Ticket #3410: ghc-tweak-ParsePkgConf.dpatch

File ghc-tweak-ParsePkgConf.dpatch, 27.7 KB (added by int-e, 5 years ago)

Fix: Extend ghc's parser to cover versioned licenses.

Line 
1Sat Aug  1 23:28:50 CEST 2009  Bertram Felgenhauer <int-e@gmx.de>
2  * Parse versioned GPL and LGPL licenses in package config
3
4
5New patches:
6
7[Parse versioned GPL and LGPL licenses in package config
8Bertram Felgenhauer <int-e@gmx.de>**20090801212850] {
9hunk ./compiler/main/ParsePkgConf.y 34
10  '}'           { L _ ITccurly }
11  '['           { L _ ITobrack }
12  ']'           { L _ ITcbrack }
13+ '('           { L _ IToparen }
14+ ')'           { L _ ITcparen }
15  ','           { L _ ITcomma }
16  '='           { L _ ITequal }
17  VARID         { L _ (ITvarid    $$) }
18hunk ./compiler/main/ParsePkgConf.y 89
19        | VARID '=' CONID STRING        { id }
20                -- another case of license
21 
22+       | VARID '=' CONID license_ver   { id }
23+               -- another case of license!
24+
25        | VARID '=' strlist             
26        {\p -> case unpackFS $1 of
27                "exposedModules"    -> p{exposedModules    = map mkModuleNameFS $3}
28hunk ./compiler/main/ParsePkgConf.y 121
29                        _             -> happyError
30                }
31 
32+license_ver :: { () }
33+       : CONID
34+               {% case unpackFS $1 of
35+                       "Nothing" -> return ()
36+                       _         -> happyError
37+               }
38+
39+       | '(' CONID '(' version ')' ')'
40+               {% case unpackFS $2 of
41+                       "Just"    -> return ()
42+                       _         -> happyError
43+               }
44+
45 pkgid  :: { PackageIdentifier }
46        : CONID '{' VARID '=' CONID STRING ',' VARID '=' version '}'
47             {% case unpackFS $5 of
48}
49
50Context:
51
52[Fix cleaning the integer package
53Ian Lynagh <igloo@earth.li>**20090730143129]
54[Fix warnings when building with the HEAD
55Ian Lynagh <igloo@earth.li>**20090730143031]
56[Remove some redundant fromIntegral's
57Ian Lynagh <igloo@earth.li>**20090730105532]
58[Fix space problems in ghci
59Ian Lynagh <igloo@earth.li>**20090730105351
60 We were making arrays with range (0, n-1) which is bad if n == 0 now
61 that we are using Word types.
62]
63[Make the types we use when creating GHCi bytecode better match reality
64Ian Lynagh <igloo@earth.li>**20090729130911
65 We were keeping things as Int, and then converting them to Word16 at
66 the last minute, when really they ought to have been Word16 all along.
67]
68[Add an (Outputable Word16) instance
69Ian Lynagh <igloo@earth.li>**20090728134436]
70[Fix whitespace in ByteCodeAsm.lhs
71Ian Lynagh <igloo@earth.li>**20090728123444]
72[Add CHECK(p), like ASSERT(p) but works even when !defined(DEBUG)
73Simon Marlow <marlowsd@gmail.com>**20090729075433
74 Ignore-this: a723f456e4b1eaeaa617a675da276aa2
75 For inexpensive assertions
76]
77[enable the x86-specific versions of atomic_inc()/atomic_dec()
78Simon Marlow <marlowsd@gmail.com>**20090729075307
79 Ignore-this: a271b7ade5502ec4d8444aac19f4b4c3
80]
81[fix warning
82Simon Marlow <marlowsd@gmail.com>**20090728103855
83 Ignore-this: ee626aa5de0a2aed8f44ae1131cc341d
84]
85[fix warning
86Simon Marlow <marlowsd@gmail.com>**20090728102707
87 Ignore-this: ad4f07a163921d17d242a9452b4bb578
88]
89[fix a warning
90Simon Marlow <marlowsd@gmail.com>**20090728101731
91 Ignore-this: 73ead0a0004723757c0a51b56681c991
92]
93[Be a bit more sensible about choosing external OccNames
94Simon Marlow <marlowsd@gmail.com>**20090728100434
95 Ignore-this: 4adcd661e76440deb2b4ab498ebd2d1e
96 
97 Instead of chr_$wchr, we now just get $wchr.  In general, when an
98 OccName is system-generated, we leave it out of the final external
99 name, preferring to use the name of the exported parent instead (which
100 is necessarily a user-written name).
