[project @ 2002-04-26 13:34:05 by simonmar]
[packages/old-locale.git] / Text / PrettyPrint / HughesPJ.hs
1 -----------------------------------------------------------------------------
2 -- |
3 -- Module : Text.PrettyPrint.HughesPJ
4 -- Copyright : (c) The University of Glasgow 2001
5 -- License : BSD-style (see the file libraries/core/LICENSE)
6 --
7 -- Maintainer : libraries@haskell.org
8 -- Stability : provisional
9 -- Portability : portable
10 --
11 -- John Hughes's and Simon Peyton Jones's Pretty Printer Combinators
12 --
13 -- Based on "The Design of a Pretty-printing Library"
14 -- in Advanced Functional Programming,
15 -- Johan Jeuring and Erik Meijer (eds), LNCS 925
16 -- http://www.cs.chalmers.se/~rjmh/Papers/pretty.ps
17 --
18 -- Heavily modified by Simon Peyton Jones, Dec 96
19 --
20 -----------------------------------------------------------------------------
21
22 {-
23 Version 3.0 28 May 1997
24 * Cured massive performance bug. If you write
25
26 foldl <> empty (map (text.show) [1..10000])
27
28 you get quadratic behaviour with V2.0. Why? For just the same
29 reason as you get quadratic behaviour with left-associated (++)
30 chains.
31
32 This is really bad news. One thing a pretty-printer abstraction
33 should certainly guarantee is insensivity to associativity. It
34 matters: suddenly GHC's compilation times went up by a factor of
35 100 when I switched to the new pretty printer.
36
37 I fixed it with a bit of a hack (because I wanted to get GHC back
38 on the road). I added two new constructors to the Doc type, Above
39 and Beside:
40
41 <> = Beside
42 $$ = Above
43
44 Then, where I need to get to a "TextBeside" or "NilAbove" form I
45 "force" the Doc to squeeze out these suspended calls to Beside and
46 Above; but in so doing I re-associate. It's quite simple, but I'm
47 not satisfied that I've done the best possible job. I'll send you
48 the code if you are interested.
49
50 * Added new exports:
51 punctuate, hang
52 int, integer, float, double, rational,
53 lparen, rparen, lbrack, rbrack, lbrace, rbrace,
54
55 * fullRender's type signature has changed. Rather than producing a
56 string it now takes an extra couple of arguments that tells it how
57 to glue fragments of output together:
58
59 fullRender :: Mode
60 -> Int -- Line length
61 -> Float -- Ribbons per line
62 -> (TextDetails -> a -> a) -- What to do with text
63 -> a -- What to do at the end
64 -> Doc
65 -> a -- Result
66
67 The "fragments" are encapsulated in the TextDetails data type:
68
69 data TextDetails = Chr Char
70 | Str String
71 | PStr FAST_STRING
72
73 The Chr and Str constructors are obvious enough. The PStr
74 constructor has a packed string (FAST_STRING) inside it. It's
75 generated by using the new "ptext" export.
76
77 An advantage of this new setup is that you can get the renderer to
78 do output directly (by passing in a function of type (TextDetails
79 -> IO () -> IO ()), rather than producing a string that you then
80 print.
81
82
83 Version 2.0 24 April 1997
84 * Made empty into a left unit for <> as well as a right unit;
85 it is also now true that
86 nest k empty = empty
87 which wasn't true before.
88
89 * Fixed an obscure bug in sep that occassionally gave very wierd behaviour
90
91 * Added $+$
92
93 * Corrected and tidied up the laws and invariants
94
95 ======================================================================
96 Relative to John's original paper, there are the following new features:
97
98 1. There's an empty document, "empty". It's a left and right unit for
99 both <> and $$, and anywhere in the argument list for
100 sep, hcat, hsep, vcat, fcat etc.
101
102 It is Really Useful in practice.
103
104 2. There is a paragraph-fill combinator, fsep, that's much like sep,
105 only it keeps fitting things on one line until itc can't fit any more.
106
107 3. Some random useful extra combinators are provided.
