glasgow_exts.rst 578 KB
Newer Older
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
.. index::
   single: language, GHC extensions

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

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

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

.. _options-language:

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

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

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

Language options can be controlled in two ways:

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

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

47
48
49
50
GHC supports these language options:

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

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

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

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

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

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

.. _primitives:

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

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

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

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

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

.. _glasgow-unboxed:

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

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

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

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

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

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

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

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

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

There are some restrictions on the use of primitive types:
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187

-  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
188
189
190
   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:
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212

   ::

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

213
214
.. extension:: UnboxedTuples
    :shortdesc: Enable the use of unboxed tuple syntax.
215
216

    :since: 6.8.1
217

218
219

Unboxed tuples aren't really exported by ``GHC.Exts``; they are a
220
syntactic extension (:extension:`UnboxedTuples`). An
221
unboxed tuple looks like this: ::
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
270

    (# 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.

271
272
273
274
275
.. _unboxed-sums:

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

276
277
.. extension:: UnboxedSums
    :shortdesc: Enable unboxed sums.
278
279

    :since: 8.2.1
280
281
282
283
284
285
286
287

    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 #)

288
289
where ``t_1`` ... ``t_N`` are types (which can be unlifted, including unboxed
tuples and sums).
290
291
292
293
294

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

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

295
296
The term level syntax is similar. Leading and preceding bars (`|`) indicate which
alternative it is. Here are two terms of the type shown above: ::
297
298
299
300
301

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

    (# | True #) -- second alternative

302
The pattern syntax reflects the term syntax: ::
303
304
305
306
307
308
309
310

    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
311
312
313
314
315
316
317
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:
318
319
320
321
322

- 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
323
324
325
  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.
326
327

- Layout fields are then overlapped so that the final layout will be as compact
328
  as possible. For example, suppose we have the unboxed sum: ::
329

330
331
    (# (# Word32#, String, Float# #)
    |  (# Float#, Float#, Maybe Int #) #)
332

333
  The final layout will be something like ::
334
335
336

    Int32, Float32, Float32, Word32, Pointer

337
338
339
340
341
342
  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.
343

344
345
  As another example, this is the layout for the unboxed version of ``Maybe a``
  type, ``(# (# #) | a #)``: ::
346
347
348

    Int32, Pointer

349
350
351
352
353
354
355
356
357
358
359
360
  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).
361

362
363
364
365
366
367
368
369
370
371
.. _syntax-extns:

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

.. _unicode-syntax:

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

372
373
.. extension:: UnicodeSyntax
    :shortdesc: Enable unicode syntax.
374
375

    :since: 6.8.1
376
377
378
379

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

380
The language extension :extension:`UnicodeSyntax` enables
381
382
383
Unicode characters to be used to stand for certain ASCII character
sequences. The following alternatives are provided:

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
415
+--------------+---------------+-------------+-----------------------------------------+
| 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 |
+--------------+---------------+-------------+-----------------------------------------+
416
417
418
419
420
421

.. _magic-hash:

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

422
423
.. extension:: MagicHash
    :shortdesc: Allow ``#`` as a postfix modifier on identifiers.
424
425

    :since: 6.8.1
426

427
    Enables the use of the hash character (``#``) as an identifier suffix.
428

429
The language extension :extension:`MagicHash` allows ``#`` as a postfix modifier
430
to identifiers. Thus, ``x#`` is a valid variable, and ``T#`` is a valid type
431
432
433
434
435
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
436
variables. Nor does the :extension:`MagicHash` extension bring anything into
437
scope. For example, to bring ``Int#`` into scope you must import
438
``GHC.Prim`` (see :ref:`primitives`); the :extension:`MagicHash` extension then
439
440
441
442
443
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``.

444
The :extension:`MagicHash` also enables some new forms of literals (see
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
: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
-----------------

467
468
.. extension:: NegativeLiterals
    :shortdesc: Enable support for negative literals.
469

470
471
    :since: 7.8.1

472
473
    Enable the use of un-parenthesized negative numeric literals.

