glasgow_exts.rst 559 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
151
152
153
154
155
156
157
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
158
floating point. In contrast, the kind ``*`` is actually just a synonym
159
for ``TYPE 'PtrRepLifted``. More details of the ``TYPE`` mechanisms appear in
160
161
the `section on runtime representation polymorphism <#runtime-rep>`__.

162
Given that ``Int#``'s kind is not ``*``, it then it follows that
163
164
165
166
167
168
169
170
``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:
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186

-  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
187
188
189
   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:
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211

   ::

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

212
213
.. extension:: UnboxedTuples
    :shortdesc: Enable the use of unboxed tuple syntax.
214
215

    :since: 6.8.1
216

217
218

Unboxed tuples aren't really exported by ``GHC.Exts``; they are a
219
syntactic extension (:extension:`UnboxedTuples`). An
220
unboxed tuple looks like this: ::
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
259
260
261
262
263
264
265
266
267
268
269

    (# 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.

270
271
272
273
274
.. _unboxed-sums:

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

275
276
.. extension:: UnboxedSums
    :shortdesc: Enable unboxed sums.
277
278

    :since: 8.2.1
279
280
281
282
283
284
285
286

    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 #)

287
288
where ``t_1`` ... ``t_N`` are types (which can be unlifted, including unboxed
tuples and sums).
289
290
291
292
293

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

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

294
295
The term level syntax is similar. Leading and preceding bars (`|`) indicate which
alternative it is. Here are two terms of the type shown above: ::
296
297
298
299
300

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

    (# | True #) -- second alternative

301
The pattern syntax reflects the term syntax: ::
302
303
304
305
306
307
308
309

    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
310
311
312
313
314
315
316
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:
317
318
319
320
321

- 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
322
323
324
  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.
325
326

- Layout fields are then overlapped so that the final layout will be as compact
327
  as possible. For example, suppose we have the unboxed sum: ::
328

329
330
    (# (# Word32#, String, Float# #)
    |  (# Float#, Float#, Maybe Int #) #)
331

332
  The final layout will be something like ::
333
334
335

    Int32, Float32, Float32, Word32, Pointer

336
337
338
339
340
341
  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.
342

343
344
  As another example, this is the layout for the unboxed version of ``Maybe a``
  type, ``(# (# #) | a #)``: ::
345
346
347

    Int32, Pointer

348
349
350
351
352
353
354
355
356
357
358
359
  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).
360

361
362
363
364
365
366
367
368
369
370
.. _syntax-extns:

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

.. _unicode-syntax:

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

371
372
.. extension:: UnicodeSyntax
    :shortdesc: Enable unicode syntax.
373
374

    :since: 6.8.1
375
376
377
378

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

379
The language extension :extension:`UnicodeSyntax` enables
380
381
382
Unicode characters to be used to stand for certain ASCII character
sequences. The following alternatives are provided:

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
+--------------+---------------+-------------+-----------------------------------------+
| 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 |
+--------------+---------------+-------------+-----------------------------------------+
415
416
417
418
419
420

.. _magic-hash:

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

421
422
.. extension:: MagicHash
    :shortdesc: Allow ``#`` as a postfix modifier on identifiers.
423
424

    :since: 6.8.1
425

426
    Enables the use of the hash character (``#``) as an identifier suffix.
427

428
The language extension :extension:`MagicHash` allows ``#`` as a postfix modifier
429
to identifiers. Thus, ``x#`` is a valid variable, and ``T#`` is a valid type
430
431
432
433
434
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
435
variables. Nor does the :extension:`MagicHash` extension bring anything into
436
scope. For example, to bring ``Int#`` into scope you must import
437
``GHC.Prim`` (see :ref:`primitives`); the :extension:`MagicHash` extension then
438
439
440
441
442
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``.

