Ticket #4122: monoid-deps-revised.dpatch

File monoid-deps-revised.dpatch, 33.8 KB (added by tibbe, 7 years ago)

Adds -XNoImplicitPrelude

Line 
11 patch for repository /usr/local/google/src/base-event:
2
3Tue Jun  8 10:02:36 CEST 2010  Johan Tibell <johan.tibell@gmail.com>
4  * Remove Prelude dependency from Data.Monoid
5 
6  This removes an import cycle involving the new I/O manager code.
7
8New patches:
9
10[Remove Prelude dependency from Data.Monoid
11Johan Tibell <johan.tibell@gmail.com>**20100608080236
12 Ignore-this: 4486a8c7d7d641d2302a316cef2ec7c0
13 
14 This removes an import cycle involving the new I/O manager code.
15] {
16hunk ./Data/Monoid.hs 1
17+{-# OPTIONS_GHC -XNoImplicitPrelude #-}
18 -----------------------------------------------------------------------------
19 -- |
20 -- Module      :  Data.Monoid
21hunk ./Data/Monoid.hs 34
22         Last(..)
23   ) where
24 
25+-- Push down the module in the dependency hierarchy.
26+#if defined(__GLASGOW_HASKELL__)
27+import GHC.Base hiding (Any)
28+import GHC.Enum
29+import GHC.Num
30+import GHC.Read
31+import GHC.Show
32+import Data.Maybe
33+#else
34 import Prelude
35hunk ./Data/Monoid.hs 44
36+#endif
37 
38 {-
39 -- just for testing
40}
41
42Context:
43
44[Attempt to untangle part of the module mess
45Bryan O'Sullivan <bos@serpentine.com>**20100607012643
46 Ignore-this: 9b599d3270e901476e7a907e532e0121
47]
48[Switch to Maybe, remove some dead code
49Bryan O'Sullivan <bos@serpentine.com>**20100607003039
50 Ignore-this: da4278964fde5881516b257762867565
51]
52[Remove ambiguous copy of startIOManagerThread
53Bryan O'Sullivan <bos@serpentine.com>**20100607002345
54 Ignore-this: edc37101dbff89ceeae5cfe2532d602c
55]
56[Fix parts of the build
57Bryan O'Sullivan <bos@serpentine.com>**20100606210421
58 Ignore-this: 5145c1c6bcc23d781e15deb1fed46698
59 
60 We're still stuck:
61 
62 libraries/base/GHC/Conc.lhs:1:1:
63     Circular imports: module `GHC.Conc' depends on itself
64 
65]
66[Patch event code in
67Bryan O'Sullivan <bos@serpentine.com>**20100606210344
68 Ignore-this: bbded0f57b793a2c2e3559ac2c6ab025
69]
70[Move files around
71Bryan O'Sullivan <bos@serpentine.com>**20100606204929
72 Ignore-this: 94052d7f77a90ab0e945e90ed79fc2e
73]
74[bump version to 4.3.0.0, added instance MonadPlus STM
75Simon Marlow <marlowsd@gmail.com>**20100601144831
76 Ignore-this: 7c3cf7574499c4267372493f2636dc0
77]
78[Moved MonadPlus instance for STM from Control.Monad.STM to GHC.Conc to avoid an orphaned instance
79Bas van Dijk <v.dijk.bas@gmail.com>**20100516160651
80 Ignore-this: 651b852942b2fae2b93f996e39239b8f
81]
82[Added Applicative and Alternative instances for STM
83Bas van Dijk <v.dijk.bas@gmail.com>**20100516171756
84 Ignore-this: 567003bc4040bc97105cda4d31ebf04a
85]
86[expand Foldable instance for Array
87Ross Paterson <ross@soi.city.ac.uk>**20100602212154
88 Ignore-this: 9bd9e9666a9400431eb92352244fe7e7
89]
90[doc comment illustrating Foldable(foldr)
91Ross Paterson <ross@soi.city.ac.uk>**20100527150833
92 Ignore-this: 8f27d889379803f3ba86d6e928428f3c
93]
94[fix syntax in doc comments
95Ross Paterson <ross@soi.city.ac.uk>**20100527150757
96 Ignore-this: cb78da51d60ff6863dc395f1a892c103
97]
98[export hGetBufSome (#4046)
99Simon Marlow <marlowsd@gmail.com>**20100520093538
100 Ignore-this: f467fad9722e27edfad6b3dd75290e7b
101]
102[hWaitForInput: don't try to read from the device (#4078)
103Simon Marlow <marlowsd@gmail.com>**20100517133741
104 Ignore-this: 55ec33b03397380259b91e4ca62207a6
105 readTextDeviceNonBlocking is not non-blocking on Windows
106]
107[hSetEncoding: change the encoding on both read and write sides (#4066)
108Simon Marlow <marlowsd@gmail.com>**20100514124628
109 Ignore-this: 5b9e9caef06356d0296c584159709ebb
110]
111[Correct haddock formatting.