474
475
The literal ``-123`` is, according to Haskell98 and Haskell 2010,
desugared as ``negate (fromInteger 123)``. The language extension
476
:extension:`NegativeLiterals` means that it is instead desugared as
477
478
479
480
481
482
483
484
485
486
487
488
``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
-----------------------------------

489
.. extension:: NumDecimals
490
    :shortdesc: Enable support for 'fractional' integer literals.
491

492
493
    :since: 7.8.1

494
495
    Allow the use of floating-point literal syntax for integral types.

496
497
498
Haskell 2010 and Haskell 98 define floating literals with the syntax
``1.2e6``. These literals have the type ``Fractional a => a``.

499
The language extension :extension:`NumDecimals` allows you to also use the
500
501
502
503
504
505
506
507
floating literal syntax for instances of ``Integral``, and have values
like ``(1.2e6 :: Num a => a)``

.. _binary-literals:

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

508
509
.. extension:: BinaryLiterals
    :shortdesc: Enable support for binary literals.
510

511
512
    :since: 7.10.1

513
514
    Allow the use of binary notation in integer literals.

515
516
517
518
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``).

519
The language extension :extension:`BinaryLiterals` adds support for expressing
520
521
integer literals in binary notation with the prefix ``0b`` or ``0B``. For
instance, the binary integer literal ``0b11001001`` will be desugared into
522
``fromInteger 201`` when :extension:`BinaryLiterals` is enabled.
523

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

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

529
.. extension:: HexFloatLiterals
Douglas Wilson's avatar
Douglas Wilson committed
530
    :shortdesc: Enable support for :ref:`hexadecimal floating point literals <hex-float-literals>`.
531
532
533
534
535
536
537
538
539
540
541
542

    :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
543
powers of 16, while the ones to the right correspond to negative ones.
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564

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``




565
566
567
568
569
.. _numeric-underscores:

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

570
.. extension:: NumericUnderscores
571
572
573
574
575
576
577
578
579
    :shortdesc: Enable support for :ref:`numeric underscores <numeric-underscores>`.

    :since: 8.6.1

    Allow the use of underscores in numeric literals.

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

580
The language extension :extension:`NumericUnderscores` adds support for expressing
581
582
underscores in numeric literals.
For instance, the numeric literal ``1_000_000`` will be parsed into
583
``1000000`` when :extension:`NumericUnderscores` is enabled.
584
That is, underscores in numeric literals are ignored when
585
:extension:`NumericUnderscores` is enabled.
586
587
See also :ghc-ticket:`14473`.

588
589
590
For example:

.. code-block:: none
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

    -- 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

622
623
624
About validity:

.. code-block:: none
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

    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

653
654
655
656
657
.. _pattern-guards:

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

658
659
660
.. extension:: NoPatternGuards
    :shortdesc: Disable pattern guards.
        Implied by :extension:`Haskell98`.
661

662
    :implied by: :extension:`Haskell98`
663
664
    :since: 6.8.1

665
Disable `pattern guards
666
<http://www.haskell.org/onlinereport/haskell2010/haskellch3.html#x8-460003.13>`__.
667

668
669
670
671
672
.. _view-patterns:

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

673
674
.. extension:: ViewPatterns
    :shortdesc: Enable view patterns.
675
676

    :since: 6.10.1
677
678
679

    Allow use of view pattern syntax.

680
View patterns are enabled by the language extension :extension:`ViewPatterns`. More
681
682
683
684
685
686
687
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
688
language as follows: ::
689
690
691
692
693
694
695
696
697
698
699
700

    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
701
Without view patterns, using this signature is a little inconvenient: ::
702
703
704
705
706
707
708
709
710
711
712

    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
713
matching against the result: ::
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734

    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
735
   :ref:`pattern-guards` can be written using view patterns as follows: ::
736
737
738
739
740
741
742

       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
743
      view pattern expression are in scope. For example: ::
744
745
746
747
748
749

          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
750
      in later arguments: ::
751
752
753
754
755
756
757
758
759
760
761

          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
762
      allowed: ::
763
764
765
766
767
768
769
770
771
772

          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
773
   Report <http://www.haskell.org/onlinereport/>`__, add the following: ::
774
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

