ``do`` expression, those introduced by ``mdo`` and ``rec``
are recursively defined, much like in an ordinary let-expression. Due to
the new keyword ``mdo``, we also call this notation the *mdo-notation*.
941

Simon Peyton Jones's avatar
Simon Peyton Jones committed
942
Here is a simple (albeit contrived) example:
943

Simon Peyton Jones's avatar
Simon Peyton Jones committed
944
::
945

Simon Peyton Jones's avatar
Simon Peyton Jones committed
946 947 948
    {-# LANGUAGE RecursiveDo #-}
    justOnes = mdo { xs <- Just (1:xs)
                   ; return (map negate xs) }
949

Simon Peyton Jones's avatar
Simon Peyton Jones committed
950
or equivalently
951

Simon Peyton Jones's avatar
Simon Peyton Jones committed
952
::
953

Simon Peyton Jones's avatar
Simon Peyton Jones committed
954 955 956
    {-# LANGUAGE RecursiveDo #-}
    justOnes = do { rec { xs <- Just (1:xs) }
                  ; return (map negate xs) }
957

Simon Peyton Jones's avatar
Simon Peyton Jones committed
958
As you can guess ``justOnes`` will evaluate to ``Just [-1,-1,-1,...``.
959

Simon Peyton Jones's avatar
Simon Peyton Jones committed
960 961
GHC's implementation the mdo-notation closely follows the original
translation as described in the paper `A recursive do for
niteria's avatar
niteria committed
962
Haskell <http://leventerkok.github.io/papers/recdo.pdf>`__, which
Simon Peyton Jones's avatar
Simon Peyton Jones committed
963
in turn is based on the work `Value Recursion in Monadic
niteria's avatar
niteria committed
964
Computations <http://leventerkok.github.io/papers/erkok-thesis.pdf>`__.
Simon Peyton Jones's avatar
Simon Peyton Jones committed
965 966
Furthermore, GHC extends the syntax described in the former paper with a
lower level syntax flagged by the ``rec`` keyword, as we describe next.
967

Simon Peyton Jones's avatar
Simon Peyton Jones committed
968 969
Recursive binding groups
~~~~~~~~~~~~~~~~~~~~~~~~
970

971
The extension :extension:`RecursiveDo` also introduces a new keyword ``rec``, which
Simon Peyton Jones's avatar
Simon Peyton Jones committed
972 973 974 975
wraps a mutually-recursive group of monadic statements inside a ``do``
expression, producing a single statement. Similar to a ``let`` statement
inside a ``do``, variables bound in the ``rec`` are visible throughout
the ``rec`` group, and below it. For example, compare
976

Simon Peyton Jones's avatar
Simon Peyton Jones committed
977
::
978

Simon Peyton Jones's avatar
Simon Peyton Jones committed
979 980 981 982
        do { a <- getChar            do { a <- getChar
           ; let { r1 = f a r2          ; rec { r1 <- f a r2
           ;     ; r2 = g r1 }          ;     ; r2 <- g r1 }
           ; return (r1 ++ r2) }        ; return (r1 ++ r2) }
983

Simon Peyton Jones's avatar
Simon Peyton Jones committed
984 985 986 987
In both cases, ``r1`` and ``r2`` are available both throughout the
``let`` or ``rec`` block, and in the statements that follow it. The
difference is that ``let`` is non-monadic, while ``rec`` is monadic. (In
Haskell ``let`` is really ``letrec``, of course.)
988

Simon Peyton Jones's avatar
Simon Peyton Jones committed
989 990 991 992 993
The semantics of ``rec`` is fairly straightforward. Whenever GHC finds a
``rec`` group, it will compute its set of bound variables, and will
introduce an appropriate call to the underlying monadic value-recursion
operator ``mfix``, belonging to the ``MonadFix`` class. Here is an
example:
994 995 996

::

Simon Peyton Jones's avatar
Simon Peyton Jones committed
997 998 999
    rec { b <- f a c     ===>    (b,c) <- mfix (\ ~(b,c) -> do { b <- f a c
        ; c <- f b a }                                         ; c <- f b a
                                                               ; return (b,c) })
1000

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1001 1002 1003
As usual, the meta-variables ``b``, ``c`` etc., can be arbitrary
patterns. In general, the statement ``rec ss`` is desugared to the
statement
1004 1005 1006

::

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1007
    vs <- mfix (\ ~vs -> do { ss; return vs })
1008

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1009
where ``vs`` is a tuple of the variables bound by ``ss``.
1010

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1011 1012 1013 1014
Note in particular that the translation for a ``rec`` block only
involves wrapping a call to ``mfix``: it performs no other analysis on
the bindings. The latter is the task for the ``mdo`` notation, which is
described next.
1015

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1016 1017
The ``mdo`` notation
~~~~~~~~~~~~~~~~~~~~
1018

