Ticket #4455: add_instance_Applicative_ST.dpatch

File add_instance_Applicative_ST.dpatch, 53.7 KB (added by pumpkin, 3 years ago)

add Applicative instance for ST

Line 
11 patch for repository http://darcs.haskell.org/packages/base:
2
3Sat Oct 30 01:26:49 CEST 2010  Bas van Dijk <v.dijk.bas@gmail.com>
4  * Add Applicative instance for ST
5
6New patches:
7
8[Add Applicative instance for ST
9Bas van Dijk <v.dijk.bas@gmail.com>**20101029232649
10 Ignore-this: 6f76cc5c6be938a8b26a5386502878ae
11] {
12hunk ./Control/Applicative.hs 45
13         (Arrow(arr, (&&&)), ArrowZero(zeroArrow), ArrowPlus((<+>)))
14 import Control.Monad (liftM, ap, MonadPlus(..))
15 import Control.Monad.Instances ()
16+import Control.Monad.ST (ST)
17+import qualified Control.Monad.ST.Lazy as Lazy (ST)
18 import Data.Functor ((<$>), (<$))
19 import Data.Monoid (Monoid(..))
20 
21hunk ./Control/Applicative.hs 154
22         pure = return
23         (<*>) = ap
24 
25+instance Applicative (ST s) where
26+        pure = return
27+        (<*>) = ap
28+
29+instance Applicative (Lazy.ST s) where
30+        pure = return
31+        (<*>) = ap
32+
33 #ifdef __GLASGOW_HASKELL__
34 instance Applicative STM where
35     pure = return
36}
37
38Context:
39
40[Add an INLINE pragma on fromInteger on Int
41simonpj@microsoft.com**20101027193931
42 Ignore-this: 6363b8e1338f1e5334c28e8967284ef3
43]
44[Add an INLINE pragme for fmapDefault
45simonpj@microsoft.com**20101027193859
46 Ignore-this: 5f140c8fe79bbe1fa6af933fb58366bb
47]
48[hGetBuf: fix a case of a short read being returned (#4427)
49Simon Marlow <marlowsd@gmail.com>**20101027144324
50 Ignore-this: 6aa4cf722bef8eb01dfec3e751fd3eeb
51]
52[Add showMultiLineString to GHC.Show
53simonpj@microsoft.com**20101025151655
54 Ignore-this: eacc594597387e8d965d17204b3ae35f
55 
56 This is part of the fix for #4436
57 
58   showMultiLineString :: String -> [String]
59   -- | Like 'showLitString' (expand escape characters using Haskell
60   -- escape conventions), but
61   --   * break the string into multiple lines
62   --   * wrap the entire thing in double quotes
63   -- Example:  @breakMultiLineString "hello\ngoodbye\nblah"@
64   -- returns   @["\"hello\\", "\\goodbye\\", "\\blah\"" ]@
65   -- where those "\\" are really just a single backslash
66   -- (but I'm writing them here as Haskell literals)
67]
68[CIntPtr, CUIntPtr, CIntMax, CUIntMax are new to nhc98.
69Malcolm.Wallace@me.com**20101025102644
70 Ignore-this: 32d703e70b9d0136cd68fa1987b35c2c
71]
72[Follow GHC.Bool/GHC.Types merge
73Ian Lynagh <igloo@earth.li>**20101023151510
74 Ignore-this: e8b93b702f02a4709706b130988f85a8
75]
76[Remove redundant imports, now that NoImplicitPrelude does not imply RebindableSyntax
77simonpj@microsoft.com**20101022143157
78 Ignore-this: 8d11a7ea4625d4d9cd1514e7fe158626
79]
80[FIX #4335
81Daniel Fischer <daniel.is.fischer@web.de>**20101019010109
82 Ignore-this: 3b8ad075637088df77937b923f623204
83 fromRational :: Rational -> Ratio a produced invalid results for fixed-width
84 types a. Reduce the fraction to avoid that.
85]
86[FIX #4337
87Daniel Fischer <daniel.is.fischer@web.de>**20101019003030
88 Ignore-this: e6eee4088d63e8d72d5ca7d92f708705
89 Special versions for the power functions with a Rational base and rewrite rules.
90]
91[remove trailing whitespace
92Simon Marlow <marlowsd@gmail.com>**20101021093337
93 Ignore-this: dda2815ba424a460ba2a31771a3f03fc
94]
95[FIX #4336
96Daniel Fischer <daniel.is.fischer@web.de>**20101021093246
97 Ignore-this: 76031829aff90251a284dbfa72f3b128
98 Avoid superfluous gcd calculation in recip for Ratio a because numerator
99 and denominator are known to be coprime.
100]
101[Add throwSTM :: Exception e => e -> STM a
102Bas van Dijk <v.dijk.bas@gmail.com>**20100926192144
103 Ignore-this: c6bfdae0eab9f0cf1360bc06d088bfd5
104]
105[Generalize catchSTM
106Bas van Dijk <v.dijk.bas@gmail.com>**20100926192106
107 Ignore-this: d2038494582d2cde2247293dd162671c
108]
109[FIX #4334
110Daniel Fischer <daniel.is.fischer@web.de>**20101020091111
111 Ignore-this: 1a1a406fcf4c352b5bc1f46f93f31b2a
112 Make selector thunks visible to GHC to fix a space leak in lines.
113]
114[FIX #1434
115Daniel Fischer <daniel.is.fischer@web.de>**20101020091014
116 Ignore-this: 3c7c73d3f4487d5aaa453087497d3534
117 Rewrite rules for RealFrac methods with sized Int and Word targets.
118 For all types whose range is contained in Int's range, there are now
119 rewrite rules for properFraction, truncate, floor, ceiling and round
120 from Double and Float, going through the specialised methods for Int.
121 
122 Unfortunately, we can't have a rewrite rule for Word.
123]
124[Define SpecConstrAnnotation in GHC.Exts, and import it from there
125simonpj@microsoft.com**20101018135857
126 Ignore-this: 8bf81cbc5787dbb5a3875b5622f67732
127   
128 Reason: avoid having to link the entire ghc package in modules
129 that use compile-time annotations:
130   
131        import GHC.Exts( SpecConstrAnnotation )
132        {-# ANN type T ForceSpecConstr #-}
133   
134 It's a kind of bug that the package exporting SpecConstrAnnotation
135 is linked even though it is only needed at compile time, but putting
136 the data type declaration in GHC.Exts is a simple way to sidestep
137 the problem
138   
139 See See Note [SpecConstrAnnotation] in SpecConstr
140]
141[Refer to 'mask' instead of 'block' in documentation of Control.Exception
142Bas van Dijk <v.dijk.bas@gmail.com>**20101016185312
143 Ignore-this: cd1bc58df53f3cd1078b9031c3c13f4e
144]
145[throwTo: mention interruptible foreign calls
146Simon Marlow <marlowsd@gmail.com>**20101014084220
147 Ignore-this: dbc53d85f870cf649f87186c7185465a
148]
149[remove trailing whitespace
150Simon Marlow <marlowsd@gmail.com>**20101013101906
151 Ignore-this: b8b424540cacbbb3c6d934242e3af795
152]
153[FIX #4381
154Simon Marlow <marlowsd@gmail.com>**20101013101849
155 Ignore-this: f0daa4845eeb444231451b975b71d055
156 Fix scaleFloat by clamping the scaling parameter so that
157 exponent + scale doesn't overflow.
158 
159 Patch by: Daniel Fischer <daniel.is.fischer@web.de>
160]
161[Replaced some throws to throwIOs where the type is IO
162Bas van Dijk <v.dijk.bas@gmail.com>**20100924221340
163 Ignore-this: e74191e4527ae6f7551c95fd41063335
164]
165[Added initial .authorspellings
166Bas van Dijk <v.dijk.bas@gmail.com>**20101005072701
167 Ignore-this: 63628bcabfdd0b7beda4cd37daeccd89
168]
169[Lazier intersperse
170Daniel Fischer <daniel.is.fischer@web.de>**20101002231201
171 Ignore-this: a0fed65930cf19e68b4363381a5ab576
172 A lazier implementation of intersperse, and consequentially intercalate, to
173 avoid space leaks.
174]
175[FIX #4228 (atanh (-1) returns NaN instead of -Infinity)
176ghc@cainnorris.net**20100816213654
177 Ignore-this: dee89c24493e84a02bea711a1c83a73f
178]
179[Make intersectBy lazier
180Daniel Fischer <daniel.is.fischer@web.de>**20100930191731
181 Ignore-this: ef687bc75923434e85c14b57171576aa
182 Add shortcuts to intersectBy for empty list arguments.
183 In addition to being faster in that case, more inputs yield defined results.
184 Treats ticket #4323
185]
186[doc tweak for Directory file type: file names are '\0'-separated
187Simon Marlow <marlowsd@gmail.com>**20100922113811
188 Ignore-this: 96b7b004bd6e5bc3e958ad55bf238ba1
189]
190[documentation for IODeviceType (#4317, edited by me)
191Simon Marlow <marlowsd@gmail.com>**20100915131341
192 Ignore-this: 21c50ca7a189eebcf299523b6e942bae
193]
194[Allow Data.HashTable construction with user-supplied size
195**20100722210726
196 Ignore-this: bd54880bb16a106a992f03b040dc4164
197 
198 This avoids some resizing for users who know they will be inserting a
199 lot of data.