112Adam Vogt <vogt.adam@gmail.com>**20100423022103
113 Ignore-this: d2622339302048fda48080f7d5ce4a2f
114]
115[Fix for hGetBufSome
116Simon Marlow <marlowsd@gmail.com>**20100505135637
117 Ignore-this: 2019680f8fb223956cacfcf0d046f133
118]
119[improve the documentation for throwTo and killThread (#3884)
120Simon Marlow <marlowsd@gmail.com>**20100505135600
121 Ignore-this: ce881d96ddb729acb6ca09c779975e7d
122]
123[elaborate the docs for unsafePerformIO a bit
124Simon Marlow <marlowsd@gmail.com>**20100505101249
125 Ignore-this: 1cec3f67560b672c64c5a0dcf9a79eb7
126]
127[add Typeable instance
128Simon Marlow <marlowsd@gmail.com>**20100504152815
129 Ignore-this: 6d9cf9d62f0ef17fa459bf213a04098
130]
131[Add hGetBufSome, like hGetBuf but can return short reads
132Simon Marlow <marlowsd@gmail.com>**20100504152759
133 Ignore-this: 195c905b43f8d9505029364e2c5b18e
134]
135[Add swap (#3298)
136Simon Marlow <marlowsd@gmail.com>**20100504095339
137 Ignore-this: 13b007dc4594ce252997ec6fa0bbd976
138]
139[inline allocaArray0, to fix withCString benchmark
140Simon Marlow <marlowsd@gmail.com>**20100423124729
141 Ignore-this: 35c96816acc2f3aaf9dd29f7995fa6f0
142]
143[raise asynchronous exceptions asynchronously (#3997)
144Simon Marlow <marlowsd@gmail.com>**20100421094932
145 Ignore-this: 6d987d93d382c0f69c68c326312abd6b
146]
147[add NOINLINE pragmas for stdin/stdout/stderr
148Simon Marlow <marlowsd@gmail.com>**20100421082041
149 Ignore-this: 3fc130268ec786f28d945858d6690986
150]
151[INLINE alloca and malloc
152Simon Marlow <marlowsd@gmail.com>**20100419135333
153 Ignore-this: b218bd611f18721b1505a8c0b9e6a16a
154 See discussion on glasgow-haskell-users:
155   http://www.haskell.org/pipermail/glasgow-haskell-users/2010-April/018740.html
156]
157[Move comment closer to the offending line
158Matthias Kilian <kili@outback.escape.de>**20100419155421
159 Ignore-this: b34a1d7affd66f67d210df2377b585d9
160]
161[Ignore the return code of c_fcntl_write again
162Matthias Kilian <kili@outback.escape.de>**20100415140452
163 Ignore-this: 266d8ba02cc3cb79c85629b3528261c9
164 
165 The return code has been ignored in the past on purpose, because
166 O_NONBLOCK will fail on BSDs for some special files. This fixes the
167 problem mentioned in
168 http://www.haskell.org/pipermail/glasgow-haskell-users/2010-April/018698.html
169 
170]
171[Fix bitrot in IO debugging code
172Ian Lynagh <igloo@earth.li>**20100413134339
173 Also switched to using Haskell Bools (rather than CPP) to en/disable it,
174 so it shouldn't break again in the future.
175]
176[Tiny code tidy-up
177Ian Lynagh <igloo@earth.li>**20100413011147]
178[remove old/wrong comment
179Simon Marlow <marlowsd@gmail.com>**20100325161403
180 Ignore-this: e6e377d44af48c4162d17d55bdf3f821
181]
182[withThread: block asynchronous exceptions before installing exception handler.
183Bas van Dijk <v.dijk.bas@gmail.com>**20100329131624
184 Ignore-this: be5aeb47dbd73807b5f94df11afbb81c
185 Note that I don't unblock the given io computation. Because AFAICS
186 withThread is only called with 'waitFd' which only performs an FFI
187 call which can't receive asynchronous exceptions anyway.
188]
189[runInUnboundThread: block asynchronous exceptions before installing exception handler
190Bas van Dijk <v.dijk.bas@gmail.com>**20100329131549
191 Ignore-this: a00c5e32fe3981ff87bedd367a69051e
192]
193[fix the deprecation message (GHC.IO.Handle.Base -> GHC.IO.Handle)
194Simon Marlow <marlowsd@gmail.com>**20100330121137
195 Ignore-this: 4ca8500a01ac93454507aa8f9dd001f9
196]
197[Make SampleVar an abstract newtype
198Bas van Dijk <v.dijk.bas@gmail.com>**20100318200349
199 Ignore-this: 27939e2a064b75e71cb146117346be30
200]
201[Fix bugs regarding asynchronous exceptions and laziness in Control.Concurrent.SampleVar
202Bas van Dijk <v.dijk.bas@gmail.com>**20100318200104
203 Ignore-this: 7376b2a3afe155daf233a8f1ddc0a7a
204  - Block asynchronous exceptions at the right places
205  - Force thunks before putting them in a MVar
206]
207[Write the thunk 'next' to the MVar
208Bas van Dijk <v.dijk.bas@gmail.com>**20100319125951
209 Ignore-this: dd25636cf220131385ff2fd32493d456
210]
211[change to use STM, fixing 4 things
212Simon Marlow <marlowsd@gmail.com>**20100318104436
213 Ignore-this: 551d30280a7941c08f5c3b14576bdd70
214   1. there was no async exception protection
215   2. there was a space leak (now new value is strict)
216   3. using atomicModifyIORef would be slightly quicker, but can
217      suffer from adverse scheduling issues (see #3838)
218   4. also, the STM version is faster.