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1019 1020 1021 1022 1023 1024
A ``rec``-block tells the compiler where precisely the recursive knot
should be tied. It turns out that the placement of the recursive knots
can be rather delicate: in particular, we would like the knots to be
wrapped around as minimal groups as possible. This process is known as
*segmentation*, and is described in detail in Section 3.2 of `A
recursive do for
niteria's avatar
niteria committed
1025
Haskell <http://leventerkok.github.io/papers/recdo.pdf>`__.
Simon Peyton Jones's avatar
Simon Peyton Jones committed
1026 1027 1028 1029 1030 1031 1032 1033
Segmentation improves polymorphism and reduces the size of the recursive
knot. Most importantly, it avoids unnecessary interference caused by a
fundamental issue with the so-called *right-shrinking* axiom for monadic
recursion. In brief, most monads of interest (IO, strict state, etc.) do
*not* have recursion operators that satisfy this axiom, and thus not
performing segmentation can cause unnecessary interference, changing the
termination behavior of the resulting translation. (Details can be found
in Sections 3.1 and 7.2.2 of `Value Recursion in Monadic
niteria's avatar
niteria committed
1034
Computations <http://leventerkok.github.io/papers/erkok-thesis.pdf>`__.)
1035

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1036 1037 1038 1039 1040 1041 1042
The ``mdo`` notation removes the burden of placing explicit ``rec``
blocks in the code. Unlike an ordinary ``do`` expression, in which
variables bound by statements are only in scope for later statements,
variables bound in an ``mdo`` expression are in scope for all statements
of the expression. The compiler then automatically identifies minimal
mutually recursively dependent segments of statements, treating them as
if the user had wrapped a ``rec`` qualifier around them.
1043

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1044
The definition is syntactic:
1045

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1046
-  A generator ⟨g⟩ *depends* on a textually following generator ⟨g'⟩, if
1047

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1048
   -  ⟨g'⟩ defines a variable that is used by ⟨g⟩, or
1049

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1050 1051
   -  ⟨g'⟩ textually appears between ⟨g⟩ and ⟨g''⟩, where ⟨g⟩ depends on
      ⟨g''⟩.
1052

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1053 1054 1055 1056 1057
-  A *segment* of a given ``mdo``-expression is a minimal sequence of
   generators such that no generator of the sequence depends on an
   outside generator. As a special case, although it is not a generator,
   the final expression in an ``mdo``-expression is considered to form a
   segment by itself.
1058

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1059 1060 1061
Segments in this sense are related to *strongly-connected components*
analysis, with the exception that bindings in a segment cannot be
reordered and must be contiguous.
1062

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1063 1064
Here is an example ``mdo``-expression, and its translation to ``rec``
blocks:
1065

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1066
::
1067

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1068 1069 1070 1071 1072 1073 1074
    mdo { a <- getChar      ===> do { a <- getChar
        ; b <- f a c                ; rec { b <- f a c
        ; c <- f b a                ;     ; c <- f b a }
        ; z <- h a b                ; z <- h a b
        ; d <- g d e                ; rec { d <- g d e
        ; e <- g a z                ;     ; e <- g a z }
        ; putChar c }               ; putChar c }
1075