108 <+> puts its arguments beside each other with a space between them,
109 unless either argument is empty in which case it returns the other
110
111
112 hcat is a list version of <>
113 hsep is a list version of <+>
114 vcat is a list version of $$
115
116 sep (separate) is either like hsep or like vcat, depending on what fits
117
118 cat is behaves like sep, but it uses <> for horizontal conposition
119 fcat is behaves like fsep, but it uses <> for horizontal conposition
120
121 These new ones do the obvious things:
122 char, semi, comma, colon, space,
123 parens, brackets, braces,
124 quotes, doubleQuotes
125
126 4. The "above" combinator, $$, now overlaps its two arguments if the
127 last line of the top argument stops before the first line of the
128 second begins.
129
130 For example: text "hi" $$ nest 5 "there"
131 lays out as
132 hi there
133 rather than
134 hi
135 there
136
137 There are two places this is really useful
138
139 a) When making labelled blocks, like this:
140 Left -> code for left
141 Right -> code for right
142 LongLongLongLabel ->
143 code for longlonglonglabel
144 The block is on the same line as the label if the label is
145 short, but on the next line otherwise.
146
147 b) When laying out lists like this:
148 [ first
149 , second
150 , third
151 ]
152 which some people like. But if the list fits on one line
153 you want [first, second, third]. You can't do this with
154 John's original combinators, but it's quite easy with the
155 new $$.
156
157 The combinator $+$ gives the original "never-overlap" behaviour.
158
159 5. Several different renderers are provided:
160 * a standard one
161 * one that uses cut-marks to avoid deeply-nested documents
162 simply piling up in the right-hand margin
163 * one that ignores indentation (fewer chars output; good for machines)
164 * one that ignores indentation and newlines (ditto, only more so)
165
166 6. Numerous implementation tidy-ups
167 Use of unboxed data types to speed up the implementation
168 -}
169
170 module Text.PrettyPrint.HughesPJ (
171 Doc, -- Abstract
172 Mode(..), TextDetails(..),
173
174 empty, isEmpty, nest,
175
176 text, char, ptext,
177 int, integer, float, double, rational,
178 parens, brackets, braces, quotes, doubleQuotes,
179 semi, comma, colon, space, equals,
180 lparen, rparen, lbrack, rbrack, lbrace, rbrace,
181
182 (<>), (<+>), hcat, hsep,
183 ($$), ($+$), vcat,
184 sep, cat,
185 fsep, fcat,
186
187 hang, punctuate,
188
189 -- renderStyle, -- Haskell 1.3 only
190 render, fullRender
191 ) where
192
193
194 import Prelude
195
196 infixl 6 <>
197 infixl 6 <+>
198 infixl 5 $$, $+$
199
200 -- ---------------------------------------------------------------------------
201 -- The interface
202
203 -- The primitive Doc values
204
205 empty :: Doc
206 isEmpty :: Doc -> Bool
207 text :: String -> Doc
208 char :: Char -> Doc
209
210 semi, comma, colon, space, equals :: Doc
211 lparen, rparen, lbrack, rbrack, lbrace, rbrace :: Doc
212
213 parens, brackets, braces :: Doc -> Doc
214 quotes, doubleQuotes :: Doc -> Doc
215
216 int :: Int -> Doc
217 integer :: Integer -> Doc
218 float :: Float -> Doc
219 double :: Double -> Doc
220 rational :: Rational -> Doc
221
222
223 -- Combining @Doc@ values
224
225 (<>) :: Doc -> Doc -> Doc -- Beside
226 hcat :: [Doc] -> Doc -- List version of <>
227 (<+>) :: Doc -> Doc -> Doc -- Beside, separated by space
228 hsep :: [Doc] -> Doc -- List version of <+>
229
230 ($$) :: Doc -> Doc -> Doc -- Above; if there is no
231 -- overlap it "dovetails" the two
232 vcat :: [Doc] -> Doc -- List version of $$
233
234 cat :: [Doc] -> Doc -- Either hcat or vcat
235 sep :: [Doc] -> Doc -- Either hsep or vcat
236 fcat :: [Doc] -> Doc -- ``Paragraph fill'' version of cat
237 fsep :: [Doc] -> Doc -- ``Paragraph fill'' version of sep
238
239 nest :: Int -> Doc -> Doc -- Nested
240
241
242 -- GHC-specific ones.
243
244 hang :: Doc -> Int -> Doc -> Doc
245 punctuate :: Doc -> [Doc] -> [Doc] -- punctuate p [d1, ... dn] = [d1 <> p, d2 <> p, ... dn-1 <> p, dn]
246
247
248 -- Displaying @Doc@ values.
249
250 instance Show Doc where
251 showsPrec prec doc cont = showDoc doc cont
252
253 render :: Doc -> String -- Uses default style
254 fullRender :: Mode
255 -> Int -- Line length
256 -> Float -- Ribbons per line
257 -> (TextDetails -> a -> a) -- What to do with text
258 -> a -- What to do at the end
259 -> Doc
260 -> a -- Result
261
262 {- When we start using 1.3
263 renderStyle :: Style -> Doc -> String
264 data Style = Style { lineLength :: Int, -- In chars
265 ribbonsPerLine :: Float, -- Ratio of ribbon length to line length
266 mode :: Mode
267 }
268 style :: Style -- The default style
269 style = Style { lineLength = 100, ribbonsPerLine = 2.5, mode = PageMode }
270 -}
271
272 data Mode = PageMode -- Normal
273 | ZigZagMode -- With zig-zag cuts
274 | LeftMode -- No indentation, infinitely long lines
275 | OneLineMode -- All on one line
276
277
278 -- ---------------------------------------------------------------------------
279 -- The Doc calculus
280
281 -- The Doc combinators satisfy the following laws:
282
283 {-
284 Laws for $$
285 ~~~~~~~~~~~
286 <a1> (x $$ y) $$ z = x $$ (y $$ z)
287 <a2> empty $$ x = x
288 <a3> x $$ empty = x
289
290 ...ditto $+$...
291
292 Laws for <>
293 ~~~~~~~~~~~
294 <b1> (x <> y) <> z = x <> (y <> z)
295 <b2> empty <> x = empty
296 <b3> x <> empty = x
297
298 ...ditto <+>...
299
300 Laws for text
301 ~~~~~~~~~~~~~
302 <t1> text s <> text t = text (s++t)
303 <t2> text "" <> x = x, if x non-empty
304
305 Laws for nest
306 ~~~~~~~~~~~~~
307 <n1> nest 0 x = x
308 <n2> nest k (nest k' x) = nest (k+k') x
309 <n3> nest k (x <> y) = nest k z <> nest k y
310 <n4> nest k (x $$ y) = nest k x $$ nest k y
311 <n5> nest k empty = empty
312 <n6> x <> nest k y = x <> y, if x non-empty
313
314 ** Note the side condition on <n6>! It is this that
315 ** makes it OK for empty to be a left unit for <>.
316
317 Miscellaneous
318 ~~~~~~~~~~~~~
319 <m1> (text s <> x) $$ y = text s <> ((text "" <> x)) $$
320 nest (-length s) y)
321
322 <m2> (x $$ y) <> z = x $$ (y <> z)
323 if y non-empty
324
325
326 Laws for list versions
327 ~~~~~~~~~~~~~~~~~~~~~~
328 <l1> sep (ps++[empty]++qs) = sep (ps ++ qs)
329 ...ditto hsep, hcat, vcat, fill...
330
331 <l2> nest k (sep ps) = sep (map (nest k) ps)
332 ...ditto hsep, hcat, vcat, fill...
333
334 Laws for oneLiner
335 ~~~~~~~~~~~~~~~~~
336 <o1> oneLiner (nest k p) = nest k (oneLiner p)
337 <o2> oneLiner (x <> y) = oneLiner x <> oneLiner y
338
339 You might think that the following verion of <m1> would
340 be neater:
341
342 <3 NO> (text s <> x) $$ y = text s <> ((empty <> x)) $$
343 nest (-length s) y)
344
345 But it doesn't work, for if x=empty, we would have
346
347 text s $$ y = text s <> (empty $$ nest (-length s) y)
348 = text s <> nest (-length s) y
349 -}
350
351 -- ---------------------------------------------------------------------------
352 -- Simple derived definitions
353
354 semi = char ';'
355 colon = char ':'
356 comma = char ','
357 space = char ' '
358 equals = char '='
359 lparen = char '('
360 rparen = char ')'
361 lbrack = char '['
362 rbrack = char ']'
363 lbrace = char '{'
364 rbrace = char '}'
365
366 int n = text (show n)
367 integer n = text (show n)
368 float n = text (show n)
369 double n = text (show n)
370 rational n = text (show n)
371 -- SIGBJORN wrote instead:
372 -- rational n = text (show (fromRationalX n))
373
374 quotes p = char '`' <> p <> char '\''
375 doubleQuotes p = char '"' <> p <> char '"'
376 parens p = char '(' <> p <> char ')'
377 brackets p = char '[' <> p <> char ']'
378 braces p = char '{' <> p <> char '}'
379
380
381 hcat = foldr (<>) empty
382 hsep = foldr (<+>) empty
383 vcat = foldr ($$) empty
384
385 hang d1 n d2 = sep [d1, nest n d2]
386
387 punctuate p [] = []
388 punctuate p (d:ds) = go d ds
389 where
390 go d [] = [d]
391 go d (e:es) = (d <> p) : go e es
392
393 -- ---------------------------------------------------------------------------
394 -- The Doc data type
395
396 -- A Doc represents a *set* of layouts. A Doc with
397 -- no occurrences of Union or NoDoc represents just one layout.
398
399 data Doc
400 = Empty -- empty
401 | NilAbove Doc -- text "" $$ x
402 | TextBeside TextDetails !Int Doc -- text s <> x
403 | Nest !Int Doc -- nest k x
404 | Union Doc Doc -- ul `union` ur
405 | NoDoc -- The empty set of documents
406 | Beside Doc Bool Doc -- True <=> space between
407 | Above Doc Bool Doc -- True <=> never overlap
408
409 type RDoc = Doc -- RDoc is a "reduced Doc", guaranteed not to have a top-level Above or Beside
410
411
412 reduceDoc :: Doc -> RDoc
413 reduceDoc (Beside p g q) = beside p g (reduceDoc q)
414 reduceDoc (Above p g q) = above p g (reduceDoc q)
415 reduceDoc p = p
416
417
418 data TextDetails = Chr Char
419 | Str String
420 | PStr String
421 space_text = Chr ' '
422 nl_text = Chr '\n'
423
424 -- Here are the invariants:
425
426 -- * The argument of NilAbove is never Empty. Therefore
427 -- a NilAbove occupies at least two lines.
428 --
429 -- * The arugment of @TextBeside@ is never @Nest@.
430 --
431 --
432 -- * The layouts of the two arguments of @Union@ both flatten to the same
433 -- string.
434 --
435 -- * The arguments of @Union@ are either @TextBeside@, or @NilAbove@.
436 --
437 -- * The right argument of a union cannot be equivalent to the empty set
438 -- (@NoDoc@). If the left argument of a union is equivalent to the
439 -- empty set (@NoDoc@), then the @NoDoc@ appears in the first line.
440
441 -- * An empty document is always represented by @Empty@. It can't be
442 -- hidden inside a @Nest@, or a @Union@ of two @Empty@s.
443
444 -- * The first line of every layout in the left argument of @Union@ is
445 -- longer than the first line of any layout in the right argument.
446 -- (1) ensures that the left argument has a first line. In view of
447 -- (3), this invariant means that the right argument must have at
448 -- least two lines.