101 
102 Names should be no less deterministic, but should be shorter and more
103 readable.
104]
105[Remove old 'foreign import dotnet' code
106Simon Marlow <marlowsd@gmail.com>**20090727144524
107 Ignore-this: 821ebea2c3897415195318f107421472
108 It still lives in darcs, if anyone wants to revive it sometime.
109]
110[remove a couple of ToDos
111Simon Marlow <marlowsd@gmail.com>**20090727094719
112 Ignore-this: ef00fc481821dff4381ba9efcd792708
113]
114[buildinfo files need a $$(wildcard)
115Simon Marlow <marlowsd@gmail.com>**20090727091012
116 Ignore-this: b4c3201dfa81fef32ee254dd9c955b2d
117]
118[Slight tweak to avoid overflowing the command-line size in bindist
119Simon Marlow <marlowsd@gmail.com>**20090727090946
120 Ignore-this: 1e2ff207d03fed08576ac59f0b46c08c
121 Not a real fix: if this bites us again we'll have to rethink
122]
123[avoid (benign) error about overriding rules for binary-dist
124Simon Marlow <marlowsd@gmail.com>**20090727090903
125 Ignore-this: ffb7f7bf1290f2faf96ac177f76a1422
126]
127[Give a better error message for hidden packages when building Cabal package
128Ian Lynagh <igloo@earth.li>**20090726194915
129 Fixes #3168
130]
131[Add a -fbuilding-cabal-package flag
132Ian Lynagh <igloo@earth.li>**20090726181934
133 This means GHC knows whether it's building a Cabal package, or just
134 Haskell sources. For example, it may wish to give different error
135 messages when building a Cabal package.
136]
137[Add an extension to disable n+k patterns
138Ian Lynagh <igloo@earth.li>**20090725134703]
139[Fix a warning on Windows
140Ian Lynagh <igloo@earth.li>**20090724221244]
141[Remove GHC's haskell98 dependency
142Ian Lynagh <igloo@earth.li>**20090724210825]
143[add number of bytes to +RTS -DS leak reports
144Simon Marlow <marlowsd@gmail.com>**20090724150010
145 Ignore-this: 3a66585c8fd2b58ce96abab1e154fb6e
146]
147[free the gc_thread structures during shutdown
148Simon Marlow <marlowsd@gmail.com>**20090724145956
149 Ignore-this: 25efeb189cbfb549af4550d266604f0e
150]
151[Add atomic_inc()/atomic_dec(), and use them to replace gc_running_mutex
152Simon Marlow <marlowsd@gmail.com>**20090724142620
153 Ignore-this: d775eeaf85fd0f9064d87a0909134bc0
154 This also fixes a memory leak on Windows with -threaded, because we
155 were calling initMutex(&gc_running_mutex) for each GC, which allocates
156 memory.
157]
158[Rewrite the foreign import string parser using ReadP
159Simon Marlow <marlowsd@gmail.com>**20090723152138
160 Ignore-this: 1c7db770a29d48710b05e2a3d216b2a8
161 And kill the duplicate one in HsSyn.Convert
162]
163[point to the wiki
164Simon Marlow <marlowsd@gmail.com>**20090723132345
165 Ignore-this: c11300bac62ce2f56d7fc271aa26dbcd
166]
167[Remove note about avoiding use of #def in libraries
168Simon Marlow <marlowsd@gmail.com>**20090723111026
169 Ignore-this: 8a027ed37b2d10094f7a31548aee2535
170 It should be safe to use now that we aren't relying on C prototypes
171 for foreign functions in via-C code.
172]
173[refactorings
174Simon Marlow <marlowsd@gmail.com>**20090723091230
175 Ignore-this: 836feb0e819127603dd56623af6e48dc
176]
177[Fix Trac #3391: make generic to/from bindings only for newly-declared types
178simonpj@microsoft.com**20090723155803
179 Ignore-this: bb56c2ec054397d421dce13d5eb6c73f
180 
181 Before this patch we were bogusly making to/from bindings for all data types
182 in the TcGblEnv.  But that is wrong when we have multiple "chunks" of
183 bindings in Template Haskell.  We should start from the declarations
184 themselves.  Easy.
185 
186]
187[Print explicit braces and semicolons in do-notation
188simonpj@microsoft.com**20090723152411
189 Ignore-this: a97ddf19774d27d15a01d63787708b20
190 
191 By printing explicit braces we make it more likely that pretty-printed
192 code will be acceptable if fed back into GHC.