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1076 1077 1078 1079
Note that a given ``mdo`` expression can cause the creation of multiple
``rec`` blocks. If there are no recursive dependencies, ``mdo`` will
introduce no ``rec`` blocks. In this latter case an ``mdo`` expression
is precisely the same as a ``do`` expression, as one would expect.
1080

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1081 1082 1083 1084 1085 1086
In summary, given an ``mdo`` expression, GHC first performs
segmentation, introducing ``rec`` blocks to wrap over minimal recursive
groups. Then, each resulting ``rec`` is desugared, using a call to
``Control.Monad.Fix.mfix`` as described in the previous section. The
original ``mdo``-expression typechecks exactly when the desugared
version would do so.
1087

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1088
Here are some other important points in using the recursive-do notation:
1089

1090 1091
-  It is enabled with the extension :extension:`RecursiveDo`, or the
   ``LANGUAGE RecursiveDo`` pragma. (The same extension enables both
Simon Peyton Jones's avatar
Simon Peyton Jones committed
1092 1093
   ``mdo``-notation, and the use of ``rec`` blocks inside ``do``
   expressions.)
1094

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1095 1096 1097
-  ``rec`` blocks can also be used inside ``mdo``-expressions, which
   will be treated as a single statement. However, it is good style to
   either use ``mdo`` or ``rec`` blocks in a single expression.
1098

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1099 1100
-  If recursive bindings are required for a monad, then that monad must
   be declared an instance of the ``MonadFix`` class.
1101

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1102 1103 1104 1105 1106
-  The following instances of ``MonadFix`` are automatically provided:
   List, Maybe, IO. Furthermore, the ``Control.Monad.ST`` and
   ``Control.Monad.ST.Lazy`` modules provide the instances of the
   ``MonadFix`` class for Haskell's internal state monad (strict and
   lazy, respectively).
1107

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1108 1109 1110 1111
-  Like ``let`` and ``where`` bindings, name shadowing is not allowed
   within an ``mdo``-expression or a ``rec``-block; that is, all the
   names bound in a single ``rec`` must be distinct. (GHC will complain
   if this is not the case.)
1112

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1113
.. _applicative-do:
1114

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1115 1116
Applicative do-notation
-----------------------
1117

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1118 1119 1120
.. index::
   single: Applicative do-notation
   single: do-notation; Applicative
1121

1122 1123
.. extension:: ApplicativeDo
    :shortdesc: Enable Applicative do-notation desugaring
1124

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1125
    :since: 8.0.1
1126

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1127
    Allow use of ``Applicative`` ``do`` notation.
1128

1129
The language option :extension:`ApplicativeDo` enables an alternative translation for
Simon Peyton Jones's avatar
Simon Peyton Jones committed
1130 1131
the do-notation, which uses the operators ``<$>``, ``<*>``, along with ``join``
as far as possible. There are two main reasons for wanting to do this:
1132

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1133 1134 1135 1136
-  We can use do-notation with types that are an instance of ``Applicative`` and
   ``Functor``, but not ``Monad``
-  In some monads, using the applicative operators is more efficient than monadic
   bind. For example, it may enable more parallelism.
1137

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1138 1139 1140
Applicative do-notation desugaring preserves the original semantics, provided
that the ``Applicative`` instance satisfies ``<*> = ap`` and ``pure = return``
(these are true of all the common monadic types). Thus, you can normally turn on
1141
:extension:`ApplicativeDo` without fear of breaking your program. There is one pitfall
Simon Peyton Jones's avatar
Simon Peyton Jones committed
1142
to watch out for; see :ref:`applicative-do-pitfall`.
1143

1144
There are no syntactic changes with :extension:`ApplicativeDo`. The only way it shows
Simon Peyton Jones's avatar
Simon Peyton Jones committed
1145 1146
up at the source level is that you can have a ``do`` expression that doesn't
require a ``Monad`` constraint. For example, in GHCi: ::
1147

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1148 1149 1150 1151
    Prelude> :set -XApplicativeDo
    Prelude> :t \m -> do { x <- m; return (not x) }
    \m -> do { x <- m; return (not x) }
      :: Functor f => f Bool -> f Bool
1152