200 
201 http://hackage.haskell.org/trac/ghc/ticket/4193
202]
203[some fixes for hGetBufSome
204Simon Marlow <marlowsd@gmail.com>**20100916113732
205 Ignore-this: 3e596a606c180dc4859ea8f4c9132ca1
206  - fix one case where it was blocking when it shouldn't
207  - a couple of error-message tweaks
208]
209[Windows: map ERROR_NO_DATA to EPIPE, rather than EINVAL
210Simon Marlow <marlowsd@gmail.com>**20100915142618
211 Ignore-this: 9023e5f0542419f225aef26cb6b1d88d
212 WriteFile() returns ERROR_NO_DATA when writing to a pipe that is
213 "closing", however by default the write() wrapper in the CRT maps this
214 to EINVAL so we get confusing things like
215 
216   hPutChar: invalid argument (Invalid Argumnet)
217 
218 when piping the output of a Haskell program into something that closes
219 the pipe early.  This was happening in the testsuite in a few place.
220 
221 The solution is to map ERROR_NO_DATA to EPIPE correctly, as we
222 explicitly check for EPIPE on stdout (in GHC.TopHandler) so we can
223 exit without an error in this case.
224]
225[tighten up parsing of numbers (#1579)
226Simon Marlow <marlowsd@gmail.com>**20100913214733
227 Ignore-this: 3411bf3d2e98cfacb9e0afd11d79e722
228]
229[Add absentError. 
230simonpj@microsoft.com**20100914134639
231 Ignore-this: d0eef5a87e1def4cdbde92a55241c8c4
232 
233 This patch accompanies the HEAD patch:
234 
235   Tue Sep 14 12:38:27 BST 2010  simonpj@microsoft.com
236     * Make absent-arg wrappers work for unlifted types (fix Trac #4306)
237     
238     Previously we were simply passing arguments of unlifted
239     type to a wrapper, even if they were absent, which was
240     stupid.
241     
242     See Note [Absent error Id] in WwLib.
243]
244[Add missing import, fixes build on windows
245simonpj@microsoft.com**20100914122750
246 Ignore-this: 12ece15ef94982ddfbf5f9f7900619da
247]
248[Add a suitable Show instance for TextEncoding (#4273)
249Simon Marlow <marlowsd@gmail.com>**20100913154459
250 Ignore-this: 77f2235460895debd2827f34c42c3435
251]
252[don't fill a finalized handle with an error (see comment)
253Simon Marlow <marlowsd@gmail.com>**20100913153350
254 Ignore-this: c72cdb6898dffa88eca1d781171b2943
255]
256[deriving (Eq, Ord, Read, Show) for Newline and NewlineMode
257Simon Marlow <marlowsd@gmail.com>**20100913153031
258 Ignore-this: 9b9b29bfb7abf5550cfbfa7788f81bf
259]
260[fix warning on Windows
261Simon Marlow <marlowsd@gmail.com>**20100913111536
262 Ignore-this: dacc5448c452daad60ed37a1a5ed096e
263]
264[Put the state-token argument on fill, done, adjust on the RHS
265simonpj@microsoft.com**20100913101832
266 Ignore-this: d228b492de7d4635c026ed24cbc17e34
267 
268 This is so that the functions will inline when
269 applied to their normal (non-state-token) aguments.
270 
271 I forget why I did this, but it seems like the right thing anyway.
272]
273[avoid Foreign.unsafePerformIO
274Ross Paterson <ross@soi.city.ac.uk>**20100909125521
275 Ignore-this: b698101119ffd1bc6311cce0736f745d
276]
277[Remove debugging code accidentally left in
278Simon Marlow <marlowsd@gmail.com>**20100909113331
279 Ignore-this: 906a14176dd37030b8203782a687936b
280]
281[Fix Windows build; patches frmo ezyang
282Ian Lynagh <igloo@earth.li>**20100908123037
283 Ignore-this: 2f02986087edd7da8382221012c27cd0
284]
285[More accurate isatty test for MinGW.
286Edward Z. Yang <ezyang@mit.edu>**20100907154144
287 Ignore-this: 93bdc2b2a8e65a7c4c7d3906bdda01db
288]
289[Fix the build when HAVE_KQUEUE but not HAVE_KEVENT64
290Ian Lynagh <igloo@earth.li>**20100904223703]
291[Fix warnings
292benl@ouroborus.net**20100830044741
293 Ignore-this: 8397aaec7c36046c9ace403e65f32d32
294]
295[fix cache variable name used by FP_SEARCH_LIBS_PROTO
296Ross Paterson <ross@soi.city.ac.uk>**20100819204858
297 Ignore-this: b8113cb3c6f0e03c507297c99d3d82b7
298]
299[Add a missing castPtr (only shows up in -DDEBUG)
300simonpj@microsoft.com**20100815145127
301 Ignore-this: 30b9c42cd3ce7837bdabd254fe66078d
302]
303[Fixed a rounding error in threadDelay
304Johan Tibell <johan.tibell@gmail.com>**20100813124043
305 Ignore-this: 1cb77d0852233ffffb144b134064ee3c
306]
307[export allocaBytesAligned; make allocaArray use the correct alignment (#2917)
308Simon Marlow <marlowsd@gmail.com>**20100812105524
309 Ignore-this: deb6495f7b7b84deaf02b88927a5ba8c
310]
311[deprecate unGetChan and isEmptyChan (see #4154)
312Simon Marlow <marlowsd@gmail.com>**20100705125952
313 Ignore-this: b4e769959f131b2d0001eb7202bc1b92
314]
315[Add type signatures to cope with lack of local generalisation
316simonpj@microsoft.com**20100728124847
317 Ignore-this: d3af9a47c2821c6081bde05a135a92fb
318]
319[Add type signature in local where
320simonpj@microsoft.com**20100727151532
321 Ignore-this: 1c57063ad32d13e0d1ec8daf968bf055
322]
323[Integrated new I/O manager
324Simon Marlow <marlowsd@gmail.com>**20100810082248
325 Ignore-this: ed70a9066ac9b676a446fe99978fef7a
326 (patch originally by Johan Tibell <johan.tibell@gmail.com>, minor merging by me)
327]
328[Add mfilter to Control.Monad
329jon.fairbairn@cl.cam.ac.uk**20090917145616
330 Ignore-this: de4240b60684f3065b29378df3ea98f2
331 
332 Straightforward MonadPlus version of List.filter. I would
333 prefer to call it filter, but the current naming scheme for
334 Control.Monad implies mfilter.
335 
336]
337[move Monad and MonadFix instances for Either from mtl (proposal #4159)
338Ross Paterson <ross@soi.city.ac.uk>**20100729122449
339 Ignore-this: b0f8cd8643679948d1da43bd7c08c5aa
340 
341 The Monad and MonadFix instances for Either (formerly in the mtl
342 package) are moved to Control.Monad.Instances and Control.Monad.Fix
343 respectively.  The Monad instance is still an orphan, to retain Haskell
344 98 compatibility, but the MonadFix instance is together with its class.
345 The Error constraint is removed from both instances, and the default
346 definition of fail is used.
347]
348[Remove egregious ghc-ish from Foreign.Marshal
349Malcolm.Wallace@me.com**20100722075449]
350[add numSparks :: IO Int (#4167)
351Simon Marlow <marlowsd@gmail.com>**20100720153858
352 Ignore-this: 4543f57a7f137f8cae1c3efc5c023a9b
353]
354[add unsafeLocalState from Haskell 2010, and docs
355Simon Marlow <marlowsd@gmail.com>**20100720082819
356 Ignore-this: dcd79fb546ebe29ddff4df279ec2f38
357]
358[docs: mention that Foreign.unsafePerformIO is deprecated
359Simon Marlow <marlowsd@gmail.com>**20100720082804
360 Ignore-this: 4cfebb8f2a1cddc7d15e94e31b2befa4
361 We can't actually deprecate it without introducing a name clash
362 between Foreign.unsafePerformIO and System.IO.Unsafe.unsafePerformIO
363]
364[doc formatting fix
365Simon Marlow <marlowsd@gmail.com>**20100714151347
366 Ignore-this: 255edef607dcd290e198015240b5d125
367]
368[add module intro from Haskell 2010
369Simon Marlow <marlowsd@gmail.com>**20100714115853
370 Ignore-this: 59b5a07507a059ccccdff2dfb6490a27
371]
372[document exception-overriding behaviour in withFile
373Simon Marlow <marlowsd@gmail.com>**20100714104107
374 Ignore-this: f99e641ea2f46d872cb7420a62fa50dc
375]
376[doc: use "finalizer" consistently
377Simon Marlow <marlowsd@gmail.com>**20100714103649
378 Ignore-this: bdfea40f31dc5045fdbc6e12266dda93
379]
380[clarify meaning of bit
381Simon Marlow <marlowsd@gmail.com>**20100714103310
382 Ignore-this: 521b031f1e83ef34ca03d9aa9273df8a
383]
384[note shortcutting behaviour of any/all/elem
385Simon Marlow <marlowsd@gmail.com>**20100714103304
386 Ignore-this: 1605f362ba0712ad1cea1309636f3ea1
387]
388[add cast{C,U}CharToChar and castCharTo{C,U}Char, from Haskell 2010
389Simon Marlow <marlowsd@gmail.com>**20100713132515
390 Ignore-this: 9b1da827016c7b08668078b45964e9de
391]
392[mention that IntPtr and WordPtr can be marshalled to/from intptr_t and uintptr_t
393Simon Marlow <marlowsd@gmail.com>**20100713132403
394 Ignore-this: dcc112a72746ba117a84fa29e71b6800
395]
396[Partial fix for Trac #4136
397simonpj@microsoft.com**20100707135725
398 Ignore-this: 9548eeb3187d9779d4e5c858a0f35354
399 
400 In 'choose' (which is a library function designed specifically
401 to support derived instances of Read), we must match Symbol
402 as well as Ident, for nullary constructors that (wierdly) are
403 symbols.