193 
194 See http://www.haskell.org/pipermail/glasgow-haskell-users/2009-July/017554.html
195 
196]
197[Documentation for stand-alone deriving (Trac #3012)
198simonpj@microsoft.com**20090723132558
199 Ignore-this: 54445c5984594eb7f82151b2ac118695
200]
201[Windows only: set the encoding on stdin to utf8
202Simon Marlow <marlowsd@gmail.com>**20090723121913
203 Ignore-this: d65115d9711b5fb68e77786565ef6de
204 Otherwise it defaults to latin1.
205]
206[Fix Trac #3012: allow more free-wheeling in standalone deriving
207simonpj@microsoft.com**20090723130145
208 Ignore-this: 357580b9388ccbe1da3c1da3ba90e456
209 
210 In standalone deriving, we now do *not* check side conditions.
211 We simply generate the code and typecheck it.  If there's a type
212 error, it's the programmer's problem.
213 
214 This means that you can do 'deriving instance Show (T a)', where
215 T is a GADT, for example, provided of course that the boilerplate
216 code does in fact typecheck.
217 
218 I put some work into getting a decent error message.  In particular
219 if there's a type error in a method, GHC will show the entire code
220 for that method (since, after all, the user did not write it).
221 Most of the changes are to achieve that goal.
222 
223 Still to come: changes in the documentation.
224 
225]
226[Use the ErrMsg record type
227simonpj@microsoft.com**20090723130108
228 Ignore-this: 7fb6dd78d3185da0c33901b8aac8d108
229]
230[Stop generating redundant parens in 'deriving' code
231simonpj@microsoft.com**20090723125903
232 Ignore-this: 6fc82df9648a82bcf7bf6fdfa9b4dad3
233 
234 This makes the code printed by -ddump-deriv look prettier
235 
236]
237[Wibble to printing tuple sections
238simonpj@microsoft.com**20090723125756
239 Ignore-this: af2a1b9784f6447fea0e11d454cf082f
240]
241[Fix Trac #3193: improve line number reporting for equality constraints
242simonpj@microsoft.com**20090723065504
243 Ignore-this: b45a68071bcaca48cad7855dccb9c9eb
244 
245 When reporting an error from a failed equality constraint, we were
246 setting the *context* but not the *line number* in TcTyFuns.eqInstMisMatch
247 As a result, the line number didn't match the context at all.  It's
248 trivial to fix.
249 
250 I'm 99% certain this fixes #3193, but it's too complicated to
251 reproduce, so I have not actually tested it.
252 
253]
254[Add fmapM_maybe :: Monad m => (a -> m b) -> Maybe a -> m (Maybe b)
255simonpj@microsoft.com**20090723064932
256 Ignore-this: db5f6319b52a5e6b5f85d76985f2a7c9
257 
258 This function isn't used at the moment, but Max added it, and it
259 looks useful.
260 
261]
262[Add tuple sections as a new feature
263simonpj@microsoft.com**20090723063859
264 Ignore-this: d42a26fc1efff112b852b5c1135c1746
265 
266 This patch adds tuple sections, so that
267 
268        (x,,z)  means   \y -> (x,y,z)
269 
270 Thanks for Max Bolinbroke for doing the hard work.
271 
272 In the end, instead of using two constructors in HsSyn, I used
273 just one (still called ExplicitTuple) whose arguments can be
274        Present (LHsExpr id)
275 or     Missing PostTcType
276 
277 While I was at it, I did a bit of refactoring too.
278 
279]
280[NetBSD defines _REENTRANT in its header files, so compiling ghc gives
281Simon Marlow <marlowsd@gmail.com>**20090723075030
282 Ignore-this: 4722c4ff0541c6080de8f433e498684
283 redefinition warnings for all files that are including includes/Rts.h.
284 
285 Contributed by: Krister Walfridsson <krister.walfridsson@gmail.com>
286]
287[includes/TSO.h: kill trailing whitespace
288Samuel Bronson <naesten@gmail.com>**20090722170354
289 Ignore-this: 36d4afd1a21188d604ed6b432942dcdc
290]
291[Say what StgTSOBlockInfo is for, where to read about it.