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1153 1154
This example only requires ``Functor``, because it is translated into ``(\x ->
not x) <$> m``. A more complex example requires ``Applicative``, ::
1155

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1156 1157 1158
    Prelude> :t \m -> do { x <- m 'a'; y <- m 'b'; return (x || y) }
    \m -> do { x <- m 'a'; y <- m 'b'; return (x || y) }
      :: Applicative f => (Char -> f Bool) -> f Bool
1159

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1160
Here GHC has translated the expression into ::
1161

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1162
    (\x y -> x || y) <$> m 'a' <*> m 'b'
1163

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1164 1165
It is possible to see the actual translation by using :ghc-flag:`-ddump-ds`, but be
warned, the output is quite verbose.
1166

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1167 1168 1169 1170
Note that if the expression can't be translated into uses of ``<$>``, ``<*>``
only, then it will incur a ``Monad`` constraint as usual. This happens when
there is a dependency on a value produced by an earlier statement in the
``do``-block: ::
1171

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1172 1173 1174
    Prelude> :t \m -> do { x <- m True; y <- m x; return (x || y) }
    \m -> do { x <- m True; y <- m x; return (x || y) }
      :: Monad m => (Bool -> m Bool) -> m Bool
1175

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1176 1177
Here, ``m x`` depends on the value of ``x`` produced by the first statement, so
the expression cannot be translated using ``<*>``.
1178

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1179 1180
In general, the rule for when a ``do`` statement incurs a ``Monad`` constraint
is as follows. If the do-expression has the following form: ::
1181

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1182
    do p1 <- E1; ...; pn <- En; return E
1183

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1184
where none of the variables defined by ``p1...pn`` are mentioned in ``E1...En``,
1185
and ``p1...pn`` are all variables or lazy patterns,
Simon Peyton Jones's avatar
Simon Peyton Jones committed
1186
then the expression will only require ``Applicative``. Otherwise, the expression
1187 1188
will require ``Monad``. The block may return a pure expression ``E`` depending
upon the results ``p1...pn`` with either ``return`` or ``pure``.
1189

1190 1191 1192 1193 1194 1195 1196 1197 1198 1199 1200 1201 1202 1203 1204
Note: the final statement must match one of these patterns exactly:

- ``return E``
- ``return $ E``
- ``pure E``
- ``pure $ E``

otherwise GHC cannot recognise it as a ``return`` statement, and the
transformation to use ``<$>`` that we saw above does not apply.  In
particular, slight variations such as ``return . Just $ x`` or ``let x
= e in return x`` would not be recognised.

If the final statement is not of one of these forms, GHC falls back to
standard ``do`` desugaring, and the expression will require a
``Monad`` constraint.
1205

Simon Marlow's avatar
Simon Marlow committed
1206 1207 1208 1209 1210 1211 1212 1213
When the statements of a ``do`` expression have dependencies between
them, and ``ApplicativeDo`` cannot infer an ``Applicative`` type, it
uses a heuristic algorithm to try to use ``<*>`` as much as possible.
This algorithm usually finds the best solution, but in rare complex
cases it might miss an opportunity.  There is an algorithm that finds
the optimal solution, provided as an option:

.. ghc-flag:: -foptimal-applicative-do
1214 1215 1216 1217
    :shortdesc: Use a slower but better algorithm for ApplicativeDo
    :type: dynamic
    :reverse: -fno-optimal-applicative-do
    :category: optimization
Simon Marlow's avatar
Simon Marlow committed
1218 1219 1220 1221 1222 1223 1224 1225 1226

    :since: 8.0.1

    Enables an alternative algorithm for choosing where to use ``<*>``
    in conjunction with the ``ApplicativeDo`` language extension.
    This algorithm always finds the optimal solution, but it is
    expensive: ``O(n^3)``, so this option can lead to long compile
    times when there are very large ``do`` expressions (over 100
    statements).  The default ``ApplicativeDo`` algorithm is ``O(n^2)``.
1227