Ticket #3939: remove-obsolete-todo-item.dpatch

File remove-obsolete-todo-item.dpatch, 27.7 KB (added by dagit, 6 years ago)

Patch to remove the todo item

11 patch for repository http://darcs.haskell.org/packages/base:
3Thu Mar 25 11:13:05 PDT 2010  Jason Dagit <[email protected]>
4  * remove obsolete todo item
6  Simon M. confirmed this comment is no longer relevant:
7    http://www.haskell.org/pipermail/haskell-cafe/2010-March/075117.html
9New patches:
11[remove obsolete todo item
12Jason Dagit <[email protected]>**20100325181305
13 Ignore-this: 79fed527918bbe5243a58d38a8225c31
15 Simon M. confirmed this comment is no longer relevant:
16   http://www.haskell.org/pipermail/haskell-cafe/2010-March/075117.html
17] hunk ./GHC/IO/Handle/Text.hs 164
18 -- ---------------------------------------------------------------------------
19 -- hGetLine
21--- ToDo: the unbuffered case is wrong: it doesn't lock the handle for
22--- the duration.
24 -- | Computation 'hGetLine' @hdl@ reads a line from the file or
25 -- channel managed by @hdl@.
26 --
30[Write the thunk 'next' to the MVar
31Bas van Dijk <[email protected]>**20100319125951
32 Ignore-this: dd25636cf220131385ff2fd32493d456
34[change to use STM, fixing 4 things
35Simon Marlow <[email protected]>**20100318104436
36 Ignore-this: 551d30280a7941c08f5c3b14576bdd70
37   1. there was no async exception protection
38   2. there was a space leak (now new value is strict)
39   3. using atomicModifyIORef would be slightly quicker, but can
40      suffer from adverse scheduling issues (see #3838)
41   4. also, the STM version is faster.
43[Tweak docs
44Ian Lynagh <[email protected]>**20100312214129]
45[Fixed dead links in documentation of forkIO
46Bas van Dijk <[email protected]>**20100308222415
47 Ignore-this: 7deb8fd064c867fbede2a6b2e9da4f15
49[Documentation fixes in Control.Exception
50Bas van Dijk <[email protected]>**20100301220442
51 Ignore-this: 761fcba401cbd1f47276ddfc9b5b80f2
53[Plug two race conditions that could lead to deadlocks in the IO manager
54Simon Marlow <[email protected]>**20100225120255
55 Ignore-this: e6983d6b953104d370278ab3e4617e8b
57[FIX #3866: improve documentation of Data.Data.Constr
58[email protected]**20100224125506
59 Ignore-this: 3818c5d8fee012a3cf322fb455b6e5dc
61[UNDO: Handle NaN, -Infinity and Infinity in the toRational for Float/Double (#3676)
62Simon Marlow <[email protected]>**20100223101603
63 Ignore-this: 78becb2d39b3cd9a1a473a5811ca7d92
65[Put the complexity in the length docs. Fixes trac #3680
66Ian Lynagh <[email protected]>**20100221191425]
67[nhc98 should build Data.Functor.
68[email protected]**20100221163218]
69[Update the exitWith docs
70Ian Lynagh <[email protected]>**20100213140004
71 Error pointed out by Volker Wysk <[email protected]>
73[Handle NaN, -Infinity and Infinity in the toRational for Float/Double (#3676)
74Simon Marlow <[email protected]>**20100211101955
75 Ignore-this: 261415363303efca265e80290eac5f28
77[For nhc98, import unsafeInterleaveIO rather than defining it here.
78[email protected]**20100204171021]
79[Stifle warning about unused return value
80[email protected]**20100203025537]
81[fix #3832: use the locale encoding in openTempFile
82Simon Marlow <[email protected]>**20100120211830
83 Ignore-this: df4f778cc5fefb32290c798db722632c
84 Also while I was here fix an XXX: the Handle contained an
85 uninformative string like <fd: 4> for error messages rather than the
86 real file path.