       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
809
.. _n-k-patterns:
810

Simon Peyton Jones's avatar
Simon Peyton Jones committed
811
812
n+k patterns
------------
813

814
.. extension:: NPlusKPatterns
815
    :shortdesc: Enable support for ``n+k`` patterns.
816
        Implied by :extension:`Haskell98`.
817

818
    :implied by: :extension:`Haskell98`
819
    :since: 6.12.1
Ben Gamari's avatar
Ben Gamari committed
820

Simon Peyton Jones's avatar
Simon Peyton Jones committed
821
    Enable use of ``n+k`` patterns.
822

Simon Peyton Jones's avatar
Simon Peyton Jones committed
823
.. _recursive-do-notation:
824

Simon Peyton Jones's avatar
Simon Peyton Jones committed
825
826
The recursive do-notation
-------------------------
827

828
829
.. extension:: RecursiveDo
    :shortdesc: Enable recursive do (mdo) notation.
830
831

    :since: 6.8.1
832

Simon Peyton Jones's avatar
Simon Peyton Jones committed
833
    Allow the use of recursive ``do`` notation.
834

Simon Peyton Jones's avatar
Simon Peyton Jones committed
835
836
837
838
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.
839

Simon Peyton Jones's avatar
Simon Peyton Jones committed
840
841
842
843
844
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: ::
845

Simon Peyton Jones's avatar
Simon Peyton Jones committed
846
847
    class Monad m => MonadFix m where
       mfix :: (a -> m a) -> m a
848

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

Simon Peyton Jones's avatar
Simon Peyton Jones committed
854
855
For monads that do belong to the ``MonadFix`` class, GHC provides an
extended version of the do-notation that allows recursive bindings. The
856
:extension:`RecursiveDo` (language pragma: ``RecursiveDo``) provides the
Simon Peyton Jones's avatar
Simon Peyton Jones committed
857
858
859
860
861
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*.
862

Simon Peyton Jones's avatar
Simon Peyton Jones committed
863
Here is a simple (albeit contrived) example:
864

Simon Peyton Jones's avatar
Simon Peyton Jones committed
865
::
866

Simon Peyton Jones's avatar
Simon Peyton Jones committed
867
868
869
    {-# LANGUAGE RecursiveDo #-}
    justOnes = mdo { xs <- Just (1:xs)
                   ; return (map negate xs) }
870

Simon Peyton Jones's avatar
Simon Peyton Jones committed
871
or equivalently
872

Simon Peyton Jones's avatar
Simon Peyton Jones committed
873
::
874

Simon Peyton Jones's avatar
Simon Peyton Jones committed
875
876
877
    {-# LANGUAGE RecursiveDo #-}
    justOnes = do { rec { xs <- Just (1:xs) }
                  ; return (map negate xs) }
878

Simon Peyton Jones's avatar
Simon Peyton Jones committed
879
As you can guess ``justOnes`` will evaluate to ``Just [-1,-1,-1,...``.
880

Simon Peyton Jones's avatar
Simon Peyton Jones committed
881
882
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
883
Haskell <http://leventerkok.github.io/papers/recdo.pdf>`__, which
Simon Peyton Jones's avatar
Simon Peyton Jones committed
884
in turn is based on the work `Value Recursion in Monadic
niteria's avatar
niteria committed
885
Computations <http://leventerkok.github.io/papers/erkok-thesis.pdf>`__.
Simon Peyton Jones's avatar
Simon Peyton Jones committed
886
887
Furthermore, GHC extends the syntax described in the former paper with a
lower level syntax flagged by the ``rec`` keyword, as we describe next.
888

Simon Peyton Jones's avatar
Simon Peyton Jones committed
889
890
Recursive binding groups
~~~~~~~~~~~~~~~~~~~~~~~~
891

892
The extension :extension:`RecursiveDo` also introduces a new keyword ``rec``, which
Simon Peyton Jones's avatar
Simon Peyton Jones committed
893
894
895
896
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
897

Simon Peyton Jones's avatar
Simon Peyton Jones committed
898
::
899

Simon Peyton Jones's avatar
Simon Peyton Jones committed
900
901
902
903
        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) }
904

Simon Peyton Jones's avatar
Simon Peyton Jones committed
905
906
907
908
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.)
909

Simon Peyton Jones's avatar
Simon Peyton Jones committed
910
911
912
913
914
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:
915
916
917

::

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

Simon Peyton Jones's avatar
Simon Peyton Jones committed
922
923
924
As usual, the meta-variables ``b``, ``c`` etc., can be arbitrary
patterns. In general, the statement ``rec ss`` is desugared to the
statement
925
926
927

::

Simon Peyton Jones's avatar
Simon Peyton Jones committed
928
    vs <- mfix (\ ~vs -> do { ss; return vs })
929

Simon Peyton Jones's avatar
Simon Peyton Jones committed
930
where ``vs`` is a tuple of the variables bound by ``ss``.
931

Simon Peyton Jones's avatar
Simon Peyton Jones committed
932
933
934
935
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.
936

Simon Peyton Jones's avatar
Simon Peyton Jones committed
937
938
The ``mdo`` notation
~~~~~~~~~~~~~~~~~~~~
939

Simon Peyton Jones's avatar
Simon Peyton Jones committed
940
941
942
943
944
945
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
946
Haskell <http://leventerkok.github.io/papers/recdo.pdf>`__.
Simon Peyton Jones's avatar
Simon Peyton Jones committed
947
948
949
950
951
952
953
954
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
955
Computations <http://leventerkok.github.io/papers/erkok-thesis.pdf>`__.)
956

Simon Peyton Jones's avatar
Simon Peyton Jones committed
957
958
959
960
961
962
963
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.
964

Simon Peyton Jones's avatar
Simon Peyton Jones committed
965
The definition is syntactic:
966

Simon Peyton Jones's avatar
Simon Peyton Jones committed
967
-  A generator ⟨g⟩ *depends* on a textually following generator ⟨g'⟩, if
968

Simon Peyton Jones's avatar
Simon Peyton Jones committed
969
   -  ⟨g'⟩ defines a variable that is used by ⟨g⟩, or
970

Simon Peyton Jones's avatar
Simon Peyton Jones committed
971
972
   -  ⟨g'⟩ textually appears between ⟨g⟩ and ⟨g''⟩, where ⟨g⟩ depends on
      ⟨g''⟩.
973

Simon Peyton Jones's avatar
Simon Peyton Jones committed
974
975
976
977
978
-  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.
979

Simon Peyton Jones's avatar
Simon Peyton Jones committed
980
981
982
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.
983

Simon Peyton Jones's avatar
Simon Peyton Jones committed
984
985
Here is an example ``mdo``-expression, and its translation to ``rec``
blocks:
986

Simon Peyton Jones's avatar
Simon Peyton Jones committed
987
::
988

Simon Peyton Jones's avatar
Simon Peyton Jones committed
989
990
991
992
993
994
995
    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 }
996

Simon Peyton Jones's avatar
Simon Peyton Jones committed
997
998
999
1000
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.
1001

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1002
1003
1004
1005
1006
1007
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.
1008

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1009
Here are some other important points in using the recursive-do notation:
1010

1011
1012
-  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
1013
1014
   ``mdo``-notation, and the use of ``rec`` blocks inside ``do``
   expressions.)
1015

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1016
1017
1018
-  ``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.
1019

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1020
1021
-  If recursive bindings are required for a monad, then that monad must
   be declared an instance of the ``MonadFix`` class.
1022

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1023
1024
1025
1026
1027
-  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).
1028

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1029
1030
1031
1032
-  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.)
1033

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1034
.. _applicative-do:
1035

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1036
1037
Applicative do-notation
-----------------------
1038

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1039
1040
1041
.. index::
   single: Applicative do-notation
   single: do-notation; Applicative
1042

1043
1044
.. extension:: ApplicativeDo
    :shortdesc: Enable Applicative do-notation desugaring
1045

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1046
    :since: 8.0.1
1047

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1048
    Allow use of ``Applicative`` ``do`` notation.
1049

1050
The language option :extension:`ApplicativeDo` enables an alternative translation for
Simon Peyton Jones's avatar
Simon Peyton Jones committed
1051
1052
the do-notation, which uses the operators ``<$>``, ``<*>``, along with ``join``
as far as possible. There are two main reasons for wanting to do this:
1053

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1054
1055
1056
1057
-  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.
1058

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1059
1060
1061
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
1062
:extension:`ApplicativeDo` without fear of breaking your program. There is one pitfall
Simon Peyton Jones's avatar
Simon Peyton Jones committed
1063
to watch out for; see :ref:`applicative-do-pitfall`.
1064

1065
There are no syntactic changes with :extension:`ApplicativeDo`. The only way it shows
Simon Peyton Jones's avatar
Simon Peyton Jones committed
1066
1067
up at the source level is that you can have a ``do`` expression that doesn't
require a ``Monad`` constraint. For example, in GHCi: ::
1068

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1069
1070
1071
1072
    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
1073

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1074
1075
This example only requires ``Functor``, because it is translated into ``(\x ->
not x) <$> m``. A more complex example requires ``Applicative``, ::
1076

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1077
1078
1079
    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
1080

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1081
Here GHC has translated the expression into ::
1082

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1083
    (\x y -> x || y) <$> m 'a' <*> m 'b'
1084

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1085
1086
It is possible to see the actual translation by using :ghc-flag:`-ddump-ds`, but be
warned, the output is quite verbose.
1087

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1088
1089
1090
1091
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: ::
1092

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1093
1094
1095
    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
1096

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1097
1098
Here, ``m x`` depends on the value of ``x`` produced by the first statement, so
the expression cannot be translated using ``<*>``.
1099

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1100
1101
In general, the rule for when a ``do`` statement incurs a ``Monad`` constraint
is as follows. If the do-expression has the following form: ::
1102

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1103
    do p1 <- E1; ...; pn <- En; return E
1104

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1105
where none of the variables defined by ``p1...pn`` are mentioned in ``E1...En``,
1106
and ``p1...pn`` are all variables or lazy patterns,
Simon Peyton Jones's avatar
Simon Peyton Jones committed
1107
then the expression will only require ``Applicative``. Otherwise, the expression
1108
1109
will require ``Monad``. The block may return a pure expression ``E`` depending
upon the results ``p1...pn`` with either ``return`` or ``pure``.
1110

1111
1112
1113
1114
1115
1116
1117
1118
1119
1120
1121
1122
1123
1124
1125
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.
1126

Simon Marlow's avatar
Simon Marlow committed
1127
1128
1129
1130
1131
1132
1133
1134
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
1135
1136
1137
1138
    :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
1139
1140
1141
1142
1143
1144
1145
1146
1147

    :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)``.
1148

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
.. _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
1185
.. _applicative-do-pitfall:
1186

