glasgow_exts.rst 506 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 34 35 36 37 38 39 40 41 42 43 44 45 46 47
.. 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

The language option flags control what variation of the language are
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`).


48 49
Although not recommended, the deprecated :ghc-flag:`-fglasgow-exts` flag enables
a large swath of the extensions supported by GHC at once.
50

51
.. ghc-flag:: -fglasgow-exts
52

53 54 55 56 57 58 59
    The flag ``-fglasgow-exts`` is equivalent to enabling the following extensions:

    .. include:: what_glasgow_exts_does.gen.rst

    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.
60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81

.. _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
:ghc-prim-ref:`detailed online documentation <GHC-Prim.html>`. (This
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
82
you need the :ghc-flag:`-XMagicHash` extension (:ref:`magic-hash`).
83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107

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.
108 109 110
(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.)
111 112 113 114 115 116 117 118
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
119 120
object. Examples include ``Array#``, the type of primitive arrays. Thus,
``Array#`` is an unlifted, boxed type. A
121 122 123 124
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:
125
no unevaluated thunks, no indirections. Nothing can be at the other end
126 127 128 129
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.

130 131
Unboxed type kinds
------------------
132

133
Because unboxed types are represented without the use of pointers, we
134 135
cannot store them in use a polymorphic datatype at an unboxed type.
For example, the ``Just`` node
136 137 138 139 140 141 142 143 144 145 146 147 148
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.

The Haskell Report describes that ``*`` is the kind of ordinary datatypes,
such as ``Int``. Furthermore, type constructors can have kinds with arrows;
for example, ``Maybe`` has kind ``* -> *``. 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
Gabor Greif's avatar
Gabor Greif committed
149
floating point. In contrast, the kind ``*`` is actually just a synonym
150
for ``TYPE 'PtrRepLifted``. More details of the ``TYPE`` mechanisms appear in
151 152
the `section on runtime representation polymorphism <#runtime-rep>`__.

153
Given that ``Int#``'s kind is not ``*``, it then it follows that
154 155 156 157 158 159 160 161
``Maybe Int#`` is disallowed. Similarly, because type variables tend
to be of kind ``*`` (for example, in ``(.) :: (b -> c) -> (a -> b) -> a -> c``,
all the type variables have kind ``*``), 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.

There are some restrictions on the use of primitive types:
162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177

-  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#

-  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
178 179 180
   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:
181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202

   ::

         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
--------------

203
.. ghc-flag:: -XUnboxedTuples
204

205 206 207 208 209
    Enable the use of unboxed tuple syntax.

Unboxed tuples aren't really exported by ``GHC.Exts``; they are a
syntactic extension enabled by the language flag :ghc-flag:`-XUnboxedTuples`. An
unboxed tuple looks like this: ::
210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258

    (# 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.

259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335
.. _unboxed-sums:

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

.. ghc-flag:: -XUnboxedSums

    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 #)

where `t_1` ... `t_N` are types (which can be unlifted, including unboxed tuple
and sums).

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

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

Term level syntax is similar. Leading and preceding bars (`|`) indicate which
alternative it is. Here is two terms of the type shown above: ::

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

    (# | True #) -- second alternative

Pattern syntax reflects the term syntax: ::

    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
of a sum type. Code generator tries to generate as compact layout as possible.
In the best case, size of an unboxed sum is size of its biggest alternative +
one word (for tag). The algorithm for generating memory layout for a sum type
works like this:

- 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
  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.

- Layout fields are then overlapped so that the final layout will be as compact
  as possible. E.g. say two alternatives have these fields: ::

    Word32, String, Float#
    Float#, Float#, Maybe Int

  Final layout will be something like ::

    Int32, Float32, Float32, Word32, Pointer

  First `Int32` is for the tag. It has 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, and second alternative
  needs two `Float32` fields. `Word32` field is for the `Word32` in the first
  alternative. `Pointer` field is shared between `String` and `Maybe Int` values
  of the alternatives.

  In the case of enumeration types (like `Bool`), the unboxed sum layout only
  has an `Int32` field (i.e. the whole thing is represented by an integer).

In the example above, a value of this type is thus represented as 5 values. As
an another example, this is the layout for unboxed version of `Maybe a` type: ::

    Int32, Pointer

The `Pointer` field is not used when tag says that it's `Nothing`. Otherwise
`Pointer` points to the value in `Just`.