292Samuel Bronson <naesten@gmail.com>**20090722163011
293 Ignore-this: 6c09e11f23594251cdc2db1bc642edc9
294]
295[Make the Integer library used directly configurable in GHC and base
296Ian Lynagh <igloo@earth.li>**20090722151048
297 Rather than indirecting through an integer package
298]
299[Fix cleaning with the new integer changes
300Ian Lynagh <igloo@earth.li>**20090722142545]
301[Add integer-simple as a build option
302Ian Lynagh <igloo@earth.li>**20090722013137]
303[Tweak whitespace
304Ian Lynagh <igloo@earth.li>**20090719221303]
305[thenIO, bindIO, returnIO moved to GHC.Base
306Simon Marlow <marlowsd@gmail.com>**20090722102219
307 Ignore-this: 5d6c5988e0abab2c5169540aa8ddedb9
308]
309[remove unused $(HscIfaceFileVersion)
310Simon Marlow <marlowsd@gmail.com>**20090720145053
311 Ignore-this: d725cbbde3c68673e2342b370460d87f
312]
313[Choose external names more predictably
314Simon Marlow <marlowsd@gmail.com>**20090720144751
315 Ignore-this: 17513bc93af646108f21bbed1c8f4a3a
316 
317 Now, for a compiler-generated binding "x", if "x" is referred to by
318 the exported "f", then it will be named "f_x" rather than something
319 like "x23".  This means that hopefully
320 
321   - compilation will more often product the same results given the
322     same input (the choice of names is not dependent on the
323     non-deterministic order of bindings within the compiler).
324 
325   - less recompilation will be necessary after making changes
326 
327   - navigating Core might be a bit easier.
328 
329 unfortunately, compilation with -O still does not consistently produce
330 the same ABI.  The simplifier sometimes does different things,
331 apparently.
332 
333 Names will be longer, but I can't see a way around that.
334]
335[Use stable ordering in the dependencies
336Simon Marlow <marlowsd@gmail.com>**20090717123449
337 Ignore-this: e20bac233cf6f834e69c027ff60b5b50
338 Fixes another cause of wobbly interface files and unnecessary recompilation.
339]
340[fall back on libffi for 'foreign import "wrapper"' if necessary
341Simon Marlow <marlowsd@gmail.com>**20090716134549
342 Ignore-this: e1073e1ad77e720326865a6d3c4f3790
343]
344[Take account of GADTs when reporting patterm-match overlap
345simonpj@microsoft.com**20090722050933
346 Ignore-this: 7dcbdcb91021e83e6e6208a2e68c50c9
347 
348 When matching against a GADT, some of the constructors may be impossible.
349 For example
350        data T a where
351           T1 :: T Int
352           T2 :: T Bool
353           T3 :: T a
354 
355         f :: T Int -> Int
356         f T1 = 3
357         f T3 = 4
358 
359 Here, does not have any missing cases, despite omittting T2, because
360 T2 :: T Bool.
361 
362 This patch teaches the overlap checker about GADTs, which happily
363 turned out to be rather easy even though the overlap checker needs
364 a serious rewrite.
365 
366]
367[Fix Trac #3382: desugaring of NPats
368simonpj@microsoft.com**20090720061226
369 Ignore-this: 4dccdaf2b7d6428141dcf174cb455a20
370 
371 Max spotted that the short-cut rules for desugaring NPats (where
372 we compare against a literal) were wrong now that we have overloaded
373 strings.
374 
375]
376[Add a -fwarn-dodgy-exports flag; fixes #1911
377Ian Lynagh <igloo@earth.li>**20090719200124
378 This is used to control warnings that were previously unconditional.
379]
380[Build terminfo if we /aren't/ on Windows, not if we /are/
381Ian Lynagh <igloo@earth.li>**20090719111709]
382[Change how PACKAGES is constructed, so that everything gets cleaned properly
383Ian Lynagh <igloo@earth.li>**20090718210058
384 If Windows wasn't defined properly then the Win32 package wasn't being
385 cleaned, as it wasn't added to PACKAGES. Now we always add everything to
386 PACKAGES when CLEANING=YES.
387]
388[temporarily turn off unused import warnings for the time library
389Ian Lynagh <igloo@earth.li>**20090718183445]
390[Follow the split directory rename in the GHC build system rules
391Ian Lynagh <igloo@earth.li>**20090718155618]
392[Add osuf to the name we use for the split dir
393Ian Lynagh <igloo@earth.li>**20090718145522
394 This avoids a collision between the directories we use when compiling
395 multiple ways, which in turn leads to a race condition in parallel
396 builds.