443
The :extension:`MagicHash` also enables some new forms of literals (see
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
: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
-----------------

466
467
.. extension:: NegativeLiterals
    :shortdesc: Enable support for negative literals.
468

469
470
    :since: 7.8.1

471
472
    Enable the use of un-parenthesized negative numeric literals.

473
474
The literal ``-123`` is, according to Haskell98 and Haskell 2010,
desugared as ``negate (fromInteger 123)``. The language extension
475
:extension:`NegativeLiterals` means that it is instead desugared as
476
477
478
479
480
481
482
483
484
485
486
487
``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
-----------------------------------

488
.. extension:: NumDecimals
489
    :shortdesc: Enable support for 'fractional' integer literals.
490

491
492
    :since: 7.8.1

493
494
    Allow the use of floating-point literal syntax for integral types.

495
496
497
Haskell 2010 and Haskell 98 define floating literals with the syntax
``1.2e6``. These literals have the type ``Fractional a => a``.

498
The language extension :extension:`NumDecimals` allows you to also use the
499
500
501
502
503
504
505
506
floating literal syntax for instances of ``Integral``, and have values
like ``(1.2e6 :: Num a => a)``

.. _binary-literals:

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

507
508
.. extension:: BinaryLiterals
    :shortdesc: Enable support for binary literals.
509

510
511
    :since: 7.10.1

512
513
    Allow the use of binary notation in integer literals.

514
515
516
517
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``).

518
The language extension :extension:`BinaryLiterals` adds support for expressing
519
520
integer literals in binary notation with the prefix ``0b`` or ``0B``. For
instance, the binary integer literal ``0b11001001`` will be desugared into
521
``fromInteger 201`` when :extension:`BinaryLiterals` is enabled.
522

523
524
525
526
527
528
.. _hex-float-literals:

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

.. ghc-flag:: -XHexFloatLiterals
Douglas Wilson's avatar
Douglas Wilson committed
529
    :shortdesc: Enable support for :ref:`hexadecimal floating point literals <hex-float-literals>`.
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
    :type: dynamic
    :reverse: -XNoHexFloatLIterals
    :category:

    :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
powers of 16, while the ones to the right correspond to negaitve ones.

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``




567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
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
640
641
642
643
644
645
646
647
648
649
650
651
652
653
.. _numeric-underscores:

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

.. ghc-flag:: -XNumericUnderscores
    :shortdesc: Enable support for :ref:`numeric underscores <numeric-underscores>`.
    :type: dynamic
    :reverse: -XNoNumericUnderscores
    :category:

    :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.

The language extension :ghc-flag:`-XNumericUnderscores` adds support for expressing
underscores in numeric literals.
For instance, the numeric literal ``1_000_000`` will be parsed into
``1000000`` when :ghc-flag:`-XNumericUnderscores` is enabled.
That is, underscores in numeric literals are ignored when
:ghc-flag:`-XNumericUnderscores` is enabled.
See also :ghc-ticket:`14473`.

For example: ::

    -- 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

About validity: ::

    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

654
655
656
657
658
.. _pattern-guards:

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

659
660
661
.. extension:: NoPatternGuards
    :shortdesc: Disable pattern guards.
        Implied by :extension:`Haskell98`.
662

663
    :implied by: :extension:`Haskell98`
664
665
    :since: 6.8.1

666
Disable `pattern guards
667
<http://www.haskell.org/onlinereport/haskell2010/haskellch3.html#x8-460003.13>`__.
668

669
670
671
672
673
.. _view-patterns:

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

674
675
.. extension:: ViewPatterns
    :shortdesc: Enable view patterns.
676
677

    :since: 6.10.1
678
679
680

    Allow use of view pattern syntax.

681
View patterns are enabled by the language extension :extension:`ViewPatterns`. More
682
683
684
685
686
687
688
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
689
language as follows: ::
690
691
692
693
694
695
696
697
698
699
700
701

    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
702
Without view patterns, using this signature is a little inconvenient: ::
703
704
705
706
707
708
709
710
711
712
713

    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
714
matching against the result: ::
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735

    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
736
   :ref:`pattern-guards` can be written using view patterns as follows: ::
737
738
739
740
741
742
743

       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
744
      view pattern expression are in scope. For example: ::
745
746
747
748
749
750

          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
751
      in later arguments: ::
752
753
754
755
756
757
758
759
760
761
762

          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
763
      allowed: ::
764
765
766
767
768
769
770
771
772
773

          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
774
   Report <http://www.haskell.org/onlinereport/>`__, add the following: ::
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809

       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
810
.. _n-k-patterns:
811

Simon Peyton Jones's avatar
Simon Peyton Jones committed
812
813
n+k patterns
------------
814

815
.. extension:: NPlusKPatterns
816
    :shortdesc: Enable support for ``n+k`` patterns.
817
        Implied by :extension:`Haskell98`.
818

819
    :implied by: :extension:`Haskell98`
820
    :since: 6.12.1
Ben Gamari's avatar
Ben Gamari committed
821

Simon Peyton Jones's avatar
Simon Peyton Jones committed
822
    Enable use of ``n+k`` patterns.
823

Simon Peyton Jones's avatar
Simon Peyton Jones committed
824
.. _recursive-do-notation:
825

Simon Peyton Jones's avatar
Simon Peyton Jones committed
826
827
The recursive do-notation
-------------------------
828

829
830
.. extension:: RecursiveDo
    :shortdesc: Enable recursive do (mdo) notation.
831
832

    :since: 6.8.1
833

Simon Peyton Jones's avatar
Simon Peyton Jones committed
834
    Allow the use of recursive ``do`` notation.
835

Simon Peyton Jones's avatar
Simon Peyton Jones committed
836
837
838
839
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.
840

Simon Peyton Jones's avatar
Simon Peyton Jones committed
841
842
843
844
845
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: ::
846

Simon Peyton Jones's avatar
Simon Peyton Jones committed
847
848
    class Monad m => MonadFix m where
       mfix :: (a -> m a) -> m a
849

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

Simon Peyton Jones's avatar
Simon Peyton Jones committed
855
856
For monads that do belong to the ``MonadFix`` class, GHC provides an
extended version of the do-notation that allows recursive bindings. The
857
:extension:`RecursiveDo` (language pragma: ``RecursiveDo``) provides the
Simon Peyton Jones's avatar
Simon Peyton Jones committed
858
859
860
861
862
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*.
863

Simon Peyton Jones's avatar
Simon Peyton Jones committed
864
Here is a simple (albeit contrived) example:
865

Simon Peyton Jones's avatar
Simon Peyton Jones committed
866
::
867

Simon Peyton Jones's avatar
Simon Peyton Jones committed
868
869
870
    {-# LANGUAGE RecursiveDo #-}
    justOnes = mdo { xs <- Just (1:xs)
                   ; return (map negate xs) }
871

Simon Peyton Jones's avatar
Simon Peyton Jones committed
872
or equivalently
873

Simon Peyton Jones's avatar
Simon Peyton Jones committed
874
::
875

Simon Peyton Jones's avatar
Simon Peyton Jones committed
876
877
878
    {-# LANGUAGE RecursiveDo #-}
    justOnes = do { rec { xs <- Just (1:xs) }
                  ; return (map negate xs) }
879

Simon Peyton Jones's avatar
Simon Peyton Jones committed
880
As you can guess ``justOnes`` will evaluate to ``Just [-1,-1,-1,...``.
881

Simon Peyton Jones's avatar
Simon Peyton Jones committed
882
883
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
884
Haskell <http://leventerkok.github.io/papers/recdo.pdf>`__, which
Simon Peyton Jones's avatar
Simon Peyton Jones committed
885
in turn is based on the work `Value Recursion in Monadic
niteria's avatar
niteria committed
886
Computations <http://leventerkok.github.io/papers/erkok-thesis.pdf>`__.
Simon Peyton Jones's avatar
Simon Peyton Jones committed
887
888
Furthermore, GHC extends the syntax described in the former paper with a
lower level syntax flagged by the ``rec`` keyword, as we describe next.
889

Simon Peyton Jones's avatar
Simon Peyton Jones committed
890
891
Recursive binding groups
~~~~~~~~~~~~~~~~~~~~~~~~
892

893
The extension :extension:`RecursiveDo` also introduces a new keyword ``rec``, which
Simon Peyton Jones's avatar
Simon Peyton Jones committed
894
895
896
897
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
898

Simon Peyton Jones's avatar
Simon Peyton Jones committed
899
::
900

Simon Peyton Jones's avatar
Simon Peyton Jones committed
901
902
903
904
        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) }
905

Simon Peyton Jones's avatar
Simon Peyton Jones committed
906
907
908
909
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.)
910

Simon Peyton Jones's avatar
Simon Peyton Jones committed
911
912
913
914
915
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:
916
917
918

::

Simon Peyton Jones's avatar
Simon Peyton Jones committed
919
920
921
    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) })
922

Simon Peyton Jones's avatar
Simon Peyton Jones committed
923
924
925
As usual, the meta-variables ``b``, ``c`` etc., can be arbitrary
patterns. In general, the statement ``rec ss`` is desugared to the
statement
926
927
928

::

Simon Peyton Jones's avatar
Simon Peyton Jones committed
929
    vs <- mfix (\ ~vs -> do { ss; return vs })
930

Simon Peyton Jones's avatar
Simon Peyton Jones committed
931
where ``vs`` is a tuple of the variables bound by ``ss``.
932

Simon Peyton Jones's avatar
Simon Peyton Jones committed
933
934
935
936
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.
937

Simon Peyton Jones's avatar
Simon Peyton Jones committed
938
939
The ``mdo`` notation
~~~~~~~~~~~~~~~~~~~~
940

Simon Peyton Jones's avatar
Simon Peyton Jones committed
941
942
943
944
945
946
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
947
Haskell <http://leventerkok.github.io/papers/recdo.pdf>`__.
Simon Peyton Jones's avatar
Simon Peyton Jones committed
948
949
950
951
952
953
954
955
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
956
Computations <http://leventerkok.github.io/papers/erkok-thesis.pdf>`__.)
957

Simon Peyton Jones's avatar
Simon Peyton Jones committed
958
959
960
961
962
963
964
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.
965

Simon Peyton Jones's avatar
Simon Peyton Jones committed
966
The definition is syntactic:
967

Simon Peyton Jones's avatar
Simon Peyton Jones committed
968
-  A generator ⟨g⟩ *depends* on a textually following generator ⟨g'⟩, if
969

Simon Peyton Jones's avatar
Simon Peyton Jones committed
970
   -  ⟨g'⟩ defines a variable that is used by ⟨g⟩, or
971

Simon Peyton Jones's avatar
Simon Peyton Jones committed
972
973
   -  ⟨g'⟩ textually appears between ⟨g⟩ and ⟨g''⟩, where ⟨g⟩ depends on
      ⟨g''⟩.
974

Simon Peyton Jones's avatar
Simon Peyton Jones committed
975
976
977
978
979
-  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.
980

Simon Peyton Jones's avatar
Simon Peyton Jones committed
981
982
983
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.
984

Simon Peyton Jones's avatar
Simon Peyton Jones committed
985
986
Here is an example ``mdo``-expression, and its translation to ``rec``
blocks:
987

Simon Peyton Jones's avatar
Simon Peyton Jones committed
988
::
989

Simon Peyton Jones's avatar
Simon Peyton Jones committed
990
991
992
993
994
995
996
    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 }
997

Simon Peyton Jones's avatar
Simon Peyton Jones committed
998
999
1000
1001
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.
1002

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1003
1004
1005
1006
1007
1008
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.
1009

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1010
Here are some other important points in using the recursive-do notation:
1011

1012
1013
-  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
1014
1015
   ``mdo``-notation, and the use of ``rec`` blocks inside ``do``
   expressions.)
1016

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1017
1018
1019
-  ``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.
1020

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1021
1022
-  If recursive bindings are required for a monad, then that monad must
   be declared an instance of the ``MonadFix`` class.
1023

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1024
1025
1026
1027
1028
-  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).
1029

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1030
1031
1032
1033
-  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.)
1034

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1035
.. _applicative-do:
1036

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1037
1038
Applicative do-notation
-----------------------
1039

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1040
1041
1042
.. index::
   single: Applicative do-notation
   single: do-notation; Applicative
1043

1044
1045
.. extension:: ApplicativeDo
    :shortdesc: Enable Applicative do-notation desugaring
1046

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1047
    :since: 8.0.1
1048

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1049
    Allow use of ``Applicative`` ``do`` notation.
1050

1051
The language option :extension:`ApplicativeDo` enables an alternative translation for
Simon Peyton Jones's avatar
Simon Peyton Jones committed
1052
1053
the do-notation, which uses the operators ``<$>``, ``<*>``, along with ``join``
as far as possible. There are two main reasons for wanting to do this:
1054

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1055
1056
1057
1058
-  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.
1059

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1060
1061
1062
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
1063
:extension:`ApplicativeDo` without fear of breaking your program. There is one pitfall
Simon Peyton Jones's avatar
Simon Peyton Jones committed
1064
to watch out for; see :ref:`applicative-do-pitfall`.
1065

1066
There are no syntactic changes with :extension:`ApplicativeDo`. The only way it shows
Simon Peyton Jones's avatar
Simon Peyton Jones committed
1067
1068
up at the source level is that you can have a ``do`` expression that doesn't
require a ``Monad`` constraint. For example, in GHCi: ::
1069

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1070
1071
1072
1073
    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
1074

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1075
1076
This example only requires ``Functor``, because it is translated into ``(\x ->
not x) <$> m``. A more complex example requires ``Applicative``, ::
1077

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1078
1079
1080
    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
1081

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1082
Here GHC has translated the expression into ::
1083

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1084
    (\x y -> x || y) <$> m 'a' <*> m 'b'
1085

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1086
1087
It is possible to see the actual translation by using :ghc-flag:`-ddump-ds`, but be
warned, the output is quite verbose.
1088

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1089
1090
1091
1092
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: ::
1093

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1094
1095
1096
    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
1097

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1098
1099
Here, ``m x`` depends on the value of ``x`` produced by the first statement, so
the expression cannot be translated using ``<*>``.
1100

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1101
1102
In general, the rule for when a ``do`` statement incurs a ``Monad`` constraint
is as follows. If the do-expression has the following form: ::
1103

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1104
    do p1 <- E1; ...; pn <- En; return E
1105

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1106
where none of the variables defined by ``p1...pn`` are mentioned in ``E1...En``,
1107
and ``p1...pn`` are all variables or lazy patterns,
Simon Peyton Jones's avatar
Simon Peyton Jones committed
1108
then the expression will only require ``Applicative``. Otherwise, the expression
1109
1110
will require ``Monad``. The block may return a pure expression ``E`` depending
upon the results ``p1...pn`` with either ``return`` or ``pure``.
1111

1112
1113
1114
1115
1116
1117
1118
1119
1120
1121
1122
1123
1124
1125
1126
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.
1127

Simon Marlow's avatar
Simon Marlow committed
1128
1129
1130
1131
1132
1133
1134
1135
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
1136
1137
1138
1139
    :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
1140
1141
1142
1143
1144
1145
1146
1147
1148

    :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)``.
1149

1150

1151
1152
1153
1154
1155
1156
1157
1158
1159
1160
1161
1162
1163
1164
1165
1166
1167
1168
1169
1170
1171
1172
1173
1174
1175
1176
1177
1178
1179
1180
1181
1182
1183
1184
1185
.. _applicative-do-strict:

Strict patterns
~~~~~~~~~~~~~~~


A strict pattern match in a bind statement prevents
``ApplicativeDo`` from transforming that statement to use
``Applicative``.  This is because the transformation would change the
semantics by making the expression lazier.

For example, this code will require a ``Monad`` constraint::

    > :t \m -> do { (x:xs) <- m; return x }
    \m -> do { (x:xs) <- m; return x } :: Monad m => m [b] -> m b

but making the pattern match lazy allows it to have a ``Functor`` constraint::

    > :t \m -> do { ~(x:xs) <- m; return x }
    \m -> do { ~(x:xs) <- m; return x } :: Functor f => f [b] -> f b

A "strict pattern match" is any pattern match that can fail.  For
example, ``()``, ``(x:xs)``, ``!z``, and ``C x`` are strict patterns,
but ``x`` and ``~(1,2)`` are not.  For the purposes of
``ApplicativeDo``, a pattern match against a ``newtype`` constructor
is considered strict.

When there's a strict pattern match in a sequence of statements,
``ApplicativeDo`` places a ``>>=`` between that statement and the one
that follows it.  The sequence may be transformed to use ``<*>``
elsewhere, but the strict pattern match and the following statement
will always be connected with ``>>=``, to retain the same strictness
semantics as the standard do-notation.  If you don't want this, simply
put a ``~`` on the pattern match to make it lazy.

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1186
.. _applicative-do-pitfall:
1187

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1188
1189
Things to watch out for
~~~~~~~~~~~~~~~~~~~~~~~
1190

1191
Your code should just work as before when :extension:`ApplicativeDo` is enabled,
Simon Peyton Jones's avatar
Simon Peyton Jones committed
1192
1193
1194
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: ::
1195

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1196
1197
    instance Functor MyType where
        fmap f m = do x <- m; return (f x)
1198

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1199
Then applicative desugaring will turn it into ::
1200

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1201
1202
    instance Functor MyType where
        fmap f m = fmap (\x -> f x) m
1203

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1204
1205
And the program will loop at runtime. Similarly, an ``Applicative`` instance
like this ::
1206

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1207
1208
1209
    instance Applicative MyType where
        pure = return
        x <*> y = do f <- x; a <- y; return (f a)
1210

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1211
will result in an infinte loop when ``<*>`` is called.
1212

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1213
1214
1215
1216
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. ::
1217

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1218
1219
    instance Functor MyType where
        fmap f m = m >>= return . f
1220

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1221
1222
1223
    instance Applicative MyType where
        pure = return
        (<*>) = ap
1224
1225


Simon Peyton Jones's avatar
Simon Peyton Jones committed
1226
.. _parallel-list-comprehensions:
1227

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1228
1229
Parallel List Comprehensions
----------------------------
1230

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1231
1232
1233
.. index::
   single: list comprehensions; parallel
   single: parallel list comprehensions
1234

1235
1236
1237
.. extension:: ParallelListComp
    :shortdesc: Enable parallel list comprehensions.
        Implied by :extension:`ParallelArrays`.
1238
1239

    :since: 6.8.1
1240

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1241
    Allow parallel list comprehension syntax.
1242

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1243
1244
1245
1246
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.
1247

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1248
1249
1250
A parallel list comprehension has multiple independent branches of
qualifier lists, each separated by a ``|`` symbol. For example, the
following zips together two lists: ::
1251

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1252
       [ (x, y) | x <- xs | y <- ys ]
1253

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1254
1255
1256
The behaviour of parallel list comprehensions follows that of zip, in
that the resulting list will have the same length as the shortest
branch.
1257