336 337 338 339 340 341 342 343 344 345
.. _syntax-extns:

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

.. _unicode-syntax:

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

346 347 348 349 350 351
.. ghc-flag:: -XUnicodeSyntax

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

The language extension :ghc-flag:`-XUnicodeSyntax` enables
352 353 354
Unicode characters to be used to stand for certain ASCII character
sequences. The following alternatives are provided:

355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386
+--------------+---------------+-------------+-----------------------------------------+
| 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 |
+--------------+---------------+-------------+-----------------------------------------+
387 388 389 390 391 392

.. _magic-hash:

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

393 394 395 396 397 398
.. ghc-flag:: -XMagicHash

    Enable the use of the hash character (``#``) as an identifier suffix.

The language extension :ghc-flag:`-XMagicHash` allows ``#`` as a postfix modifier
to identifiers. Thus, ``x#`` is a valid variable, and ``T#`` is a valid type
399 400 401 402 403
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
404
variables. Nor does the :ghc-flag:`-XMagicHash` extension bring anything into
405
scope. For example, to bring ``Int#`` into scope you must import
406
``GHC.Prim`` (see :ref:`primitives`); the :ghc-flag:`-XMagicHash` extension then
407 408 409 410 411
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``.

412
The :ghc-flag:`-XMagicHash` also enables some new forms of literals (see
413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434
: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
-----------------

435 436
.. ghc-flag:: -XNegativeLiterals

437 438
    :since: 7.8.1

439 440
    Enable the use of un-parenthesized negative numeric literals.

441 442
The literal ``-123`` is, according to Haskell98 and Haskell 2010,
desugared as ``negate (fromInteger 123)``. The language extension
443
:ghc-flag:`-XNegativeLiterals` means that it is instead desugared as
444 445 446 447 448 449 450 451 452 453 454 455
``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
-----------------------------------

456 457
.. ghc-flag:: -XNumDecimals

458 459
    :since: 7.8.1

460 461
    Allow the use of floating-point literal syntax for integral types.

462 463 464
Haskell 2010 and Haskell 98 define floating literals with the syntax
``1.2e6``. These literals have the type ``Fractional a => a``.

465
The language extension :ghc-flag:`-XNumDecimals` allows you to also use the
466 467 468 469 470 471 472 473
floating literal syntax for instances of ``Integral``, and have values
like ``(1.2e6 :: Num a => a)``

.. _binary-literals:

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

474 475
.. ghc-flag:: -XBinaryLiterals

476 477
    :since: 7.10.1

478 479
    Allow the use of binary notation in integer literals.

480 481 482 483
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``).

484 485 486 487
The language extension :ghc-flag:`-XBinaryLiterals` adds support for expressing
integer literals in binary notation with the prefix ``0b`` or ``0B``. For
instance, the binary integer literal ``0b11001001`` will be desugared into
``fromInteger 201`` when :ghc-flag:`-XBinaryLiterals` is enabled.
488 489 490 491 492 493

.. _pattern-guards:

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

494
.. ghc-flag:: -XNoPatternGuards
495

496 497 498
    :implied by: :ghc-flag:`-XHaskell98`
    :since: 6.8.1

499
Disable `pattern guards
500
<http://www.haskell.org/onlinereport/haskell2010/haskellch3.html#x8-460003.13>`__.
501

502 503 504 505 506
.. _view-patterns:

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

507 508 509 510 511
.. ghc-flag:: -XViewPatterns

    Allow use of view pattern syntax.

View patterns are enabled by the flag :ghc-flag:`-XViewPatterns`. More
512 513 514 515 516 517 518
information and examples of view patterns can be found on the
:ghc-wiki:`Wiki page <ViewPatterns>`.