397]
398[Temporarily turn off unused-do-bind warnings for the time package
399Ian Lynagh <igloo@earth.li>**20090718134536]
400[Make ghc-cabal handle "Custom" Setup.hs files that have a configure script
401Ian Lynagh <igloo@earth.li>**20090718131555]
402[Add the time library, and support for libraries in tarballs
403Ian Lynagh <igloo@earth.li>**20090718121649]
404[Always serialise Int as 64bit values; fixes trac #3041
405Ian Lynagh <igloo@earth.li>**20090717224203
406 This means that, provided the values are small enough, files
407 serialized are portable between architectures. In particular,
408 .haddock files are portable.
409]
410[Remove some code that has always been commented out
411Ian Lynagh <igloo@earth.li>**20090717224100]
412[Fix Trac #3346: tcSimplify for LHS of RULES with type equalities
413simonpj@microsoft.com**20090717155722
414 Ignore-this: dfdd0f9a62d78d63276a4d558831099c
415]
416[Allow mixed case in the LINE pragma; patch from squadette; fixes #1817
417Ian Lynagh <igloo@earth.li>**20090717133522]
418[Comment only
419simonpj@microsoft.com**20090717120154
420 Ignore-this: f96b11e602fe4b311c1e466af9aa1908
421]
422[Add missing case for eq_note.
423t-peterj@microsoft.com**20090624134407]
424[Rename parameters to make debugging code compile.
425t-peterj@microsoft.com**20090626105440]
426[Comment fix: use the same variable names in the conclusion as in the premise.
427t-peterj@microsoft.com**20090618092235]
428[Typo fixes, from Alexey Mahotkin
429Ian Lynagh <igloo@earth.li>**20090717010817]
430[Use names like '$fOrdInt' for dfuns (and TF instances), rather than '$f21'
431Simon Marlow <marlowsd@gmail.com>**20090716125643
432 Ignore-this: d0b4632cf8ed9e05b67a19aa19ab3e19
433 
434 2 reasons for this:
435   - compilation is more predictable.  Adding or removing an instance
436     is less likely to force unnecessary recompilation due to
437     renumbering other dfun names.
438   - it makes it easier to read Core / C-- / asm
439 
440 The names aren't completely deterministic.  To do that, we'd have to
441 include package and module names, which would make the symbol names
442 long and reduce readability.  So the compromise is that if there's a
443 clash, we disambiguate by adding an integer suffix.  This is fairly
444 unlikely in practice unless you're using overlapping instances.
445 
446 Type family instances are handled in the same way, with the same
447 disambiguation strategy.
448]
449[Use a stable ordering for the export list in the interface
450Simon Marlow <marlowsd@gmail.com>**20090716122601
451 Ignore-this: 847dd7adc8b52e56f28d2478c78c925
452 The export list was ordered according to the whim of FastStrings,
453 which meant that interface fingerprints could change for no good
454 reason, causing spurious recompilation.
455]
456[Don't put all of $CFLAGS into $SRC_CC_OPTS
457Ian Lynagh <igloo@earth.li>**20090716131309
458 Instead, we just put the flags we need in there (e.g. -m64 on OS X 64).
459 This fixes a problem found by Simon M, where we were compiling
460 everything with -g, leading to a bloated RTS.
461]
462[Move showOpt into DynFlags
463Ian Lynagh <igloo@earth.li>**20090716005314]
464[Make the --info values printable with "ghc --print-foo"; trac #3122
465Ian Lynagh <igloo@earth.li>**20090716001718
466 Also, libdir is now part of the --info output, so this subsumes the old
467 --print-libdir flag.
468 The mode parsing was getting rather adhoc, so I've tidied it up a bit
469 in the process.
470]
471[whitespace only
472Simon Marlow <marlowsd@gmail.com>**20090716104217
473 Ignore-this: 38cff291d9ef15c30e3ed685ffc3c9f9
474]
475[refactor: use packageConfigId in place of mkPackageId . package
476Simon Marlow <marlowsd@gmail.com>**20090716104145
477 Ignore-this: f3d73e7bd1b307a67d26585c49f3d89f
478]
479[Fix a flag name in the docs
480Ian Lynagh <igloo@earth.li>**20090714165943]
481[Add the -fno-shared-implib flag
482Ian Lynagh <igloo@earth.li>**20090714165631
483 Patch from
484     Max Bolingbroke <batterseapower@hotmail.com>
485 Rerecorded to avoid conflicts.
486]
487[Derived Foldable instances should use Data.Foldable.foldr
488m.niloc@gmail.com**20090711130647
489 Ignore-this: e3eb841e9535a842a98bb1ae0532c6e8
490]
491[remove Solaris-specific hacks, now unnecessary
492Simon Marlow <marlowsd@gmail.com>**20090713083524
493 Ignore-this: 500077008e463532e0677ee82f5284bb
494]
495[Simplify timestamp restoration
496Matthias Kilian <kili@outback.escape.de>**20090711100244
497 Ignore-this: 7eaede224befa6b5368c91b92366211
498]
499[FIX #3272
500Manuel M T Chakravarty <chak@cse.unsw.edu.au>**20090714054559
501 Ignore-this: 225fe4d82d4eed02e9b1377687661bac
502]
503[Fix warnings
504Roman Leshchinskiy <rl@cse.unsw.edu.au>**20090713092032
505 Ignore-this: 3631b87164fc54d82e3a02875dc08f7d
506]
507[Separate length from data in DPH arrays
508Roman Leshchinskiy <rl@cse.unsw.edu.au>**20090713044212
509 Ignore-this: aa2cc3b5ae43bd2c493ce4b330c883cd
510]
511[Stop using -fno-warn-unused-do-bind when compiling the libraries
512Ian Lynagh <igloo@earth.li>**20090709160422
513 They're now fixed to not generate those warnings
514]
515[Remove maybePrefixMatch, using stripPrefix instead
516Ian Lynagh <igloo@earth.li>**20090709160412
517 We already require GHC 6.8 to build, and that included stripPrefix
518 in Data.List.
519]
520[TFs: FIX #2203 (second half)
521Manuel M T Chakravarty <chak@cse.unsw.edu.au>**20090710064834
522 Ignore-this: 46a46feaa73f74feb08524b9e7547414
523]
524[TFs: Fix should_compile/Simple8
525Manuel M T Chakravarty <chak@cse.unsw.edu.au>**20090710042728
526 Ignore-this: 471ab67e3df1c5245921be5286a45f93
527]
528[workaround new Cygwin bash CRLF behaviour
529Simon Marlow <marlowsd@gmail.com>**20090709132850
530 Ignore-this: 5cfa2cc9d776ebe315c0f6ad7ab56d98
531]
532[Use /usr/bin/test if it exists, and fix test syntax.
533Simon Marlow <marlowsd@gmail.com>**20090709124616
534 Ignore-this: 83a75ba7c3ce2a1d02bddb7bfe414bfe
535 Should fix Solaris build failures
536]
537[Allow mixed case pragmas; #1817. Patch from squadette
538Ian Lynagh <igloo@earth.li>**20090709153737
539 This patch allow you to use "Language CPP", or even "LaNgUaGe CPP",
540 if you wish, as the manual claims you can.
541]
542[don't create inplace/bin/ghc-<version>
543Simon Marlow <marlowsd@gmail.com>**20090706092031
544 Ignore-this: 2584d7bf56e77b27ca5b7b557c152c5e
545]
546[Fix ignored-monadic-result warnings
547Ian Lynagh <igloo@earth.li>**20090707181857]
548[Fix an unused import warning
549Ian Lynagh <igloo@earth.li>**20090707144706]
550[Fix unused import warnings
551Ian Lynagh <igloo@earth.li>**20090707143216]
552[Fix unused import warnings
553Ian Lynagh <igloo@earth.li>**20090707133537]
554[When exporting F(..), all the children of F are also exported
555Ian Lynagh <igloo@earth.li>**20090707133427
556 This fixes the unused imports warning when
557     Foo (F(x,y,z))
558 is imported and
559     Foo (F(..))
560 is exported.
561]
562[Remove unused imports
563Ian Lynagh <igloo@earth.li>**20090707121548]
564[Major patch to fix reporting of unused imports
565simonpj@microsoft.com**20090706112503
566 Ignore-this: 3b5ecdd880474493d73bdbdc0fa0b782
567 
568 This patch, joint work between and Ian and Simon, fixes Trac #1074
569 by reporting unused import declarations much more accuratly than
570 before.  The specification is described at
571 
572 http://hackage.haskell.org/trac/ghc/wiki/Commentary/Compiler/UnusedImports
573 
574 The implementation is both easier to understand than before, and shorter
575 too. 