449
450
451 -- Arg of a NilAbove is always an RDoc
452 nilAbove_ p = NilAbove p
453
454 -- Arg of a TextBeside is always an RDoc
455 textBeside_ s sl p = TextBeside s sl p
456
457 -- Arg of Nest is always an RDoc
458 nest_ k p = Nest k p
459
460 -- Args of union are always RDocs
461 union_ p q = Union p q
462
463
464 -- Notice the difference between
465 -- * NoDoc (no documents)
466 -- * Empty (one empty document; no height and no width)
467 -- * text "" (a document containing the empty string;
468 -- one line high, but has no width)
469
470
471 -- ---------------------------------------------------------------------------
472 -- @empty@, @text@, @nest@, @union@
473
474 empty = Empty
475
476 isEmpty Empty = True
477 isEmpty _ = False
478
479 char c = textBeside_ (Chr c) 1 Empty
480 text s = case length s of {sl -> textBeside_ (Str s) sl Empty}
481 ptext s = case length s of {sl -> textBeside_ (PStr s) sl Empty}
482
483 nest k p = mkNest k (reduceDoc p) -- Externally callable version
484
485 -- mkNest checks for Nest's invariant that it doesn't have an Empty inside it
486 mkNest k _ | k `seq` False = undefined
487 mkNest k (Nest k1 p) = mkNest (k + k1) p
488 mkNest k NoDoc = NoDoc
489 mkNest k Empty = Empty
490 mkNest 0 p = p -- Worth a try!
491 mkNest k p = nest_ k p
492
493 -- mkUnion checks for an empty document
494 mkUnion Empty q = Empty
495 mkUnion p q = p `union_` q
496
497 -- ---------------------------------------------------------------------------
498 -- Vertical composition @$$@
499
500 p $$ q = Above p False q
501 p $+$ q = Above p True q
502
503 above :: Doc -> Bool -> RDoc -> RDoc
504 above (Above p g1 q1) g2 q2 = above p g1 (above q1 g2 q2)
505 above p@(Beside _ _ _) g q = aboveNest (reduceDoc p) g 0 (reduceDoc q)
506 above p g q = aboveNest p g 0 (reduceDoc q)
507
508 aboveNest :: RDoc -> Bool -> Int -> RDoc -> RDoc
509 -- Specfication: aboveNest p g k q = p $g$ (nest k q)
510
511 aboveNest _ _ k _ | k `seq` False = undefined
512 aboveNest NoDoc g k q = NoDoc
513 aboveNest (p1 `Union` p2) g k q = aboveNest p1 g k q `union_`
514 aboveNest p2 g k q
515
516 aboveNest Empty g k q = mkNest k q
517 aboveNest (Nest k1 p) g k q = nest_ k1 (aboveNest p g (k - k1) q)
518 -- p can't be Empty, so no need for mkNest
519
520 aboveNest (NilAbove p) g k q = nilAbove_ (aboveNest p g k q)
521 aboveNest (TextBeside s sl p) g k q = k1 `seq` textBeside_ s sl rest
522 where
523 k1 = k - sl
524 rest = case p of
525 Empty -> nilAboveNest g k1 q
526 other -> aboveNest p g k1 q
527
528
529 nilAboveNest :: Bool -> Int -> RDoc -> RDoc
530 -- Specification: text s <> nilaboveNest g k q
531 -- = text s <> (text "" $g$ nest k q)
532
533 nilAboveNest _ k _ | k `seq` False = undefined
534 nilAboveNest g k Empty = Empty -- Here's why the "text s <>" is in the spec!
535 nilAboveNest g k (Nest k1 q) = nilAboveNest g (k + k1) q
536
537 nilAboveNest g k q | (not g) && (k > 0) -- No newline if no overlap
538 = textBeside_ (Str (spaces k)) k q
539 | otherwise -- Put them really above
540 = nilAbove_ (mkNest k q)
541
542 -- ---------------------------------------------------------------------------
543 -- Horizontal composition @<>@
544
545 p <> q = Beside p False q
546 p <+> q = Beside p True q
547
548 beside :: Doc -> Bool -> RDoc -> RDoc
549 -- Specification: beside g p q = p <g> q
550
551 beside NoDoc g q = NoDoc
552 beside (p1 `Union` p2) g q = (beside p1 g q) `union_` (beside p2 g q)
553 beside Empty g q = q
554 beside (Nest k p) g q = nest_ k (beside p g q) -- p non-empty
555 beside p@(Beside p1 g1 q1) g2 q2
556 {- (A `op1` B) `op2` C == A `op1` (B `op2` C) iff op1 == op2
557 [ && (op1 == <> || op1 == <+>) ] -}
558 | g1 == g2 = beside p1 g1 (beside q1 g2 q2)
559 | otherwise = beside (reduceDoc p) g2 q2
560 beside p@(Above _ _ _) g q = beside (reduceDoc p) g q
561 beside (NilAbove p) g q = nilAbove_ (beside p g q)
562 beside (TextBeside s sl p) g q = textBeside_ s sl rest
563 where
564 rest = case p of
565 Empty -> nilBeside g q
566 other -> beside p g q
567
568
569 nilBeside :: Bool -> RDoc -> RDoc
570 -- Specification: text "" <> nilBeside g p
571 -- = text "" <g> p
572
573 nilBeside g Empty = Empty -- Hence the text "" in the spec
574 nilBeside g (Nest _ p) = nilBeside g p
575 nilBeside g p | g = textBeside_ space_text 1 p
576 | otherwise = p
577
578 -- ---------------------------------------------------------------------------
579 -- Separate, @sep@, Hughes version
580
581 -- Specification: sep ps = oneLiner (hsep ps)
582 -- `union`
583 -- vcat ps
584
585 sep = sepX True -- Separate with spaces
586 cat = sepX False -- Don't
587
588 sepX x [] = empty
589 sepX x (p:ps) = sep1 x (reduceDoc p) 0 ps
590
591
592 -- Specification: sep1 g k ys = sep (x : map (nest k) ys)
593 -- = oneLiner (x <g> nest k (hsep ys))
594 -- `union` x $$ nest k (vcat ys)
595
596 sep1 :: Bool -> RDoc -> Int -> [Doc] -> RDoc
597 sep1 g _ k ys | k `seq` False = undefined
598 sep1 g NoDoc k ys = NoDoc
599 sep1 g (p `Union` q) k ys = sep1 g p k ys
600 `union_`
601 (aboveNest q False k (reduceDoc (vcat ys)))
602
603 sep1 g Empty k ys = mkNest k (sepX g ys)
604 sep1 g (Nest n p) k ys = nest_ n (sep1 g p (k - n) ys)
605
606 sep1 g (NilAbove p) k ys = nilAbove_ (aboveNest p False k (reduceDoc (vcat ys)))
607 sep1 g (TextBeside s sl p) k ys = textBeside_ s sl (sepNB g p (k - sl) ys)
608
609 -- Specification: sepNB p k ys = sep1 (text "" <> p) k ys
610 -- Called when we have already found some text in the first item
611 -- We have to eat up nests
612
613 sepNB g (Nest _ p) k ys = sepNB g p k ys
614
615 sepNB g Empty k ys = oneLiner (nilBeside g (reduceDoc rest))
616 `mkUnion`
617 nilAboveNest False k (reduceDoc (vcat ys))
618 where
619 rest | g = hsep ys
620 | otherwise = hcat ys
621
622 sepNB g p k ys = sep1 g p k ys
623
624 -- ---------------------------------------------------------------------------
625 -- @fill@
626
627 fsep = fill True
628 fcat = fill False
629
630 -- Specification:
631 -- fill [] = empty
632 -- fill [p] = p
633 -- fill (p1:p2:ps) = oneLiner p1 <#> nest (length p1)
634 -- (fill (oneLiner p2 : ps))
635 -- `union`
636 -- p1 $$ fill ps
637
638 fill g [] = empty
639 fill g (p:ps) = fill1 g (reduceDoc p) 0 ps
640
641
642 fill1 :: Bool -> RDoc -> Int -> [Doc] -> Doc
643 fill1 g _ k ys | k `seq` False = undefined
644 fill1 g NoDoc k ys = NoDoc
645 fill1 g (p `Union` q) k ys = fill1 g p k ys
646 `union_`
647 (aboveNest q False k (fill g ys))
648
649 fill1 g Empty k ys = mkNest k (fill g ys)
650 fill1 g (Nest n p) k ys = nest_ n (fill1 g p (k - n) ys)
651
652 fill1 g (NilAbove p) k ys = nilAbove_ (aboveNest p False k (fill g ys))
653 fill1 g (TextBeside s sl p) k ys = textBeside_ s sl (fillNB g p (k - sl) ys)
654
655 fillNB g _ k ys | k `seq` False = undefined
656 fillNB g (Nest _ p) k ys = fillNB g p k ys
657 fillNB g Empty k [] = Empty
658 fillNB g Empty k (y:ys) = nilBeside g (fill1 g (oneLiner (reduceDoc y)) k1 ys)
659 `mkUnion`
660 nilAboveNest False k (fill g (y:ys))
661 where
662 k1 | g = k - 1
663 | otherwise = k
664
665 fillNB g p k ys = fill1 g p k ys
666
667
668 -- ---------------------------------------------------------------------------
669 -- Selecting the best layout
670
671 best :: Mode
672 -> Int -- Line length
673 -> Int -- Ribbon length
674 -> RDoc
675 -> RDoc -- No unions in here!
676
677 best OneLineMode w r p
678 = get p
679 where
680 get Empty = Empty
681 get NoDoc = NoDoc
682 get (NilAbove p) = nilAbove_ (get p)
683 get (TextBeside s sl p) = textBeside_ s sl (get p)
684 get (Nest k p) = get p -- Elide nest
685 get (p `Union` q) = first (get p) (get q)
686
687 best mode w r p
688 = get w p
689 where
690 get :: Int -- (Remaining) width of line
691 -> Doc -> Doc
692 get w _ | w==0 && False = undefined
693 get w Empty = Empty
694 get w NoDoc = NoDoc
695 get w (NilAbove p) = nilAbove_ (get w p)
696 get w (TextBeside s sl p) = textBeside_ s sl (get1 w sl p)
697 get w (Nest k p) = nest_ k (get (w - k) p)
698 get w (p `Union` q) = nicest w r (get w p) (get w q)
699
700 get1 :: Int -- (Remaining) width of line
701 -> Int -- Amount of first line already eaten up
702 -> Doc -- This is an argument to TextBeside => eat Nests
703 -> Doc -- No unions in here!
704
705 get1 w _ _ | w==0 && False = undefined
706 get1 w sl Empty = Empty
707 get1 w sl NoDoc = NoDoc
708 get1 w sl (NilAbove p) = nilAbove_ (get (w - sl) p)
709 get1 w sl (TextBeside t tl p) = textBeside_ t tl (get1 w (sl + tl) p)
710 get1 w sl (Nest k p) = get1 w sl p
711 get1 w sl (p `Union` q) = nicest1 w r sl (get1 w sl p)
712 (get1 w sl q)
713
714 nicest w r p q = nicest1 w r 0 p q
715 nicest1 w r sl p q | fits ((w `minn` r) - sl) p = p
716 | otherwise = q
717
718 fits :: Int -- Space available
719 -> Doc
720 -> Bool -- True if *first line* of Doc fits in space available
721
722 fits n p | n < 0 = False
723 fits n NoDoc = False
724 fits n Empty = True
725 fits n (NilAbove _) = True
726 fits n (TextBeside _ sl p) = fits (n - sl) p
727
728 minn x y | x < y = x
729 | otherwise = y
730
731 -- @first@ and @nonEmptySet@ are similar to @nicest@ and @fits@, only simpler.
732 -- @first@ returns its first argument if it is non-empty, otherwise its second.
733
734 first p q | nonEmptySet p = p
735 | otherwise = q
736
737 nonEmptySet NoDoc = False
738 nonEmptySet (p `Union` q) = True
739 nonEmptySet Empty = True
740 nonEmptySet (NilAbove p) = True -- NoDoc always in first line
741 nonEmptySet (TextBeside _ _ p) = nonEmptySet p
742 nonEmptySet (Nest _ p) = nonEmptySet p
743
744 -- @oneLiner@ returns the one-line members of the given set of @Doc@s.