88[Fix the build: export void, so it doesn't give an unused binding warning
89Ian Lynagh <[email protected]>**20100116174451]
90[hIsEOF: don't do any decoding (#3808)
91Simon Marlow <[email protected]>**20100112230317
92 Ignore-this: 6a384dd2d547ffe3ad3762920e5c1671
94[Control.Monad: +void :: f a -> f ()
95[email protected]**20100108214455
96 Ignore-this: 4dc07452315f2d1b4941903ff42fc45f
97 See http://hackage.haskell.org/trac/ghc/ticket/3292
98 Turns m a -> m (). Lets one call functions for their side-effects without
99 having to get rid of their return values with '>> return ()'. Very useful
100 in many contexts (parsing, IO etc.); particularly good for 'forkIO' and 'forM_',
101 as they demand return types of 'IO ()' though most interesting IO functions
102 return non-().
104[Replace the implementation of mergesort with a 2x faster one.
105[email protected]**20091224152014
106 See ticket http://hackage.haskell.org/trac/ghc/ticket/2143.
108[Restore previous Data.Typeable.typeOf*Default implementations for non-ghc.
109[email protected]**20091223142625
110 Not all compilers have ScopedTypeVariables.
112[Add comments about double bounds-checking, and fast paths for rectangular arrays
113[email protected]**20091218165655
114 Ignore-this: ea0849419dc00927aba4bd410b1cc58d
116 See Note [Double bounds-checking of index values] for the details.
118 The fast paths omit the doubled checks for cases we know about
120[Fix Trac #3245: memoising typeOf
121[email protected]**20091218155117
122 Ignore-this: 5a178a7f2222293c5ee0c3c43bd1b625
124 The performance bug in #3245 was caused by computing the typeRep
125 once for each call of typeOf, rather than once for each dictionary
126 contruction.  (Computing TypeReps is reasonably expensive, because
127 of the hash-consing machinery.)
129 This is readily fixed by putting the TypeRep construction outside
130 the lambda.  (Arguably GHC might have worked that out itself,
131 but it involves floating something between a type lambda and a
132 value lambda, which GHC doesn't currently do. If it happens a lot
133 we could fix that.)
135[Mark 'index' as INLINE in GHC.Arr
136[email protected]**20091216170441
137 Ignore-this: a4df9d8acf496c8e0e9ce5a520509a2a
139 This makes indexing much faster. See Trac #1216
141[Comment the remaining orphan instance modules
142Ian Lynagh <[email protected]>**20091206125021]
143[De-orphan Eq/Ord Float/Double
144Ian Lynagh <[email protected]>**20091205181238]
145[Add comments to "OPTIONS_GHC -fno-warn-orphans" pragmas
146Ian Lynagh <[email protected]>**20091205165854]
147[Data.Either.partitionEithers was insufficiently lazy.
148[email protected]**20091202032807
149 Ignore-this: 77e1b3288f66608c71458d8a91bcbe12
150 Spotted by Daniel Fischer.
152[fix the docs regarding finalizer guarantees
153Simon Marlow <[email protected]>**20091130144409
154 Ignore-this: d1ab9532c74a002b8075ff60febcbe2d
156[x86_64 requires more stack
157[email protected]**20091201033745]
158[check for size < 0 in mallocForeignPtrBytes and friends (#3514)
159Simon Marlow <[email protected]>**20091125143822
160 Ignore-this: 91077d01da2bbe1dfed5155e8b40da9
162[hGetContents: close the handle properly on error
163Simon Marlow <[email protected]>**20091125123435
164 Ignore-this: bc37ff678acc6e547dc390285e056eb9
166 When hGetContents caught an error it was closing the handle and then
167 throwing the exception, without updating the handle with the new
168 closed state.  This lead to a double-closed, which was the cause of
170 *** glibc detected *** ./Setup: double free or corruption
172 when iconv_close was called twice on the decoder.