404]
405[Fix typo in documentation
406Simon Hengel <simon.hengel@wiktory.org>**20100711141648
407 Ignore-this: c052dd8a681832ef598a323ad55eae3a
408]
409[Remove duplicated word in documentation
410Simon Hengel <simon.hengel@wiktory.org>**20100711072703
411 Ignore-this: fb3732dc57be55f14168792f923433
412]
413[Allow nhc98 to cope with recent changes to Control.Exception.
414Malcolm.Wallace@me.com**20100710170940]
415[ New asynchronous exception control API (base parts)
416Simon Marlow <marlowsd@gmail.com>**20100708152735
417 Ignore-this: 71a4811804f04259f1fe739f8863beaf
418   
419 As discussed on the libraries/haskell-cafe mailing lists
420   http://www.haskell.org/pipermail/libraries/2010-April/013420.html
421 
422 This is a replacement for block/unblock in the asychronous exceptions
423 API to fix a problem whereby a function could unblock asynchronous
424 exceptions even if called within a blocked context.
425 
426 The new terminology is "mask" rather than "block" (to avoid confusion
427 due to overloaded meanings of the latter).
428 
429 The following is the new API; the old API is deprecated but still
430 available for the time being.
431 
432 Control.Exception
433 -----------------
434 
435 mask  :: ((forall a. IO a -> IO a) -> IO b) -> IO b
436 mask_ :: IO a -> IO a
437 
438 uninterruptibleMask :: ((forall a. IO a -> IO a) -> IO b) -> IO b
439 uninterruptibleMask_ :: IO a -> IO
440 
441 getMaskingState :: IO MaskingState
442 
443 data MaskingState
444   = Unmasked
445   | MaskedInterruptible
446   | MaskedUninterruptible
447 
448 
449 Control.Concurrent
450 ------------------
451 
452 forkIOUnmasked :: IO () -> IO ThreadId
453]
454[Async-exception safety, and avoid space leaks
455Simon Marlow <marlowsd@gmail.com>**20100708145819
456 Ignore-this: dbfd0e61551e9e7b4fc1c6fe9b9a83de
457 Patch submitted by: Bas van Dijk <v.dijk.bas@gmail.com>
458 Modified slightly by me to remove non-functional changes.
459]
460[Async-exception safety, and avoid space leaks
461Simon Marlow <marlowsd@gmail.com>**20100708103154
462 Ignore-this: 190c3ac8f6633231624da8cf1316588
463 Patch submitted by: Bas van Dijk <v.dijk.bas@gmail.com>
464 Modified slightly by me to remove non-functional changes.
465]
466[Fix a few places where we forgot to close the text codecs (#4029)
467Simon Marlow <marlowsd@gmail.com>**20100702130210
468 Ignore-this: 2e81a4b4cb343181cef34b0f9e2ded47
469 Each time you invoke :load in GHCi it resets the CAFs, including
470 stdin/stdout/stderr, and each of these was allocating a new iconv_t.
471]
472[remove docs from Monad that belonged on the instance for MonadPlus IO
473Simon Marlow <marlowsd@gmail.com>**20100701154203
474 Ignore-this: 59df02542a7ac9421552a2155d848d27
475]
476[docs: unqualify Prelude.IO
477Simon Marlow <marlowsd@gmail.com>**20100701153817
478 Ignore-this: 73b0202876c827e7a5b4a5ce74e724c4
479]
480[unqualify Float and Double
481Simon Marlow <marlowsd@gmail.com>**20100701142727
482 Ignore-this: cbe89d31a00bf49996a33933324fca17
483]
484[extract information about Data.Time from docs for CTime
485Simon Marlow <marlowsd@gmail.com>**20100701142415
486 Ignore-this: c48c9609b8d36e43e033a7bea81d6f17
487]
488[doc typo
489Simon Marlow <marlowsd@gmail.com>**20100701142354
490 Ignore-this: 17a1fd703831c888975ff63fbfa3a9b2
491]
492[peekArray docs: remove mentions of "this version" and "previous version"
493Simon Marlow <marlowsd@gmail.com>**20100701125333
494 Ignore-this: 39a744874258670bd935ba9e38390939
495]
496[doc typo
497Simon Marlow <marlowsd@gmail.com>**20100701124154
498 Ignore-this: 98f5c286e38c2c34c96b05d5e8bc5ad9
499]
500[doc typo
501Simon Marlow <marlowsd@gmail.com>**20100701124128
502 Ignore-this: 10a4314ec7aed336701fc616fb574ebc
503]
504[doc typo
505Simon Marlow <marlowsd@gmail.com>**20100701123715
506 Ignore-this: c4909a7bf7163460ee5d32f58812041e
507]
508[doc wibble: Haskell 98 I/O Error -> 'IOError'
509Simon Marlow <marlowsd@gmail.com>**20100701123612
510 Ignore-this: bf373df781acbc575e4ffe3b7e6059ae
511]
512[doc typo
513Simon Marlow <marlowsd@gmail.com>**20100701123014
514 Ignore-this: 16aaccae48ef3101adf78ea5b0d5a8fd
515]
516[Haddock hacks to fix whitespace consistency
517Simon Marlow <marlowsd@gmail.com>**20100701121631
518 Ignore-this: 61c58dec52a31fd2d3f331a87d2f903f
519]
520[use '==' consistently rather than '->' in examples
521Simon Marlow <marlowsd@gmail.com>**20100701121616
522 Ignore-this: 472b0a05a85d34d9712186040e1636d9
523]
524[doc wibble: remove confusing mention of "Prelude"
525Simon Marlow <marlowsd@gmail.com>**20100701113308
526 Ignore-this: 232283d0096d01cd45e9b3c5c1e63a6d
527]
528[doc wibble: nonstrict -> non-strict
529Simon Marlow <marlowsd@gmail.com>**20100701113253
530 Ignore-this: 4264f0ab23a0835fc13c6e8601d6b743
531]
532[doc whitespace
533Simon Marlow <marlowsd@gmail.com>**20100701112242
534 Ignore-this: 777a95b1d1140c61d3ab95d5eb5809e7
535]
536[move the doc for 'Char' to its new home in ghc-prim:GHC.Types
537Simon Marlow <marlowsd@gmail.com>**20100629134150
538 Ignore-this: 7687db0077a29498349bfb4b44983985
539]
540[doc wibble
541Simon Marlow <marlowsd@gmail.com>**20100629122608
542 Ignore-this: 9a909e5d015332dc445bd9592e6e386d
543]
544[doc updates in System.IO
545Simon Marlow <marlowsd@gmail.com>**20100629122118
546 Ignore-this: 2257ec1cc4cdb8b7804cfa1f3cf32753
547]
548[doc wibble
549Simon Marlow <marlowsd@gmail.com>**20100625134858
550 Ignore-this: 64c50f29df6c389273b818918fe7033a
551]
552[doc wibbles
553Simon Marlow <marlowsd@gmail.com>**20100624154614
554 Ignore-this: b364aad53beea6e741fee2824459b6e8
555]
556[Fix haddock formatting
557Ian Lynagh <igloo@earth.li>**20100625222623]
558[Give nub's complexity in the haddock docs; fixes #4086
559Ian Lynagh <igloo@earth.li>**20100625222059]
560[correct docs for exitWith: only stdout/stderr are flushed, not all Handles
561Simon Marlow <marlowsd@gmail.com>**20100624130506
562 Ignore-this: 33a938dad8f0bc061572e2ec571cacc7
563]
564[fix docs for isSpace
565Simon Marlow <marlowsd@gmail.com>**20100624130444
566 Ignore-this: b35ff080dbb9833176f08e39dbd9ff6d
567]
568[make the hGetBuf/hPutBuf family work with non-FD Handles (#4144)
569Simon Marlow <marlowsd@gmail.com>**20100624130425
570 Ignore-this: 8200f0208a9b1b1cf4824f343d75819a
571]
572[nit in docs for accumArray
573Simon Marlow <marlowsd@gmail.com>**20100622121131
574 Ignore-this: c066a456c40907e767df10c3990f35ff
575]
576[add doc for the ExitCode type
577Simon Marlow <marlowsd@gmail.com>**20100622120930
578 Ignore-this: 99c34332be7f3565da844528b470054a
579]
580[remove extraneous info from docs for Array
581Simon Marlow <marlowsd@gmail.com>**20100622120921
582 Ignore-this: e2a3f5e84fc23eb7bae911f0680e805e
583]
584[add an INLINE to the list version of traverse, to enable fusion
585Simon Marlow <marlowsd@gmail.com>**20100608082531
586 Ignore-this: ea98cdc3308b406bb04c0f7a38c4424b
587]
588[Don't define the C localeEncoding on Windows
589Ian Lynagh <igloo@earth.li>**20100620202342
590 Ignore-this: c4992f6832a391b0cccc5a9b7d643976
591 (it causes warnings, and isn't used)
592]
593[add Applicative instance for Either (proposal #4095)
594Ross Paterson <ross@soi.city.ac.uk>**20100617225110
595 Ignore-this: 50262ec4700dc16efec5755be5b308c5
596 
597 This is not the only possible instance for Either, but this one is
598 compatible with the usual Monad instance.