745
746 oneLiner :: Doc -> Doc
747 oneLiner NoDoc = NoDoc
748 oneLiner Empty = Empty
749 oneLiner (NilAbove p) = NoDoc
750 oneLiner (TextBeside s sl p) = textBeside_ s sl (oneLiner p)
751 oneLiner (Nest k p) = nest_ k (oneLiner p)
752 oneLiner (p `Union` q) = oneLiner p
753
754
755 -- ---------------------------------------------------------------------------
756 -- Displaying the best layout
757
758 {-
759 renderStyle Style{mode, lineLength, ribbonsPerLine} doc
760 = fullRender mode lineLength ribbonsPerLine doc ""
761 -}
762
763 render doc = showDoc doc ""
764 showDoc doc rest = fullRender PageMode 100 1.5 string_txt rest doc
765
766 string_txt (Chr c) s = c:s
767 string_txt (Str s1) s2 = s1 ++ s2
768 string_txt (PStr s1) s2 = s1 ++ s2
769
770
771 fullRender OneLineMode _ _ txt end doc = easy_display space_text txt end (reduceDoc doc)
772 fullRender LeftMode _ _ txt end doc = easy_display nl_text txt end (reduceDoc doc)
773
774 fullRender mode line_length ribbons_per_line txt end doc
775 = display mode line_length ribbon_length txt end best_doc
776 where
777 best_doc = best mode hacked_line_length ribbon_length (reduceDoc doc)
778
779 hacked_line_length, ribbon_length :: Int
780 ribbon_length = round (fromIntegral line_length / ribbons_per_line)
781 hacked_line_length = case mode of { ZigZagMode -> maxBound; other -> line_length }
782
783 display mode page_width ribbon_width txt end doc
784 = case page_width - ribbon_width of { gap_width ->
785 case gap_width `quot` 2 of { shift ->
786 let
787 lay k _ | k `seq` False = undefined
788 lay k (Nest k1 p) = lay (k + k1) p
789 lay k Empty = end
790
791 lay k (NilAbove p) = nl_text `txt` lay k p
792
793 lay k (TextBeside s sl p)
794 = case mode of
795 ZigZagMode | k >= gap_width
796 -> nl_text `txt` (
797 Str (multi_ch shift '/') `txt` (
798 nl_text `txt` (
799 lay1 (k - shift) s sl p)))
800
801 | k < 0
802 -> nl_text `txt` (
803 Str (multi_ch shift '\\') `txt` (
804 nl_text `txt` (
805 lay1 (k + shift) s sl p )))
806
807 other -> lay1 k s sl p
808
809 lay1 k _ sl _ | k+sl `seq` False = undefined
810 lay1 k s sl p = Str (indent k) `txt` (s `txt` lay2 (k + sl) p)
811
812 lay2 k _ | k `seq` False = undefined
813 lay2 k (NilAbove p) = nl_text `txt` lay k p
814 lay2 k (TextBeside s sl p) = s `txt` (lay2 (k + sl) p)
815 lay2 k (Nest _ p) = lay2 k p
816 lay2 k Empty = end
817 in
818 lay 0 doc
819 }}
820
821 cant_fail = error "easy_display: NoDoc"
822 easy_display nl_text txt end doc
823 = lay doc cant_fail
824 where
825 lay NoDoc no_doc = no_doc
826 lay (Union p q) no_doc = {- lay p -} (lay q cant_fail) -- Second arg can't be NoDoc
827 lay (Nest k p) no_doc = lay p no_doc
828 lay Empty no_doc = end
829 lay (NilAbove p) no_doc = nl_text `txt` lay p cant_fail -- NoDoc always on first line
830 lay (TextBeside s sl p) no_doc = s `txt` lay p no_doc
831
832 indent n | n >= 8 = '\t' : indent (n - 8)
833 | otherwise = spaces n
834
835 multi_ch 0 ch = ""
836 multi_ch n ch = ch : multi_ch (n - 1) ch
837
838 spaces 0 = ""
839 spaces n = ' ' : spaces (n - 1)
840