174 See http://hackage.haskell.org/trac/hackage/ticket/609
176[Fix arities of mapFB and zipFB
177Roman Leshchinskiy <[email protected]>**20091126232219
178 Ignore-this: c4e14cd0a92622549c86e67237a40865
180[Remove an unnecessary -fno-warn-orphans flag
181Ian Lynagh <[email protected]>**20091126123404]
182[Tweak layout to work with alternative layout rule
183Ian Lynagh <[email protected]>**20091125232349]
184[Tweak layout to be accepted by the alternative layout rul
185Ian Lynagh <[email protected]>**20091125194147]
186[Make sure zipWithFB has arity 2
187Roman Leshchinskiy <[email protected]>**20091125010003
188 Ignore-this: 4cf60c55666f03d22a9f5a6e07f52d36
190 It gets 2 arguments in the "zipWith" rule but its arity was higher and the new
191 inliner didn't inline it sometimes, for instance here:
193 mpp ::  [Double] -> [Double] -> [Double] -> [Double] -> [Double]
194 mpp as bs cs ds = zipWith (*) (zipWith (+) as bs) (zipWith (+) cs ds)
196 This was a regression vs. 6.10.
198[Remove an old comment
199Ian Lynagh <[email protected]>**20091124134647]
200[De-orphan the Eq/Ord Integer instances
201Ian Lynagh <[email protected]>**20091124133639]
202[Whitespace only
203Ian Lynagh <[email protected]>**20091124133421]
204[Derive some more instances, rather than writing them by hand
205Ian Lynagh <[email protected]>**20091124011747]
206[We can now derive Ord ()
207Ian Lynagh <[email protected]>**20091124011416]
208[De-orphan tuple Eq/Ord instances
209Ian Lynagh <[email protected]>**20091123233343]
210[Control.Exception.Base no longer has any orphans
211Ian Lynagh <[email protected]>**20091123224905]
212[De-orphan the MonadFix ST instance for GHC
213Ian Lynagh <[email protected]>**20091123223544]
214[Rearrange the contents of Control.Monad.ST; no functionality changes
215Ian Lynagh <[email protected]>**20091123222702]
216[De-orphan the Eq/Ord [a] instances
217Ian Lynagh <[email protected]>**20091123215635]
218[De-orphan the Eq/Ord Char instances
219Ian Lynagh <[email protected]>**20091123202253]
220[De-orphan the Eq/Ord Bool instances
221Ian Lynagh <[email protected]>**20091123201817]
222[Move Eq/Ord Ordering instances to de-orphan them
223Ian Lynagh <[email protected]>**20091123194310]
224[Remove ffi warnings for nhc98.
225[email protected]**20091123063743]
226[Second attempt to fix #1185 (forkProcess and -threaded)
227Simon Marlow <[email protected]>**20091111151915
228 Ignore-this: fa5f5d5e4e080d4b612a37244f937f9c
230 Patch 2/2: first patch is to ghc
232 This time without dynamic linker hacks, instead I've expanded the
233 existing rts/Globals.c to cache more CAFs, specifically those in
234 GHC.Conc.  We were already using this trick for signal handlers, I
235 should have realised before.
237 It's still quite unsavoury, but we can do away with rts/Globals.c in
238 the future when we switch to a dynamically-linked GHCi.
240[Rollback #1185 fix
241Simon Marlow <[email protected]>**20091106140629
242 Ignore-this: cd5667e8474e37e01ba26a1984274811
244 rolling back:
246 Tue Nov  3 16:05:40 GMT 2009  Simon Marlow <[email protected]>
247   * Fix #1185: restart the IO manager after fork()
249   This is the libraries/base part of the patch; there is a corresponding
250   patch to GHC itself.
252   The main change is that we now keep track of the IO manager's ThreadId
253   in a top-level MVar, and ensureIOManagerIsRunning checks whether a
254   previous IO manager thread is alive before starting one.  In the child
255   of fork(), we can hence call ensureIOManagerIsRunning to restart the
256   IO manager.
258     M ./GHC/Conc.lhs -46 +44
260 Wed Nov  4 17:49:45 GMT 2009  Ian Lynagh <[email protected]>
261   * Fix the build on Windows
263     M ./GHC/Conc.lhs -6 +4
265[Fix the build on Windows
266Ian Lynagh <[email protected]>**20091104174945]
267[Fix #1185: restart the IO manager after fork()
268Simon Marlow <[email protected]>**20091103160540
269 Ignore-this: 6dc05464f1500104554637f4759738cc
271 This is the libraries/base part of the patch; there is a corresponding
272 patch to GHC itself.