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
519
language as follows: ::
520 521 522 523 524 525 526 527 528 529 530 531

    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
532
Without view patterns, using this signature is a little inconvenient: ::
533 534 535 536 537 538 539 540 541 542 543

    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
544
matching against the result: ::
545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565

    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
566
   :ref:`pattern-guards` can be written using view patterns as follows: ::
567 568 569 570 571 572 573

       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
574
      view pattern expression are in scope. For example: ::
575 576 577 578 579 580

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

      Additionally, in function definitions, variables bound by matching
      earlier curried arguments may be used in view pattern expressions
581
      in later arguments: ::
582 583 584 585 586 587 588 589 590 591 592

          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
593
      allowed: ::
594 595 596 597 598 599 600 601 602 603

          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
604
   Report <http://www.haskell.org/onlinereport/>`__, add the following: ::
605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639

       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
640
.. _n-k-patterns:
641

Simon Peyton Jones's avatar
Simon Peyton Jones committed
642 643
n+k patterns
------------
644

Simon Peyton Jones's avatar
Simon Peyton Jones committed
645
.. ghc-flag:: -XNPlusKPatterns
646

647
    :implied by: :ghc-flag:`-XHaskell98`
Ben Gamari's avatar
Ben Gamari committed
648 649
    :since: 6.12

Simon Peyton Jones's avatar
Simon Peyton Jones committed
650
    Enable use of ``n+k`` patterns.
651

Simon Peyton Jones's avatar
Simon Peyton Jones committed
652
.. _recursive-do-notation:
653

Simon Peyton Jones's avatar
Simon Peyton Jones committed
654 655
The recursive do-notation
-------------------------
656

Simon Peyton Jones's avatar
Simon Peyton Jones committed
657
.. ghc-flag:: -XRecursiveDo
658

Simon Peyton Jones's avatar
Simon Peyton Jones committed
659
    Allow the use of recursive ``do`` notation.
660

Simon Peyton Jones's avatar
Simon Peyton Jones committed
661 662 663 664
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.
665

Simon Peyton Jones's avatar
Simon Peyton Jones committed
666 667 668 669 670
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: ::
671

Simon Peyton Jones's avatar
Simon Peyton Jones committed
672 673
    class Monad m => MonadFix m where
       mfix :: (a -> m a) -> m a
674

Simon Peyton Jones's avatar
Simon Peyton Jones committed
675 676 677 678
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.
679

Simon Peyton Jones's avatar
Simon Peyton Jones committed
680 681 682 683 684 685 686 687
For monads that do belong to the ``MonadFix`` class, GHC provides an
extended version of the do-notation that allows recursive bindings. The
:ghc-flag:`-XRecursiveDo` (language pragma: ``RecursiveDo``) provides the
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*.
688

Simon Peyton Jones's avatar
Simon Peyton Jones committed
689
Here is a simple (albeit contrived) example:
690

Simon Peyton Jones's avatar
Simon Peyton Jones committed
691
::
692

Simon Peyton Jones's avatar
Simon Peyton Jones committed
693 694 695
    {-# LANGUAGE RecursiveDo #-}
    justOnes = mdo { xs <- Just (1:xs)
                   ; return (map negate xs) }
696

Simon Peyton Jones's avatar
Simon Peyton Jones committed
697
or equivalently
698

Simon Peyton Jones's avatar
Simon Peyton Jones committed
699
::
700

Simon Peyton Jones's avatar
Simon Peyton Jones committed
701 702 703
    {-# LANGUAGE RecursiveDo #-}
    justOnes = do { rec { xs <- Just (1:xs) }
                  ; return (map negate xs) }
704

Simon Peyton Jones's avatar
Simon Peyton Jones committed
705
As you can guess ``justOnes`` will evaluate to ``Just [-1,-1,-1,...``.
706

Simon Peyton Jones's avatar
Simon Peyton Jones committed
707 708
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
709
Haskell <http://leventerkok.github.io/papers/recdo.pdf>`__, which
Simon Peyton Jones's avatar
Simon Peyton Jones committed
710
in turn is based on the work `Value Recursion in Monadic
niteria's avatar
niteria committed
711
Computations <http://leventerkok.github.io/papers/erkok-thesis.pdf>`__.
Simon Peyton Jones's avatar
Simon Peyton Jones committed
712 713
Furthermore, GHC extends the syntax described in the former paper with a
lower level syntax flagged by the ``rec`` keyword, as we describe next.
714