599]
600[Use libcharset instead of nl_langinfo(CODESET) if possible.
601pho@cielonegro.org**20100519013112
602 Ignore-this: 4c1e278e022a3d276848afc1dcba4425
603 
604 nl_langinfo(CODESET) doesn't always return standardized variations of the encoding names. Use libcharset if possible, which is shipped together with GNU libiconv.
605]
606[Add a note about the interruptibility of throwTo.
607Simon Marlow <marlowsd@gmail.com>**20100615112720
608 Ignore-this: ae9fabe95310d7c364e95f7784793485
609]
610[docs: note that hGetBufNonBlocking isn't non-blocking on Windows
611Simon Marlow <marlowsd@gmail.com>**20100615112547
612 Ignore-this: 4f3e5213e142149affe08c5123d6efea
613]
614[don't depend on Prelude (#4122)
615Simon Marlow <marlowsd@gmail.com>**20100615105631
616 Ignore-this: 1a3fd49b103fe31cbb453f302c18767f
617]
618[Don't depend on Prelude (#4123)
619Simon Marlow <marlowsd@gmail.com>**20100615105401
620 Ignore-this: cc7616d85a1637bc7621b4f2bc181c0e
621]
622[bump version to 4.3.0.0, added instance MonadPlus STM
623Simon Marlow <marlowsd@gmail.com>**20100601144831
624 Ignore-this: 7c3cf7574499c4267372493f2636dc0
625]
626[Moved MonadPlus instance for STM from Control.Monad.STM to GHC.Conc to avoid an orphaned instance
627Bas van Dijk <v.dijk.bas@gmail.com>**20100516160651
628 Ignore-this: 651b852942b2fae2b93f996e39239b8f
629]
630[expand Foldable instance for Array
631Ross Paterson <ross@soi.city.ac.uk>**20100602212154
632 Ignore-this: 9bd9e9666a9400431eb92352244fe7e7
633]
634[doc comment illustrating Foldable(foldr)
635Ross Paterson <ross@soi.city.ac.uk>**20100527150833
636 Ignore-this: 8f27d889379803f3ba86d6e928428f3c
637]
638[fix syntax in doc comments
639Ross Paterson <ross@soi.city.ac.uk>**20100527150757
640 Ignore-this: cb78da51d60ff6863dc395f1a892c103
641]
642[export hGetBufSome (#4046)
643Simon Marlow <marlowsd@gmail.com>**20100520093538
644 Ignore-this: f467fad9722e27edfad6b3dd75290e7b
645]
646[hWaitForInput: don't try to read from the device (#4078)
647Simon Marlow <marlowsd@gmail.com>**20100517133741
648 Ignore-this: 55ec33b03397380259b91e4ca62207a6
649 readTextDeviceNonBlocking is not non-blocking on Windows
650]
651[Added Applicative and Alternative instances for STM
652Bas van Dijk <v.dijk.bas@gmail.com>**20100516171756
653 Ignore-this: 567003bc4040bc97105cda4d31ebf04a
654]
655[hSetEncoding: change the encoding on both read and write sides (#4066)
656Simon Marlow <marlowsd@gmail.com>**20100514124628
657 Ignore-this: 5b9e9caef06356d0296c584159709ebb
658]
659[Correct haddock formatting.
660Adam Vogt <vogt.adam@gmail.com>**20100423022103
661 Ignore-this: d2622339302048fda48080f7d5ce4a2f
662]
663[Fix for hGetBufSome
664Simon Marlow <marlowsd@gmail.com>**20100505135637
665 Ignore-this: 2019680f8fb223956cacfcf0d046f133
666]
667[improve the documentation for throwTo and killThread (#3884)
668Simon Marlow <marlowsd@gmail.com>**20100505135600
669 Ignore-this: ce881d96ddb729acb6ca09c779975e7d
670]
671[elaborate the docs for unsafePerformIO a bit
672Simon Marlow <marlowsd@gmail.com>**20100505101249
673 Ignore-this: 1cec3f67560b672c64c5a0dcf9a79eb7
674]
675[add Typeable instance
676Simon Marlow <marlowsd@gmail.com>**20100504152815
677 Ignore-this: 6d9cf9d62f0ef17fa459bf213a04098
678]
679[Add hGetBufSome, like hGetBuf but can return short reads
680Simon Marlow <marlowsd@gmail.com>**20100504152759
681 Ignore-this: 195c905b43f8d9505029364e2c5b18e
682]
683[Add swap (#3298)
684Simon Marlow <marlowsd@gmail.com>**20100504095339
685 Ignore-this: 13b007dc4594ce252997ec6fa0bbd976
686]
687[inline allocaArray0, to fix withCString benchmark
688Simon Marlow <marlowsd@gmail.com>**20100423124729
689 Ignore-this: 35c96816acc2f3aaf9dd29f7995fa6f0
690]
691[raise asynchronous exceptions asynchronously (#3997)
692Simon Marlow <marlowsd@gmail.com>**20100421094932
693 Ignore-this: 6d987d93d382c0f69c68c326312abd6b
694]
695[add NOINLINE pragmas for stdin/stdout/stderr
696Simon Marlow <marlowsd@gmail.com>**20100421082041
697 Ignore-this: 3fc130268ec786f28d945858d6690986
698]
699[INLINE alloca and malloc
700Simon Marlow <marlowsd@gmail.com>**20100419135333
701 Ignore-this: b218bd611f18721b1505a8c0b9e6a16a
702 See discussion on glasgow-haskell-users:
703   http://www.haskell.org/pipermail/glasgow-haskell-users/2010-April/018740.html
704]
705[Move comment closer to the offending line
706Matthias Kilian <kili@outback.escape.de>**20100419155421
707 Ignore-this: b34a1d7affd66f67d210df2377b585d9
708]
709[Ignore the return code of c_fcntl_write again
710Matthias Kilian <kili@outback.escape.de>**20100415140452
711 Ignore-this: 266d8ba02cc3cb79c85629b3528261c9
712 
713 The return code has been ignored in the past on purpose, because
714 O_NONBLOCK will fail on BSDs for some special files. This fixes the
715 problem mentioned in
716 http://www.haskell.org/pipermail/glasgow-haskell-users/2010-April/018698.html
717 
718]
719[Fix bitrot in IO debugging code
720Ian Lynagh <igloo@earth.li>**20100413134339
721 Also switched to using Haskell Bools (rather than CPP) to en/disable it,
722 so it shouldn't break again in the future.
723]
724[Tiny code tidy-up
725Ian Lynagh <igloo@earth.li>**20100413011147]
726[remove old/wrong comment
727Simon Marlow <marlowsd@gmail.com>**20100325161403
728 Ignore-this: e6e377d44af48c4162d17d55bdf3f821
729]
730[withThread: block asynchronous exceptions before installing exception handler.
731Bas van Dijk <v.dijk.bas@gmail.com>**20100329131624
732 Ignore-this: be5aeb47dbd73807b5f94df11afbb81c
733 Note that I don't unblock the given io computation. Because AFAICS
734 withThread is only called with 'waitFd' which only performs an FFI
735 call which can't receive asynchronous exceptions anyway.