274 The main change is that we now keep track of the IO manager's ThreadId
275 in a top-level MVar, and ensureIOManagerIsRunning checks whether a
276 previous IO manager thread is alive before starting one.  In the child
277 of fork(), we can hence call ensureIOManagerIsRunning to restart the
278 IO manager.
280[improve the documentation for throwErrnoIfRetry
281Simon Marlow <[email protected]>**20091016112404
282 Ignore-this: b77275cacf730e15757946027168f63e
284[Don't inline unpackFoldrCString ever
285[email protected]**20091029135350
286 Ignore-this: 85d672649b1b776efc7e97500b05d4f9
288[Inline more default methods
289[email protected]**20091029135330
290 Ignore-this: 289c44b0afd6d5631c2a4e0664275ca9
292 Namely Monad: (>>)
293        Eq:    (==), (/=)
294        Num:   (-), negate
295        Real:  quot, rem, div, mod, recip, (/), truncate
296        Float: (**), logBase, sqrt, tan, tanh
298[Move error messages out of INLINEd default methods
299[email protected]**20091029135118
300 Ignore-this: 9e35dc947f94827a3529eb53a41575fd
302 No need to duplicate the error generation!
304[Exploit now-working default-method INLINE pragmas for Data.Bits
305[email protected]**20091029135041
306 Ignore-this: 8adf225f31ca7a3181ee087e9e4fe535
308 * Add INLINE pragmas to default methods for class Bits
310 * Remove redundant instance methods elsewhere, now that
311   the default method will do the job
313[Tidy up and comment imports
314[email protected]**20091029134414
315 Ignore-this: bf2be31035de975d8995e988933cc940
317[Inline foldr and (.) when applied to two arguments not three
318[email protected]**20091029134335
319 Ignore-this: fccb6f3e90e15f44cb465814be85ede2
321 The new INLINE story is (by design) arity-sensitive, so we must
322 put fewer argument on the LHS for foldr and (.)
324[dirUtils.c no longer available
325[email protected]**20091013093833]
326[Make hGetContents throw an exception if an error is encountered
327Simon Marlow <[email protected]>**20091012152955
328 Ignore-this: 9f7a7176193eab25c9daaacd9261f2de
330 Strictly speaking this breaks Haskell 98 compatibility, which requires
331 hGetContents to just end the lazy stream silently if an error is
332 encountered.  However, for a few reasons we think it will make
333 everyone's life a bit easier if we make this change
335  1. Errors will be a lot more common in GHC 6.12.1, in the form
336     of Unicode decoding errors.
338  2. When Haskell 98 was designed, we didn't know how to throw
339     exceptions from inside lazy I/O, but now we do.
341  3. If anyone is actually relying on the previous behaviour, their
342     code is arguably broken.
344[Re-instate System.Console.Getopt for nhc98 builds.
345[email protected]**20091013092843
346 Although it was split out of base a while back, that change was
347 reverted for ghc soon afterwards, but nhc98 never noticed.
349[Roll back "Another instance of nhc98's strange import semantics."
350Ian Lynagh <[email protected]>**20091009185618
351 Fri Oct  9 14:50:51 BST 2009  [email protected]
352 GHC (correctly) warns about the unused import, which breaks the validate
353 build.
355[Roll back "Cope with nhc98's (occasionally-strange) import semantics"
356Ian Lynagh <[email protected]>**20091009184704
357 Fri Oct  9 14:43:51 BST 2009  [email protected]
358 GHC (correctly) warns about the unused import, which breaks the validate
359 build.
361[It seems that nhc98 needs defaulting in Data.Fixed.
362[email protected]**20091009135242]
363[Another instance of nhc98's strange import semantics.
364[email protected]**20091009135051]
365[Make Data.Functor compatible with non-GHC compilers.
366[email protected]**20091009134821]
367[Cope with nhc98's (occasionally-strange) import semantics.
368[email protected]**20091009134351]
369[Fix gratuitous breakage of nhc98 in System.IO.
370[email protected]**20091009134001]
371[Fix gratuitous breakage of nhc98 in Control.Exception.Base.
372[email protected]**20091009133615]
373[Fix gratuitous breakage of non-GHC in Data.Fixed.