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1258
1259
We can define parallel list comprehensions by translation to regular
comprehensions. Here's the basic idea:
1260

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1261
Given a parallel comprehension of the form: ::
1262

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1263
1264
1265
1266
       [ e | p1 <- e11, p2 <- e12, ...
           | q1 <- e21, q2 <- e22, ...
           ...
       ]
1267

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1268
This will be translated to: ::
1269

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1270
1271
1272
1273
       [ e | ((p1,p2), (q1,q2), ...) <- zipN [(p1,p2) | p1 <- e11, p2 <- e12, ...]
                                             [(q1,q2) | q1 <- e21, q2 <- e22, ...]
                                             ...
       ]
1274

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1275
where ``zipN`` is the appropriate zip for the given number of branches.
1276

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1277
.. _generalised-list-comprehensions:
1278

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1279
1280
Generalised (SQL-like) List Comprehensions
------------------------------------------
1281

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1282
1283
1284
1285
1286
.. index::
   single: list comprehensions; generalised
   single: extended list comprehensions
   single: group
   single: SQL
1287

1288
1289
.. extension:: TransformListComp
    :shortdesc: Enable generalised list comprehensions.
1290
1291

    :since: 6.10.1
1292

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1293
1294
    Allow use of generalised list (SQL-like) comprehension syntax. This
    introduces the ``group``, ``by``, and ``using`` keywords.
1295

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1296
1297
1298
1299
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
1300
"group by" <https://www.microsoft.com/en-us/research/wp-content/uploads/2007/09/list-comp.pdf>`__,
Simon Peyton Jones's avatar
Simon Peyton Jones committed
1301
except that the syntax we use differs slightly from the paper.
1302

1303
The extension is enabled with the extension :extension:`TransformListComp`.
1304

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1305
Here is an example:
1306

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1307
::
1308

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1309
1310
1311
1312
1313
    employees = [ ("Simon", "MS", 80)
                , ("Erik", "MS", 100)
                , ("Phil", "Ed", 40)
                , ("Gordon", "Ed", 45)
                , ("Paul", "Yale", 60) ]
1314

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1315
1316
1317
1318
1319
    output = [ (the dept, sum salary)
             | (name, dept, salary) <- employees
             , then group by dept using groupWith
             , then sortWith by (sum salary)
             , then take 5 ]
1320

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1321
In this example, the list ``output`` would take on the value:
1322

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1323
::
1324

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1325
    [("Yale", 60), ("Ed", 85), ("MS", 180)]
1326

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1327
1328
1329
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``.)
1330

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1331
1332
There are five new forms of comprehension qualifier, all introduced by
the (existing) keyword ``then``:
1333

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1334
-  ::
1335

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1336
       then f
1337

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1338
1339
1340
1341
1342
1343
1344
1345
1346
   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
   .
-  ::
1347

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1348
       then f by e
1349

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1350
1351
1352
1353
1354
1355
   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.
1356

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1357
1358
1359
   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.
1360

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1361
-  ::
1362

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1363
       then group by e using f
1364

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1365
1366
1367
1368
1369
1370
1371
1372
1373
1374
   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:
1375

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1376
   ::
1377

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1378
1379
1380
       -- 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)
1381

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1382
1383
1384
1385
       output = [ (the x, y)
       | x <- ([1..3] ++ [1..2])
       , y <- [4..6]
       , then group by x using groupRuns ]
1386

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1387
   This results in the variable ``output`` taking on the value below:
1388

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1389
   ::
1390

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1391
       [(1, [4, 5, 6]), (2, [4, 5, 6]), (3, [4, 5, 6]), (1, [4, 5, 6]), (2, [4, 5, 6])]
1392

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1393
1394
1395
1396
   Note that we have used the ``the`` function to change the type of x
   from a list to its original numeric type. The variable y, in
   contrast, is left unchanged from the list form introduced by the
   grouping.
1397