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1187
1188
Things to watch out for
~~~~~~~~~~~~~~~~~~~~~~~
1189

1190
Your code should just work as before when :extension:`ApplicativeDo` is enabled,
Simon Peyton Jones's avatar
Simon Peyton Jones committed
1191
1192
1193
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: ::
1194

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1195
1196
    instance Functor MyType where
        fmap f m = do x <- m; return (f x)
1197

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1198
Then applicative desugaring will turn it into ::
1199

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1200
1201
    instance Functor MyType where
        fmap f m = fmap (\x -> f x) m
1202

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1203
1204
And the program will loop at runtime. Similarly, an ``Applicative`` instance
like this ::
1205

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1206
1207
1208
    instance Applicative MyType where
        pure = return
        x <*> y = do f <- x; a <- y; return (f a)
1209

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1210
will result in an infinte loop when ``<*>`` is called.
1211

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1212
1213
1214
1215
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. ::
1216

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1217
1218
    instance Functor MyType where
        fmap f m = m >>= return . f
1219

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1220
1221
1222
    instance Applicative MyType where
        pure = return
        (<*>) = ap
1223
1224


Simon Peyton Jones's avatar
Simon Peyton Jones committed
1225
.. _parallel-list-comprehensions:
1226

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1227
1228
Parallel List Comprehensions
----------------------------
1229

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1230
1231
1232
.. index::
   single: list comprehensions; parallel
   single: parallel list comprehensions
1233

1234
1235
.. extension:: ParallelListComp
    :shortdesc: Enable parallel list comprehensions.
1236
1237

    :since: 6.8.1
1238

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1239
    Allow parallel list comprehension syntax.
1240

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1241
1242
1243
1244
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.
1245

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1246
1247
1248
A parallel list comprehension has multiple independent branches of
qualifier lists, each separated by a ``|`` symbol. For example, the
following zips together two lists: ::
1249

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1250
       [ (x, y) | x <- xs | y <- ys ]
1251

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1252
1253
1254
The behaviour of parallel list comprehensions follows that of zip, in
that the resulting list will have the same length as the shortest
branch.
1255

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1256
1257
We can define parallel list comprehensions by translation to regular
comprehensions. Here's the basic idea:
1258

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1259
Given a parallel comprehension of the form: ::
1260

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1261
1262
1263
1264
       [ e | p1 <- e11, p2 <- e12, ...
           | q1 <- e21, q2 <- e22, ...
           ...
       ]
1265

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1266
This will be translated to: ::
1267

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1268
1269
1270
1271
       [ e | ((p1,p2), (q1,q2), ...) <- zipN [(p1,p2) | p1 <- e11, p2 <- e12, ...]
                                             [(q1,q2) | q1 <- e21, q2 <- e22, ...]
                                             ...
       ]
1272

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1273
where ``zipN`` is the appropriate zip for the given number of branches.
1274

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1275
.. _generalised-list-comprehensions:
1276

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1277
1278
Generalised (SQL-like) List Comprehensions
------------------------------------------
1279

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1280
1281
1282
1283
1284
.. index::
   single: list comprehensions; generalised
   single: extended list comprehensions
   single: group
   single: SQL
1285

1286
1287
.. extension:: TransformListComp
    :shortdesc: Enable generalised list comprehensions.
1288
1289

    :since: 6.10.1
1290

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1291
1292
    Allow use of generalised list (SQL-like) comprehension syntax. This
    introduces the ``group``, ``by``, and ``using`` keywords.
1293

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1294
1295
1296
1297
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
1298
"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
1299
except that the syntax we use differs slightly from the paper.
1300

1301
The extension is enabled with the extension :extension:`TransformListComp`.
1302

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1303
Here is an example:
1304

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1305
::
1306

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1307
1308
1309
1310
1311
    employees = [ ("Simon", "MS", 80)
                , ("Erik", "MS", 100)
                , ("Phil", "Ed", 40)
                , ("Gordon", "Ed", 45)
                , ("Paul", "Yale", 60) ]
1312

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1313
1314
1315
1316
1317
    output = [ (the dept, sum salary)
             | (name, dept, salary) <- employees
             , then group by dept using groupWith
             , then sortWith by (sum salary)
             , then take 5 ]
1318

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1319
In this example, the list ``output`` would take on the value:
1320

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1321
::
1322

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1323
    [("Yale", 60), ("Ed", 85), ("MS", 180)]
1324

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1325
1326
1327
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``.)
1328

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1329
1330
There are five new forms of comprehension qualifier, all introduced by
the (existing) keyword ``then``:
1331

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1332
-  ::
1333

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1334
       then f
1335

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1336
1337
1338
1339
1340
1341
1342
1343
1344
   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
   .
-  ::
1345

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1346
       then f by e
1347

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1348
1349
1350
1351
1352
1353
   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.
1354

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1355
1356
1357
   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.
1358

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1359
-  ::
1360

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1361
       then group by e using f
1362

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1363
1364
1365
1366
1367
1368
1369
1370
1371
1372
   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:
1373

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1374
   ::
1375

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1376
1377
1378
       -- 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)
1379

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1380
1381
1382
1383
       output = [ (the x, y)
       | x <- ([1..3] ++ [1..2])
       , y <- [4..6]
       , then group by x using groupRuns ]
1384

Simon Peyton Jones's avatar
Simon Peyton Jones committed
1385
   This results in the variable ``output`` taking on the value below:
1386