736]
737[runInUnboundThread: block asynchronous exceptions before installing exception handler
738Bas van Dijk <v.dijk.bas@gmail.com>**20100329131549
739 Ignore-this: a00c5e32fe3981ff87bedd367a69051e
740]
741[fix the deprecation message (GHC.IO.Handle.Base -> GHC.IO.Handle)
742Simon Marlow <marlowsd@gmail.com>**20100330121137
743 Ignore-this: 4ca8500a01ac93454507aa8f9dd001f9
744]
745[Make SampleVar an abstract newtype
746Bas van Dijk <v.dijk.bas@gmail.com>**20100318200349
747 Ignore-this: 27939e2a064b75e71cb146117346be30
748]
749[Fix bugs regarding asynchronous exceptions and laziness in Control.Concurrent.SampleVar
750Bas van Dijk <v.dijk.bas@gmail.com>**20100318200104
751 Ignore-this: 7376b2a3afe155daf233a8f1ddc0a7a
752  - Block asynchronous exceptions at the right places
753  - Force thunks before putting them in a MVar
754]
755[Write the thunk 'next' to the MVar
756Bas van Dijk <v.dijk.bas@gmail.com>**20100319125951
757 Ignore-this: dd25636cf220131385ff2fd32493d456
758]
759[change to use STM, fixing 4 things
760Simon Marlow <marlowsd@gmail.com>**20100318104436
761 Ignore-this: 551d30280a7941c08f5c3b14576bdd70
762   1. there was no async exception protection
763   2. there was a space leak (now new value is strict)
764   3. using atomicModifyIORef would be slightly quicker, but can
765      suffer from adverse scheduling issues (see #3838)
766   4. also, the STM version is faster.
767]
768[Tweak docs
769Ian Lynagh <igloo@earth.li>**20100312214129]
770[Fixed dead links in documentation of forkIO
771Bas van Dijk <v.dijk.bas@gmail.com>**20100308222415
772 Ignore-this: 7deb8fd064c867fbede2a6b2e9da4f15
773]
774[Documentation fixes in Control.Exception
775Bas van Dijk <v.dijk.bas@gmail.com>**20100301220442
776 Ignore-this: 761fcba401cbd1f47276ddfc9b5b80f2
777]
778[Plug two race conditions that could lead to deadlocks in the IO manager
779Simon Marlow <marlowsd@gmail.com>**20100225120255
780 Ignore-this: e6983d6b953104d370278ab3e4617e8b
781]
782[FIX #3866: improve documentation of Data.Data.Constr
783jpm@cs.uu.nl**20100224125506
784 Ignore-this: 3818c5d8fee012a3cf322fb455b6e5dc
785]
786[UNDO: Handle NaN, -Infinity and Infinity in the toRational for Float/Double (#3676)
787Simon Marlow <marlowsd@gmail.com>**20100223101603
788 Ignore-this: 78becb2d39b3cd9a1a473a5811ca7d92
789]
790[Put the complexity in the length docs. Fixes trac #3680
791Ian Lynagh <igloo@earth.li>**20100221191425]
792[nhc98 should build Data.Functor.
793Malcolm.Wallace@cs.york.ac.uk**20100221163218]
794[Update the exitWith docs
795Ian Lynagh <igloo@earth.li>**20100213140004
796 Error pointed out by Volker Wysk <vw@volker-wysk.de>
797]
798[Handle NaN, -Infinity and Infinity in the toRational for Float/Double (#3676)
799Simon Marlow <marlowsd@gmail.com>**20100211101955
800 Ignore-this: 261415363303efca265e80290eac5f28
801]
802[For nhc98, import unsafeInterleaveIO rather than defining it here.
803Malcolm.Wallace@cs.york.ac.uk**20100204171021]
804[Stifle warning about unused return value
805benl@cse.unsw.edu.au**20100203025537]
806[fix #3832: use the locale encoding in openTempFile
807Simon Marlow <marlowsd@gmail.com>**20100120211830
808 Ignore-this: df4f778cc5fefb32290c798db722632c
809 Also while I was here fix an XXX: the Handle contained an
810 uninformative string like <fd: 4> for error messages rather than the
811 real file path.
812]
813[Fix the build: export void, so it doesn't give an unused binding warning
814Ian Lynagh <igloo@earth.li>**20100116174451]
815[hIsEOF: don't do any decoding (#3808)
816Simon Marlow <marlowsd@gmail.com>**20100112230317
817 Ignore-this: 6a384dd2d547ffe3ad3762920e5c1671
818]
819[Control.Monad: +void :: f a -> f ()
820gwern0@gmail.com**20100108214455
821 Ignore-this: 4dc07452315f2d1b4941903ff42fc45f
822 See http://hackage.haskell.org/trac/ghc/ticket/3292
823 Turns m a -> m (). Lets one call functions for their side-effects without
824 having to get rid of their return values with '>> return ()'. Very useful
825 in many contexts (parsing, IO etc.); particularly good for 'forkIO' and 'forM_',
826 as they demand return types of 'IO ()' though most interesting IO functions
827 return non-().
828]
829[Replace the implementation of mergesort with a 2x faster one.
830Malcolm.Wallace@cs.york.ac.uk**20091224152014
831 See ticket http://hackage.haskell.org/trac/ghc/ticket/2143.
832]
833[Restore previous Data.Typeable.typeOf*Default implementations for non-ghc.
834Malcolm.Wallace@cs.york.ac.uk**20091223142625
835 Not all compilers have ScopedTypeVariables.
836]
837[Add comments about double bounds-checking, and fast paths for rectangular arrays
838simonpj@microsoft.com**20091218165655
839 Ignore-this: ea0849419dc00927aba4bd410b1cc58d
840 
841 See Note [Double bounds-checking of index values] for the details.
842 
843 The fast paths omit the doubled checks for cases we know about
844]
845[Fix Trac #3245: memoising typeOf
846simonpj@microsoft.com**20091218155117
847 Ignore-this: 5a178a7f2222293c5ee0c3c43bd1b625
848 
849 The performance bug in #3245 was caused by computing the typeRep
850 once for each call of typeOf, rather than once for each dictionary
851 contruction.  (Computing TypeReps is reasonably expensive, because
852 of the hash-consing machinery.)
853 
854 This is readily fixed by putting the TypeRep construction outside
855 the lambda.  (Arguably GHC might have worked that out itself,
856 but it involves floating something between a type lambda and a
857 value lambda, which GHC doesn't currently do. If it happens a lot
858 we could fix that.)
859]
860[Mark 'index' as INLINE in GHC.Arr
861simonpj@microsoft.com**20091216170441
862 Ignore-this: a4df9d8acf496c8e0e9ce5a520509a2a
863 
864 This makes indexing much faster. See Trac #1216
865]
866[Comment the remaining orphan instance modules
867Ian Lynagh <igloo@earth.li>**20091206125021]
868[De-orphan Eq/Ord Float/Double
869Ian Lynagh <igloo@earth.li>**20091205181238]
870[Add comments to "OPTIONS_GHC -fno-warn-orphans" pragmas
871Ian Lynagh <igloo@earth.li>**20091205165854]
872[Data.Either.partitionEithers was insufficiently lazy.
873Malcolm.Wallace@cs.york.ac.uk**20091202032807
874 Ignore-this: 77e1b3288f66608c71458d8a91bcbe12
875 Spotted by Daniel Fischer.
876]
877[fix the docs regarding finalizer guarantees
878Simon Marlow <marlowsd@gmail.com>**20091130144409
879 Ignore-this: d1ab9532c74a002b8075ff60febcbe2d
880]
881[x86_64 requires more stack
882Malcolm.Wallace@cs.york.ac.uk**20091201033745]
883[check for size < 0 in mallocForeignPtrBytes and friends (#3514)
884Simon Marlow <marlowsd@gmail.com>**20091125143822
885 Ignore-this: 91077d01da2bbe1dfed5155e8b40da9
886]
887[hGetContents: close the handle properly on error
888Simon Marlow <marlowsd@gmail.com>**20091125123435
889 Ignore-this: bc37ff678acc6e547dc390285e056eb9
890 
891 When hGetContents caught an error it was closing the handle and then
892 throwing the exception, without updating the handle with the new
893 closed state.  This lead to a double-closed, which was the cause of
894 
895 *** glibc detected *** ./Setup: double free or corruption
896 
897 when iconv_close was called twice on the decoder.