374[email protected]**20091009133330]
375[Fix gratuitous breakage for non-GHC in Data.Bits.
376[email protected]**20091009133257]
377[Use UTF-32LE instead of UTF32LE
378Simon Marlow <[email protected]>**20091006100207
379 Ignore-this: 7f881e36543d250ef848c9f60d67655a
380 The latter is not recognised by some iconv implementations.
382[Strip any Byte Order Mark (BOM) from the front of decoded streams.
383[email protected]*-20090930084229
384 Ignore-this: d0d0c3ae87b31d71ef1627c8e1786445
385 When decoding to UTF-32, Solaris iconv inserts a BOM at the front
386 of the stream, but Linux iconv doesn't.
388[use UTF32BE/UTF32LE instead of UCS-4/UCS-4LE
389Simon Marlow <[email protected]>**20091005101554
390 Ignore-this: 2aef5e9bec421e714953b7aa1bdfc1b3
392[Strip any Byte Order Mark (BOM) from the front of decoded streams.
393[email protected]**20090930084229
394 Ignore-this: d0d0c3ae87b31d71ef1627c8e1786445
395 When decoding to UTF-32, Solaris iconv inserts a BOM at the front
396 of the stream, but Linux iconv doesn't.
398[Add traceEvent :: String -> IO ()
399Simon Marlow <[email protected]>**20090925141257
400 Ignore-this: 8b1888bbf9682ffba13f815b6000e4b1
401 For emitting an event via the RTS tracing framework
403[Fix the error message when flushing the read buffer of a non-seekable Handle
404Simon Marlow <[email protected]>**20090923090536
405 Ignore-this: 4342026df93759d99480f4e13f80a492
407[Fix #3534: No need to flush the byte buffer when setting binary mode
408Simon Marlow <[email protected]>**20090923090445
409 Ignore-this: 625817ed7ae2c12291eb993a99dc640a
411[Use let !y = x in .. x .. instead of seq in $! and evaluate (#2273)
412Simon Marlow <[email protected]>**20090916140454]
413[make some Applicative functions into methods, and split off Data.Functor (proposal #3335)
414Ross Paterson <[email protected]>**20090915173109
415 Ignore-this: a0cff4de6dfdbcbd56a66101bc4855a9
417 The following functions
419     (<$) :: Functor f => a -> f b -> f a
420     (*>) :: Applicative f => f a -> f b -> f b
421     (<*) :: Applicative f => f a -> f b -> f a
422     some :: Alternative f => f a -> f [a]
423     many :: Alternative f => f a -> f [a]
425 are moved into the corresponding classes, with the existing implementations
426 as default definitions.  This gives people creating instances the option of
427 defining specialized implementations of these functions, though they should
428 be equivalent to the default definitions.
430 Although (<$) is now a method of the Functor class, it is hidden in the
431 re-export by the Prelude, Control.Monad and Monad.  The new module
432 Data.Functor exposes the full class, plus the function (<$>).  These are
433 also re-exported by Control.Applicative.
435[On Windows, use the console code page for text file encoding/decoding.
436Judah Jacobson <[email protected]>**20090913022126
437 Ignore-this: 86c2f2db8ef92b751599795d3195187b
439 We keep all of the code page tables in the module
440 GHC.IO.Encoding.CodePage.Table.  That file was generated automatically
441 by running codepages/MakeTable.hs; more details are in the comments at the
442 start of that script.
444 Storing the lookup tables adds about 40KB to each statically linked executable;
445 this only increases the size of a "hello world" program by about 7%.
447 Currently we do not support double-byte encodings (Chinese/Japanese/Korean), since
448 including those codepages would increase the table size to 400KB.  It will be
449 straightforward to implement them once the work on library DLLs is finished.