219]
220[Tweak docs
221Ian Lynagh <igloo@earth.li>**20100312214129]
222[Fixed dead links in documentation of forkIO
223Bas van Dijk <v.dijk.bas@gmail.com>**20100308222415
224 Ignore-this: 7deb8fd064c867fbede2a6b2e9da4f15
225]
226[Documentation fixes in Control.Exception
227Bas van Dijk <v.dijk.bas@gmail.com>**20100301220442
228 Ignore-this: 761fcba401cbd1f47276ddfc9b5b80f2
229]
230[Plug two race conditions that could lead to deadlocks in the IO manager
231Simon Marlow <marlowsd@gmail.com>**20100225120255
232 Ignore-this: e6983d6b953104d370278ab3e4617e8b
233]
234[FIX #3866: improve documentation of Data.Data.Constr
235jpm@cs.uu.nl**20100224125506
236 Ignore-this: 3818c5d8fee012a3cf322fb455b6e5dc
237]
238[UNDO: Handle NaN, -Infinity and Infinity in the toRational for Float/Double (#3676)
239Simon Marlow <marlowsd@gmail.com>**20100223101603
240 Ignore-this: 78becb2d39b3cd9a1a473a5811ca7d92
241]
242[Put the complexity in the length docs. Fixes trac #3680
243Ian Lynagh <igloo@earth.li>**20100221191425]
244[nhc98 should build Data.Functor.
245Malcolm.Wallace@cs.york.ac.uk**20100221163218]
246[Update the exitWith docs
247Ian Lynagh <igloo@earth.li>**20100213140004
248 Error pointed out by Volker Wysk <vw@volker-wysk.de>
249]
250[Handle NaN, -Infinity and Infinity in the toRational for Float/Double (#3676)
251Simon Marlow <marlowsd@gmail.com>**20100211101955
252 Ignore-this: 261415363303efca265e80290eac5f28
253]
254[For nhc98, import unsafeInterleaveIO rather than defining it here.
255Malcolm.Wallace@cs.york.ac.uk**20100204171021]
256[Stifle warning about unused return value
257benl@cse.unsw.edu.au**20100203025537]
258[fix #3832: use the locale encoding in openTempFile
259Simon Marlow <marlowsd@gmail.com>**20100120211830
260 Ignore-this: df4f778cc5fefb32290c798db722632c
261 Also while I was here fix an XXX: the Handle contained an
262 uninformative string like <fd: 4> for error messages rather than the
263 real file path.
264]
265[Fix the build: export void, so it doesn't give an unused binding warning
266Ian Lynagh <igloo@earth.li>**20100116174451]
267[hIsEOF: don't do any decoding (#3808)
268Simon Marlow <marlowsd@gmail.com>**20100112230317
269 Ignore-this: 6a384dd2d547ffe3ad3762920e5c1671
270]
271[Control.Monad: +void :: f a -> f ()
272gwern0@gmail.com**20100108214455
273 Ignore-this: 4dc07452315f2d1b4941903ff42fc45f
274 See http://hackage.haskell.org/trac/ghc/ticket/3292
275 Turns m a -> m (). Lets one call functions for their side-effects without
276 having to get rid of their return values with '>> return ()'. Very useful
277 in many contexts (parsing, IO etc.); particularly good for 'forkIO' and 'forM_',
278 as they demand return types of 'IO ()' though most interesting IO functions
279 return non-().
280]
281[Replace the implementation of mergesort with a 2x faster one.
282Malcolm.Wallace@cs.york.ac.uk**20091224152014
283 See ticket http://hackage.haskell.org/trac/ghc/ticket/2143.
284]
285[Restore previous Data.Typeable.typeOf*Default implementations for non-ghc.
286Malcolm.Wallace@cs.york.ac.uk**20091223142625
287 Not all compilers have ScopedTypeVariables.
288]
289[Add comments about double bounds-checking, and fast paths for rectangular arrays
290simonpj@microsoft.com**20091218165655
291 Ignore-this: ea0849419dc00927aba4bd410b1cc58d
292 
293 See Note [Double bounds-checking of index values] for the details.
294 
295 The fast paths omit the doubled checks for cases we know about
296]
297[Fix Trac #3245: memoising typeOf
298simonpj@microsoft.com**20091218155117
299 Ignore-this: 5a178a7f2222293c5ee0c3c43bd1b625
300 
301 The performance bug in #3245 was caused by computing the typeRep
302 once for each call of typeOf, rather than once for each dictionary
303 contruction.  (Computing TypeReps is reasonably expensive, because
304 of the hash-consing machinery.)
305 
306 This is readily fixed by putting the TypeRep construction outside
307 the lambda.  (Arguably GHC might have worked that out itself,
308 but it involves floating something between a type lambda and a
309 value lambda, which GHC doesn't currently do. If it happens a lot
310 we could fix that.)
311]
312[Mark 'index' as INLINE in GHC.Arr
313simonpj@microsoft.com**20091216170441
314 Ignore-this: a4df9d8acf496c8e0e9ce5a520509a2a
315 
316 This makes indexing much faster. See Trac #1216
317]
318[Comment the remaining orphan instance modules
319Ian Lynagh <igloo@earth.li>**20091206125021]
320[De-orphan Eq/Ord Float/Double
321Ian Lynagh <igloo@earth.li>**20091205181238]
322[Add comments to "OPTIONS_GHC -fno-warn-orphans" pragmas
323Ian Lynagh <igloo@earth.li>**20091205165854]
324[Data.Either.partitionEithers was insufficiently lazy.
325Malcolm.Wallace@cs.york.ac.uk**20091202032807
326 Ignore-this: 77e1b3288f66608c71458d8a91bcbe12
327 Spotted by Daniel Fischer.
328]
329[fix the docs regarding finalizer guarantees
330Simon Marlow <marlowsd@gmail.com>**20091130144409
331 Ignore-this: d1ab9532c74a002b8075ff60febcbe2d
332]
333[x86_64 requires more stack
334Malcolm.Wallace@cs.york.ac.uk**20091201033745]
335[check for size < 0 in mallocForeignPtrBytes and friends (#3514)
336Simon Marlow <marlowsd@gmail.com>**20091125143822
337 Ignore-this: 91077d01da2bbe1dfed5155e8b40da9
338]
339[hGetContents: close the handle properly on error
340Simon Marlow <marlowsd@gmail.com>**20091125123435
341 Ignore-this: bc37ff678acc6e547dc390285e056eb9
342 
343 When hGetContents caught an error it was closing the handle and then
344 throwing the exception, without updating the handle with the new
345 closed state.  This lead to a double-closed, which was the cause of
346 
347 *** glibc detected *** ./Setup: double free or corruption
348 
349 when iconv_close was called twice on the decoder.
350 
351 See http://hackage.haskell.org/trac/hackage/ticket/609
352]
353[Fix arities of mapFB and zipFB
354Roman Leshchinskiy <rl@cse.unsw.edu.au>**20091126232219
355 Ignore-this: c4e14cd0a92622549c86e67237a40865
356]
357[Remove an unnecessary -fno-warn-orphans flag
358Ian Lynagh <igloo@earth.li>**20091126123404]
359[Tweak layout to work with alternative layout rule
360Ian Lynagh <igloo@earth.li>**20091125232349]
361[Tweak layout to be accepted by the alternative layout rul
362Ian Lynagh <igloo@earth.li>**20091125194147]
363[Make sure zipWithFB has arity 2
364Roman Leshchinskiy <rl@cse.unsw.edu.au>**20091125010003
365 Ignore-this: 4cf60c55666f03d22a9f5a6e07f52d36
366 
367 It gets 2 arguments in the "zipWith" rule but its arity was higher and the new
368 inliner didn't inline it sometimes, for instance here:
369 
370 mpp ::  [Double] -> [Double] -> [Double] -> [Double] -> [Double]
371 mpp as bs cs ds = zipWith (*) (zipWith (+) as bs) (zipWith (+) cs ds)
372 
373 This was a regression vs. 6.10.
374]
375[Remove an old comment
376Ian Lynagh <igloo@earth.li>**20091124134647]
377[De-orphan the Eq/Ord Integer instances
378Ian Lynagh <igloo@earth.li>**20091124133639]
379[Whitespace only
380Ian Lynagh <igloo@earth.li>**20091124133421]
381[Derive some more instances, rather than writing them by hand
382Ian Lynagh <igloo@earth.li>**20091124011747]
383[We can now derive Ord ()
384Ian Lynagh <igloo@earth.li>**20091124011416]
385[De-orphan tuple Eq/Ord instances
386Ian Lynagh <igloo@earth.li>**20091123233343]
387[Control.Exception.Base no longer has any orphans
388Ian Lynagh <igloo@earth.li>**20091123224905]
389[De-orphan the MonadFix ST instance for GHC
390Ian Lynagh <igloo@earth.li>**20091123223544]
391[Rearrange the contents of Control.Monad.ST; no functionality changes
392Ian Lynagh <igloo@earth.li>**20091123222702]
393[De-orphan the Eq/Ord [a] instances
394Ian Lynagh <igloo@earth.li>**20091123215635]
395[De-orphan the Eq/Ord Char instances
396Ian Lynagh <igloo@earth.li>**20091123202253]
397[De-orphan the Eq/Ord Bool instances
398Ian Lynagh <igloo@earth.li>**20091123201817]
399[Move Eq/Ord Ordering instances to de-orphan them
400Ian Lynagh <igloo@earth.li>**20091123194310]
401[Remove ffi warnings for nhc98.
402Malcolm.Wallace@cs.york.ac.uk**20091123063743]
403[Second attempt to fix #1185 (forkProcess and -threaded)
404Simon Marlow <marlowsd@gmail.com>**20091111151915
405 Ignore-this: fa5f5d5e4e080d4b612a37244f937f9c
406 
407 Patch 2/2: first patch is to ghc
408 
409 This time without dynamic linker hacks, instead I've expanded the
410 existing rts/Globals.c to cache more CAFs, specifically those in
411 GHC.Conc.  We were already using this trick for signal handlers, I
412 should have realised before.
413 
414 It's still quite unsavoury, but we can do away with rts/Globals.c in
415 the future when we switch to a dynamically-linked GHCi.
416]
417[Rollback #1185 fix
418Simon Marlow <marlowsd@gmail.com>**20091106140629
419 Ignore-this: cd5667e8474e37e01ba26a1984274811
420 
421 rolling back:
422 
423 Tue Nov  3 16:05:40 GMT 2009  Simon Marlow <marlowsd@gmail.com>
424   * Fix #1185: restart the IO manager after fork()
425   
426   This is the libraries/base part of the patch; there is a corresponding
427   patch to GHC itself.
428   
429   The main change is that we now keep track of the IO manager's ThreadId
430   in a top-level MVar, and ensureIOManagerIsRunning checks whether a
431   previous IO manager thread is alive before starting one.  In the child
432   of fork(), we can hence call ensureIOManagerIsRunning to restart the
433   IO manager.
434 
435     M ./GHC/Conc.lhs -46 +44
436 
437 Wed Nov  4 17:49:45 GMT 2009  Ian Lynagh <igloo@earth.li>
438   * Fix the build on Windows
439 
440     M ./GHC/Conc.lhs -6 +4
441]
442[Fix the build on Windows
443Ian Lynagh <igloo@earth.li>**20091104174945]
444[Fix #1185: restart the IO manager after fork()
445Simon Marlow <marlowsd@gmail.com>**20091103160540
446 Ignore-this: 6dc05464f1500104554637f4759738cc
447 
448 This is the libraries/base part of the patch; there is a corresponding
449 patch to GHC itself.
450 
451 The main change is that we now keep track of the IO manager's ThreadId
452 in a top-level MVar, and ensureIOManagerIsRunning checks whether a
453 previous IO manager thread is alive before starting one.  In the child
454 of fork(), we can hence call ensureIOManagerIsRunning to restart the
455 IO manager.
456]
457[improve the documentation for throwErrnoIfRetry
458Simon Marlow <marlowsd@gmail.com>**20091016112404
459 Ignore-this: b77275cacf730e15757946027168f63e
460]
461[Don't inline unpackFoldrCString ever
462simonpj@microsoft.com**20091029135350
463 Ignore-this: 85d672649b1b776efc7e97500b05d4f9
464]
465[Inline more default methods
466simonpj@microsoft.com**20091029135330
467 Ignore-this: 289c44b0afd6d5631c2a4e0664275ca9
468 
469 Namely Monad: (>>)
470        Eq:    (==), (/=)
471        Num:   (-), negate
472        Real:  quot, rem, div, mod, recip, (/), truncate
473        Float: (**), logBase, sqrt, tan, tanh
474]
475[Move error messages out of INLINEd default methods
476simonpj@microsoft.com**20091029135118
477 Ignore-this: 9e35dc947f94827a3529eb53a41575fd
478 
479 No need to duplicate the error generation!
480]
481[Exploit now-working default-method INLINE pragmas for Data.Bits
482simonpj@microsoft.com**20091029135041
483 Ignore-this: 8adf225f31ca7a3181ee087e9e4fe535
484 
485 * Add INLINE pragmas to default methods for class Bits
486 
487 * Remove redundant instance methods elsewhere, now that
488   the default method will do the job
489]
490[Tidy up and comment imports
491simonpj@microsoft.com**20091029134414
492 Ignore-this: bf2be31035de975d8995e988933cc940
493]
494[Inline foldr and (.) when applied to two arguments not three
495simonpj@microsoft.com**20091029134335
496 Ignore-this: fccb6f3e90e15f44cb465814be85ede2
497 
498 The new INLINE story is (by design) arity-sensitive, so we must
499 put fewer argument on the LHS for foldr and (.)
500]
501[dirUtils.c no longer available
502Malcolm.Wallace@cs.york.ac.uk**20091013093833]
503[Make hGetContents throw an exception if an error is encountered
504Simon Marlow <marlowsd@gmail.com>**20091012152955
505 Ignore-this: 9f7a7176193eab25c9daaacd9261f2de
506 
507 Strictly speaking this breaks Haskell 98 compatibility, which requires
508 hGetContents to just end the lazy stream silently if an error is
509 encountered.  However, for a few reasons we think it will make
510 everyone's life a bit easier if we make this change
511 
512  1. Errors will be a lot more common in GHC 6.12.1, in the form
513     of Unicode decoding errors.
514 
515  2. When Haskell 98 was designed, we didn't know how to throw
516     exceptions from inside lazy I/O, but now we do.
517 
518  3. If anyone is actually relying on the previous behaviour, their
519     code is arguably broken.
520]
521[Re-instate System.Console.Getopt for nhc98 builds.
522Malcolm.Wallace@cs.york.ac.uk**20091013092843
523 Although it was split out of base a while back, that change was
524 reverted for ghc soon afterwards, but nhc98 never noticed.
525]
526[Roll back "Another instance of nhc98's strange import semantics."
527Ian Lynagh <igloo@earth.li>**20091009185618
528 Fri Oct  9 14:50:51 BST 2009  Malcolm.Wallace@cs.york.ac.uk
529 GHC (correctly) warns about the unused import, which breaks the validate
530 build.
531]
532[Roll back "Cope with nhc98's (occasionally-strange) import semantics"
533Ian Lynagh <igloo@earth.li>**20091009184704
534 Fri Oct  9 14:43:51 BST 2009  Malcolm.Wallace@cs.york.ac.uk
535 GHC (correctly) warns about the unused import, which breaks the validate
536 build.
537]
538[It seems that nhc98 needs defaulting in Data.Fixed.
539Malcolm.Wallace@cs.york.ac.uk**20091009135242]
540[Another instance of nhc98's strange import semantics.
541Malcolm.Wallace@cs.york.ac.uk**20091009135051]
542[Make Data.Functor compatible with non-GHC compilers.
543Malcolm.Wallace@cs.york.ac.uk**20091009134821]
544[Cope with nhc98's (occasionally-strange) import semantics.
545Malcolm.Wallace@cs.york.ac.uk**20091009134351]
546[Fix gratuitous breakage of nhc98 in System.IO.
547Malcolm.Wallace@cs.york.ac.uk**20091009134001]
548[Fix gratuitous breakage of nhc98 in Control.Exception.Base.
549Malcolm.Wallace@cs.york.ac.uk**20091009133615]
550[Fix gratuitous breakage of non-GHC in Data.Fixed.
551Malcolm.Wallace@cs.york.ac.uk**20091009133330]
552[Fix gratuitous breakage for non-GHC in Data.Bits.
553Malcolm.Wallace@cs.york.ac.uk**20091009133257]
554[Use UTF-32LE instead of UTF32LE
555Simon Marlow <marlowsd@gmail.com>**20091006100207
556 Ignore-this: 7f881e36543d250ef848c9f60d67655a
557 The latter is not recognised by some iconv implementations.
558]
559[Strip any Byte Order Mark (BOM) from the front of decoded streams.
560Ben.Lippmeier@anu.edu.au*-20090930084229
561 Ignore-this: d0d0c3ae87b31d71ef1627c8e1786445
562 When decoding to UTF-32, Solaris iconv inserts a BOM at the front
563 of the stream, but Linux iconv doesn't.
564]
565[use UTF32BE/UTF32LE instead of UCS-4/UCS-4LE
566Simon Marlow <marlowsd@gmail.com>**20091005101554
567 Ignore-this: 2aef5e9bec421e714953b7aa1bdfc1b3
568]
569[Strip any Byte Order Mark (BOM) from the front of decoded streams.
570Ben.Lippmeier@anu.edu.au**20090930084229
571 Ignore-this: d0d0c3ae87b31d71ef1627c8e1786445
572 When decoding to UTF-32, Solaris iconv inserts a BOM at the front
573 of the stream, but Linux iconv doesn't.
574]
575[Add traceEvent :: String -> IO ()
576Simon Marlow <marlowsd@gmail.com>**20090925141257
577 Ignore-this: 8b1888bbf9682ffba13f815b6000e4b1
578 For emitting an event via the RTS tracing framework
579]
580[Fix the error message when flushing the read buffer of a non-seekable Handle
581Simon Marlow <marlowsd@gmail.com>**20090923090536
582 Ignore-this: 4342026df93759d99480f4e13f80a492
583]
584[Fix #3534: No need to flush the byte buffer when setting binary mode
585Simon Marlow <marlowsd@gmail.com>**20090923090445
586 Ignore-this: 625817ed7ae2c12291eb993a99dc640a
587]
588[Use let !y = x in .. x .. instead of seq in $! and evaluate (#2273)
589Simon Marlow <marlowsd@gmail.com>**20090916140454]
590[make some Applicative functions into methods, and split off Data.Functor (proposal #3335)
591Ross Paterson <ross@soi.city.ac.uk>**20090915173109
592 Ignore-this: a0cff4de6dfdbcbd56a66101bc4855a9
593 
594 The following functions
595 
596     (<$) :: Functor f => a -> f b -> f a
597     (*>) :: Applicative f => f a -> f b -> f b
598     (<*) :: Applicative f => f a -> f b -> f a
599     some :: Alternative f => f a -> f [a]
600     many :: Alternative f => f a -> f [a]
601 
602 are moved into the corresponding classes, with the existing implementations
603 as default definitions.  This gives people creating instances the option of
604 defining specialized implementations of these functions, though they should
605 be equivalent to the default definitions.
606 
607 Although (<$) is now a method of the Functor class, it is hidden in the
608 re-export by the Prelude, Control.Monad and Monad.  The new module
609 Data.Functor exposes the full class, plus the function (<$>).  These are
610 also re-exported by Control.Applicative.
611]
612[On Windows, use the console code page for text file encoding/decoding.
613Judah Jacobson <judah.jacobson@gmail.com>**20090913022126
614 Ignore-this: 86c2f2db8ef92b751599795d3195187b
615 
616 We keep all of the code page tables in the module
617 GHC.IO.Encoding.CodePage.Table.  That file was generated automatically
618 by running codepages/MakeTable.hs; more details are in the comments at the
619 start of that script.
620 
621 Storing the lookup tables adds about 40KB to each statically linked executable;
622 this only increases the size of a "hello world" program by about 7%.
623 
624 Currently we do not support double-byte encodings (Chinese/Japanese/Korean), since
625 including those codepages would increase the table size to 400KB.  It will be
626 straightforward to implement them once the work on library DLLs is finished.
627]
628[Fix "init" docs: the input list need not be finite. Fixes trac #3465
629Ian Lynagh <igloo@earth.li>**20090911210437]
630[Bump base version to 4.2.0.0
631Ian Lynagh <igloo@earth.li>**20090911153913]
632[Address #3310
633Simon Marlow <marlowsd@gmail.com>**20090830152850
634 Ignore-this: 40c7f7c171ee299a83092fd360a952b7
635 
636  - Rename BlockedOnDeadMVar   -> BlockedIndefinitelyOnMVar
637  - Rename BlockedIndefinitely -> BlockedIndefinitelyOnSTM
638  - instance Show BlockedIndefinitelyOnMVar is now
639      "blocked indefinitely in an MVar operation"
640  - instance Show BlockedIndefinitelyOnSTM is now
641      "blocked indefinitely in an STM transaction"
642 
643 clients using Control.OldException will be unaffected (the new
644 exceptions are mapped to the old names).  However, for base4-compat
645 we'll need to make a version of catch/try that does a similar
646 mapping.
647]
648[Fix unicode conversion for MSB architectures
649Ben.Lippmeier@anu.edu.au**20090830130028
650 This fixes the SPARC/Solaris build.
651]
652[Fix #3441: detect errors in partial sequences
653Simon Marlow <marlowsd@gmail.com>**20090830075909
654 Ignore-this: d12a75d95e0cae5eb1555266810ec281
655]
656[Fix hWaitForInput
657Simon Marlow <marlowsd@gmail.com>**20090827152116
658 Ignore-this: 2550e911f1a4d4357a5aa8d1764238ce
659 It was erroneously waiting when there were bytes to decode waiting in
660 the byte buffer.
661]
662[fix debugging code
663Simon Marlow <marlowsd@gmail.com>**20090827150628
664 Ignore-this: e1c82fdc19a22e247cd69ff6fa11921d
665]
666[Allow for configurable iconv include and library locations.
667Matthias Kilian <kili@outback.escape.de>**20090826154406
668 Ignore-this: be95fab611a5534cf184b508964ed498
669 This should help to fix the build on OpenBSD.
670]
671[typo in comment
672Simon Marlow <marlowsd@gmail.com>**20090826085252
673 Ignore-this: 1903ee0f354157a6ed3871c100f6b1b9
674]
675[un-hide some modules from the Haddock docs
676Simon Marlow <marlowsd@gmail.com>**20090825152457
677 Ignore-this: dce6606f93cf977fb24ebe99082dfa62
678]
679[Apply fix for #1548, from squadette@gmail.com
680Simon Marlow <marlowsd@gmail.com>**20090819120700
681 Ignore-this: 31c237c46a6445f588ed4b8c51bb6231
682]
683[improvements to Data.Fixed: instances for Typeable and Data, more predefined types
684Ashley Yakeley <ashley@semantic.org>**20090812055058
685 Ignore-this: feeece36d5632f02a05d137d2a39ab78
686]
687[Fix "Cabal check" warnings
688Ian Lynagh <igloo@earth.li>**20090811215856]
689[Add a GHC.Constants module; fixes trac #3094
690Ian Lynagh <igloo@earth.li>**20090809183252]
691[Apply proposal #3393
692Ian Lynagh <igloo@earth.li>**20090809134717
693 Add openTempFileWithDefaultPermissions and
694 openBinaryTempFileWithDefaultPermissions.
695]
696[Add some more C wrappers; patch from Krister Walfridsson
697Ian Lynagh <igloo@earth.li>**20090807200631
698 Fixes 21 testsuite errors on NetBSD 5.99.
699]
700[Fixing configure for autoconf 2.64
701Alexander Dunlap <alexander.dunlap@gmail.com>**20090805060748
702 Ignore-this: 992ab91ae3d68c12dbb265776e33e243
703]
704[add INLINE toList
705Ross Paterson <ross@soi.city.ac.uk>**20090806142853
706 Ignore-this: aba16aabb17d5dca44f15d188945680e
707 
708 In anticipation of the fixing of #2353.
709]
710[fix a copyright
711Simon Marlow <marlowsd@gmail.com>**20090805134045
712 Ignore-this: b0ffbdd38fbba121e8bcba37c4082a60
713]
714[Tweak the BufferedIO class to enable a memory-mapped file implementation
715Simon Marlow <marlowsd@gmail.com>**20090805134036
716 Ignore-this: ec67d7a0a6d977438deaa342503f77e0
717 We have to eliminate the assumption that an empty write buffer can be
718 constructed by setting the buffer pointers to zero: this isn't
719 necessarily the case when the buffer corresponds to a memory-mapped
720 file, or other in-memory device implementation.
721]
722[Deprecate Control.OldException
723Ian Lynagh <igloo@earth.li>**20090804143910]
724[Windows build fix, following RTS tidyup
725Simon Marlow <marlowsd@gmail.com>**20090803131121
726 Ignore-this: ce862fb91c2b234211a8757f98690778
727]
728[Updates to follow the RTS tidyup
729Simon Marlow <marlowsd@gmail.com>**20090801220743
730 Ignore-this: 6e92412df93a66c12d75344053d5634
731 C functions like isDoubleNaN moved here (primFloat.c)
732]
733[Add integer-simple as a build option
734Ian Lynagh <igloo@earth.li>**20090722013151]
735[Use shift[LR]Integer in the Bits Integer instance
736Ian Lynagh <igloo@earth.li>**20090721222440]
737[depend directly on integer-gmp, rather than indirecting through integer
738Ian Lynagh <igloo@earth.li>**20090721185228]
739[Move the instances of Functor and Monad IO to GHC.Base, to avoid orphans
740Simon Marlow <marlowsd@gmail.com>**20090722102130
741 Ignore-this: a7d85ac0025d559674249de0108dbcf4
742]
743[move "instance Exception Dynamic" so it isn't an orphan
744Simon Marlow <marlowsd@gmail.com>**20090721093854
745 Ignore-this: 5ede91ecfec2112c91b699d4de87cd02
746]
747[Improve the index checking for array accesses; fixes #2120 #2669
748Ian Lynagh <igloo@earth.li>**20090719153228
749 As well as checking that offset we are reading is actually inside the
750 array, we now also check that it is "in range" as defined by the Ix
751 instance. This fixes confusing behaviour (#2120) and improves some error
752 messages (#2669).
753]
754[Make chr say what its argument was, if it's a bad argument
755Ian Lynagh <igloo@earth.li>**20090718151049]
756[remove unused warning
757Simon Marlow <marlowsd@gmail.com>**20090715124416
758 Ignore-this: 31f613654089d0f4a44363946087b41e
759]
760[warning fix: -fno-implicit-prelude -> -XNoImplicitPrelude
761Simon Marlow <marlowsd@gmail.com>**20090715122839
762 Ignore-this: dc8957249731d5bcb71c01899e5adf2b
763]
764[Add hGetEncoding :: Handle -> IO (Maybe TextEncoding)
765Simon Marlow <marlowsd@gmail.com>**20090715122519
766 Ignore-this: 14c3eff996db062da1199739781e4708
767 as suggested during the discussion on the libraries list
768]
769[Add more documentation to mkTextEncoding
770Simon Marlow <marlowsd@gmail.com>**20090715122414
771 Ignore-this: 97253b2624267df3a246a18121e8ea81
772 noting that "//IGNORE" and "//TRANSLIT" suffixes can be used with GNU
773 iconv.
774]
775[Add the utf8_bom codec
776Simon Marlow <marlowsd@gmail.com>**20090715122257
777 Ignore-this: 1c9396cd805201fe873a39382ced79c7
778 as suggested during the discussion on the libraries list.
779]
780[Export Unicode and newline functionality from System.IO; update Haddock docs
781Simon Marlow <marlowsd@gmail.com>**20090713113104
782 Ignore-this: c3f017a555335aa55d106253393f72e2
783]
784[add a comment about the non-workingness of CHARBUF_UTF16
785Simon Marlow <marlowsd@gmail.com>**20090707124406
786 Ignore-this: 98d00411b68d688b3b4cffc9507b1f35
787]
788[Fix build on Windows
789Ian Lynagh <igloo@earth.li>**20090711004351]
790[Fix some "warn-unused-do-bind" warnings where we want to ignore the value
791Ian Lynagh <igloo@earth.li>**20090710204513]
792[Use throwErrnoIfMinus1_ when calling getrusage
793Ian Lynagh <igloo@earth.li>**20090710204221]
794[Remove an unused import
795Ian Lynagh <igloo@earth.li>**20090710153345]
796[reportStackOverflow now returns IO ()
797Ian Lynagh <igloo@earth.li>**20090710153257
798 It used to do "return undefined" to return IO a.
799]
800[GHC.Conc.reportError now returns IO ()
801Ian Lynagh <igloo@earth.li>**20090710152646
802 It used to return IO a, by "return undefined".
803]
804[Fix some "warn-unused-do-bind" warnings where we want to ignore the value
805Ian Lynagh <igloo@earth.li>**20090710152526]
806[Minor SampleVar refactoring
807Ian Lynagh <igloo@earth.li>**20090710151438]
808[Fix "warn-unused-do-bind" warnings in GHC/IO/Handle/Text.hs
809Ian Lynagh <igloo@earth.li>**20090710122905]
810[Fix some "warn-unused-do-bind" warnings where we just want to ignore the result
811Ian Lynagh <igloo@earth.li>**20090710005638]
812[Use the result of writeCharBuf in GHC/IO/Encoding/Latin1.hs too
813Ian Lynagh <igloo@earth.li>**20090710004032]
814[Minor code tidyups in GHC.Conc
815Ian Lynagh <igloo@earth.li>**20090710003801]
816[Fix "warn-unused-do-bind" warning in GHC.Conc
817Ian Lynagh <igloo@earth.li>**20090710003530
818 If we fail to communicate with the IO manager then we print a warning
819 using debugErrLn from the ghc-prim package.
820]
821[Fix "warn-unused-do-bind" warnings in System.Posix.Internals
822Ian Lynagh <igloo@earth.li>**20090709164546]
823[Fix "warn-unused-do-bind" warnings where we really do want to ignore the result
824Ian Lynagh <igloo@earth.li>**20090709163912]
825[Add back imports needed on Windows
826Ian Lynagh <igloo@earth.li>**20090707181924]
827[Remove unused imports
828Ian Lynagh <igloo@earth.li>**20090707115810]
829[Remove unused imports from base
830simonpj@microsoft.com**20090706111842
831 Ignore-this: f9b5f353e3bb820f787c56d615b28765
832 
833 These unused imports are detected by the new unused-import code
834 
835]
836[Use the result of writeCharBuf
837Simon Marlow <marlowsd@gmail.com>**20090706133303
838 Ignore-this: 52288dd559bf4c4f313df6197091d935
839   
840 This only makes a difference when CHARBUF_UTF16 is in use, which it
841 normally isn't.  I suspect CHARBUF_UTF16 doesn't currently work for
842 other reasons (CHARBUF_UTF16 was an experiment before I wrote the
843 GHC.IO.Encoding.UTF* codecs), but this patch at least makes it
844 slightly closer to working.
845]
846[Remove some cruft from Data.HashTable
847Ian Lynagh <igloo@earth.li>**20090706181630]
848[Add 'eof' to Text.ParserCombinators.ReadP
849simonpj@microsoft.com**20090706111801
850 Ignore-this: 2aea7b848e00c894761bc4011adaa95d
851 
852 Add a ReadP parser that succeeds at the end of input. Very useful!
853 
854]
855[Don't export CLDouble for GHC; fixes trac #2793
856Ian Lynagh <igloo@earth.li>**20090705155120
857 We never really supported CLDouble (it was a plain old double underneath),
858 and pretending that we do does more harm than good.
859]
860[a byte between 0x80 and 0xBF is illegal immediately (#3341)
861Simon Marlow <marlowsd@gmail.com>**20090702081415
862 Ignore-this: dc19ef59a1a21118d5a7dd38aa2f611c
863]
864[avoid a warning
865Simon Marlow <marlowsd@gmail.com>**20090630084134
866 Ignore-this: c92a45ee216faf01327feae9fe06d6e2
867]
868[Add a wrapper for libiconv.
869Matthias Kilian <kili@outback.escape.de>**20090629183634
870 Ignore-this: 23c6047c0d71b745b495cc223574a47f
871]
872[#include <sys/times.h> if we have it (should fix build problems)
873Simon Marlow <marlowsd@gmail.com>**20090629085351
874 Ignore-this: a35e93b37ca9595c73460243180f4b9d
875]
876[set binary mode for existing FDs on Windows (fixes some GHCi test failures)
877Simon Marlow <marlowsd@gmail.com>**20090626120522
878 Ignore-this: 580cf636e9c77d8427aff6861d089481
879]
880[Move directory-related stuff to the unix package
881Simon Marlow <marlowsd@gmail.com>**20090625120325
882 Ignore-this: b997b3cbce0a46ca87ad825bbdc0a411
883 now that it isn't used on Windows any more.
884]
885[TAG 2009-06-25
886Ian Lynagh <igloo@earth.li>**20090625160056]
887Patch bundle hash:
8881187464f899e283a7f4f5bc9b766623c1bb25729