898 
899 See http://hackage.haskell.org/trac/hackage/ticket/609
900]
901[Fix arities of mapFB and zipFB
902Roman Leshchinskiy <rl@cse.unsw.edu.au>**20091126232219
903 Ignore-this: c4e14cd0a92622549c86e67237a40865
904]
905[Remove an unnecessary -fno-warn-orphans flag
906Ian Lynagh <igloo@earth.li>**20091126123404]
907[Tweak layout to work with alternative layout rule
908Ian Lynagh <igloo@earth.li>**20091125232349]
909[Tweak layout to be accepted by the alternative layout rul
910Ian Lynagh <igloo@earth.li>**20091125194147]
911[Make sure zipWithFB has arity 2
912Roman Leshchinskiy <rl@cse.unsw.edu.au>**20091125010003
913 Ignore-this: 4cf60c55666f03d22a9f5a6e07f52d36
914 
915 It gets 2 arguments in the "zipWith" rule but its arity was higher and the new
916 inliner didn't inline it sometimes, for instance here:
917 
918 mpp ::  [Double] -> [Double] -> [Double] -> [Double] -> [Double]
919 mpp as bs cs ds = zipWith (*) (zipWith (+) as bs) (zipWith (+) cs ds)
920 
921 This was a regression vs. 6.10.
922]
923[Remove an old comment
924Ian Lynagh <igloo@earth.li>**20091124134647]
925[De-orphan the Eq/Ord Integer instances
926Ian Lynagh <igloo@earth.li>**20091124133639]
927[Whitespace only
928Ian Lynagh <igloo@earth.li>**20091124133421]
929[Derive some more instances, rather than writing them by hand
930Ian Lynagh <igloo@earth.li>**20091124011747]
931[We can now derive Ord ()
932Ian Lynagh <igloo@earth.li>**20091124011416]
933[De-orphan tuple Eq/Ord instances
934Ian Lynagh <igloo@earth.li>**20091123233343]
935[Control.Exception.Base no longer has any orphans
936Ian Lynagh <igloo@earth.li>**20091123224905]
937[De-orphan the MonadFix ST instance for GHC
938Ian Lynagh <igloo@earth.li>**20091123223544]
939[Rearrange the contents of Control.Monad.ST; no functionality changes
940Ian Lynagh <igloo@earth.li>**20091123222702]
941[De-orphan the Eq/Ord [a] instances
942Ian Lynagh <igloo@earth.li>**20091123215635]
943[De-orphan the Eq/Ord Char instances
944Ian Lynagh <igloo@earth.li>**20091123202253]
945[De-orphan the Eq/Ord Bool instances
946Ian Lynagh <igloo@earth.li>**20091123201817]
947[Move Eq/Ord Ordering instances to de-orphan them
948Ian Lynagh <igloo@earth.li>**20091123194310]
949[Remove ffi warnings for nhc98.
950Malcolm.Wallace@cs.york.ac.uk**20091123063743]
951[Second attempt to fix #1185 (forkProcess and -threaded)
952Simon Marlow <marlowsd@gmail.com>**20091111151915
953 Ignore-this: fa5f5d5e4e080d4b612a37244f937f9c
954 
955 Patch 2/2: first patch is to ghc
956 
957 This time without dynamic linker hacks, instead I've expanded the
958 existing rts/Globals.c to cache more CAFs, specifically those in
959 GHC.Conc.  We were already using this trick for signal handlers, I
960 should have realised before.
961 
962 It's still quite unsavoury, but we can do away with rts/Globals.c in
963 the future when we switch to a dynamically-linked GHCi.
964]
965[Rollback #1185 fix
966Simon Marlow <marlowsd@gmail.com>**20091106140629
967 Ignore-this: cd5667e8474e37e01ba26a1984274811
968 
969 rolling back:
970 
971 Tue Nov  3 16:05:40 GMT 2009  Simon Marlow <marlowsd@gmail.com>
972   * Fix #1185: restart the IO manager after fork()
973   
974   This is the libraries/base part of the patch; there is a corresponding
975   patch to GHC itself.
976   
977   The main change is that we now keep track of the IO manager's ThreadId
978   in a top-level MVar, and ensureIOManagerIsRunning checks whether a
979   previous IO manager thread is alive before starting one.  In the child
980   of fork(), we can hence call ensureIOManagerIsRunning to restart the
981   IO manager.
982 
983     M ./GHC/Conc.lhs -46 +44
984 
985 Wed Nov  4 17:49:45 GMT 2009  Ian Lynagh <igloo@earth.li>
986   * Fix the build on Windows
987 
988     M ./GHC/Conc.lhs -6 +4
989]
990[Fix the build on Windows
991Ian Lynagh <igloo@earth.li>**20091104174945]
992[Fix #1185: restart the IO manager after fork()
993Simon Marlow <marlowsd@gmail.com>**20091103160540
994 Ignore-this: 6dc05464f1500104554637f4759738cc
995 
996 This is the libraries/base part of the patch; there is a corresponding
997 patch to GHC itself.
998 
999 The main change is that we now keep track of the IO manager's ThreadId
1000 in a top-level MVar, and ensureIOManagerIsRunning checks whether a
1001 previous IO manager thread is alive before starting one.  In the child
1002 of fork(), we can hence call ensureIOManagerIsRunning to restart the
1003 IO manager.
1004]
1005[improve the documentation for throwErrnoIfRetry
1006Simon Marlow <marlowsd@gmail.com>**20091016112404
1007 Ignore-this: b77275cacf730e15757946027168f63e
1008]
1009[Don't inline unpackFoldrCString ever
1010simonpj@microsoft.com**20091029135350
1011 Ignore-this: 85d672649b1b776efc7e97500b05d4f9
1012]
1013[Inline more default methods
1014simonpj@microsoft.com**20091029135330
1015 Ignore-this: 289c44b0afd6d5631c2a4e0664275ca9
1016 
1017 Namely Monad: (>>)
1018        Eq:    (==), (/=)
1019        Num:   (-), negate
1020        Real:  quot, rem, div, mod, recip, (/), truncate
1021        Float: (**), logBase, sqrt, tan, tanh
1022]
1023[Move error messages out of INLINEd default methods
1024simonpj@microsoft.com**20091029135118
1025 Ignore-this: 9e35dc947f94827a3529eb53a41575fd
1026 
1027 No need to duplicate the error generation!
1028]
1029[Exploit now-working default-method INLINE pragmas for Data.Bits
1030simonpj@microsoft.com**20091029135041
1031 Ignore-this: 8adf225f31ca7a3181ee087e9e4fe535
1032 
1033 * Add INLINE pragmas to default methods for class Bits
1034 
1035 * Remove redundant instance methods elsewhere, now that
1036   the default method will do the job
1037]
1038[Tidy up and comment imports
1039simonpj@microsoft.com**20091029134414
1040 Ignore-this: bf2be31035de975d8995e988933cc940
1041]
1042[Inline foldr and (.) when applied to two arguments not three
1043simonpj@microsoft.com**20091029134335
1044 Ignore-this: fccb6f3e90e15f44cb465814be85ede2
1045 
1046 The new INLINE story is (by design) arity-sensitive, so we must
1047 put fewer argument on the LHS for foldr and (.)
1048]
1049[dirUtils.c no longer available
1050Malcolm.Wallace@cs.york.ac.uk**20091013093833]
1051[Make hGetContents throw an exception if an error is encountered
1052Simon Marlow <marlowsd@gmail.com>**20091012152955
1053 Ignore-this: 9f7a7176193eab25c9daaacd9261f2de
1054 
1055 Strictly speaking this breaks Haskell 98 compatibility, which requires
1056 hGetContents to just end the lazy stream silently if an error is
1057 encountered.  However, for a few reasons we think it will make
1058 everyone's life a bit easier if we make this change
1059 
1060  1. Errors will be a lot more common in GHC 6.12.1, in the form
1061     of Unicode decoding errors.
1062 
1063  2. When Haskell 98 was designed, we didn't know how to throw
1064     exceptions from inside lazy I/O, but now we do.
1065 
1066  3. If anyone is actually relying on the previous behaviour, their
1067     code is arguably broken.
1068]
1069[Re-instate System.Console.Getopt for nhc98 builds.
1070Malcolm.Wallace@cs.york.ac.uk**20091013092843
1071 Although it was split out of base a while back, that change was
1072 reverted for ghc soon afterwards, but nhc98 never noticed.
1073]
1074[Roll back "Another instance of nhc98's strange import semantics."
1075Ian Lynagh <igloo@earth.li>**20091009185618
1076 Fri Oct  9 14:50:51 BST 2009  Malcolm.Wallace@cs.york.ac.uk
1077 GHC (correctly) warns about the unused import, which breaks the validate
1078 build.
1079]
1080[Roll back "Cope with nhc98's (occasionally-strange) import semantics"
1081Ian Lynagh <igloo@earth.li>**20091009184704
1082 Fri Oct  9 14:43:51 BST 2009  Malcolm.Wallace@cs.york.ac.uk
1083 GHC (correctly) warns about the unused import, which breaks the validate
1084 build.
1085]
1086[It seems that nhc98 needs defaulting in Data.Fixed.
1087Malcolm.Wallace@cs.york.ac.uk**20091009135242]
1088[Another instance of nhc98's strange import semantics.
1089Malcolm.Wallace@cs.york.ac.uk**20091009135051]
1090[Make Data.Functor compatible with non-GHC compilers.
1091Malcolm.Wallace@cs.york.ac.uk**20091009134821]
1092[Cope with nhc98's (occasionally-strange) import semantics.
1093Malcolm.Wallace@cs.york.ac.uk**20091009134351]
1094[Fix gratuitous breakage of nhc98 in System.IO.
1095Malcolm.Wallace@cs.york.ac.uk**20091009134001]
1096[Fix gratuitous breakage of nhc98 in Control.Exception.Base.
1097Malcolm.Wallace@cs.york.ac.uk**20091009133615]
1098[Fix gratuitous breakage of non-GHC in Data.Fixed.
1099Malcolm.Wallace@cs.york.ac.uk**20091009133330]
1100[Fix gratuitous breakage for non-GHC in Data.Bits.
1101Malcolm.Wallace@cs.york.ac.uk**20091009133257]
1102[Use UTF-32LE instead of UTF32LE
1103Simon Marlow <marlowsd@gmail.com>**20091006100207
1104 Ignore-this: 7f881e36543d250ef848c9f60d67655a
1105 The latter is not recognised by some iconv implementations.
1106]
1107[Strip any Byte Order Mark (BOM) from the front of decoded streams.
1108Ben.Lippmeier@anu.edu.au*-20090930084229
1109 Ignore-this: d0d0c3ae87b31d71ef1627c8e1786445
1110 When decoding to UTF-32, Solaris iconv inserts a BOM at the front
1111 of the stream, but Linux iconv doesn't.
1112]
1113[use UTF32BE/UTF32LE instead of UCS-4/UCS-4LE
1114Simon Marlow <marlowsd@gmail.com>**20091005101554
1115 Ignore-this: 2aef5e9bec421e714953b7aa1bdfc1b3
1116]
1117[Strip any Byte Order Mark (BOM) from the front of decoded streams.
1118Ben.Lippmeier@anu.edu.au**20090930084229
1119 Ignore-this: d0d0c3ae87b31d71ef1627c8e1786445
1120 When decoding to UTF-32, Solaris iconv inserts a BOM at the front
1121 of the stream, but Linux iconv doesn't.
1122]
1123[Add traceEvent :: String -> IO ()
1124Simon Marlow <marlowsd@gmail.com>**20090925141257
1125 Ignore-this: 8b1888bbf9682ffba13f815b6000e4b1
1126 For emitting an event via the RTS tracing framework
1127]
1128[Fix the error message when flushing the read buffer of a non-seekable Handle
1129Simon Marlow <marlowsd@gmail.com>**20090923090536
1130 Ignore-this: 4342026df93759d99480f4e13f80a492
1131]
1132[Fix #3534: No need to flush the byte buffer when setting binary mode
1133Simon Marlow <marlowsd@gmail.com>**20090923090445
1134 Ignore-this: 625817ed7ae2c12291eb993a99dc640a
1135]
1136[Use let !y = x in .. x .. instead of seq in $! and evaluate (#2273)
1137Simon Marlow <marlowsd@gmail.com>**20090916140454]
1138[make some Applicative functions into methods, and split off Data.Functor (proposal #3335)
1139Ross Paterson <ross@soi.city.ac.uk>**20090915173109
1140 Ignore-this: a0cff4de6dfdbcbd56a66101bc4855a9
1141 
1142 The following functions
1143 
1144     (<$) :: Functor f => a -> f b -> f a
1145     (*>) :: Applicative f => f a -> f b -> f b
1146     (<*) :: Applicative f => f a -> f b -> f a
1147     some :: Alternative f => f a -> f [a]
1148     many :: Alternative f => f a -> f [a]
1149 
1150 are moved into the corresponding classes, with the existing implementations
1151 as default definitions.  This gives people creating instances the option of
1152 defining specialized implementations of these functions, though they should
1153 be equivalent to the default definitions.
1154 
1155 Although (<$) is now a method of the Functor class, it is hidden in the
1156 re-export by the Prelude, Control.Monad and Monad.  The new module
1157 Data.Functor exposes the full class, plus the function (<$>).  These are
1158 also re-exported by Control.Applicative.
1159]
1160[On Windows, use the console code page for text file encoding/decoding.
1161Judah Jacobson <judah.jacobson@gmail.com>**20090913022126
1162 Ignore-this: 86c2f2db8ef92b751599795d3195187b
1163 
1164 We keep all of the code page tables in the module
1165 GHC.IO.Encoding.CodePage.Table.  That file was generated automatically
1166 by running codepages/MakeTable.hs; more details are in the comments at the
1167 start of that script.
1168 
1169 Storing the lookup tables adds about 40KB to each statically linked executable;
1170 this only increases the size of a "hello world" program by about 7%.
1171 
1172 Currently we do not support double-byte encodings (Chinese/Japanese/Korean), since
1173 including those codepages would increase the table size to 400KB.  It will be
1174 straightforward to implement them once the work on library DLLs is finished.
1175]
1176[Fix "init" docs: the input list need not be finite. Fixes trac #3465
1177Ian Lynagh <igloo@earth.li>**20090911210437]
1178[Bump base version to 4.2.0.0
1179Ian Lynagh <igloo@earth.li>**20090911153913]
1180[Address #3310
1181Simon Marlow <marlowsd@gmail.com>**20090830152850
1182 Ignore-this: 40c7f7c171ee299a83092fd360a952b7
1183 
1184  - Rename BlockedOnDeadMVar   -> BlockedIndefinitelyOnMVar
1185  - Rename BlockedIndefinitely -> BlockedIndefinitelyOnSTM
1186  - instance Show BlockedIndefinitelyOnMVar is now
1187      "blocked indefinitely in an MVar operation"
1188  - instance Show BlockedIndefinitelyOnSTM is now
1189      "blocked indefinitely in an STM transaction"
1190 
1191 clients using Control.OldException will be unaffected (the new
1192 exceptions are mapped to the old names).  However, for base4-compat
1193 we'll need to make a version of catch/try that does a similar
1194 mapping.
1195]
1196[Fix unicode conversion for MSB architectures
1197Ben.Lippmeier@anu.edu.au**20090830130028
1198 This fixes the SPARC/Solaris build.
1199]
1200[Fix #3441: detect errors in partial sequences
1201Simon Marlow <marlowsd@gmail.com>**20090830075909
1202 Ignore-this: d12a75d95e0cae5eb1555266810ec281
1203]
1204[Fix hWaitForInput
1205Simon Marlow <marlowsd@gmail.com>**20090827152116
1206 Ignore-this: 2550e911f1a4d4357a5aa8d1764238ce
1207 It was erroneously waiting when there were bytes to decode waiting in
1208 the byte buffer.
1209]
1210[fix debugging code
1211Simon Marlow <marlowsd@gmail.com>**20090827150628
1212 Ignore-this: e1c82fdc19a22e247cd69ff6fa11921d
1213]
1214[Allow for configurable iconv include and library locations.
1215Matthias Kilian <kili@outback.escape.de>**20090826154406
1216 Ignore-this: be95fab611a5534cf184b508964ed498
1217 This should help to fix the build on OpenBSD.
1218]
1219[typo in comment
1220Simon Marlow <marlowsd@gmail.com>**20090826085252
1221 Ignore-this: 1903ee0f354157a6ed3871c100f6b1b9
1222]
1223[un-hide some modules from the Haddock docs
1224Simon Marlow <marlowsd@gmail.com>**20090825152457
1225 Ignore-this: dce6606f93cf977fb24ebe99082dfa62
1226]
1227[Apply fix for #1548, from squadette@gmail.com
1228Simon Marlow <marlowsd@gmail.com>**20090819120700
1229 Ignore-this: 31c237c46a6445f588ed4b8c51bb6231
1230]
1231[improvements to Data.Fixed: instances for Typeable and Data, more predefined types
1232Ashley Yakeley <ashley@semantic.org>**20090812055058
1233 Ignore-this: feeece36d5632f02a05d137d2a39ab78
1234]
1235[Fix "Cabal check" warnings
1236Ian Lynagh <igloo@earth.li>**20090811215856]
1237[Add a GHC.Constants module; fixes trac #3094
1238Ian Lynagh <igloo@earth.li>**20090809183252]
1239[Apply proposal #3393
1240Ian Lynagh <igloo@earth.li>**20090809134717
1241 Add openTempFileWithDefaultPermissions and
1242 openBinaryTempFileWithDefaultPermissions.
1243]
1244[Add some more C wrappers; patch from Krister Walfridsson
1245Ian Lynagh <igloo@earth.li>**20090807200631
1246 Fixes 21 testsuite errors on NetBSD 5.99.
1247]
1248[Fixing configure for autoconf 2.64
1249Alexander Dunlap <alexander.dunlap@gmail.com>**20090805060748
1250 Ignore-this: 992ab91ae3d68c12dbb265776e33e243
1251]
1252[add INLINE toList
1253Ross Paterson <ross@soi.city.ac.uk>**20090806142853
1254 Ignore-this: aba16aabb17d5dca44f15d188945680e
1255 
1256 In anticipation of the fixing of #2353.
1257]
1258[fix a copyright
1259Simon Marlow <marlowsd@gmail.com>**20090805134045
1260 Ignore-this: b0ffbdd38fbba121e8bcba37c4082a60
1261]
1262[Tweak the BufferedIO class to enable a memory-mapped file implementation
1263Simon Marlow <marlowsd@gmail.com>**20090805134036
1264 Ignore-this: ec67d7a0a6d977438deaa342503f77e0
1265 We have to eliminate the assumption that an empty write buffer can be
1266 constructed by setting the buffer pointers to zero: this isn't
1267 necessarily the case when the buffer corresponds to a memory-mapped
1268 file, or other in-memory device implementation.
1269]
1270[Deprecate Control.OldException
1271Ian Lynagh <igloo@earth.li>**20090804143910]
1272[Windows build fix, following RTS tidyup
1273Simon Marlow <marlowsd@gmail.com>**20090803131121
1274 Ignore-this: ce862fb91c2b234211a8757f98690778
1275]
1276[Updates to follow the RTS tidyup
1277Simon Marlow <marlowsd@gmail.com>**20090801220743
1278 Ignore-this: 6e92412df93a66c12d75344053d5634
1279 C functions like isDoubleNaN moved here (primFloat.c)
1280]
1281[Add integer-simple as a build option
1282Ian Lynagh <igloo@earth.li>**20090722013151]
1283[Use shift[LR]Integer in the Bits Integer instance
1284Ian Lynagh <igloo@earth.li>**20090721222440]
1285[depend directly on integer-gmp, rather than indirecting through integer
1286Ian Lynagh <igloo@earth.li>**20090721185228]
1287[Move the instances of Functor and Monad IO to GHC.Base, to avoid orphans
1288Simon Marlow <marlowsd@gmail.com>**20090722102130
1289 Ignore-this: a7d85ac0025d559674249de0108dbcf4
1290]
1291[move "instance Exception Dynamic" so it isn't an orphan
1292Simon Marlow <marlowsd@gmail.com>**20090721093854
1293 Ignore-this: 5ede91ecfec2112c91b699d4de87cd02
1294]
1295[Improve the index checking for array accesses; fixes #2120 #2669
1296Ian Lynagh <igloo@earth.li>**20090719153228
1297 As well as checking that offset we are reading is actually inside the
1298 array, we now also check that it is "in range" as defined by the Ix
1299 instance. This fixes confusing behaviour (#2120) and improves some error
1300 messages (#2669).
1301]
1302[Make chr say what its argument was, if it's a bad argument
1303Ian Lynagh <igloo@earth.li>**20090718151049]
1304[remove unused warning
1305Simon Marlow <marlowsd@gmail.com>**20090715124416
1306 Ignore-this: 31f613654089d0f4a44363946087b41e
1307]
1308[warning fix: -fno-implicit-prelude -> -XNoImplicitPrelude
1309Simon Marlow <marlowsd@gmail.com>**20090715122839
1310 Ignore-this: dc8957249731d5bcb71c01899e5adf2b
1311]
1312[Add hGetEncoding :: Handle -> IO (Maybe TextEncoding)
1313Simon Marlow <marlowsd@gmail.com>**20090715122519
1314 Ignore-this: 14c3eff996db062da1199739781e4708
1315 as suggested during the discussion on the libraries list
1316]
1317[Add more documentation to mkTextEncoding
1318Simon Marlow <marlowsd@gmail.com>**20090715122414
1319 Ignore-this: 97253b2624267df3a246a18121e8ea81
1320 noting that "//IGNORE" and "//TRANSLIT" suffixes can be used with GNU
1321 iconv.
1322]
1323[Add the utf8_bom codec
1324Simon Marlow <marlowsd@gmail.com>**20090715122257
1325 Ignore-this: 1c9396cd805201fe873a39382ced79c7
1326 as suggested during the discussion on the libraries list.
1327]
1328[Export Unicode and newline functionality from System.IO; update Haddock docs
1329Simon Marlow <marlowsd@gmail.com>**20090713113104
1330 Ignore-this: c3f017a555335aa55d106253393f72e2
1331]
1332[add a comment about the non-workingness of CHARBUF_UTF16
1333Simon Marlow <marlowsd@gmail.com>**20090707124406
1334 Ignore-this: 98d00411b68d688b3b4cffc9507b1f35
1335]
1336[Fix build on Windows
1337Ian Lynagh <igloo@earth.li>**20090711004351]
1338[Fix some "warn-unused-do-bind" warnings where we want to ignore the value
1339Ian Lynagh <igloo@earth.li>**20090710204513]
1340[Use throwErrnoIfMinus1_ when calling getrusage
1341Ian Lynagh <igloo@earth.li>**20090710204221]
1342[Remove an unused import
1343Ian Lynagh <igloo@earth.li>**20090710153345]
1344[reportStackOverflow now returns IO ()
1345Ian Lynagh <igloo@earth.li>**20090710153257
1346 It used to do "return undefined" to return IO a.
1347]
1348[GHC.Conc.reportError now returns IO ()
1349Ian Lynagh <igloo@earth.li>**20090710152646
1350 It used to return IO a, by "return undefined".
1351]
1352[Fix some "warn-unused-do-bind" warnings where we want to ignore the value
1353Ian Lynagh <igloo@earth.li>**20090710152526]
1354[Minor SampleVar refactoring
1355Ian Lynagh <igloo@earth.li>**20090710151438]
1356[Fix "warn-unused-do-bind" warnings in GHC/IO/Handle/Text.hs
1357Ian Lynagh <igloo@earth.li>**20090710122905]
1358[Fix some "warn-unused-do-bind" warnings where we just want to ignore the result
1359Ian Lynagh <igloo@earth.li>**20090710005638]
1360[Use the result of writeCharBuf in GHC/IO/Encoding/Latin1.hs too
1361Ian Lynagh <igloo@earth.li>**20090710004032]
1362[Minor code tidyups in GHC.Conc
1363Ian Lynagh <igloo@earth.li>**20090710003801]
1364[Fix "warn-unused-do-bind" warning in GHC.Conc
1365Ian Lynagh <igloo@earth.li>**20090710003530
1366 If we fail to communicate with the IO manager then we print a warning
1367 using debugErrLn from the ghc-prim package.
1368]
1369[Fix "warn-unused-do-bind" warnings in System.Posix.Internals
1370Ian Lynagh <igloo@earth.li>**20090709164546]
1371[Fix "warn-unused-do-bind" warnings where we really do want to ignore the result
1372Ian Lynagh <igloo@earth.li>**20090709163912]
1373[Add back imports needed on Windows
1374Ian Lynagh <igloo@earth.li>**20090707181924]
1375[Remove unused imports
1376Ian Lynagh <igloo@earth.li>**20090707115810]
1377[Remove unused imports from base
1378simonpj@microsoft.com**20090706111842
1379 Ignore-this: f9b5f353e3bb820f787c56d615b28765
1380 
1381 These unused imports are detected by the new unused-import code
1382 
1383]
1384[Use the result of writeCharBuf
1385Simon Marlow <marlowsd@gmail.com>**20090706133303
1386 Ignore-this: 52288dd559bf4c4f313df6197091d935
1387   
1388 This only makes a difference when CHARBUF_UTF16 is in use, which it
1389 normally isn't.  I suspect CHARBUF_UTF16 doesn't currently work for
1390 other reasons (CHARBUF_UTF16 was an experiment before I wrote the
1391 GHC.IO.Encoding.UTF* codecs), but this patch at least makes it
1392 slightly closer to working.
1393]
1394[Remove some cruft from Data.HashTable
1395Ian Lynagh <igloo@earth.li>**20090706181630]
1396[Add 'eof' to Text.ParserCombinators.ReadP
1397simonpj@microsoft.com**20090706111801
1398 Ignore-this: 2aea7b848e00c894761bc4011adaa95d
1399 
1400 Add a ReadP parser that succeeds at the end of input. Very useful!
1401 
1402]
1403[Don't export CLDouble for GHC; fixes trac #2793
1404Ian Lynagh <igloo@earth.li>**20090705155120
1405 We never really supported CLDouble (it was a plain old double underneath),
1406 and pretending that we do does more harm than good.
1407]
1408[a byte between 0x80 and 0xBF is illegal immediately (#3341)
1409Simon Marlow <marlowsd@gmail.com>**20090702081415
1410 Ignore-this: dc19ef59a1a21118d5a7dd38aa2f611c
1411]
1412[avoid a warning
1413Simon Marlow <marlowsd@gmail.com>**20090630084134
1414 Ignore-this: c92a45ee216faf01327feae9fe06d6e2
1415]
1416[Add a wrapper for libiconv.
1417Matthias Kilian <kili@outback.escape.de>**20090629183634
1418 Ignore-this: 23c6047c0d71b745b495cc223574a47f
1419]
1420[#include <sys/times.h> if we have it (should fix build problems)
1421Simon Marlow <marlowsd@gmail.com>**20090629085351
1422 Ignore-this: a35e93b37ca9595c73460243180f4b9d
1423]
1424[set binary mode for existing FDs on Windows (fixes some GHCi test failures)
1425Simon Marlow <marlowsd@gmail.com>**20090626120522
1426 Ignore-this: 580cf636e9c77d8427aff6861d089481
1427]
1428[Move directory-related stuff to the unix package
1429Simon Marlow <marlowsd@gmail.com>**20090625120325
1430 Ignore-this: b997b3cbce0a46ca87ad825bbdc0a411
1431 now that it isn't used on Windows any more.
1432]
1433[TAG 2009-06-25
1434Ian Lynagh <igloo@earth.li>**20090625160056]
1435Patch bundle hash:
1436f8c4a9201d60b7fdfe30a8ff3e88dae5362a7a3e