451[Fix "init" docs: the input list need not be finite. Fixes trac #3465
452Ian Lynagh <[email protected]>**20090911210437]
453[Bump base version to
454Ian Lynagh <[email protected]>**20090911153913]
455[Address #3310
456Simon Marlow <[email protected]>**20090830152850
457 Ignore-this: 40c7f7c171ee299a83092fd360a952b7
459  - Rename BlockedOnDeadMVar   -> BlockedIndefinitelyOnMVar
460  - Rename BlockedIndefinitely -> BlockedIndefinitelyOnSTM
461  - instance Show BlockedIndefinitelyOnMVar is now
462      "blocked indefinitely in an MVar operation"
463  - instance Show BlockedIndefinitelyOnSTM is now
464      "blocked indefinitely in an STM transaction"
466 clients using Control.OldException will be unaffected (the new
467 exceptions are mapped to the old names).  However, for base4-compat
468 we'll need to make a version of catch/try that does a similar
469 mapping.
471[Fix unicode conversion for MSB architectures
472[email protected]**20090830130028
473 This fixes the SPARC/Solaris build.
475[Fix #3441: detect errors in partial sequences
476Simon Marlow <[email protected]>**20090830075909
477 Ignore-this: d12a75d95e0cae5eb1555266810ec281
479[Fix hWaitForInput
480Simon Marlow <[email protected]>**20090827152116
481 Ignore-this: 2550e911f1a4d4357a5aa8d1764238ce
482 It was erroneously waiting when there were bytes to decode waiting in
483 the byte buffer.
485[fix debugging code
486Simon Marlow <[email protected]>**20090827150628
487 Ignore-this: e1c82fdc19a22e247cd69ff6fa11921d
489[Allow for configurable iconv include and library locations.
490Matthias Kilian <[email protected]>**20090826154406
491 Ignore-this: be95fab611a5534cf184b508964ed498
492 This should help to fix the build on OpenBSD.
494[typo in comment
495Simon Marlow <[email protected]>**20090826085252
496 Ignore-this: 1903ee0f354157a6ed3871c100f6b1b9
498[un-hide some modules from the Haddock docs
499Simon Marlow <marlowsd@gmail.com>**20090825152457
500 Ignore-this: dce6606f93cf977fb24ebe99082dfa62
502[Apply fix for #1548, from squadette@gmail.com
503Simon Marlow <marlowsd@gmail.com>**20090819120700
504 Ignore-this: 31c237c46a6445f588ed4b8c51bb6231
506[improvements to Data.Fixed: instances for Typeable and Data, more predefined types
507Ashley Yakeley <ashley@semantic.org>**20090812055058
508 Ignore-this: feeece36d5632f02a05d137d2a39ab78
510[Fix "Cabal check" warnings
511Ian Lynagh <igloo@earth.li>**20090811215856]
512[Add a GHC.Constants module; fixes trac #3094
513Ian Lynagh <igloo@earth.li>**20090809183252]
514[Apply proposal #3393
515Ian Lynagh <igloo@earth.li>**20090809134717
516 Add openTempFileWithDefaultPermissions and
517 openBinaryTempFileWithDefaultPermissions.
519[Add some more C wrappers; patch from Krister Walfridsson
520Ian Lynagh <igloo@earth.li>**20090807200631
521 Fixes 21 testsuite errors on NetBSD 5.99.
523[Fixing configure for autoconf 2.64
524Alexander Dunlap <alexander.dunlap@gmail.com>**20090805060748
525 Ignore-this: 992ab91ae3d68c12dbb265776e33e243
527[add INLINE toList
528Ross Paterson <ross@soi.city.ac.uk>**20090806142853
529 Ignore-this: aba16aabb17d5dca44f15d188945680e
531 In anticipation of the fixing of #2353.
533[fix a copyright
534Simon Marlow <marlowsd@gmail.com>**20090805134045
535 Ignore-this: b0ffbdd38fbba121e8bcba37c4082a60
537[Tweak the BufferedIO class to enable a memory-mapped file implementation
538Simon Marlow <marlowsd@gmail.com>**20090805134036
539 Ignore-this: ec67d7a0a6d977438deaa342503f77e0
540 We have to eliminate the assumption that an empty write buffer can be
541 constructed by setting the buffer pointers to zero: this isn't
542 necessarily the case when the buffer corresponds to a memory-mapped
543 file, or other in-memory device implementation.