576 
577 Also fixed are #1148, #2267
578 
579 Also fixed is -ddump-minimal imports, which now works properly, fixing
580 Trac #1792.
581 
582 
583]
584[Trim unused imports detected by new unused-import code
585simonpj@microsoft.com**20090706112201
586 Ignore-this: c6ca46d3a750c1cd1d58ea2c0de9f14f
587]
588[Avoid unnecessary recompilation after ./configure (helps #3228)
589Simon Marlow <marlowsd@gmail.com>**20090707085040
590 Ignore-this: f8b3e7a2a96bc23cd29505ab9c8dbd7d
591 We cache the old versions of files generated by configure, so that if
592 configure touches the file without changing it, we can detect that and
593 restore the timestamp.
594]
595[check for tabs in compiler/ghc.cabal.in (#3344)
596Simon Marlow <marlowsd@gmail.com>**20090707081845
597 Ignore-this: 6073db47eafd52e13e76c58ef738afcf
598]
599[remove tabs
600Simon Marlow <marlowsd@gmail.com>**20090707081823
601 Ignore-this: 3d65831fc019f76cefac03291904842a
602]
603[fix cleaning of libraries (now 'make clean' in libraries/* works again)
604Simon Marlow <marlowsd@gmail.com>**20090703114638
605 Ignore-this: b3af731d50ff5bfbd453f94aa40cb92c
606]
607[FIX #2677
608Manuel M T Chakravarty <chak@cse.unsw.edu.au>**20090707055442
609 Ignore-this: e224dd09d0d1c9ec4f3b46c7accb8d57
610]
611[Update driver/Makefile for the new build system
612Ian Lynagh <igloo@earth.li>**20090705204041]
613[Fix generational GC bug (#3348)
614Simon Marlow <marlowsd@gmail.com>**20090706112227
615 Ignore-this: 5938338efa0ad1550968c664a5a76f31
616]
617[Windows fixes to build system: use the 'find' and 'sort' found by configure
618simonpj@microsoft.com**20090706103413
619 Ignore-this: a96197917f388a637118bafefb427495
620 
621 The build system should use 'find' and 'sort' that are discovered by
622 configure, not the ones in your path.  On Windows the ones in your path
623 might well be the non-Unixy Windows versions.
624 
625 This patch fixes the ones I tripped over. There may be more.
626 
627]
628[Follow Cabal changes
629Ian Lynagh <igloo@earth.li>**20090705180414]
630[Update TODO list
631Ian Lynagh <igloo@earth.li>**20090705165009]
632[Make -fext-core a dynamic flag (it was a static flag)
633Ian Lynagh <igloo@earth.li>**20090705132420]
634[Update a few points about shared libs in other sections
635Duncan Coutts <duncan@well-typed.com>**20090704212212
636 And add links to the new shared libs section.
637]
638[Document -dynload flag. Also add it and -shared to the flags reference.
639Duncan Coutts <duncan@well-typed.com>**20090704212119]
640[Add new section on using shared libs
641Duncan Coutts <duncan@well-typed.com>**20090704212003]
642[Document foreign import prim in the user guide
643Duncan Coutts <duncan@well-typed.com>**20090704180547
644 Basically just stat that it exists and refer to the ghc dev wiki
645 for the details, because we don't really want people using it.
646]
647[For now, use -fno-warn-unused-do-bind when building the libraries
648Ian Lynagh <igloo@earth.li>**20090704210654]
649[Make changes to -fwarn-unused-do-bind and -fwarn-wrong-do-bind suggested by SPJ
650Max Bolingbroke <batterseapower@hotmail.com>**20090702150943
651 Ignore-this: 595368298d2e11623c0bd280ff89d8de
652]
653[Support for -fwarn-unused-do-bind and -fwarn-wrong-do-bind, as per #3263
654Max Bolingbroke <batterseapower@hotmail.com>**20090701200344
655 Ignore-this: 511117ffc10d4b656e530b751559b8b8
656]
657[Improved infrastructure for fast-rebuilding of parts of the tree
658Simon Marlow <marlowsd@gmail.com>**20090703074527
659 Ignore-this: ab348d0988d8bbc28c2b4babbd6bbfb8
660 
661 e.g.
662 
663   cd compiler
664   make FAST=YES stage1/build/HscTypes.o
665 
666 builds just the specified .o file, without rebuilding dependencies,
667 and omitting some of the makefile phases.  FAST=YES works anywhere, to
668 omit depenencies and phases.  'make fast' is shorthand for 'make
669 all FAST=YES'.
670]
671[Fix Trac #3342: missed zonking in TcHsSyn
672simonpj@microsoft.com**20090702124331
673 Ignore-this: 9b97b2142dfc665b503f59df7c55dd17
674 
675 The type in a ViewPat wasn't being zonked.  Easily fixed.
676 
677]
678[Type synonym families may be nullary
679Manuel M T Chakravarty <chak@cse.unsw.edu.au>**20090702084826
680 Ignore-this: bcfe6ed62c901206daf5a5088890bbea
681]
682[New syntax for GADT-style record declarations, and associated refactoring
683simonpj@microsoft.com**20090702094657
684 Ignore-this: bd9817230d3773b3b01fae3d7f04c57d
685 
686 The main purpose of this patch is to fix Trac #3306, by fleshing out the
687 syntax for GADT-style record declraations so that you have a context in
688 the type.  The new form is
689    data T a where
690      MkT :: forall a. Eq a => { x,y :: !a } -> T a
691 See discussion on the Trac ticket.
692 
693 The old form is still allowed, but give a deprecation warning.
694 
695 When we remove the old form we'll also get rid of the one reduce/reduce
696 error in the grammar. Hurrah!
697 
698 While I was at it, I failed as usual to resist the temptation to do lots of
699 refactoring.  The parsing of data/type declarations is now much simpler and
700 more uniform.  Less code, less chance of errors, and more functionality.
701 Took longer than I planned, though.
702 
703 ConDecl has record syntax, but it was not being used consistently, so I
704 pushed that through the compiler.
705 
706]
707[White space only
708simonpj@microsoft.com**20090702094627
709 Ignore-this: 19f654cbf371c8dcc6517fd4934855b4
710]
711[Comments only
712simonpj@microsoft.com**20090702094531
713 Ignore-this: 384fc2729c7c50a1680775a1f9ff89e4
714]
715[Look through Notes when matching
716simonpj@microsoft.com**20090702094444
717 Ignore-this: 7daea81e905ec6061d3e0fd588d7e61b
718]
719[FIX #3197
720Manuel M T Chakravarty <chak@cse.unsw.edu.au>**20090702070905
721 Ignore-this: ebf829f0ae025e82bccdfa4345828ffe
722]
723[Fix #2197 (properly this time)
724Simon Marlow <marlowsd@gmail.com>**20090701122354
725 Ignore-this: 39b6e4b0bcdd8c2f4660f976b7db768d
726 
727 $ ./inplace/bin/ghc-stage2 --interactive
728 GHCi, version 6.11.20090701: http://www.haskell.org/ghc/  :? for help
729 ghc-stage2: GHCi cannot be used when compiled with -prof
730 [1]    32473 exit 1     ./inplace/bin/ghc-stage2 --interactive
731]
732[make GhcProfiled work, and add a "prof" flavour to build.mk
733Simon Marlow <marlowsd@gmail.com>**20090701114211
734 Ignore-this: 386d347e4ad8b6c2bd40a2ba7da31ba6
735 
736 Building a profiled GHC is as simple as adding
737 
738 GhcLibWays += p
739 GhcProfiled = YES
740 
741 to your build.mk and saying 'make'.  Then you have a profiled
742 inplace/bin/ghc-stage2.
743]
744[remove unnecessary $(RM)s
745Simon Marlow <marlowsd@gmail.com>**20090701110609
746 Ignore-this: f326ec8931d0d484a66b67ce1270cc6e
747]
748['make html' in a library builds the Haddock docs
749Simon Marlow <marlowsd@gmail.com>**20090630111137
750 Ignore-this: 781bf10e2d4bca23b7f70c6f0465d120
751]
752[fix GC bug introduced with the C finalizer support
753Simon Marlow <marlowsd@gmail.com>**20090630080834
754 Ignore-this: 3567e3adb5ae4a5dcbce81733487f348
755]
756[Add a configure test for whether or not __mingw_vfprintf exists
757Ian Lynagh <igloo@earth.li>**20090627150501]
758[Fix #3319, and do various tidyups at the same time
759Simon Marlow <marlowsd@gmail.com>**20090626095421
760 Ignore-this: ea54175f6bd49e101d7b33392764f643
761  - converting a THSyn FFI declaration to HsDecl was broken; fixed
762  - pretty-printing of FFI declarations was variously bogus; fixed
763  - there was an unused "library" field in CImport; removed
764]
765[rename cache variable to keep recent autoconfs happy
766Ross Paterson <ross@soi.city.ac.uk>**20090626131410
767 Ignore-this: 187091bbe78f2b14402162acfb98180f
768]
769[TAG 2009-06-25
770Ian Lynagh <igloo@earth.li>**20090625155528]
771Patch bundle hash:
772dc2df567959f85d4b7ae5561f152976b2718c753