Simon Peyton Jones's avatar
Simon Peyton Jones committed
715 716
Recursive binding groups
~~~~~~~~~~~~~~~~~~~~~~~~
717

Simon Peyton Jones's avatar
Simon Peyton Jones committed
718 719 720 721 722
The flag :ghc-flag:`-XRecursiveDo` also introduces a new keyword ``rec``, which
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
723

Simon Peyton Jones's avatar
Simon Peyton Jones committed
724
::
725

Simon Peyton Jones's avatar
Simon Peyton Jones committed
726 727 728 729
        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) }
730

Simon Peyton Jones's avatar
Simon Peyton Jones committed
731 732 733 734
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.)
735

Simon Peyton Jones's avatar
Simon Peyton Jones committed
736 737 738 739 740
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:
741 742 743

::

Simon Peyton Jones's avatar
Simon Peyton Jones committed
744 745 746
    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) })
747

Simon Peyton Jones's avatar
Simon Peyton Jones committed
748 749 750
As usual, the meta-variables ``b``, ``c`` etc., can be arbitrary
patterns. In general, the statement ``rec ss`` is desugared to the
statement
751 752 753

::

Simon Peyton Jones's avatar
Simon Peyton Jones committed
754
    vs <- mfix (\ ~vs -> do { ss; return vs })
755

Simon Peyton Jones's avatar
Simon Peyton Jones committed
756
where ``vs`` is a tuple of the variables bound by ``ss``.
757

Simon Peyton Jones's avatar
Simon Peyton Jones committed
758 759 760 761
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.
762

Simon Peyton Jones's avatar
Simon Peyton Jones committed
763 764
The ``mdo`` notation
~~~~~~~~~~~~~~~~~~~~
765

Simon Peyton Jones's avatar
Simon Peyton Jones committed
766 767 768 769 770 771
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
772
Haskell <http://leventerkok.github.io/papers/recdo.pdf>`__.
Simon Peyton Jones's avatar
Simon Peyton Jones committed
773 774 775 776 777 778 779 780
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
781
Computations <http://leventerkok.github.io/papers/erkok-thesis.pdf>`__.)
782

Simon Peyton Jones's avatar
Simon Peyton Jones committed
783 784 785 786 787 788 789
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.
790

Simon Peyton Jones's avatar
Simon Peyton Jones committed
791
The definition is syntactic:
792

Simon Peyton Jones's avatar
Simon Peyton Jones committed
793
-  A generator ⟨g⟩ *depends* on a textually following generator ⟨g'⟩, if
794

Simon Peyton Jones's avatar
Simon Peyton Jones committed
795
   -  ⟨g'⟩ defines a variable that is used by ⟨g⟩, or
796

Simon Peyton Jones's avatar
Simon Peyton Jones committed
797 798
   -  ⟨g'⟩ textually appears between ⟨g⟩ and ⟨g''⟩, where ⟨g⟩ depends on
      ⟨g''⟩.
799

Simon Peyton Jones's avatar
Simon Peyton Jones committed
800 801 802 803 804
-  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.
805

Simon Peyton Jones's avatar
Simon Peyton Jones committed
806 807 808
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.
809

Simon Peyton Jones's avatar
Simon Peyton Jones committed
810 811
Here is an example ``mdo``-expression, and its translation to ``rec``
blocks:
812

Simon Peyton Jones's avatar
Simon Peyton Jones committed
813
::
814

Simon Peyton Jones's avatar
Simon Peyton Jones committed
815 816 817 818 819 820 821
    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 }
822

Simon Peyton Jones's avatar
Simon Peyton Jones committed
823 824 825 826
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.
827

Simon Peyton Jones's avatar
Simon Peyton Jones committed
828 829 830 831 832 833
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.
834

Simon Peyton Jones's avatar
Simon Peyton Jones committed
835
Here are some other important points in using the recursive-do notation:
836

Simon Peyton Jones's avatar
Simon Peyton Jones committed
837 838 839 840
-  It is enabled with the flag :ghc-flag:`-XRecursiveDo`, or the
   ``LANGUAGE RecursiveDo`` pragma. (The same flag enables both
   ``mdo``-notation, and the use of ``rec`` blocks inside ``do``
   expressions.)
841

Simon Peyton Jones's avatar
Simon Peyton Jones committed
842 843 844
-  ``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.
845

Simon Peyton Jones's avatar
Simon Peyton Jones committed
846 847
-  If recursive bindings are required for a monad, then that monad must
   be declared an instance of the ``MonadFix`` class.
848

Simon Peyton Jones's avatar
Simon Peyton Jones committed
849 850 851 852 853
-  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).
854

Simon Peyton Jones's avatar
Simon Peyton Jones committed
855 856 857 858
-  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.)
859

Simon Peyton Jones's avatar
Simon Peyton Jones committed
860
.. _applicative-do:
861

Simon Peyton Jones's avatar
Simon Peyton Jones committed
862 863
Applicative do-notation
-----------------------
864

Simon Peyton Jones's avatar
Simon Peyton Jones committed
865 866 867
.. index::
   single: Applicative do-notation
   single: do-notation; Applicative
868

Simon Peyton Jones's avatar
Simon Peyton Jones committed
869
.. ghc-flag:: -XApplicativeDo
870

Simon Peyton Jones's avatar
Simon Peyton Jones committed
871
    :since: 8.0.1
872

Simon Peyton Jones's avatar
Simon Peyton Jones committed
873
    Allow use of ``Applicative`` ``do`` notation.
874

Simon Peyton Jones's avatar
Simon Peyton Jones committed
875 876 877
The language option :ghc-flag:`-XApplicativeDo` enables an alternative translation for
the do-notation, which uses the operators ``<$>``, ``<*>``, along with ``join``
as far as possible. There are two main reasons for wanting to do this:
878

Simon Peyton Jones's avatar
Simon Peyton Jones committed
879 880 881 882
-  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.
883

Simon Peyton Jones's avatar
Simon Peyton Jones committed
884 885 886 887 888
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
:ghc-flag:`-XApplicativeDo` without fear of breaking your program. There is one pitfall
to watch out for; see :ref:`applicative-do-pitfall`.
889

Simon Peyton Jones's avatar
Simon Peyton Jones committed
890 891 892
There are no syntactic changes with :ghc-flag:`-XApplicativeDo`. The only way it shows
up at the source level is that you can have a ``do`` expression that doesn't
require a ``Monad`` constraint. For example, in GHCi: ::
893

Simon Peyton Jones's avatar
Simon Peyton Jones committed
894 895 896 897
    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
898

Simon Peyton Jones's avatar
Simon Peyton Jones committed
899 900
This example only requires ``Functor``, because it is translated into ``(\x ->
not x) <$> m``. A more complex example requires ``Applicative``, ::
901

Simon Peyton Jones's avatar
Simon Peyton Jones committed
902 903 904
    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
905

Simon Peyton Jones's avatar
Simon Peyton Jones committed
906
Here GHC has translated the expression into ::
907

Simon Peyton Jones's avatar
Simon Peyton Jones committed
908
    (\x y -> x || y) <$> m 'a' <*> m 'b'
909

Simon Peyton Jones's avatar
Simon Peyton Jones committed
910 911
It is possible to see the actual translation by using :ghc-flag:`-ddump-ds`, but be
warned, the output is quite verbose.
912

Simon Peyton Jones's avatar
Simon Peyton Jones committed
913 914 915 916
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: ::
917

Simon Peyton Jones's avatar
Simon Peyton Jones committed
918 919 920
    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
921

Simon Peyton Jones's avatar
Simon Peyton Jones committed
922 923
Here, ``m x`` depends on the value of ``x`` produced by the first statement, so
the expression cannot be translated using ``<*>``.
924

Simon Peyton Jones's avatar
Simon Peyton Jones committed
925 926
In general, the rule for when a ``do`` statement incurs a ``Monad`` constraint
is as follows. If the do-expression has the following form: ::
927

Simon Peyton Jones's avatar
Simon Peyton Jones committed
928
    do p1 <- E1; ...; pn <- En; return E
929

Simon Peyton Jones's avatar
Simon Peyton Jones committed
930 931
where none of the variables defined by ``p1...pn`` are mentioned in ``E1...En``,
then the expression will only require ``Applicative``. Otherwise, the expression
932 933
will require ``Monad``. The block may return a pure expression ``E`` depending
upon the results ``p1...pn`` with either ``return`` or ``pure``.
934

935 936 937 938 939 940 941 942 943 944 945 946 947 948 949
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.
950

Simon Marlow's avatar
Simon Marlow committed
951 952 953 954 955 956 957 958 959 960 961 962 963 964 965 966 967
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

    :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)``.
968

969 970 971 972 973 974 975 976 977 978 979 980 981 982 983 984 985 986 987 988 989 990 991 992 993 994 995 996 997 998 999 1000 1001 1002 1003 1004

.. _applicative-do-existential:

Existential patterns and GADTs
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

Note that when the pattern in a statement matches a constructor with
existential type variables and/or constraints, the transformation that
``ApplicativeDo`` performs may mean that the pattern does not scope
over the statements that follow it.  This is because the rearrangement
happens before the expression is typechecked.  For example, this
program does not typecheck::

    {-# LANGUAGE RankNTypes, GADTs, ApplicativeDo #-}

    data T where A :: forall a . Eq a => a -> T

    test = do
      A x <- undefined
      _ <- return True
      return (x == x)

The reason is that the ``Eq`` constraint that would be brought into
scope from the pattern match ``A x`` is not available when
typechecking the expression ``x == x``, because ``ApplicativeDo`` has
rearranged the expression to look like this::

    test =
      (\x _ -> x == x)
        <$> do A x <- undefined; return x
        <*> return True

Turning off ``ApplicativeDo`` lets the program typecheck.  This is
something to bear in mind when using ``ApplicativeDo`` in combination
with :ref:`existential-quantification` or :ref:`gadt`.

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1005
.. _applicative-do-pitfall:
1006

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1007 1008
Things to watch out for
~~~~~~~~~~~~~~~~~~~~~~~
1009

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1010 1011 1012 1013
Your code should just work as before when :ghc-flag:`-XApplicativeDo` is enabled,
provided you use conventional ``Applicative`` instances. However, if you define
a ``Functor`` or ``Applicative`` instance using do-notation, then it will likely
get turned into an infinite loop by GHC. For example, if you do this: ::
1014

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1015 1016
    instance Functor MyType where
        fmap f m = do x <- m; return (f x)
1017

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1018
Then applicative desugaring will turn it into ::
1019

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1020 1021
    instance Functor MyType where
        fmap f m = fmap (\x -> f x) m
1022

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1023 1024
And the program will loop at runtime. Similarly, an ``Applicative`` instance
like this ::
1025

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1026 1027 1028
    instance Applicative MyType where
        pure = return
        x <*> y = do f <- x; a <- y; return (f a)
1029

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1030
will result in an infinte loop when ``<*>`` is called.
1031

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1032 1033 1034 1035
Just as you wouldn't define a ``Monad`` instance using the do-notation, you
shouldn't define ``Functor`` or ``Applicative`` instance using do-notation (when
using ``ApplicativeDo``) either. The correct way to define these instances in
terms of ``Monad`` is to use the ``Monad`` operations directly, e.g. ::
1036

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1037 1038
    instance Functor MyType where
        fmap f m = m >>= return . f
1039

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1040 1041 1042
    instance Applicative MyType where
        pure = return
        (<*>) = ap
1043 1044


Simon Peyton Jones's avatar
Simon Peyton Jones committed
1045
.. _parallel-list-comprehensions:
1046

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1047 1048
Parallel List Comprehensions
----------------------------
1049

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1050 1051 1052
.. index::
   single: list comprehensions; parallel
   single: parallel list comprehensions
1053

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1054
.. ghc-flag:: -XParallelListComp
1055

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1056
    Allow parallel list comprehension syntax.
1057

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1058 1059 1060 1061
Parallel list comprehensions are a natural extension to list
comprehensions. List comprehensions can be thought of as a nice syntax
for writing maps and filters. Parallel comprehensions extend this to
include the ``zipWith`` family.
1062

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1063 1064 1065
A parallel list comprehension has multiple independent branches of
qualifier lists, each separated by a ``|`` symbol. For example, the
following zips together two lists: ::
1066

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1067
       [ (x, y) | x <- xs | y <- ys ]
1068

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1069 1070 1071
The behaviour of parallel list comprehensions follows that of zip, in
that the resulting list will have the same length as the shortest
branch.
1072

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1073 1074
We can define parallel list comprehensions by translation to regular
comprehensions. Here's the basic idea:
1075

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1076
Given a parallel comprehension of the form: ::
1077

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1078 1079 1080 1081
       [ e | p1 <- e11, p2 <- e12, ...
           | q1 <- e21, q2 <- e22, ...
           ...
       ]
1082

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1083
This will be translated to: ::
1084

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1085 1086 1087 1088
       [ e | ((p1,p2), (q1,q2), ...) <- zipN [(p1,p2) | p1 <- e11, p2 <- e12, ...]
                                             [(q1,q2) | q1 <- e21, q2 <- e22, ...]
                                             ...
       ]
1089

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1090
where ``zipN`` is the appropriate zip for the given number of branches.
1091

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1092
.. _generalised-list-comprehensions:
1093

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1094 1095
Generalised (SQL-like) List Comprehensions
------------------------------------------
1096

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1097 1098 1099 1100 1101
.. index::
   single: list comprehensions; generalised
   single: extended list comprehensions
   single: group
   single: SQL
1102

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1103
.. ghc-flag:: -XTransformListComp
1104

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1105 1106
    Allow use of generalised list (SQL-like) comprehension syntax. This
    introduces the ``group``, ``by``, and ``using`` keywords.
1107

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1108 1109 1110 1111 1112 1113
Generalised list comprehensions are a further enhancement to the list
comprehension syntactic sugar to allow operations such as sorting and
grouping which are familiar from SQL. They are fully described in the
paper `Comprehensive comprehensions: comprehensions with "order by" and
"group by" <http://research.microsoft.com/~simonpj/papers/list-comp>`__,
except that the syntax we use differs slightly from the paper.
1114

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1115
The extension is enabled with the flag :ghc-flag:`-XTransformListComp`.
1116

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1117
Here is an example:
1118

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1119
::
1120

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1121 1122 1123 1124 1125
    employees = [ ("Simon", "MS", 80)
                , ("Erik", "MS", 100)
                , ("Phil", "Ed", 40)
                , ("Gordon", "Ed", 45)
                , ("Paul", "Yale", 60) ]
1126

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1127 1128 1129 1130 1131
    output = [ (the dept, sum salary)
             | (name, dept, salary) <- employees
             , then group by dept using groupWith
             , then sortWith by (sum salary)
             , then take 5 ]
1132

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1133
In this example, the list ``output`` would take on the value:
1134

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1135
::
1136

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1137
    [("Yale", 60), ("Ed", 85), ("MS", 180)]
1138

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1139 1140 1141
There are three new keywords: ``group``, ``by``, and ``using``. (The
functions ``sortWith`` and ``groupWith`` are not keywords; they are
ordinary functions that are exported by ``GHC.Exts``.)
1142

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1143 1144
There are five new forms of comprehension qualifier, all introduced by
the (existing) keyword ``then``:
1145

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1146
-  ::
1147

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1148
       then f
1149

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1150 1151 1152 1153 1154 1155 1156 1157 1158
   This statement requires that
   f
   have the type
   forall a. [a] -> [a]
   . You can see an example of its use in the motivating example, as
   this form is used to apply
   take 5
   .
-  ::
1159

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1160
       then f by e
1161

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1162 1163 1164 1165 1166 1167
   This form is similar to the previous one, but allows you to create a
   function which will be passed as the first argument to f. As a
   consequence f must have the type
   ``forall a. (a -> t) -> [a] -> [a]``. As you can see from the type,
   this function lets f "project out" some information from the elements
   of the list it is transforming.
1168

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1169 1170 1171
   An example is shown in the opening example, where ``sortWith`` is
   supplied with a function that lets it find out the ``sum salary`` for
   any item in the list comprehension it transforms.
1172

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1173
-  ::
1174

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1175
       then group by e using f
1176

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1177 1178 1179 1180 1181 1182 1183 1184 1185 1186
   This is the most general of the grouping-type statements. In this
   form, f is required to have type
   ``forall a. (a -> t) -> [a] -> [[a]]``. As with the ``then f by e``
   case above, the first argument is a function supplied to f by the
   compiler which lets it compute e on every element of the list being
   transformed. However, unlike the non-grouping case, f additionally
   partitions the list into a number of sublists: this means that at
   every point after this statement, binders occurring before it in the
   comprehension refer to *lists* of possible values, not single values.
   To help understand this, let's look at an example:
1187

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1188
   ::
1189

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1190 1191 1192
       -- This works similarly to groupWith in GHC.Exts, but doesn't sort its input first
       groupRuns :: Eq b => (a -> b) -> [a] -> [[a]]
       groupRuns f = groupBy (\x y -> f x == f y)
1193

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1194 1195 1196 1197
       output = [ (the x, y)
       | x <- ([1..3] ++ [1..2])
       , y <- [4..6]
       , then group by x using groupRuns ]
1198