545[Deprecate Control.OldException
546Ian Lynagh <igloo@earth.li>**20090804143910]
547[Windows build fix, following RTS tidyup
548Simon Marlow <marlowsd@gmail.com>**20090803131121
549 Ignore-this: ce862fb91c2b234211a8757f98690778
551[Updates to follow the RTS tidyup
552Simon Marlow <marlowsd@gmail.com>**20090801220743
553 Ignore-this: 6e92412df93a66c12d75344053d5634
554 C functions like isDoubleNaN moved here (primFloat.c)
556[Add integer-simple as a build option
557Ian Lynagh <igloo@earth.li>**20090722013151]
558[Use shift[LR]Integer in the Bits Integer instance
559Ian Lynagh <igloo@earth.li>**20090721222440]
560[depend directly on integer-gmp, rather than indirecting through integer
561Ian Lynagh <igloo@earth.li>**20090721185228]
562[Move the instances of Functor and Monad IO to GHC.Base, to avoid orphans
563Simon Marlow <marlowsd@gmail.com>**20090722102130
564 Ignore-this: a7d85ac0025d559674249de0108dbcf4
566[move "instance Exception Dynamic" so it isn't an orphan
567Simon Marlow <marlowsd@gmail.com>**20090721093854
568 Ignore-this: 5ede91ecfec2112c91b699d4de87cd02
570[Improve the index checking for array accesses; fixes #2120 #2669
571Ian Lynagh <igloo@earth.li>**20090719153228
572 As well as checking that offset we are reading is actually inside the
573 array, we now also check that it is "in range" as defined by the Ix
574 instance. This fixes confusing behaviour (#2120) and improves some error
575 messages (#2669).
577[Make chr say what its argument was, if it's a bad argument
578Ian Lynagh <igloo@earth.li>**20090718151049]
579[remove unused warning
580Simon Marlow <marlowsd@gmail.com>**20090715124416
581 Ignore-this: 31f613654089d0f4a44363946087b41e
583[warning fix: -fno-implicit-prelude -> -XNoImplicitPrelude
584Simon Marlow <marlowsd@gmail.com>**20090715122839
585 Ignore-this: dc8957249731d5bcb71c01899e5adf2b
587[Add hGetEncoding :: Handle -> IO (Maybe TextEncoding)
588Simon Marlow <marlowsd@gmail.com>**20090715122519
589 Ignore-this: 14c3eff996db062da1199739781e4708
590 as suggested during the discussion on the libraries list
592[Add more documentation to mkTextEncoding
593Simon Marlow <marlowsd@gmail.com>**20090715122414
594 Ignore-this: 97253b2624267df3a246a18121e8ea81
595 noting that "//IGNORE" and "//TRANSLIT" suffixes can be used with GNU
596 iconv.
598[Add the utf8_bom codec
599Simon Marlow <marlowsd@gmail.com>**20090715122257
600 Ignore-this: 1c9396cd805201fe873a39382ced79c7
601 as suggested during the discussion on the libraries list.
603[Export Unicode and newline functionality from System.IO; update Haddock docs
604Simon Marlow <marlowsd@gmail.com>**20090713113104
605 Ignore-this: c3f017a555335aa55d106253393f72e2
607[add a comment about the non-workingness of CHARBUF_UTF16
608Simon Marlow <marlowsd@gmail.com>**20090707124406
609 Ignore-this: 98d00411b68d688b3b4cffc9507b1f35
611[Fix build on Windows
612Ian Lynagh <igloo@earth.li>**20090711004351]
613[Fix some "warn-unused-do-bind" warnings where we want to ignore the value
614Ian Lynagh <igloo@earth.li>**20090710204513]
615[Use throwErrnoIfMinus1_ when calling getrusage
616Ian Lynagh <igloo@earth.li>**20090710204221]
617[Remove an unused import
618Ian Lynagh <igloo@earth.li>**20090710153345]
619[reportStackOverflow now returns IO ()
620Ian Lynagh <igloo@earth.li>**20090710153257
621 It used to do "return undefined" to return IO a.
623[GHC.Conc.reportError now returns IO ()
624Ian Lynagh <igloo@earth.li>**20090710152646
625 It used to return IO a, by "return undefined".
627[Fix some "warn-unused-do-bind" warnings where we want to ignore the value
628Ian Lynagh <igloo@earth.li>**20090710152526]
629[Minor SampleVar refactoring
630Ian Lynagh <igloo@earth.li>**20090710151438]
631[Fix "warn-unused-do-bind" warnings in GHC/IO/Handle/Text.hs
632Ian Lynagh <igloo@earth.li>**20090710122905]
633[Fix some "warn-unused-do-bind" warnings where we just want to ignore the result
634Ian Lynagh <igloo@earth.li>**20090710005638]
635[Use the result of writeCharBuf in GHC/IO/Encoding/Latin1.hs too
636Ian Lynagh <igloo@earth.li>**20090710004032]
637[Minor code tidyups in GHC.Conc
638Ian Lynagh <igloo@earth.li>**20090710003801]
639[Fix "warn-unused-do-bind" warning in GHC.Conc
640Ian Lynagh <igloo@earth.li>**20090710003530
641 If we fail to communicate with the IO manager then we print a warning
642 using debugErrLn from the ghc-prim package.
644[Fix "warn-unused-do-bind" warnings in System.Posix.Internals
645Ian Lynagh <igloo@earth.li>**20090709164546]
646[Fix "warn-unused-do-bind" warnings where we really do want to ignore the result
647Ian Lynagh <igloo@earth.li>**20090709163912]
648[Add back imports needed on Windows
649Ian Lynagh <igloo@earth.li>**20090707181924]
650[Remove unused imports
651Ian Lynagh <igloo@earth.li>**20090707115810]
652[Remove unused imports from base
654 Ignore-this: f9b5f353e3bb820f787c56d615b28765
656 These unused imports are detected by the new unused-import code
659[Use the result of writeCharBuf
660Simon Marlow <marlowsd@gmail.com>**20090706133303
661 Ignore-this: 52288dd559bf4c4f313df6197091d935
663 This only makes a difference when CHARBUF_UTF16 is in use, which it
664 normally isn't.  I suspect CHARBUF_UTF16 doesn't currently work for
665 other reasons (CHARBUF_UTF16 was an experiment before I wrote the
666 GHC.IO.Encoding.UTF* codecs), but this patch at least makes it
667 slightly closer to working.
669[Remove some cruft from Data.HashTable
670Ian Lynagh <igloo@earth.li>**20090706181630]
671[Add 'eof' to Text.ParserCombinators.ReadP
673 Ignore-this: 2aea7b848e00c894761bc4011adaa95d
675 Add a ReadP parser that succeeds at the end of input. Very useful!
678[Don't export CLDouble for GHC; fixes trac #2793
679Ian Lynagh <igloo@earth.li>**20090705155120
680 We never really supported CLDouble (it was a plain old double underneath),
681 and pretending that we do does more harm than good.
683[a byte between 0x80 and 0xBF is illegal immediately (#3341)
684Simon Marlow <marlowsd@gmail.com>**20090702081415
685 Ignore-this: dc19ef59a1a21118d5a7dd38aa2f611c
687[avoid a warning
688Simon Marlow <marlowsd@gmail.com>**20090630084134
689 Ignore-this: c92a45ee216faf01327feae9fe06d6e2
691[Add a wrapper for libiconv.
692Matthias Kilian <kili@outback.escape.de>**20090629183634
693 Ignore-this: 23c6047c0d71b745b495cc223574a47f
695[#include <sys/times.h> if we have it (should fix build problems)
696Simon Marlow <marlowsd@gmail.com>**20090629085351
697 Ignore-this: a35e93b37ca9595c73460243180f4b9d
699[set binary mode for existing FDs on Windows (fixes some GHCi test failures)
700Simon Marlow <marlowsd@gmail.com>**20090626120522
701 Ignore-this: 580cf636e9c77d8427aff6861d089481
703[Move directory-related stuff to the unix package
704Simon Marlow <marlowsd@gmail.com>**20090625120325
705 Ignore-this: b997b3cbce0a46ca87ad825bbdc0a411
706 now that it isn't used on Windows any more.
708[TAG 2009-06-25
709Ian Lynagh <igloo@earth.li>**20090625160056]
710Patch bundle hash: