Patch that fixes the long standing sdist file permission bug

John D. Ramsdell ramsdell at
Mon Dec 20 19:27:04 CET 2010

Mon Dec 20 13:13:08 EST 2010  ramsdell at
  * sdist perms fix

New patches:

[sdist perms fix
ramsdell at**20101220181308
 Ignore-this: ca0ff71dc7318d12c4f64134c636a08e
] {
hunk ./Distribution/Simple/SrcDist.hs 81
          ( Version(versionBranch) )
 import Distribution.Simple.Utils
          ( createDirectoryIfMissingVerbose, withUTF8FileContents, writeUTF8File
-         , copyFiles, copyFileVerbose
+         , installOrdinaryFile, installOrdinaryFiles
          , findFile, findFileWithExtension, matchFileGlob
          , withTempDirectory, defaultPackageDesc
          , die, warn, notice, setupMessage )
hunk ./Distribution/Simple/SrcDist.hs 344
   createDirectoryIfMissingVerbose verbosity True (takeDirectory targetFile)
   copyFileVerbose verbosity file targetFile
+copyFileVerbose :: Verbosity -> FilePath -> FilePath -> IO ()
+copyFileVerbose = installOrdinaryFile
+copyFiles :: Verbosity -> FilePath -> [(FilePath, FilePath)] -> IO ()
+copyFiles = installOrdinaryFiles
 printPackageProblems :: Verbosity -> PackageDescription -> IO ()
 printPackageProblems verbosity pkg_descr = do
   ioChecks      <- checkPackageFiles pkg_descr "."


[Replace uses of the old try function with the new one
Ian Lynagh <igloo at>**20101218230920
 Ignore-this: e0a9db2f97bc3f90ceb5772944f4e25b
[Replace uses of the old catch function with the new one
Ian Lynagh <igloo at>**20101218213608] 
[Add GADTSyntax to extension list
Ian Lynagh <igloo at>**20101218150259] 
[Allow CABAL_VERSION to be defined when bootstrapping
Ian Lynagh <igloo at>**20101212213359
 Needed for the GHC build system
[Add NondecreasingIndentation to Language.Haskell.Extension
Ian Lynagh <igloo at>**20101124223953
 Ignore-this: 2d072a45d76770cf97553b0e5df9e998
[Export commandParseArgs
Duncan Coutts <duncan at>**20101105140655
 Ignore-this: f452f13b130e886ecd8262c6f24a0c52
[Add some more extensions that GHC knows about
Ian Lynagh <igloo at>**20101026220409
 Ignore-this: 92bbbb269985fe576798cd2cdba4ceca
[Whitespace only
Ian Lynagh <igloo at>**20101026220403
 Ignore-this: 8746b070884de3ed45db486a0a49db3e
[Fix parsing of empty default-language fields
Ian Lynagh <igloo at>**20101026220312
 Ignore-this: 99011d6c9af431e7c8d90c3cee364606
[Add support for general extension en/disabling
Ian Lynagh <igloo at>**20101026145503] 
[Tweak a couple of types
Ian Lynagh <igloo at>**20101026131250] 
[Generalise deprecatedExtensions
Ian Lynagh <igloo at>**20101026131121
 It now works with Extensions rather than KnownExtensions
[Split known-extensions off into their own type, and remove knownExtensions
Ian Lynagh <igloo at>**20101025232624
 The knownExtensions list had already got out of sync with the datatype.
 Now we use an Enum instance to get the list of constructors.
 Also, having the actual extensions in a separate type than UnknownExtension
 and EnableExtension makes it easier to add extension disabling support
[Add a test program to look for ghc extensions that are not registered
Duncan Coutts <duncan at>**20101025163536
 Ignore-this: cf38e2434eba386b83b818a29fd5ac14
 Should live in the ghc testsuite but no harm having it here too.
[Make the .cabal file fully 1.10-compliant
Ian Lynagh <igloo at>**20101020133037
 Ignore-this: 66409062c8d4b698f69aa58a83b85ef1
 Add Default-Language field, and change Extensions to Default-Extensions.
[Require "cabal-version: >= 1.10" as we use the new testsuite syntax
Ian Lynagh <igloo at>**20101020131829
 Ignore-this: 6aecdd77f78234f0359af0509f1ba636
[Fix warnings: Remove duplicate imports
Ian Lynagh <igloo at>**20101020125710
 Ignore-this: dea9d889078c729459e1ab92a8e54c08
[Merge conflicts
Ian Lynagh <igloo at>**20101020112200
 Ignore-this: d96cab403511f914d9d8df1a29ca58ee
[Add final newline
Simon Marlow <marlowsd at>**20101013125337
 Ignore-this: 99e055c7186219369902a63c63c3fd76
 Something in the Windows build was complaining about the lack of a
 final newline.
[Update the pretty printer to the current testsuite interface
Duncan Coutts <duncan at>**20101020120506
 Ignore-this: 480a349e93be8228f81f93632cabe2c7
[Improve error messages about programs not being found
Duncan Coutts <duncan at>**20101019074436
 Ignore-this: 9b4daf9a09179482049146669af59f76
 Make it clear it is a program we are talking about, not a library.
[Add a note about the remaining issue with runhugs scripts
Duncan Coutts <duncan at>**20101018232248
 Ignore-this: 3e83a3238d08cdb3ad9e353d43fbf7fa
[Add package QA checks for the new language and extensions fields
Duncan Coutts <duncan at>**20101018180343
 Ignore-this: a063fd21e86e4e19b8152258c5a0711f
[Check at configure time that languages are supported by the compiler
Duncan Coutts <duncan at>**20101018180300
 Ignore-this: e3988c2eff46129b1f2732cd9647c935
[Add compiler support for using the new languages and extensions fields
Duncan Coutts <duncan at>**20101018180151
 Ignore-this: ec1b61767d492dcd2dfa6a07ace3b982
[Add hugs support for the languages and extensions fields
Duncan Coutts <duncan at>**20101018175647
 Ignore-this: 9b39c392119aebbbad22a64dcf992f88
 Incidentally, should fixe ticket #43.
[Add parsing for the new language and extensions fields
Duncan Coutts <duncan at>**20101018091918
 Ignore-this: c59c8ef96c83f35e293c8f495b89fda6
[Add new language and extensions fields (internal data structures)
Duncan Coutts <duncan at>**20101018090620
 Ignore-this: f49dd9278b0053bc441a37beeb2f6145
 New fields default-language and other-languages for specifying the
 base languages for the package, e.g. Haskell98, Haskell2010
 New fields default-extensions and other-extensions for the language
 extensions. Separate from the old extensions field.
 The separation lets us express the difference between declaring to
 the outside world that a package uses certain languages or extensions
 and whether certain languages or extensions should be applied to
 all modules in the package component.
[Extend the Compiler interface to list the supported languages
Duncan Coutts <duncan at>**20101018082608
 Ignore-this: 8b1ab7cc14f35dd6604c2a85173ad48
[Add a Language type, starting with Haskell98 and Haskell2010
Duncan Coutts <duncan at>**20101018082208
 Ignore-this: de80df5317b5d44900e30c947a86364b
[Merge and tweaks following pretty printer changes
Duncan Coutts <duncan at>**20101018195344
 Ignore-this: 513a2b0ce02fed98f5beba0dd893eb16
[pretty printer fixes (FreeText starting with ., version tags, version range parens, option order).
jnf at**20101011094517
 Ignore-this: 400ac5be014f1529632bd16ffd878a92
[reentered accidentally lost exports
jnf at**20100929105852
 Ignore-this: 2ad4ecfac6feba4c2a59a131c05c8a40
[new cabal pretty printer.
jnf at**20100929103653
 Ignore-this: 43f954ec31373e327f2c29fcbf3f0865
[Fix old doc links
Duncan Coutts <duncan at>**20100727012425
 Ignore-this: 9e9d0f2045adebe7bb38b5d1a7d2d71a
[Add note about ticket #689 about deprecated extensions
Duncan Coutts <duncan at>**20100622151328
 Ignore-this: 56cf0027a4c10f4c07045e2ccdbb819c
[Head version needs cabal-version: >=1.8 due to test stanza
Duncan Coutts <duncan at>**20101017155521
 Ignore-this: 6652a529071ccb2eb1cdeda3451aac30
[Remove unused cpp version definition
Duncan Coutts <duncan at>**20101017155218
 Ignore-this: 8dc3425bd811d60d28c2eaf365ec1e66
[Change the way we handle the version number during bootstrapping
Duncan Coutts <duncan at>**20101016191252
 Ignore-this: e21559acc28b7bd811dc4fe147e645b2
 Means we only need to have the version in one place now. Yay.
[Restore compatability with ghc-6.8 and 6.6 but drop support for ghc-6.4
Duncan Coutts <duncan at>**20101016182714
 Ignore-this: 6aef933e4bfb4a9c47021d17370805ea
[Add the Cabal unit tests as a test-suite stanza
Duncan Coutts <duncan at>**20101016165656
 Ignore-this: b9abbfe867754b8bf5b02919c1f25509
 Still some other tests that need hooking up to this mechanism
[Update the unit tests of the testsuite feature
Duncan Coutts <duncan at>**20101016165615
 Ignore-this: bc08d2a8609943f40e901c42916072c2
[Rename test --human-log to just --log
Duncan Coutts <duncan at>**20101016153441
 Ignore-this: a2ccb95759a67ae2888a2db2d2ba678d
 Still have --machine-log as a separate flag
[Remove the test --append-human-log and --replay features
Duncan Coutts <duncan at>**20101016153235
 Ignore-this: 23d9c6431f929cd4078c54e03928cef2
[Add instance Text TestShowDetails parsing and use it for --show-details flag
Duncan Coutts <duncan at>**20101016153015
 Ignore-this: 1018e874b8acbf51bd7081df1210285
 Rather than Read/Show
[Fix a cabal-version check so it accepts the field missing entirely
Duncan Coutts <duncan at>**20101016151929
 Ignore-this: e30f6fe6e7f54cc5c791a3b6761e93de
[Add testsuites to the allBuildInfo function, used mainly in checks
Duncan Coutts <duncan at>**20101013172229
 Ignore-this: be876b609bf1266f3928f25e94f87703
[Update the message for a package check
Duncan Coutts <duncan at>**20101013172206
 Ignore-this: 875cc7c6a763321f2e3251195c9dfb1e
[Add a few TODOs about package checks
Duncan Coutts <duncan at>**20101013172128
 Ignore-this: 837bda043d740f0b92549f3379d43909
[Check test-suite sections even when they have been disabled at configure time
Duncan Coutts <duncan at>**20101013171851
 Ignore-this: 938113ec297b2e5c511a72c5ac8a86d
[Default to Haskell98 for GHC 7.x
Duncan Coutts <duncan at>**20101013022258
 Ignore-this: c34b946f55e5ec054cc842914f58b5a5
 GHC 7 defaults to Haskell2010, we stick with 98 for the moment.
 We will later introduce a new language field to control this.
[Change the syntax and behaviour of the cabal-version field
Duncan Coutts <duncan at>**20101013014933
 Ignore-this: c4cb33360d623ff312b5c3f2d78f730c
 For historical reasons the cabal-version is specified with a version range,
 to indicate the range of versions of tools that the package will work with.
 We now think it makes more sense to specify the version of the Cabal spec
 that the package follows. Old Cabal versions will not be able to parse simple
 versions in this field. So we initially make the parser allow plain versions
 but then we add a check to warn about using it prior to Cabal-1.12 at which
 point it will be allowed.
 Added a check about using version ranges that are not of the form '>= x.y'.
 Also change behaviour to ignore upper bounds in the given version range.
[Fix duplicate import warnings
Ian Lynagh <igloo at>**20101012131227
 Ignore-this: a9b51a864f95206f4d972f1e7506be55
[Remove the ghc Makefile stuff for the old docbook xml
Duncan Coutts <duncan at>**20101010210026
 Ignore-this: 6c714ba98b60e15e44577b64d2de3e1c
 Not woth keeping it, none of it can sensibly be reused. The GHC
 devs will need to add something new for the markdown user guide.
[Update Makefile for new markdown user guide
Duncan Coutts <duncan at>**20101010205938
 Ignore-this: dcb5bd072c619516e23329f170458d72
[Add a FIXME about parsing cabal files
Duncan Coutts <duncan at>**20101010203421
 Ignore-this: 12804d54e81b5ea5878e52628717bc1d
[Refactor the missing VCS info QA check
Duncan Coutts <duncan at>**20101010192346
 Ignore-this: fa6a50a9bd43cf9069e7cd16e23a2b4e
 Starting with Marc Weber's code and just moving
 it about and making it fit in a bit nicer.
[Remove redundant import
Duncan Coutts <duncan at>**20101010190724
 Ignore-this: a978a6a257a31a57d07d57471f04af49
[Rename LocalBuildInfo extraArgs to extraConfigArgs to avoid name clashes
Duncan Coutts <duncan at>**20101010190036
 Ignore-this: 817061002399a51cbf31da2ba002737b
[Add a bunch of TODOs about the test feature
Duncan Coutts <duncan at>**20101010173245
 Ignore-this: f2acc005fef4996a2c5260356eaf219f
[Change how the test suite interfaces are represented, parsed and used
Duncan Coutts <duncan at>**20101010162526
 Ignore-this: 8773e7756fc23bc04c9e5b2b14031512
 Also, stick to version 0.9 of the "detailed" test interface, since it's
 not quite finalised yet. Misc other minor code cleanups.
[Added test suites to Cabal User Guide
Thomas Tuegel <ttuegel at>**20100811141332
 Ignore-this: 3975acc803fdba809ca1c8abeef21677
 Ticket #215 (Overhaul support for packages' tests).
[Removed "$stdio" from acceptable template variables
Thomas Tuegel <ttuegel at>**20100810201828
 Ignore-this: c50e05e9a73c726567eff0a364f29750
 Ticket #215 (Overhaul support for packages' tests).  In the usage message, the
 template variable "$stdio" was incorrectly listed as being available in the
 "--human-log" and "--machine-log" flags.  The variable has been removed.
[Added --test-option(s) flags
Thomas Tuegel <ttuegel at>**20100809161341
 Ignore-this: cea82267bbc0b16d5f21bfc086285905
 Ticket #215 (Overhaul support for packages' tests).  This patch adds the
 --test-option(s) flags for passing command-line options to test executables.
[Respect verbosity flag when outputting test log
Thomas Tuegel <ttuegel at>**20100809151517
 Ignore-this: 73668e49eeea216c27b5233c7e3fe2cb
 Ticket #215 (Overhaul support for packages' tests).  This patch corrects the
 printing of the human-readable test log to the terminal so the setting of the
 verbosity flag is respected.
[Added --replay option to cabal test
Thomas Tuegel <ttuegel at>**20100803164932
 Ignore-this: c97d70e21d3847aa4d889304a7e94451
 Ticket #215 (Overhaul support for packages' tests).  This patch adds support for
 using machine logs from previous runs to replay tests with the same options.
 When using --replay, Cabal will replay all test suites listed in the machine log
 specified; test suite names specified on the command line are ignored.
[Renamed "library" test suite type to "detailed"
Thomas Tuegel <ttuegel at>**20100803141125
 Ignore-this: 457c1a155020303962ae55b2bcd8415c
 Ticket #215 (Overhaul support for packages' tests).
[Fixed human test log output for failing tests
Thomas Tuegel <ttuegel at>**20100803030246
 Ignore-this: f9d3ef127cdb849762af79f31a0c80c9
 Ticket #215 (Overhaul support for packages' tests).  This patch corrects the bug
 in the human test log output to terminal where the line wrapping code in Cabal
 caused terminal control codes output by the test framework to be placed
 incorrectly.  Line wrapping is no longer performed.
[Displaying human-readable test log when requested
Thomas Tuegel <ttuegel at>**20100730151818
 Ignore-this: 98f0adb8e47a52a59ccb3581982157ed
 Ticket #215 (Overhaul support for packages' tests).  This patch causes Cabal to
 display the contents of the human-readable log file on the terminal when run 
 with --show-details=always or when run with --show-details=failures and a 
 failure is detected.  The structure of the test logging code has been changed 
 for clarity to debug a problem where test executables that write to their stderr 
 channel disrupt terminal output.
[Added debugging output to test summary
Thomas Tuegel <ttuegel at>**20100730134008
 Ignore-this: 826d20639f17ae0650d1c9b8a56b43a7
 Ticket #215 (Overhaul support for packages' tests).
[Using correct name of log file in human-readable log
Thomas Tuegel <ttuegel at>**20100730041720
 Ignore-this: 3c6df44f5d6414ce1fe2b58c50590726
 Ticket #215 (Overhaul support for packages' tests).  Previously, human-readable
 logs listed the name of the temporary file where test suite output is
 initially logged.
[Renamed option '--test-filter' to '--show-details'
Thomas Tuegel <ttuegel at>**20100730023026
 Ignore-this: 8b78eed5ccf9cb7ed6a55b86d886e5cc
 Ticket #215 (Overhaul support for packages' tests).
[Displaying location of test suite log on terminal
Thomas Tuegel <ttuegel at>**20100729141159
 Ignore-this: e98a67180c6ff1511b86e442f9acf3c1
 Ticket #215 (Overhaul support for packages' tests).  It was decided that
 indicating the location of the human-readable test suite log made Cabal's test
 runner easier to use.
[Passing names of test suites to run to test stage
Thomas Tuegel <ttuegel at>**20100726150811
 Ignore-this: bf556a0a06fe26b132f1eb5caec21805
 Ticket #215 (Overhaul support for packages' tests).
[Fixed deprecation warning in Distribution.TestSuite
Thomas Tuegel <ttuegel at>**20100726141448
 Ignore-this: 64cd6a5a936efd6b0ee0f50564440a9d
 Ticket #215 (Overhaul support for packages' tests).  Warning resulted from use
 of Control.OldException.
[Fixed help message for machine-log path template
Thomas Tuegel <ttuegel at>**20100724164652
 Ignore-this: 52205830166a307eedce807e908f7a0a
 Ticket #215 (Overhaul support for packages' tests).  The message previously
 indicated that $test-suite is acceptable in the machine log path template,
 which is not true.
[Catching exceptions when lifting pure tests to IO
Thomas Tuegel <ttuegel at>**20100724134336
 Ignore-this: 89a2265a94ee0082935d236dd64c12d4
 Ticket #215 (Overhaul support for packages' tests).
[Using common function to name path environment variables in PathsModule and Test
Thomas Tuegel <ttuegel at>**20100722145840
 Ignore-this: 5cea1a3e77acb84a162de3d1c85a3fe6
 Ticket #215 (Overhaul support for packages' tests).  The names of environment
 variables used to set package paths at runtime were previously hard-coded into
 the function that generates the paths module.  A function generating the
 variable names is now exported from Distribution.Simple.Build.PathsModule and
 used to set the datadir in Distribution.Simple.Test to prevent breakage if the
 naming scheme changes.
[Added documentation for machine-readable test log types
Thomas Tuegel <ttuegel at>**20100722140017
 Ignore-this: 60f934e11b1c3ee4df9f48677528af47
 Ticket #215 (Overhaul support for packages' tests).
[Improved documentation for Distribution.TestSuite
Thomas Tuegel <ttuegel at>**20100722124239
 Ignore-this: f7b1261270c4815b0691ce33c664908c
 Ticket #215 (Overhaul support for packages' tests).
[Fixed documentation errors in Distribution.Simple.Test
Thomas Tuegel <ttuegel at>**20100721221844
 Ignore-this: b111e727b4a556b17c2a1eb4dfd6971b
 Ticket #215 (Overhaul support for packages' tests).
[Renamed Distribution.TestSuite.optionLookup to lookupOption
Thomas Tuegel <ttuegel at>**20100721170724
 Ignore-this: 3a5e79fd0a14d974e664a74af5cb83d1
 Ticket #215 (Overhaul support for packages' tests.)
[Added console-style test summary information to human-readable logs
Thomas Tuegel <ttuegel at>**20100721163516
 Ignore-this: 494c2220285ef4bda036ecdfd7242adf
 Ticket #215 (Overhaul support for packages' tests).
[Added flag allowing human-readable test logs to be appended instead of overwritten
Thomas Tuegel <ttuegel at>**20100721160421
 Ignore-this: f157a4830f82a4a978c1cfb1931a0258
 Ticket #215 (Overhaul support for packages' tests).
[Added clarifying comments to Distribution.Simple.Test
Thomas Tuegel <ttuegel at>**20100721160417
 Ignore-this: 100000890351a75557a6dfa0f71c419f
 Ticket #215 (Overhaul support for packages' tests).
[Using a separate PathTemplateEnv for machine test logs
Thomas Tuegel <ttuegel at>**20100720204724
 Ignore-this: 4920c6e88b8d3d3d788ba42d6b5268af
 Ticket #215 (Overhaul support for packages' tests).  Previously, the use of a
 dummy test suite name would lead to unintuitive expansions of the PathTemplate
 governing the location of the machine-readable package test log.
[Exporting machine log types from Distribution.Simple.Test
Thomas Tuegel <ttuegel at>**20100720204558
 Ignore-this: 6920621dbabf471cef7d688904b9b1b8
 Ticket #215 (Overhaul support for packages' tests).  These types were exported
 to enable parsing of test logs by external utilities.
[Added 'check' and 'optionLookup' to Distribution.TestSuite
Thomas Tuegel <ttuegel at>**20100720134707
 Ignore-this: 21ef44dc0087ff8333b1722309f8fbdd
 Ticket #215 (Overhaul support for packages' tests).  These functions were added
 for the benefit of test runner authors.
[Saving Options from test runs for reproducability
Thomas Tuegel <ttuegel at>**20100715180003
 Ignore-this: 858c387d83e93e193f7c66df3901f6e3
 Ticket #215 (Overhaul support for packages' tests).  Saving the Options used
 required changes to the TestSuite interface, with the Option values now being
 specified as Strings and not Dynamics.  This was necessary because the lack of
 a Read instance for Dynamic.
[Improvements to test suite logging
Thomas Tuegel <ttuegel at>**20100715020549
 Ignore-this: b47f4358302b283e93ccaff81db123f2
 Ticket #215 (Overhaul support for packages' tests).  This patch includes proper
 support for both machine- and human-readable logs.
[Removed duplicate code for test suite interface version checks
Thomas Tuegel <ttuegel at>**20100713144835
 Ignore-this: e06c264351013480a66c013ca398db4b
 Ticket #215 (Overhaul support for packages' tests).  Duplicate code for checking
 the test suite interface version was replaced with a single function
 'testVersion1 :: Version -> Bool' exported from Distribution.PackageDescription.
[Added QA checks for test suite name conflicts
Thomas Tuegel <ttuegel at>**20100712154401
 Ignore-this: df13e2f16cf4e879f5d5f6538c6e7db8
 Ticket #215 (Overhaul suppport for packages' tests).
[New test suite log format
Thomas Tuegel <ttuegel at>**20100708132650
 Ignore-this: 96a300e6acedd0de63757713fbb1d832
 Ticket #215 (Overhaul support for package's tests).  This patch adds a new test
 log file format based on the TestSuiteLog data structure.  The interface between
 Cabal and the library test suite runner has consequently changed.
[Inheriting existing environment for test suite executables
Thomas Tuegel <ttuegel at>**20100707222244
 Ignore-this: 6f08245c83817a85c7da5a05f810abd6
 Ticket #215 (Overhaul support for packages' tests).  Previously, the test
 runner replaced the environment for test suite executables with one containing
 only the datadir path for package data files.  For test suites invoking other
 programs, it is necessary to preserve the system paths, so the datadir path
 variable is appended to the inherited environment.
[Improvements to library test suite runner, including documentation
Thomas Tuegel <ttuegel at>**20100624181304
 Ignore-this: 45baa7905de5423e91707f52e590bbad
 Ticket #215 (Overhaul support for packages' tests).
[Renamed 'result' and 'getResult' to 'run' and 'runM'
Thomas Tuegel <ttuegel at>**20100623184640
 Ignore-this: 7bb6dd598eaa135fcbf73e82ab0d2ce2
 Ticket #215 (Overhaul support for packages' tests).
[Setting datadir environment variables when running tests
Thomas Tuegel <ttuegel at>**20100623183201
 Ignore-this: 7a9e26c684417871609847f6e4d4883e
 Ticket #215 (Overhaul support for packages' tests).
[Added convenience functions and default instances making export of 'Test' constructors unnecessary
Thomas Tuegel <ttuegel at>**20100623151934
 Ignore-this: 1979265e345e268787b5b6fe49bdfd64
 Ticket #215 (Overhaul support for packages' tests).
[Added support for running the default stub executables Cabal creates for library test suites
Thomas Tuegel <ttuegel at>**20100623151903
 Ignore-this: 2be1bbfb07a7fc0e3a0d2c9e5bdf2252
 Ticket #215 (Overhaul support for packages' tests).
[Removed dependency on extensible-exceptions from detailed test suite interface
Thomas Tuegel <ttuegel at>**20100623150227
 Ignore-this: 50ad3ee8c2dc5f62b48aa84d0318c3e6
 Ticket #215 (Overhaul support for packages' tests).
[Added support for building detailed library test suites
Thomas Tuegel <ttuegel at>**20100623150222
 Ignore-this: 1f2a6034af9adf493088265cc8481df5
 Ticket #215 (Overhaul support for packages' tests).  This patch preprocesses
 and builds library test suites.  The fake packages are created for each test
 suite, where the fake package and test suite share the same name; the packages
 and libraries are registered in the inplace package database so that test
 agents can build stub executables against them.
[Improved security of test log file creation
Thomas Tuegel <ttuegel at>**20100621114726
 Ignore-this: 6fed3aa4ebcb587b48bb2a256fcbc61b
 Ticket #215 (Overhaul support for packages' tests).  The algorithm previously
 used to name the log files for test output suffers from a known vulnerability
 due to the predictability of chosen names.
[Added detailed test interface
Thomas Tuegel <ttuegel at>**20100617210631
 Ignore-this: 161624662d6ec7946a33415ddbff4445
 Ticket #215 (Overhaul support for packages' tests).  This patch provides the
 detailed test interface for exposing individual tests to Cabal and other test
 agents.  It also provides the simple function Cabal will provide as the default
 test runner.
[Fixed test suite breakage due to TestSuite API changes
Thomas Tuegel <ttuegel at>**20100629212935
 Ignore-this: 8de228836efb206e1adb833c841ae757
 Ticket #215 (Overhaul support for packages' tests).  The update to the
 TestSuite parser which stopped disallowed configurations during parsing also
 broke the existing test suite with API changes.
[Added --test-filter flag
Thomas Tuegel <ttuegel at>**20100624175917
 Ignore-this: e8fcaddf34a42326d0f3a1081aafb724
 Ticket #215 (Overhaul support for packages' tests).
[Qualified import of Distribution.Simple.Setup in Distribution.Simple.Haddock
Thomas Tuegel <ttuegel at>**20100623193755
 Ignore-this: ec5750f56b22f67e5862036fcdd8ecee
[Using path templates to name test log files
Thomas Tuegel <ttuegel at>**20100622162317
 Ignore-this: af6564bf6154e29e363ee343c9fc5806
 Ticket #215 (Overhaul support for packages' tests).
[More docs about the meaning of the cabal-version field
Duncan Coutts <duncan at>**20101010154251
 Ignore-this: 381ede9227f7a9db78f1007364660648
[Document how Cabal-Version affects behaviour of build-depends
Ben Millwood <haskell at>**20100926025550
 Ignore-this: ba6367db93c15906331457a0468db436
[initial support for building executables with UHC
Atze Dijkstra <atze at>**20100923214130
 Ignore-this: bbbf1adcec2fcfe87ce1db18c804f21a
[Added flags related to UHC, uhcbase package
Atze Dijkstra <atze at>**20100706115341
 Ignore-this: f7dd2b14e3146f8844635ddcb70ac3b9
[Minor changes to the auto-reconfiguration feature.
Duncan Coutts <duncan at>**20101010144111
 Ignore-this: 944f595482ea42eb1907fb1150d6d4c0
 Change the messages slightly. Make configure return the
 new lbi, rather than having to re-read the lbi from file
 (avoiding potential infinite loop if the IO failed).
[Auto-reconfiguration when .cabal is newer than setup-config
Dmitry Astapov <dastapov at>**20100825131106
 Ignore-this: 22ab2b6de0251f6cf1da7c2538544d4
 This patch adds "ConfigFlags" to the "LocalBuildInfo" and reuses them to
 perform "configureAction" when .cabal file is changed. This has
 the same effect as re-running "configure" with the most recent used
 set of options, which should be the sensible thing to do.
 Closes #294, #477, #309 and #518.
[Fix processing of ghc-shared-options
Duncan Coutts <duncan at>**20101009204809
 Ignore-this: 571b3d70fbc705282b9fdfdafdc2f009
 Original patch was:
   Sun Oct  7 13:41:53 BST 2007  Thorkil Naur <naur at>
     * Fix processing of shared options
 Re-recorded due to code churn
[Correct spelling of 'transative'
Duncan Coutts <duncan at>**20101009202836
 Ignore-this: fe7ec5ae621135024403ae0aa42094c2
 Original patch by:
   Thu Aug 21 21:19:51 MDT 2008  dbueno at
     * Correct spelling of 'transative'.
 Re-recorded due to conflict.
[print a warning if repository location isn't specified and the cabal project looks like being tracked by a version control system
marco-oweber at**20091129192013
 Ignore-this: 5ce5073f1793193e437353490eff0276
[Bump Cabal HEAD version to 1.11.0
Duncan Coutts <duncan at>**20101010154518
 Ignore-this: 407e2b1c0de8c10f399841b3fbea1dd3
[TAG 1.10 branch forked
Duncan Coutts <duncan at>**20101010155050
 Ignore-this: 7b0241166f919e2a374a2a69669b2e6b
[Bump Cabal version number to
Ian Lynagh <igloo at>**20100923141731
 Ignore-this: 8b1d551dacebdd3f1b8ebdc4e040825c
[Bump dependencies
Ian Lynagh <igloo at>**20100917183849
 Ignore-this: 4b0979a3f1d16c779abc3c94876ed36c
[add a separator before the pid suffix to the tmpdir template
Jens Petersen <petersen at>**20100823012755
 Ignore-this: 3327cb2d1d1f199efd5c0a9e55f8cc38
 Currently Cabal just appends the pid to the package name-version
 which makes it hard to see the actual package version number.
 This patch just inserts a '-' between name-version and pid.
[Mention self-referential dependency feature in Cabal >=
ezyang at**20100611022749
 Ignore-this: 64289dc45e9b6cd83a549636c062a8a4
[Pass additional parameters in generated shell scripts
Matthias Kilian <kili at>**20100516161230
 Ignore-this: 73d7f4da34a1de1bb90f592e4febe23f
 While here, simplify the argument list construction in
 invocationAsShellScript a little bit.
[update for containers-0.4
Simon Marlow <marlowsd at>**20100903105147
 Ignore-this: eefb1a1b230812397bee0a3cd2aa723
[Add a note explaining the logic of Igloo's previous preprocessing change
Duncan Coutts <duncan at>**20100820120031
 Ignore-this: ae227a082ffa2890ae5e9c0261c464c4
 We previously checked during preprocessing that there is some source
 file for every module and produce a nice error message if not (rather
 than having ghc --make fail later). We only looked for source files in
 the source directories, ie not the target/output build dir, since the
 build dir is where pre-processed files go, not original source files.
 Igloo's change should allow builds where there is no original source file
 at all, so long as there is some generated file in the build dir. In
 particular this allows custom Setup.hs files to generate modules out of
 thin air and for things to mostly work, even though the rest of the build
 system is unaware about how the file got built.
[When looking for already pre-processed source files, check the build directory
Ian Lynagh <igloo at>**20100820001755
 Fixes hscolour'ing Config.hs in GHC
[Type signature for local ioerror
simonpj at**20100816101000
 Ignore-this: d6ec567b79cfbdc3aa86f541b5e9c737
[Add a space to the VERSION_pkg #defines
Ian Lynagh <igloo at>**20100805125306
 We were generating things like
     #define VERSION_ghc_prim""
 rather than
     #define VERSION_ghc_prim ""
[Add two local type signatures
simonpj at microsoft/com**20100728163539
 Ignore-this: 5c89b52c6c906429f9dc6c4c712c5206
 I'm adding these type signatures to satisfy the "do not generalise
 local let/where" rule that GHC is taking on.
 The signatures are clearly correct, but I was surprised at the 
 polymorphism needed.  For example
 parseOptVersion :: ReadP r Version
 parseOptVersion = parseQuoted ver <++ ver
   where ver :: ReadP r Version
         ver = parse <++ return noVersion
         noVersion = Version{ versionBranch=[], versionTags=[] }
 Note that 'ver' really is called at two different types!  That 
 in turn is because of the type of (<++)
   (<++) :: ReadP a a -> ReadP r a -> ReadP r a
   (+++) :: ReadP r a -> ReadP r a -> ReadP r a
 Note the "a a" in the first arg, which is very unusual.
 For example, compare the type of (+++).
 Changing it to match the type of (+++) makes ReadP fail to compile,
 though, so I assume it's right as it stands. But surely this deserves
 a comment?!
[Fix warning
Ian Lynagh <igloo at>**20100620180455
 Ignore-this: b5913b856eacc34001eaf1198aee74df
[Removed extra "test-suite" field in test-suite stanza
Thomas Tuegel <ttuegel at>**20100616150948
 Ignore-this: 4351031606d853474aadb09e393a52b9
 Ticket #215 (Overhaul support for packages' tests).  Previously, the test-suite
 stanza was allowed to contain a test-suite field, such as the executable stanza
 parser allows.  This was removed because it is only required by legacy support
 for executables and no legacy support is required for test suites.
[Added QA check requiring cabal-version: >= 1.9.2 for packages with test-suite stanzas
Thomas Tuegel <ttuegel at>**20100616150359
 Ignore-this: 4d713596feb7aaee53d182f5ed6cb0ec
 Ticket #215 (Overhaul support for packages' tests).
[Setting exit code indicating failure when one or more test suites fails
Thomas Tuegel <ttuegel at>**20100615123631
 Ignore-this: 244d3725fd2627250887df19bd84452a
 Ticket #215 (Overhaul support for packages' tests).
[Added test suite output logging options
Thomas Tuegel <ttuegel at>**20100615123340
 Ignore-this: 28c850b6a2bf69db6aa551e0ee8999b3
 Ticket #215 (Overhaul support for packages' tests).
[--help shows first long option and added --hyperlink-sources
Vo Minh Thu <noteed at>**20100608145444
 Ignore-this: 8eea9aa3fc87c38131844444096c9fab
 The --help option output now prints only the first (if any)
 long option. Because of this, --{enable,disable}-optimisation
 (british spelling) is simply added to the list of options
 without testing showOrParseArgs. --hyperlink-sources is now
 also accepted for the haddock command.
[Do not recognise the "executable" field inside an executable stanza
Duncan Coutts <duncan at>**20100616150146
 Ignore-this: 15df99db5c5ae688fe5d1c67d24377d1
 The executable field is only for the legacy pre-Cabal-1.2 stanza syntax.
[Parsing TestSuite through intermediate data structure
Thomas Tuegel <ttuegel at>**20100609213406
 Ignore-this: 997a00dfc2128d9b7d4c8f8b3c1cd0ea
 Ticket #215 (Overhaul support for packages' tests).  By parsing the test suite
 stanza through an intermediate data structure, we can isolate errors due to
 missing required fields in the parsing stage and avoid having error handlers
 througout the code.
[Added QA checks for executable test suites
Thomas Tuegel <ttuegel at>**20100609180636
 Ignore-this: 52ccd6572f89310180372d1d3eb4393b
 Ticket #215 (Overhaul support for packages' tests).
[Updated test suite for new test suite stanza format
Thomas Tuegel <ttuegel at>**20100608161618
 Ignore-this: 7d1ec575ac5ba02eb5322becd371e4a9
 Ticket #215 (Overhaul support for packages' tests).
[Producing more useful output when running test suites
Thomas Tuegel <ttuegel at>**20100608154847
 Ignore-this: 645ea35c621dc3cd835e055228844c85
 Ticket #215 (Overhaul support for packages' tests).  The previous output format
 made it difficult to distinguish successful test suites from failed test
[Using safe 'runProcess' interface to run executable test suites
Thomas Tuegel <ttuegel at>**20100608152001
 Ignore-this: 38c0697bec7dc428a0a9d63bc0fce6da
 Ticket #215 (Overhaul support for packages' tests).  The use of the old
 interface also makes it possible to log stdout and stderr without separating
 the two.
[New test-suite stanza format with 'main-is' and 'test-module' fields
Thomas Tuegel <ttuegel at>**20100608150739
 Ignore-this: 1baf9ec7c2bcfb637c4064b9b3c6112a
 Ticket #215 (Overhaul support for packages' tests).  Using a new format for the
 test-suite stanza in the .cabal file, similar to the source-repository stanza.
[Consistently using 'test suite' instead of 'testsuite'
Thomas Tuegel <ttuegel at>**20100607205935
 Ignore-this: 242cd1e23afc0ca2f3227421a91dd62e
 Ticket #215 (Overhaul support for packages' tests).
[Fix haddock syntax
Ian Lynagh <igloo at>**20100615112848
 Ignore-this: d3c3b114f6df4eeb4dbf87b2f8f8f77c
[Fix haddock syntax
Ian Lynagh <igloo at>**20100615111829
 Ignore-this: 31e81b5887bc683fe58aa6eefcba40e5
[Use -fno-warn-deprecations in Distribution.Simple
Ian Lynagh <igloo at>**20100615105418
 Ignore-this: 78ce8578ee05e7b2fe9b23ba7449282b
 Works around:
         Warning: In the use of `runTests'
                  (imported from Distribution.Simple.UserHooks):
                  Deprecated: "Please use the new testing interface instead!"
[Merge the dylib-install-name patch
Duncan Coutts <duncan at>**20100612162152
 Ignore-this: 3fe05b2892d4bed2bdf6f23ab110e669
[Make it so cabal passes the dylib-install-name when building dynamic libraries on Mac OS/X.
Stephen Blackheath <oversensitive.pastors.stephen at>**20091001053101
 Ignore-this: 76b7a6a834389eba436e71829510e0ce
[Restored 'runTests' UserHook
Thomas Tuegel <ttuegel at>**20100604175635
 Ignore-this: 128369134d640522c9e5fc3008c121c2
 Ticket #215 (Overhaul support for packages' tests). Deprecated 'runTests'
 UserHook to maintain compatibility with old packages, but encourage authors to
 update to the new interface.
[Logging test output to file
Thomas Tuegel <ttuegel at>**20100603210204
 Ignore-this: 6f2254b3e0c330039992f01957e62a8f
 Ticket #215 (Overhaul support for packages' tests). Log test output to a
 uniquely named file in the system temporary directory. This avoids flooding the
 terminal with long error messages.
[Defined constant for matching test type 'executable == 1'
Thomas Tuegel <ttuegel at>**20100603150106
 Ignore-this: 9eacb8f34bf7f86b704bf1aa3ac94660
 Ticket #215 (Overhaul support for packages' tests).
[Corrected function 'withTest' and updated documentation
Thomas Tuegel <ttuegel at>**20100602203311
 Ignore-this: adb13ce3e294405e1a9706202465feb1
 Ticket #215 (Overhaul support for packages' tests).
[Test command runs all executable testsuites
Thomas Tuegel <ttuegel at>**20100601145827
 Ignore-this: 7e5ccf7c9061e52b82d1647bd3b5e1c0
 Ticket #215 (Overhaul support for packages' tests). The 'test' command runs all
 executable tests listed in the package description when the package is
 configured with tests enabled. The exit codes and standard output/error are
 collected and reported.
[Using more specific error messages for unsupported test types
Thomas Tuegel <ttuegel at>**20100601145430
 Ignore-this: c6fa9827e890fa37cb06c8005c14edb
 Ticket #215 (Overhaul support for packages' tests). Replaced generic error
 message about unsupported test types with specific error messages for each
 stage of the build/test process. This required changing the type of 'withTest'
 to better match 'withExe' and 'withLib'.
[Check testsuite during package sanity checks
Thomas Tuegel <ttuegel at>**20100527204242
 Ignore-this: 1a80a8da0d8d55778605c87abbaa7708
 Ticket #215 (Overhaul support for packages' tests). Check for 
 duplicate testsuite name or modules during 'cabal check'.
[Check for duplicate testsuite names
Thomas Tuegel <ttuegel at>**20100527195438
 Ignore-this: a9425d813d5e396a834c2cb33162090c
 Ticket #215 (Overhaul support for packages' tests). During 
 package configuration, check for testsuites with the same name 
 as other testsuites or executables.
[Conditional inclusion of testsuites
Thomas Tuegel <ttuegel at>**20100526195728
 Ignore-this: a342fbd013e4c1b61f1a8c0a707c7a2f
 Ticket #215 (Overhaul support for packages' tests). The
 "--enable-tests" and "--disable-tests" command-line flags are
 introduced, with "--disable-tests" being the default. If tests
 are disabled, the testsuites are stripped from the
 GenericPackageDescription during the configure stage, before
 dependencies are resolved.
[Testsuite for Test stanza parsing
Thomas Tuegel <ttuegel at>**20100526195509
 Ignore-this: 72405e3311a7b0827decd07251373176
 Ticket #215 (Overhaul support for packages' tests). Parse and
 finalize a simple, dummy .cabal file containing a Test stanza.
 Compare with the PackageDescription it which should result
 from parsing.
[Configure and build executable testsuites
Thomas Tuegel <ttuegel at>**20100526194355
 Ignore-this: ae27f422a5565778c43bd25c553b2952
 Ticket #215 (Overhaul support for packages' test). During the
 build stage, executable testsuites are handled analogously to
 ordinary executables. This means their sources are
 preprocessed and compiled. To actually compile, the build
 stage actions are performed on a dummy Executable.
[Parse Test stanzas
Thomas Tuegel <ttuegel at>**20100526194027
 Ignore-this: aba259fefc52dd6cf95e58624a527b8a
 Ticket #215 (Overhaul support for packages' tests). Test
 stanzas are parsed into the GenericPackageDescription and
 PackageDescription data structures.
[Fix QA check on version range syntax to detect use of ()'s
Duncan Coutts <duncan at>**20100602173703
 Ignore-this: ae60f158b63b458a42086f262b2bc277
 The problem was that we do the QA check on using the new version range
 syntax after parsing. The new syntax allows ()'s but previously the
 code threw them away in the parser stage. We now retain them in the
 AST and deal with them appropriately. This now allows the QA check to
 be accurate and detect things like "build-depends: base (>= 4.2)".
[Fix the wrapping of ghc options for the LHC backend.
Lemmih <lemmih at>**20100602154611
 Ignore-this: 92e42803c0db45403a5ef618fd5d31ef
[change LHC builder to use new command line interface
austin seipp <as at>**20100531173525
 Ignore-this: ffece261e28f302cfbe0d7de51ed03a
[Add QA checks for tested-with field
Duncan Coutts <duncan at>**20100601174045
 Ignore-this: 96fcaa07e52edd94dabfbbe546dfc9f
 Not allowed invalid version ranges. Also check for use of new syntax.
[Bump version to 1.9.2
Duncan Coutts <duncan at>**20100531121711
 Ignore-this: 462e00ecf4497be46209d310de16f192
[Make test suite dependencies less strict
Johan Tibell <johan.tibell at>**20100527203753
 Ignore-this: 509c4e9a5dc2fc33fc39a4c6a748fc89
[Fix warnings in LHC module
Duncan Coutts <duncan at>**20100528004559
 Ignore-this: 4def35179c2509f9cd0d393ada9ffb23
[Add a README explaining how to build and run the test suite
Johan Tibell <johan.tibell at>**20100527212701
 Ignore-this: 13b5cf62708214d5fc3b40fb3662084d
 Fixes ticket #693.
[Make ar create index for in-place libraries when building
Johan Tibell <johan.tibell at>**20100527211104
 Ignore-this: 93e79618b6f015cb069abaadcab96ff
 Should fix ticket #318
[Fix building packages with ghc-6.8 or older
Duncan Coutts <duncan at>**20100521141048
 Ignore-this: 65952598af9ac520c3d2764a8a2e63c
[There's no rts package for LHC.
Lemmih <lemmih at>**20100519192710
 Ignore-this: 65094c079459b6283fc5b00fa1f1aac2
[Fix register --global/--user
Duncan Coutts <duncan at>**20100515213204
 Ignore-this: 86a5e9dc03e47a8dc11595d6eadb5937
 It is a rather silly feature and is not guaranteed to work. Having
 configured and built using one set of dbs, there's no guarantee you
 can register into another set. We should probably just remove it.
[Add some diagnostic info to some internal errors
Ian Lynagh <igloo at>**20100515193810
 When giving an "internal error: unexpected package db stack" error, say
 what the stack is.
[Fix warnings
Ian Lynagh <igloo at>**20100509003309] 
[Allow filepath-1.2.*
Simon Marlow <marlowsd at>**20100505101313
 Ignore-this: c5bf7a356c3347b2d907e8ad7e2d7f3a
[Add the hs output dir to the C include dirs when compiling C code
Duncan Coutts <duncan at>**20100422001101
 Ignore-this: 11ff682fcceb3e0bede08f316c46c654
 This allows C files in a package to include the _stub.h files
 that ghc generates. This is needed for the C code to be able
 to call the C functions exported using the FFI.
[Add the autogen dir to the C include dirs when checking foreign libs
Duncan Coutts <duncan at>**20100422000908
 Ignore-this: f014681ebb019fa310cbf4a87d18637e
 This allows packages to generate .h files and stick them in the
 autogen dir, and still have local .h files include those generated
 header files. Used in gtk2hs.
[Do not pass the CopyFlags to the per-compiler installLib/Exe functions
Duncan Coutts <duncan at>**20100420225830
 Ignore-this: 62277568628a3eed636c06f4df57353e
[Bump version number
Duncan Coutts <duncan at>**20100420224011
 Ignore-this: 135d62045b5e6aedeeb865797b15cf53
[Better error message when installing for unsupported compilers
Duncan Coutts <duncan at>**20100420223935
 Ignore-this: 2f1369d9028e1f7639e73735aa25c62
[Pass the exe/lib to the per-compiler installLib/Exe functions
Duncan Coutts <duncan at>**20100420223652
 Ignore-this: 5ddfb4517bb0f89529256f8ae75d5854
 So the generic code iterates over the libs/exes rather than
 the per-compiler code doing that. It'll help when we have
 more complex inter-component dependencies.
 This is also the proper fix for ticket #525.
[cleaning up and removing warnings in UHC support
Andres Loeh <andres at>**20100330171045
 Ignore-this: 1adb659d170f9cb7478e0110dcc198f7
[significant progress on UHC support
Andres Loeh <andres at>**20100330170211
 Ignore-this: 193f161988c2f03032f3b73c290a8043
[started on UHC support
Andres Loeh <andres at>**20091215125422
 Ignore-this: d3dc77aaf26be3fc1696eaf4a9d8e0f1
[Use Setup.hs when calling setup
Duncan Coutts <duncan at>**20100321111750
 Ignore-this: 117881f91837aa2755549edff49063fe
 In case we've got a compiled Setup binary
[Add the new DoRec extension to Language.Haskell.Extension
Ian Lynagh <igloo at>**20100416205627
 Part of GHC trac #3968.
[Remove BSD4 from the list of known/suggested licenses
Duncan Coutts <duncan at>**20100416040538
 Ignore-this: 8903f1d478b7c034143f700ef0873c75
 We do not ever want to suggest the use of the BSD4 license.
 This will fix a QA check message and the 'cabal init' command.
 This change should be safe to push into the stable 1.8 branch.
[Add a VERSION_<package> define for each package in cabal_macros.h
Mathieu Boespflug <mboes at>**20100329104632
 Ignore-this: a3f8a43f9f15b044f7e9c5a31359717
 The MIN_VERSION_<package> macros are useful to test whether we have at
 least a given version of a dependency, but we can't extract the actual
 version of dependency using this macro. This patch makes the version
 of each dependency available to the code of the cabal package, which
 can be useful when constructing global names in Template Haskell, for
[Also check compiling C headers at configure-time
Duncan Coutts <duncan at>**20100415063657
 Ignore-this: ce1d74257f28099530eaa2cf8da4d42
 So we now both check if the headers can be pre-processed and separately
 if they can be compiled. If it fails the pre-processing check then it
 is usually (but not always) a missing header. If it fails to compile
 then it is definately borked, not missing. Error messages reflect this.
[better error messages for erroneous local C headers
mcallister.keegan at**20100409022543
 Ignore-this: d3840ee01cf2076fe89d14e7ccc2ccac
 Addresses the issue in tickets #532, #648.
 When the all-together test of C libraries and headers fails, use only the
 preprocessor to probe the headers individually.  If they all check out,
 give a more equivocal error message and suggest -v3.
[document all the language extensions
mcallister.keegan at**20100409060559
 Ignore-this: 21de95e9f614235dd2c447297c33fc0f
 Resolves ticket #344.
[Various minor improvements to c2hs support
Duncan Coutts <duncan at>**20100413083252
 Ignore-this: 37c3a2896d2d6229408f8db300817a92
   - ticket #327: require c2hs version 0.15 as minimum
   - ticket #536: pass c2hs include dirs of dependent packages
   - ticket #537: pass c2hs location of gcc -E
[Get ghc, hsc2hs and cpphs programs more robustly in preprocessing
Duncan Coutts <duncan at>**20100412023642
 Ignore-this: 14e2d1dd686dbb3256178729c4af5331
 Avoid pattern match failure in obscure circumstances
 Should fix ticket #383
[Move .hs-boot file pre-processor hack to a more sensible place
Duncan Coutts <duncan at>**20100412015127
 Ignore-this: 3608adbc406e99018862efa996c1ad96
[Also set -stubdir for haddock-stomping workaround
Duncan Coutts <duncan at>**20100411194550
 Ignore-this: f4ca351bb11ff3ffcc1f5e0332dd9db4
 Probably needed in the case of foreign exports and TH.
[Adjust cpp options passed to cpp, c2hs and hsc2hs
Duncan Coutts <duncan at>**20100411194213
 Ignore-this: 7d65b0ddcaac8bdda9954aee5a81fe1d
  * Not changed for cpp, just code refactored.
  * For c2hs add cpp-options and the sysdefines (-D${os}_BUILD_OS etc)
  * For hsc2hs add the sysdefines
[de-haskell98 wash2hs
gwern0 at**20100309015416
 Ignore-this: 3b47d4982f1ebe40d9d78d71a9ab2d4d
[de-haskell98 twoMains
gwern0 at**20100309015328
 Ignore-this: 48f03ca94606814463112fdcc8b9e98f
[Add markdown version of user guide
Duncan Coutts <duncan at>**20100411161012
 Ignore-this: 3bee08cc1631204ade2c7d8086f52dfc
 Plan is to switch over from docbook xml.
[Workaround the fact that haddock stomps on our precious .hi and .o files
Duncan Coutts <duncan at>**20100408225156
 Ignore-this: 301321a9ad31195da3cf78d8a0a72edd
 When using "haddock --optghc-XTemplateHaskell" haddock will write out .o
 and .hi files. This is bad because it replaces the ones we previously
 built. This results in broken packages later on. Of course haddock
 should not do this, it should write temp files elsewhere. The workaround
 is to tell haddock to write the files to a temp dir.
[Remove an out-of-date comment
Ian Lynagh <igloo at>**20100326131752] 
[Factorise duplicate finding in package description checks
Duncan Coutts <duncan at>**20100320215132
 Ignore-this: 9a0e068b05f611f77bee85ca5b943919
[Fix local inplace registration for ghc-6.12
Duncan Coutts <duncan at>**20100320172108
 Ignore-this: 197ec567d3bf300c6ed6430f03a5409d
 This is for the case of intra-package deps where the lib has to be
 registered into a local package db. We use "-inplace" suffix for
 the local installed package ids (rather than using an ABI hash).
[On windows, pick up ar.exe from the ghc install before looking on the $PATH
Duncan Coutts <duncan at>**20100320143643
 Ignore-this: a3931cad3a8fd821a9b1ddd893db8c32
 Some ar.exe versions floating around seem to have weird non-posix behaviour.
[Document the "manual" attribute of flags
Ian Lynagh <igloo at>**20100317203744] 
[Tweak doc Makefile
Ian Lynagh <igloo at>**20100317202418] 
[Fix mismatch between arch and os variables in D.S.InstallDirs interpolation
Andrea Vezzosi <sanzhiyan at>**20100119031642
 Ignore-this: 3a66bd9771f294fc1f52be8824ca051b
[Get the correct value of $topdir on Windows with GHC 6.12.1
Ian Lynagh <igloo at>**20091230204613] 
[Revert the change to filter out deps of non-buildable components
Duncan Coutts <duncan at>**20091229222533
 Ignore-this: 7f6f18997e28b843422d2c55a8bce302
 It does not work as intended and gives inconsistent results between
 cabal install and cabal configure. The problem with the approach was
 that we were filtering out the dependencies of non-buildable components
 at the end. But that does not help much since if one of the deps of
 the non-buildable component were not available then we would have
 failed earlier with a constraint failure. A proper solution would have
 to tackle it from the beginning, not just as a filter at the end.
 The meaning of build-depends and buildable: False needs more thought.
[Silence warning about unused paramater
Duncan Coutts <duncan at>**20091229211649] 
[Remove deprecated --copy-prefix= feature
Duncan Coutts <duncan at>**20091228203513
 It's been deprecated since Cabal 1.1.4 (over 3 years).
 The replacement since then has been --destdir=
[Fix generating Paths_pkgname module with hugs
Duncan Coutts <duncan at>**20091228202618
 In the case that the progsdir is not relative to the $prefix
[Change preprocessModule to preprocessFile
Duncan Coutts <duncan at>**20091228190125
 So we can stop pretending that "main-is: foo.hs" is a module name.
 Also allows us to deprecate ModuleName.simple which bypasses the
 ModuleName type invariant.
[Move registering to per-compiler modules, fix inplace register for hugs
Duncan Coutts <duncan at>**20091228025220] 
[Add a number of TODOs
Duncan Coutts <duncan at>**20091228024948] 
[Fix priority of the two global hugs package dbs
Duncan Coutts <duncan at>**20091228024849] 
[Fix name clashes in hugs module
Duncan Coutts <duncan at>**20091228024814] 
[Add parenthesis to macros in cabal_macros.h
Antoine Latter <aslatter at>**20091229023358
 Ignore-this: 5c5e7244d10bcd82da2cbf4432b30a6d
 Now this like "#if !MIN_VERSION_base(4,2,0)" will work
[Make the datadir follow the $prefix on Windows
Duncan Coutts <duncan at>**20091228165056
 Ignore-this: ce33a328dbf2d1e419cf6e5047bff091
 This is slightly experimental, we'll see how it goes. See ticket #466.
[Simplify getInstalledPackages and callers
Duncan Coutts <duncan at>**20091223234857
 Ignore-this: 7927e992e0b040347dc24530219eb8eb
 No longer returns Maybe, we can find installed packages for all
 the compilers we support (and this feature is a requirement for
 support for new compilers).
 This is an API change so cannot merge to Cabal-1.8.x branch.
[Implement support for hugs and nhc98 package databases
Duncan Coutts <duncan at>**20091223233557
 Ignore-this: aa12e2a278e89544ead82d7c8d3b325a
 That is, work out which packages are installed for hugs and nhc98.
 In both cases there is special support for the core packages.
 In future both should use the standard method when they supply
 proper installed package info files for the bundled libraries.
[Find the version of hugs
Duncan Coutts <duncan at>**20091222175415
 This is really hard and rather nasty.
[Convert XXX comments to TODO or FIXME as appropriate
Duncan Coutts <duncan at>**20091222160247] 
[Fixes for compiling Cabal with hugs
Duncan Coutts <duncan at>**20091223103916
 Ignore-this: f49c719d33e3909996f391e80911c51c
[Make lack of language extensions an error not a warning
Duncan Coutts <duncan at>**20091216204505
 Also improve the error message somewhat
[Specify DOCTYPE when generating userguide html
Duncan Coutts <duncan at>**20091216035321
 Ignore-this: 9d3b09e3c593a8d5f39b691ef2ceb377
 Useless docbook tools.
[Registering packages needs all the package dbs listed
Duncan Coutts <duncan at>**20091211133233
 Ignore-this: 249cc7ae1452bfa8e970f0ba24d4ff5f
 Important for the case of registering inplace when we're using a
 specific package db, e.g. when doing isolated builds.
[Switch a few distribution checks to build warnings
Duncan Coutts <duncan at>**20091202143549
 Ignore-this: bbadf449113d009b51837bc1dd3488af
 In particular the one about -prof since this leads to borked packages.
[Make it so cabal passes the dylib-install-name when building dynamic libraries on Mac OS/X.
Ian Lynagh <igloo at>**20091204144142
 This is a rerecord of
     Stephen Blackheath <oversensitive.pastors.stephen at>**20091001053101
 to avoid conflicts.
[Take line endings into account in IOEncodingUTF8 mode
Duncan Coutts <duncan at>**20091202002553
 Ignore-this: 59a60d9adaf90e94b69336bbb5619d2f
 When collecting the output from programs.
[Install shared libraries as executable files
Ian Lynagh <igloo at>**20091201145349
 Fixes GHC trac #3682. Patch from juhpetersen.
[Fix warnings
Ian Lynagh <igloo at>**20091129164643] 
[Tweak layout to work with alternative layout rule
Ian Lynagh <igloo at>**20091129163614] 
[Bump version in HEAD branch to 1.9.0
Duncan Coutts <duncan at>**20091129161734
 Ignore-this: bbbc9fbb9abdee3f51bdcf972554d93d
 The 1.8.x branch remains at
[Update changelog for 1.8.x release
Duncan Coutts <duncan at>**20091129161613
 Ignore-this: 43b797ce87ba439869c080e0a8e20eab
[Package registration files are always UTF8
Duncan Coutts <duncan at>**20091129153341
 Ignore-this: 7863e05c5514d1fee5165fda3815b116
 As is the output from ghc-pkg dump.
[Add support for text encoding when invoking a program
Duncan Coutts <duncan at>**20091129153110
 Ignore-this: c839875b970d31200d98ead9c8730215
 Can be either locale text or specifically UTF8.
 Also tidy up the rawSystemStd* variants and pass
 a text/binary mode flag for the input and output.
[Change where we add a trailing newline when showing InstalledPackageInfo
Duncan Coutts <duncan at>**20091128171755
 Ignore-this: 5e91f3e0988cf60572eefc374735727f
 Do it in the pretty-printing rather than just before writing the file.
[Update docs on a couple util functions
Duncan Coutts <duncan at>**20091128171114
 Ignore-this: f8df6d6f06dcc2f173180fa063aba091
[Update docs for class Package
Duncan Coutts <duncan at>**20091128170909
 Ignore-this: 2b36ab0d8c422465d489ca2dc7010204
[Be less verbose when checking foreign deps
Duncan Coutts <duncan at>**20091128170810
 Ignore-this: 576c714db6ac0dad7220f9824552c30c
[Add backwards compat version of findProgramOnPath
Duncan Coutts <duncan at>**20091128152444
 Ignore-this: be475bbcbb539d2f2f834a2f0ca235f9
 Break a couple fewer package's Setup.hs files
[Experimentally, change order of user-supplied program args
Duncan Coutts <duncan at>**20091120135619
 Ignore-this: f9b8b4ecc68ee8b2294768c95a2ef774
 Put them last rather than first.
[Fix building with base 3
Duncan Coutts <duncan at>**20091117141958
 Ignore-this: 5ee71207064f5298bb40bafbf537d684
[Canonicalise the package deps returned by finalizePackageDescription
Duncan Coutts <duncan at>**20091109193556
 Ignore-this: eb489503c4a9e27dd8d427cf707461f9
 The guarantee is supposed to be that each package name appears at most
 once with all the constraints for that dependency. The cabal-install
 planner relies on this property.
[Improve the error message for missing sh.exe on Windows
Duncan Coutts <duncan at>**20091109182624
 Ignore-this: 5801ce5598387ad238e61a075285ddf0
 When ettempting to run ./configure scripts. Fixes ticket #403.
[Deprecated PatternSignatures in favor of ScopedTypeVariables
Niklas Broberg <niklas.broberg at>**20090603121321
 Ignore-this: cc1cb044e3db25cb53e1830da34a317b
[JHC support requires jhc >= 0.7.2
Duncan Coutts <duncan at>**20091109113232
 Ignore-this: 97e456357263977187c2768c957b8be7
[Remove some commented out code
Duncan Coutts <duncan at>**20091106133047
 Ignore-this: f4a122d90207de861acc5af603cf55ea
[JHC.getInstalledPackages: remove check for --user flag, since JHC-0.7.2 supports local packages
haskell at**20091029103515] 
[JHC.buildLib: make this working for JHC-0.7.2
haskell at**20091028223004] 
[JHC.getInstalledPackages: adapt to package lists emitted by jhc-0.7.2
haskell at**20091028211802] 
[Bump minor version
Duncan Coutts <duncan at>**20091104180106
 Ignore-this: 6ccbdc1bfa80f66cf74584ff66018582
 A few changes since version 1.8.0 that was released with ghc-6.12rc1
[Tweak the assertion on overall package deps
Duncan Coutts <duncan at>**20091104175912
 Ignore-this: abe2735469828829219938ee4275ebda
[Exclude non-buildable components when constructing the overall package deps
Duncan Coutts <duncan at>**20091104130435
 Ignore-this: 60a11926991ce894da07f3e4c54ed257
[Pass profiling flag to ghc when compiling C files.
Bertram Felgenhauer <int-e at>**20091103165558
 Ignore-this: 422eaeca8c7246170931adecc8556f15
 The main effect of this change is that the PROFILING macro gets defined
 when compiling C files and profiling is enabled. This is useful for code
 that inspects closures.
   The change relies on the fact that Cabal does not track dependencies
   of C files but recompiles them every time. If dependency tracking is
   added, we'll need different extensions for profiling and non-profiling
   object files.
[Build with ghc-6.6
Duncan Coutts <duncan at>**20091028161849
 Ignore-this: 5042dc4f628b68d9d9e40d33837e9818
[Fix haddock markup
Duncan Coutts <duncan at>**20091021124111
 Ignore-this: 7a301d4ede384f2256b02dbf179c0d9b
[Only pass -B... to gcc for GHC < 6.11
Ian Lynagh <igloo at>**20091012144707
 It is no longer needed in 6.12.
[Use -Wl,-R, for the gcc rpath flag.
Duncan Coutts <duncan at>**20091006211326
 Ignore-this: 2323285313bad2ec91e4e27a0ae6177
 Apparently it was only gcc on Solaris that groks -R directly,
 so we have to use -Wl to pass it through to ld. Both GNU and
 Solaris ld grok -R, only GNU ld accepts -rpath.
[Don't complain if there is no ghc rts package registered
Duncan Coutts <duncan at>**20091006172618
 Ignore-this: 8634380cb75891a13988123bbbfd372b
[I was wrong, the test was correct before.
Duncan Coutts <duncan at>**20091006172046
 Ignore-this: e1967271893b0d745d25a9ee09b624cb
 rolling back:
 Mon Oct  5 17:32:02 BST 2009  Stephen Blackheath <grossly.sensitive.stephen at>
   * Fix test case InternalLibrary4 on account of a change in Cabal's behaviour.
     M ./tests/PackageTests/BuildDeps/InternalLibrary4/Check.hs -5 +4
[Unbreak configure for packages that depend on different versions of themselves
Duncan Coutts <duncan at>**20091006171845
 Ignore-this: bce724e750b48ffa7e669c6573b49a8d
[Bump version number to 1.8.0
Ian Lynagh <igloo at>**20091006160120] 
[Loosen the invariant, it was too strict
Simon Marlow <marlowsd at>**20091006133756
 Ignore-this: 6087cfc11c19d4bac2fddcf007e9bb7c
 The packages do not need to be in the same order when sorted by
 package name as when sorted by InstalledPackageId.
[Fix test case InternalLibrary4 on account of a change in Cabal's behaviour.
Stephen Blackheath <grossly.sensitive.stephen at>**20091005163202
 Ignore-this: abc7f45e5fe6eb8e28dee5bfe406b16c
[Keep asserts for testing pre-releases
Duncan Coutts <duncan at>**20091005161425
 Ignore-this: ca9406c17afcba272383eb3f66305a
[Fix configuring for packages with internal lib deps.
Duncan Coutts <duncan at>**20091005161235
 Ignore-this: 31d508508230864d43e3df033dd687d1
 Was using the wrong set of packages when doing the finalise.
[Bump minor version
Duncan Coutts <duncan at>**20091005114150
 Ignore-this: 85fe7d55bac00035cd92c4ffc3f71c2
[Stop converting between installed package id and source package id
Duncan Coutts <duncan at>**20091005111806
 Ignore-this: ed5b14ec4010ab1b8d5cde0ac5ac83e5
 In the LocalBuildInfo, for each component, for the list of component
 dependencies, keep both the InstalledPackageId and the PackageId.
 That way we don't need to keep converting the InstalledPackageId
 to the PackageId, via the package index (which is just horrible).
[Reduce the insanity in Configure
Duncan Coutts <duncan at>**20091005110957
 Ignore-this: 93fa0d351701e6317afb46df94e64777
 It was making so many different kinds of package index and converting
 back and forth between package id and installed package id that it
 made my brain hurt. Thou shalt not convert blithly between package Id
 and installed package Id for thou know not what thy doest.
[Get and merge ghc's package dbs more sensibly
Duncan Coutts <duncan at>**20091005110714
 Ignore-this: fc4b3c983dc7aa0ebc362e990c6a8a2b
 Use the merge rather than just making a big list.
[Rewrite the PackageIndex again
Duncan Coutts <duncan at>**20091005110330
 Ignore-this: ea36efcd3ee2bb5dae40de998b6b4de6
 It's a unified index again, rather than one for looking up by an
 InstalledPackageId and one for the source PackageId. The new one
 lets you look up by either. It also maintains the order of
 preference of different installed packages that share the same
 source PackageId. In configure we just pick the first preference.
[Don't use -#include flag with ghc-6.12
Duncan Coutts <duncan at>**20091004154334
 Ignore-this: b520998e5b5213155de0cba1680e1338
[Fix sdist for packages using .lhs-boot files
Duncan Coutts <duncan at>**20091002145221
 Ignore-this: eaed8d1170bc81b9baa3825f8b662adc
[Fix deadlocks when calling waitForProcess; GHC trac #3542
Ian Lynagh <igloo at>**20090928174553
 We now wait on MVars indicating that stdout and stderr have been
 closed before making the waitForProcess call.
[Update dependencies
Ian Lynagh <igloo at>**20090920152411] 
[Simplify the use of simplifyVersionRange
Duncan Coutts <duncan at>**20090920153433
 Ignore-this: fd41587284e064b185cd1db423ddf493
 Use the simplifyDependency wrapper
[Fix Integer defaulting
Duncan Coutts <duncan at>**20090920153249
 Ignore-this: a747789b708b34cf3589a4c5a9fdb9cd
[Don't simplify empty/inconsistent version ranges
Duncan Coutts <duncan at>**20090920151647
 Ignore-this: aeddcb5625dd29963036c07be5a9c2da
 They all get squashed to ">1 && <1" which while canonical is not helpful.
[Fix pretty printing of version ranges to use parens in the right places
Duncan Coutts <duncan at>**20090917042612
 Ignore-this: 198737aa806fa92774494f6e33344a84
[Fix the depth calculation for the version range expression check
Duncan Coutts <duncan at>**20090917032748
 Ignore-this: da1f0fefd17c1269f39e93e4623dc274
 Previously it was complaining about expressions like
  >= 2 && < 3
 because internally >= is represented in terms of >, == and ||.
 The fix is to use new fold that gives a view with the extra
 syntactic sugar so that <= and <= can be counted as one.
[Modify foldVersionRange and add foldVersionRange'
Duncan Coutts <duncan at>**20090917032713
 Ignore-this: cdb7ff28869dd8637cbc7d1a02620c5b
 Now foldVersionRange gives a view with no syntactic sugar
 while foldVersionRange' gives a view with the syntactic sugar.
[Fix the version-range parser to allow arbitrary expressions over constraints.
Malcolm.Wallace at**20090911111643
 Previously, only a single conjunction (&&) or disjunction (||) was
 parseable, despite an internal representation that permits arbitrary
 combinations.  Now, any sequence of (&&) and (||) combining forms is
 parsed.  (&&) binds more tightly than (||).
[Use -package-id rather than -package with GHC 6.11+
Simon Marlow <marlowsd at>**20090906112416
 Ignore-this: a9cb1e8684411ea3fa04e0d54826f76b
[Install dyn way hi files
Duncan Coutts <duncan at>**20090908142853
 Ignore-this: 9de867381e500fda321a29e137e71414
[Use -R flags for hsc2hs wherever we use -L flags on ELF platforms.
Duncan Coutts <duncan at>**20090828233704
 Ignore-this: 357aa0d1865a1fe6bf37fb6f41e2c93a
 This fixes use of hsc2hs in cases where things like libgmp
 is not on the system's dynamic linker path. It's ok to use
 rpath in the case of hsc2hs because the linked programs are
 run immediately and never installed.
[Add the ABI hash to the InstalledPackageId for inplace registrations too
Simon Marlow <marlowsd at>**20090826155157
 Ignore-this: f10a1e90492a356a8ed8ed78fd056176
 Previously, we just added a -inplace suffix, but this will cause
 problems when developing multiple packages inplace, and then
 installing them.
 Also, there was a round of refactoring: registerPackage now takes the
 InstalledPackageId as an argument, and generateRegistrationInfo is
 exposed for constructing it.  This means that callers of
 registerPackage get to munge the InstalledPackageInfo before it is
[Bump minor version to 1.7.4
Duncan Coutts <duncan at>**20090824010433
 Ignore-this: 360111493090f2cf86a65dfa2bd41da0
 due to recent API changes
[Change finalizePackageDescription to take dep test function
Duncan Coutts <duncan at>**20090822234256
 Ignore-this: 573c4d60e9c40f8814371a6c3a0b8c9
 Ths is instead of taking a set of installed packages. This insulates
 us from the representation of the installed package set which is good
 since we now have several representations. It also opens the door to
 generalising the kinds of constraints we can handle.
[Add a HACKING file
Duncan Coutts <duncan at>**20090822151636
 Ignore-this: 222a83edbe3bf27574bfe51df7f109f0
 Seems people do not know that the source guide exists.
[Add a bunch of TODOs and minor doc changes
Duncan Coutts <duncan at>**20090822134721
 Most TODOs related to the new InstalledPackageId stuff.
[Refactor and simplify the fixup in "hc-pkg dump"
Duncan Coutts <duncan at>**20090822134408
 We have to set the installedPackageId if it's missing.
 There's no need for this to depend on the version of ghc which
 is nice since this module is not supposed to be ghc specific.
[Rename the InstalledPackageInfo field package to sourcePackageId
Duncan Coutts <duncan at>**20090822134111
 The old one was always too generic and it now contrasts
 better with the new installedPackageId field.
[Add a unuque identifier for installed packages (part 9 of 9)
Simon Marlow <marlowsd at>**20090806132217
 Ignore-this: 942731e2e26cfad6c53e728b911f1912
 When registering, choose the InstalledPackageId.
  - When registering inplace, use "foo-1.0-inplace"
  - If this isn't GHC, just use "foo-1.0-installed"
  - When installing a package with GHC, call
    Distribution.Simple.GHC.libAbiHash to get the hash, and
    use "foo-1.0-<hash>".
[Add a unuque identifier for installed packages (part 7 of 9)
Simon Marlow <marlowsd at>**20090806131728
 Ignore-this: cf0e7da3e1e8e2b39336649c479c0938
 Follow changes in Distribution.Simple.LocalBuildInfo (installedPkgs is
 now an InstalledPackageIndex).
[Add a unuque identifier for installed packages (part 8 of 9)
Simon Marlow <marlowsd at>**20090806131700
 Ignore-this: 7cc5db4eb24ced8f3e8770fb8c19650f
 Distribution.Simple.Configure: follow changes to PackageIndex and
[Add a unuque identifier for installed packages (part 6 of 9)
Simon Marlow <marlowsd at>**20090806132002
 Ignore-this: 80e560a4b659edd2ec9345aa57af862a
 Add libAbiHash, which extracts a String representing a hash of
 the ABI of a built library.  It can fail if the library has not
 yet been built.
[Add a unuque identifier for installed packages (part 5 of 9)
Simon Marlow <marlowsd at>**20090806131748
 Ignore-this: 9e242223ca16314148bf92616c19838b
 Follow changes to Distribution.Simple.LocalBuildInfo.componentPackageDeps.
[Add a unuque identifier for installed packages (part 4 of 9)
Simon Marlow <marlowsd at>**20090806131829
 Ignore-this: cd1f965c30d3dbd26dd184b3fd126163
   - the LocalBuildInfo record contains an index of the installed
     packages; this has changed from PackageIndex InstalledPackageInfo
     to InstalledPackageIndex.
   - ComponentLocalBuildInfo stores the "external package dependencies"
     of the component, which was 
       componentPackageDeps :: [PackageId]
     and is now
       componentInstalledPackageDeps :: [InstalledPackageId]
   - we now export
       componentPackageDeps :: LocalBuildInfo -> [PackageId]
     (since to get the PackageId for an InstalledPackageId, you need
     to look it up in the InstalledPackageIndex, which is in the
   - similarly, previously
       externalPackageDeps :: LocalBuildInfo -> [PackageId]
     is now
       externalPackageDeps :: LocalBuildInfo -> [InstalledPackageId]
[Add a unuque identifier for installed packages (part 3 of 9)
Simon Marlow <marlowsd at>**20090806131810
 Ignore-this: 455a736ea5a3241aa6040f4c684ab0b3
 This part adds the InstalledPackageIndex type to
 Distribution.Simple.PackageIndex.  Now that packages have a unique
 identifier within a package database, it makes sense to use this as
 the key for looking up installed packages, so InstalledPackageIndex is
 a mapping from InstalledPackageId to InstalledPackageInfo.
 Distribution.Simple.PackageIndex still supports other kinds of package
 mappings: PackageIndex is a mapping from PackageName.
 All the functions in the section "Special Queries" now work on
 InstalledPackageIndex rather than PackageFixedDeps pkg => PackageIndex
[Add a unuque identifier for installed packages (part 2 of 9)
Simon Marlow <marlowsd at>**20090806131906
 Ignore-this: f3fba4465373bd4b7397384f08ca189e
 Note: this patch doesn't build on its own, you also need the rest of
 the patch series.
 Compatibility with older GHCs.  When reading a package database
 created by an older version of GHC without installedPackageIds, we
 have to fake an InstalledPackageId for the internal
 So, when reading in an InstalledPackageInfo from an older GHC, we set
 the installedPackageId to be the textual representation of the
 PackageIdentifier: i.e. <package>-<version>.  Now, previously the
 depends field of InstalledPackageInfo was [PackageIdentifier], and is
 now [InstalledPackageId], so we will read each PackageIdentifier as an
 InstalledPackageId (a String).  The dependencies will still point to
 the correct package, however, because we have chosen the
 installedPackageId to be the textual representation of the
[Add a unuque identifier for installed packages (part 1 of 9)
Simon Marlow <marlowsd at>**20090806131928
 Ignore-this: b774e5719e666baee504e1f52381cc8b
   NOTE: the patch has been split into 9 pieces for easy
   reviewing, the individual pieces won't build on their own.
 Part 1 does the following:
   - define the InstalledPackageId type as a newtype of String
   - add an installedPackageId field to InstalledPackageInfo
   - change the type of the depends field from [PackageIdentifier]
     to [InstalledPackageId]
 The idea behind this change is to add a way to uniquely identify
 installed packages, letting us decouple the identity of an installed
 package instance from its package name and version.  The benefits of
 this are
   - We get to detect when a package is broken because its
     dependencies have been recompiled, or because it is being
     used with a different package than it was compiled against.
   - We have the possibility of having multiple instances of a
     given <package>-<version> installed at the same time.  In the
     future this might be used for "ways".  It might also be
     useful during the process of upgrading/recompiling packages.
[Refactoring: fit into 80 columns
Simon Marlow <marlowsd at>**20090806115825
 Ignore-this: e4e9bdea632814842121fcf7c7d6c3a
[Use a simpler method for the check on the base-upper-bound
Duncan Coutts <duncan at>**20090812133313
 Ignore-this: 9234cc76670a135906c8b7d2e19e6199
 The key point is that we do not complain for packages
 that do not depend on base at all (like ghc-prim).
[Fix the base-upper-bound Cabal check error
Ian Lynagh <igloo at>**20090811204455] 
[Fix "unused-do-bind" warnings properly
Duncan Coutts <duncan at>**20090727161125
 Though I'm not at all sure I like the _ <- bind idiom.
[Undo incorrect fixes for "unused-do-bind" warnings.
Duncan Coutts <duncan at>**20090727160907
 We capture (and discard) the program output because we do not
 want it printed to the console. We do not currently have a
 specific variant for redirecting the output to /dev/null so
 we simply use the variant that captures the output.
 rolling back:
 Fri Jul 10 22:04:45 BST 2009  Ian Lynagh <igloo at>
   * Don't ask for the output of running ld, as we ignore it anyway
     M ./Distribution/Simple/GHC.hs -2 +2
     M ./Distribution/Simple/LHC.hs -2 +2
 Fri Jul 10 22:08:02 BST 2009  Ian Lynagh <igloo at>
   * Don't use the Stdout variant of rawSystemProgramConf to call gcc
   We ignore the output anyway
     M ./Distribution/Simple/Configure.hs -2 +3
[Pass GHC >= 6.11 the -fbuilding-cabal-package flag
Ian Lynagh <igloo at>**20090726181405] 
[Follow the change in GHC's split-objs directory naming
Ian Lynagh <igloo at>**20090723234430] 
[Fix a "warn-unused-do-bind" warning
Ian Lynagh <igloo at>**20090710212059] 
[Don't use the Stdout variant of rawSystemProgramConf to call gcc
Ian Lynagh <igloo at>**20090710210802
 We ignore the output anyway
[Don't ask for the output of running ld, as we ignore it anyway
Ian Lynagh <igloo at>**20090710210445] 
[Fix some "warn-unused-do-bind" warnings where we want to ignore the value
Ian Lynagh <igloo at>**20090710210407] 
[Fix unused import warnings
Ian Lynagh <igloo at>**20090707133559] 
[Remove unused imports
Ian Lynagh <igloo at>**20090707115824] 
[Bump version to 1.7.3 due to recent API changes
Duncan Coutts <duncan at>**20090707095901] 
[Simplify and generalise installDirsTemplateEnv 
Duncan Coutts <duncan at>**20090705205411
 Take a set of templates rather than file paths.
[Rename and export substituteInstallDirTemplates
Duncan Coutts <duncan at>**20090705205257
 This does the mutual substituition of the installation
 directory templates into each other.
[Follow changes in haddock
Ian Lynagh <igloo at>**20090705193610
 The --verbose flag is now called --verbosity
[TAG 2009-06-25
Ian Lynagh <igloo at>**20090625160144] 
[Undo a simplification in the type of absoluteInstallDirs
Duncan Coutts <duncan at>**20090705154155
 Existing Setup scripts use it so we can't change it. Fixes #563.
[Help Cabal find gcc/ld on Windows
Simon Marlow <marlowsd at>**20090626140250
 Ignore-this: bad21fe3047bc6e23165160c88dd53d9
 the layout changed in the new GHC build system
[clean up createTempDirectory, using System.Posix or System.Directory
Simon Marlow <marlowsd at>**20090625105648
 Ignore-this: 732aac57116c308198a8aaa2f67ec475
 rather than low-level System.Posix.Internals operations which are
 about to go away.
[follow change in System.Posix.Internals.c_open
Simon Marlow <marlowsd at>**20090622133654
 Ignore-this: d2c775473d6dfb1dcca40f51834a2d26
[update to work with the new GHC IO library internals (fdToHandle)
Simon Marlow <marlowsd at>**20090612095346
 Ignore-this: 2697bd2b64b3231ab4d9bb13490c124f
[Describe the autoconfUserHooks option more accurately in the user guide
Duncan Coutts <duncan at>**20090614191400] 
[Fix && entity refs in doc xml
Duncan Coutts <duncan at>**20090614191230] 
[documentation update: add a description of the syntax for 'compiler' fields in .cabal files
Brent Yorgey <byorgey at>**20090610194550] 
[use Haskell 98 import syntax
Ross Paterson <ross at>**20090610174619
 Ignore-this: 26774087968e247b41d69350c015bc30
[fix typo of exitcode
Ross Paterson <ross at>**20090610174541
 Ignore-this: e21da0e6178e69694011e5286b382d72
[Put a "%expect 0" directive in the .y file of a test
Ian Lynagh <igloo at>**20090608204035] 
[Rearrange the PathTemplateEnv stuff and export more pieces
Duncan Coutts <duncan at>**20090607224721] 
[Rewrite the Register module
Duncan Coutts <duncan at>**20090607182821
 It was getting increasingly convoluted and incomprehensible.
 Now uses the Program.HcPkg and Program.Scripts modules.
[Simplify OSX ranlib madness
Duncan Coutts <duncan at>**20090607180717] 
[Use new Program.Ld and Program.Ar in GHC module
Duncan Coutts <duncan at>**20090607180534] 
[Use the new HcPkg module in the GHC getInstalledPackages function
Duncan Coutts <duncan at>**20090607180442] 
[Add specialised modules for handling ar and ld
Duncan Coutts <duncan at>**20090607180257] 
[Add improved xargs style function
Duncan Coutts <duncan at>**20090607180214
 More flexible and based on the ProgramInvocation stuff
[Pass verbosity to hc-pkg
Duncan Coutts <duncan at>**20090607180146] 
[Use a better api for registering libs in the internal package db
Duncan Coutts <duncan at>**20090607125436] 
[Add new Program modules
Duncan Coutts <duncan at>**20090607121301] 
[New module for handling calling the hc-pkg program
Duncan Coutts <duncan at>**20090607120650] 
[New module to write program invocations as shell scripts or batch files
Duncan Coutts <duncan at>**20090607120520
 For tasks like registering where we call hc-pkg, this allows us to
 produce a single program invocation and then either run it directly
 or write it out as a script.
[Re-export the program invocation stuff from the Program module
Duncan Coutts <duncan at>**20090607120404] 
[Fix rawSystemStdin util function
Duncan Coutts <duncan at>**20090607120324
 Close the input after pushing it. Return any error message.
[Split the Program module up a bit
Duncan Coutts <duncan at>**20090607101246
 Add an explicit intermediate ProgramInvocation data type.
[Do not pass Maybe LocalBuildInfo to clean hook
Duncan Coutts <duncan at>**20090604203830
 It is a bad idea for clean to do anything different depending
 on whether the package was configured already or not. The
 actual cleaning code did not use the LocalBuildInfo so this
 only changes in the UserHooks interface. No Setup.hs scripts
 actually make of this parameter for the clean hook.
 Part of ticket #133.
[Simplify checkPackageProblems function
Duncan Coutts <duncan at>**20090604203709
 Since we now always have a GenericPackageDescription
[Change UserHooks.confHook to use simply GenericPackageDescription
Duncan Coutts <duncan at>**20090604203400
 Rather than Either GenericPackageDescription PackageDescription
 In principle this is an interface change that could break Setup.hs
 scripts but in practise the few scripts that use confHook just pass
 the arguments through and so are not sensitve to the type change.
[Change UserHooks.readDesc to use GenericPackageDescription
Duncan Coutts <duncan at>**20090604202837
 Also changes Simple.defaultMainNoRead to use GenericPackageDescription.
 This is an API change that in principle could break Setup.hs scripts
 but in practise there are no Setup.hs scripts that use either.
[Pass a verbosity flag to ghc-pkg
Ian Lynagh <igloo at>**20090605143244] 
[When build calls register, pass the verbosity level too
Ian Lynagh <igloo at>**20090605142718] 
[Fix unlit
Ian Lynagh <igloo at>**20090605130801
 The arguments to isPrefixOf were the wrong way round. We want to see if
 the line starts "\\begin{code}", not if the line is a prefix of that string.
[Tweak a comment so that it doesn't confuse haddock
Ian Lynagh <igloo at>**20090605130728] 
[Tweak new build system
Ian Lynagh <igloo at>**20090404204426] 
[GHC new build system fixes
Ian Lynagh <igloo at>**20090329153151] 
[Add for the new GHC build system
Ian Lynagh <igloo at>**20090324211819] 
[Bump version due to recent changes
Duncan Coutts <duncan at>**20090603101833] 
[Ticket #89 final: Regression tests for new dependency behaviour.
rubbernecking.trumpet.stephen at**20090601215651
 Ignore-this: 52e04d50f1d045a14706096413c19a85
[Make message "refers to a library which is defined within the same.." more grammatical
rubbernecking.trumpet.stephen at**20090601214918
 Ignore-this: 3887c33ff39105f3483ca97a7f05f3eb
[Remove a couple unused imports.
Duncan Coutts <duncan at>**20090601192932] 
[Ban upwardly open version ranges in dependencies on base
Duncan Coutts <duncan at>**20090601191629
 Fixes ticket #435. This is an approximation. It will ban most
 but not all cases where a package specifies no upper bound.
 There should be no false positives however, that is cases that
 really are always bounded above that the check flags up.
 Doing a fully precise test needs a little more work.
[Split requireProgram into two different functions
Duncan Coutts <duncan at>**20090601174846
 Now requireProgram doesn't take a version range and does not check
 the program version (indeed it doesn't need to have one). The new
 function requireProgramVersion takes a required program version
 range and returns the program version. Also update callers.
 Also fixes the check that GHC has a version number.
[Ignore a byte order mark (BOM) when reading UTF8 text files
Duncan Coutts <duncan at>**20090531225008
 Yes of course UTF8 text files should not use the BOM but
 notepad.exe does anyway. Fixes ticket #533.
[executables can now depend on a library in the same package.
Duncan Coutts <duncan at>**20090531220720
 Fixes ticket #89. The library gets registered into an inplace
 package db file which is used when building the executables.
 Based partly on an original patch by Stephen Blackheath.
[Always build ar files with indexes
Duncan Coutts <duncan at>**20090531193412
 Since we have to be able to use these inplace we always need
 the index it's not enough to just make the index on installing.
 This particularly affects OSX.
[Make rendering the ghc package db stack more lenient
Duncan Coutts <duncan at>**20090531192545
 Allow the user package db to appear after a specific one.
 No reason not to and makes some things in cabal-install more convenient.
[Simplify version ranges in configure messages and errors
Duncan Coutts <duncan at>**20090531192426
 Part of #369
[Add and export simplifyDependency
Duncan Coutts <duncan at>**20090531192332
 Just uses simplifyVersionRange on the version range in the dep
[Use the PackageDbStack in the local build info and compiler modules
Duncan Coutts <duncan at>**20090531153124
 This lets us pass a whole stack of package databases to the compiler.
 This is more flexible than passing just one and working out what
 other dbs that implies. This also lets us us more than one specific
 package db, which we need for the inplace package db use case.
[Simplify version ranges before printing in configure error message
Duncan Coutts <duncan at>**20090530213922
 Part of ticket #369. Now instead of:
   setup: At least the following dependencies are missing:
   base <3 && <4 && <3 && <3 && <4
 we get:
   setup: At least the following dependencies are missing:
   base <3
[Bump version to 1.7.1 due to recent changes
Duncan Coutts <duncan at>**20090530211320] 
[Minor renaming
Duncan Coutts <duncan at>**20090530202312
 Part of one of Stephen Blackheath's patches
[Improve an internal error message slightly
Duncan Coutts <duncan at>**20090530205540] 
[Detect intra-package build-depends
Duncan Coutts <duncan at>**20090530204447
 Based on an original patch by Stephen Blackheath
 With this change build-depends on a library within the same package
 are detected. Such deps are not full handled yet so for the moment
 they are explicitly banned, however this is another step towards
 actually supporting such dependencies. In particular deps on
 internal libs are resolved to the internal one in preference to any
 existing external version of the same lib.
[Use accurate per-component package deps
Duncan Coutts <duncan at>**20090530202350
 Based on an original patch by Stephen Blackheath
 Previously each component got built using the union of all package
 deps of all components in the entire package. Now we use exactly the
 deps specified for that component. To prevent breaking old packages
 that rely on the sloppy behaviour, package will only get the new
 behaviour if they specify they need at least cabal-version: >= 1.7.1
[Add *LBI variants of withLib and withExe that give corresponding build info
rubbernecking.trumpet.stephen at**20090528113232
 Ignore-this: 6856385f1c210e33c352da4a0b6e876a
[Register XmlSyntax and RegularPatterns as known extensions in Language.Haskell.Extension
Niklas Broberg <d00nibro at>**20090529102848
 Ignore-this: 32aacd8aeef9402a1fdf3966a213db7d
 Concrete XML syntax is used in the Haskell Server Pages extension 
 language, and a description can be found in the paper "Haskell Server 
 Pages through Dynamic Loading" by Niklas Broberg, published in Haskell 
 Workshop '05.
 Regular expression pattern matching is described in the paper "Regular 
 Expression Patterns" by Niklas Broberg, Andreas Farre and Josef 
 Svenningsson, published in ICFP '04.
[Resolve merge conflict with dynlibPref patch
Duncan Coutts <duncan at>**20090528115249
 The dynlibPref patch accidentally was only pushed to ghc's branch.
[Use dynlibdir = libdir for the moment
Duncan Coutts <duncan at>**20090519134115
 It will need more thought about how much control the user needs
 and what the default shared libs management scheme should be.
[Use componentPackageDeps, remove packageDeps, add externalPackageDeps
Duncan Coutts <duncan at>**20090527225016
 So now when building, we actually use per-component set of package deps.
 There's no actual change in behaviour yet as we're still setting each of
 the componentPackageDeps to the union of all the package deps.
[Pass ComponentLocalBuildInfo to the buildLib/Exe
Duncan Coutts <duncan at>**20090527210731
 Not yet used
[Simplify writeInstalledConfig slightly
Duncan Coutts <duncan at>**20090527204755] 
[No need to drop dist/installed-pkg-config after every build
Duncan Coutts <duncan at>**20090527204500
 We generate this file if necessary when registering.
[Make absoluteInstallDirs only take the package id
Duncan Coutts <duncan at>**20090527203112
 It doesn't need the entire PackageDescription
[Rejig calls to per-compiler build functions
Duncan Coutts <duncan at>**20090527195146
 So it's now a bit clearer what is going on in the generic build code
 Also shift info calls up to generic code
[Split nhc and hugs's build action into buildLib and buildExe
Duncan Coutts <duncan at>**20090527194206] 
[Split JHC's build into buildLib and buildExe
Duncan Coutts <duncan at>**20090527192036] 
[Sync LHC module from GHC module
Duncan Coutts <duncan at>**20090527191615] 
[Give withLib and withExe sensible types
Duncan Coutts <duncan at>**20090527185634] 
[Fix types of libModules and exeModules
Duncan Coutts <duncan at>**20090527185108
 Take a Library/Executable rather than a PackageDescription
 Means we're more precise in using it, just passing the info we need.
[Split ghc's build action into buildLib and buildExe
Duncan Coutts <duncan at>**20090527183250] 
[Remove unused ghc-only executable wrapper feature
Duncan Coutts <duncan at>**20090527183245
 Some kind of shell script wrapper feature might be useful,
 but we should design it properly.
[Fixup .cabal file with the removed modules and files
Duncan Coutts <duncan at>**20090527182344] 
[Fix warnings about unused definitions and imports
Duncan Coutts <duncan at>**20090527175253] 
[Remove the makefile generation feature
Duncan Coutts <duncan at>**20090527175002
 It was an ugly hack and ghc no longer uses it.
[Add new ComponentLocalBuildInfo
Duncan Coutts <duncan at>**20090527174418
 We want to have each component have it's own dependencies,
 rather than using the union of deps of the whole package.
[Ticket #89 part 2: Dependency-related test cases and a simple test harness
rubbernecking.trumpet.stephen at**20090526133509
 Ignore-this: 830dd56363c34d8edff65314cd8ccb2
 The purpose of these tests is mostly to pin down some existing behaviour to
 ensure it doesn't get broken by the ticket #89 changes.
[Ticket #89 part 1: add targetBuildDepends field to PackageDescription's target-specific BuildInfos
rubbernecking.trumpet.stephen at**20090526133729
 Ignore-this: 96572adfad12ef64a51dce2f7c5f738
 This provides dependencies specifically for each library and executable target.
 buildDepends is calculated as the union of the individual targetBuildDepends,
 giving a result that's exactly equivalent to the old behaviour.
[LHC: register the external core files.
Lemmih <lemmih at>**20090521021511
 Ignore-this: d4e484d7b8e541c3ec4cb35ba8aba4d0
[Update the support for LHC.
Lemmih <lemmih at>**20090515211659
 Ignore-this: 2884d3eca0596a441e3b3c008e16fd6f
[Print a more helpful message when haddock's ghc version doesn't match
Duncan Coutts <duncan at>**20090422093240
 Eg now says something like:
 cabal: Haddock's internal GHC version must match the configured GHC version.
 The GHC version is 6.8.2 but haddock is using GHC version 6.10.1
[use -D__HADDOCK__ only when preprocessing for haddock < 2
Andrea Vezzosi <sanzhiyan at>**20090302015137
 Ignore-this: d186a5dbebe6d7fdc64e6414493c6271
 haddock-2.x doesn't define any additional macros.
[Make die use an IOError that gets handled at the top level
Duncan Coutts <duncan at>**20090301195143
 Rather than printing the error there and then and throwing an
 exit exception. The top handler now catches IOErrors and
 formats and prints them before throwing an exit exception.
 Fixes ticket #512.
[rewrite of Distribution.Simple.Haddock
Andrea Vezzosi <sanzhiyan at>**20090219153738
 Ignore-this: 5b465b2b0f5ee001caa0cb19355d6fce
 In addition to (hopefully) making clear what's going on
 we now do the additional preprocessing for all the versions of haddock 
 (but not for hscolour) and we run cpp before moving the files.
[Allow --with-ghc to be specified when running Cabal
Ian Lynagh <igloo at>**20090225172249] 
[fix imports for non-GHC
Ross Paterson <ross at>**20090221164939
 Ignore-this: 12756e3863e312352d5f6c69bba63b92
[Fix user guide docs about --disable-library-vanilla
Duncan Coutts <duncan at>**20090219165539
 It is not default. Looks like it was a copy and paste error.
[Specify a temp output file for the header/lib checks
Duncan Coutts <duncan at>**20090218233928
 Otherwise we litter the current dir with a.out and *.o files.
[Final changelog updates for
Duncan Coutts <duncan at>**20090218222106] 
[Use more cc options when checking for header files and libs
Duncan Coutts <duncan at>**20090218110520
 Use -I. to simulate the search path that gets used when we tell ghc
 to -#include something. Also use the include dirs and cc options of
 dependent packages. These two changes fix about 3 packages each.
[Validate the docbook xml before processing.
Duncan Coutts <duncan at>**20090213134136
 Apparently xsltproc does not validate against the dtd.
 This should stop errors creaping back in.
[Make documentation validate
Samuel Bronson <naesten at>**20090212235057] 
[Folly the directions for docbook-xsl
Samuel Bronson <naesten at>**20090213022615
 As it says in
   - Use the base canonical URI in combination with one of the
     pathnames below. For example, for "chunked" HTML, output:
[Fix compat functions for setting file permissions on windows
Duncan Coutts <duncan at>**20090205224415
 Spotted by Dominic Steinitz
[Only print message about ignoring -threaded if its actually present
Duncan Coutts <duncan at>**20090206174707] 
[Don't build ghci lib if we're not making vanilla libs
Duncan Coutts <duncan at>**20090206173914
 As the .o files will not exist.
[Correct docdir -> mandir in InstallDirs
Samuel Bronson <naesten at>**20090203043338] 
[Fix message suggesting the --executables flag
Samuel Bronson <naesten at>**20090201010708] 
[Remove #ifdefery for windows, renameFile now works properly
Duncan Coutts <duncan at>**20090202004450
 It's even atomic on windows so we don't need the workaround.
[Make withTempDirectory create a new secure temp dir
Duncan Coutts <duncan at>**20090201233318
 Rather than taking a specific dir to create.
 Update the one use of the function.
[Add createTempDirectory to Compat.TempFile module
Duncan Coutts <duncan at>**20090201233213
 Also clean up imports
[Improve the error message for missing foreign libs and make it fatal
Duncan Coutts <duncan at>**20090131184813
 The check should now be accurate enough that we can make it an
 error rather than just a warning.
[Use the cc, cpp and ld options when checking foreign headers and libs
Duncan Coutts <duncan at>**20090131184016
 In partiular this is needed for packages that use ./configure
 scripts to write .buildinfo files since they typically do not
 split the cpp/cc/ldoptions into the more specific fields.
[Do the check for foreign libs after running configure
Duncan Coutts <duncan at>**20090131182213
 This lets us pick up build info discovered by the ./configure script
[move imports outside ifdef GHC
Ross Paterson <ross at>**20090130153505] 
[Document most of the new file utility functions
Duncan Coutts <duncan at>**20090130151640] 
[#262 iterative tests for foreign dependencies
Gleb Alexeyev <gleb.alexeev at>**20090130120228
 Optimize for succesful case. First try all libs and includes in one command, 
 proceed with further tests only if the first test fails. The same goes for libs 
 and headers: look for an offending one only when overall test fails.
[Misc minor comment and help message changes
Duncan Coutts <duncan at>**20090129233455] 
[Deprecate smartCopySources and copyDirectoryRecursiveVerbose
Duncan Coutts <duncan at>**20090129233234
 Also use simplified implementation in terms of recently added functions.
[Switch copyFileVerbose to use compat copyFile
Duncan Coutts <duncan at>**20090129233125
 All remaining uses of it do not require copying permissions
[Let the setFileExecutable function work with hugs too
Duncan Coutts <duncan at>**20090129232948] 
[Switch hugs wrapper code to use setFileExecutable
Duncan Coutts <duncan at>**20090129232542
 instead of get/setPermissions which don't really work properly.
[Switch last uses of copyFile to copyFileVerbose
Duncan Coutts <duncan at>**20090129232429] 
[Stop using smartCopySources or copyDirectoryRecursiveVerbose
Duncan Coutts <duncan at>**20090129231656
 Instead if copyDirectoryRecursiveVerbose use installDirectoryContents
 and for smartCopySources use findModuleFiles and installExecutableFiles
 In both cases the point is so that we use functions for installing
 files rather than functions to copy files.
[Use installOrdinaryFile and installExecutableFile in various places
Duncan Coutts <duncan at>**20090129231321
 instead of copyFileVerbose
[Make the Compat.CopyFile module with with old and new ghc
Duncan Coutts <duncan at>**20090129225423] 
[Add a bunch of utility functions for installing files
Duncan Coutts <duncan at>**20090129180243
 We want to separate the functions that do ordinary file copies
 from the functions that install files because in the latter
 case we have to do funky things with file permissions.
[Use setFileExecutable instead of copyPermissions
Duncan Coutts <duncan at>**20090129180130
 This lets us get rid of the Compat.Permissions module
[Export setFileOrdinary and setFileExecutable from Compat.CopyFile
Duncan Coutts <duncan at>**20090129173413] 
[Warn if C dependencies not found (kind of fixes #262)
gleb.alexeev at**20090126185832
 This is just a basic check - generate a sample program and check if it compiles and links with relevant flags. Error messages (warning messages, 
 actually) could use some improvement.
[Pass include directories to LHC
Samuel Bronson <naesten at>**20090127220021] 
[Add Distribution.Compat.CopyFile module
Duncan Coutts <duncan at>**20090128181115
 This is to work around the file permissions problems with the
 standard System.Directory.copyFile function. When installing
 files we do not want to copy permissions or attributes from the
 source files. On unix we want to use specific permissions and
 on windows we want to inherit default permissions. On unix:
 copyOrdinaryFile   sets the permissions to -rw-r--r--
 copyExecutableFile sets the permissions to -rwxr-xr-x
[Remove unused support for installing dynamic exe files
Duncan Coutts <duncan at>**20090128170421
 No idea why this was ever added, they've never been built.
[Check for ghc-options: -threaded in libraries
Duncan Coutts <duncan at>**20090125161226
 It's totally unnecessary and messes up profiling in older ghc versions.
[Filter ghc-options -threaded for libs too
Duncan Coutts <duncan at>**20090125145035] 
[New changelog entries for 1.7.x
Duncan Coutts <duncan at>**20090123175645] 
[Update changelog for
Duncan Coutts <duncan at>**20090123175629] 
[Fix openNewBinaryFile on Windows with ghc-6.6
Duncan Coutts <duncan at>**20090122172100
 fdToHandle calls fdGetMode which does not work with ghc-6.6 on
 windows, the workaround is not to call fdToHandle, but call
 openFd directly. Bug reported by Alistair Bayley, ticket #473.
[filter -threaded when profiling is on
Duncan Coutts <duncan at>**20090122014425
 Fixes #317. Based on a patch by gleb.alexeev at
[Move installDataFiles out of line to match installIncludeFiles
Duncan Coutts <duncan at>**20090122005318] 
[Fix installIncludeFiles to create target directories properly
Duncan Coutts <duncan at>**20090122004836
 Previously for 'install-includes: subdir/blah.h' we would not
 create the subdir in the target location.
[Typo in docs for source-repository
Joachim Breitner <mail at>**20090121220747] 
[Make 'ghc-options: -O0' a warning rather than an error
Duncan Coutts <duncan at>**20090118141949] 
[Improve runE parse error message
Duncan Coutts <duncan at>**20090116133214
 Only really used in parsing config files derived from command line flags.
[The Read instance for License and InstalledPackageInfo is authoritative
Duncan Coutts <duncan at>**20090113234229
 It is ghc's optimised InstalledPackageInfo parser that needs updating.
 rolling back:
 Fri Dec 12 18:36:22 GMT 2008  Ian Lynagh <igloo at>
   * Fix Show/Read for License
   We were ending up with things like
       InstalledPackageInfo {
           license = LGPL Nothing,
   i.e. "LGPL Nothing" rather than "LGPL", which we couldn't then read.
     M ./Distribution/License.hs -2 +14
[Swap the order of global usage messages
Duncan Coutts <duncan at>**20090113191810
 Put the more important one first.
[Enable the global command usage to be set
Duncan Coutts <duncan at>**20090113181303
 extend it rather than overriding it.
 Also rearrange slightly the default global --help output.
[On Windows, if gcc isn't where we expect it then keep looking
Ian Lynagh <igloo at>**20090109153507] 
[Ban ghc-options: --make
Duncan Coutts <duncan at>**20081223170621
 I dunno, some people...
[Update changelog for release
Duncan Coutts <duncan at>**20081211142202] 
[Make the compiler PackageDB stuff more flexible
Duncan Coutts <duncan at>**20081211141649
 We support using multiple package dbs, however the method for
 specifying them is very limited. We specify a single package db
 and that implicitly specifies any other needed dbs. For example
 the user or a specific db require the global db too. We now
 represent that stack explicitly. The user interface still uses
 the single value method and we convert internally.
[Fix Show/Read for License
Ian Lynagh <igloo at>**20081212183622
 We were ending up with things like
     InstalledPackageInfo {
         license = LGPL Nothing,
 i.e. "LGPL Nothing" rather than "LGPL", which we couldn't then read.
[Un-deprecate Distribution.ModuleName.simple for now
Ian Lynagh <igloo at>**20081212164540
 Distribution/Simple/PreProcess.hs uses it, so this causes build failures
 with -Werror.
[Use the first three lhc version digits
Duncan Coutts <duncan at>**20081211224048
 Rather than two, and do it in a simpler way.
[Remove obsolete test code
Duncan Coutts <duncan at>**20081211142054] 
[Update the VersionInterval properties which now all pass
Duncan Coutts <duncan at>**20081210145653] 
[Eliminate NoLowerBound, Versions do have a lower bound of 0.
Duncan Coutts <duncan at>**20081210145433
 This eliminates the duplicate representation of ">= 0" vs "-any"
 and makes VersionIntervals properly canonical.
[Update and extend the Version quickcheck properties
Duncan Coutts <duncan at>**20081210143251
 One property fails. The failure reveals that the VersionInterval type
 is not quite a canonical representation of the VersionRange semantics.
 This is because the lowest Version is [0] and not -infinity, so for
 example the intervals (.., 0] and [0,0] are equivalent.
[Add documentation for VersionRange functions
Duncan Coutts <duncan at>**20081210140632
 With properties.
[Export withinVersion and deprecate betweenVersionsInclusive
Duncan Coutts <duncan at>**20081210140411] 
[Add checking of Version validity to the VersionIntervals invariant
Duncan Coutts <duncan at>**20081210134100
 Version numbers have to be a non-empty sequence of non-negataive ints.
[Fix implementation of withinIntervals
Duncan Coutts <duncan at>**20081210000141] 
[Fix configCompilerAux to consider user-supplied program flags
Duncan Coutts <duncan at>**20081209193320
 This fixes a bug in cabal-install
[Add ModuleName.fromString and deprecate ModuleName.simple
Duncan Coutts <duncan at>**20081209151232
 Also document the functions in the ModuleName module.
[Check for absolute, outside-of-tree and dist/ paths
Duncan Coutts <duncan at>**20081208234312] 
[Export more VersionIntervals operations
Duncan Coutts <duncan at>**20081208222420
 and check internal invariants
[Check for use of cc-options: -O
Duncan Coutts <duncan at>**20081208182047] 
[Fake support for NamedFieldPuns in ghc-6.8
Duncan Coutts <duncan at>**20081208180018
 Implement it in terms of the -XRecordPuns which was accidentally
 added in ghc-6.8 and deprecates in 6.10 in favor of NamedFieldPuns
 So this is for compatability so we can tell package authors always
 to use NamedFieldPuns instead.
[Make getting ghc supported language extensions its own function
Duncan Coutts <duncan at>**20081208175815] 
[Check for use of deprecated extensions
Duncan Coutts <duncan at>**20081208175441] 
[Add a list of deprecated extenstions
Duncan Coutts <duncan at>**20081208175337
 Along with possibly another extension that replaces it.
[Change the checking of new language extensions
Duncan Coutts <duncan at>**20081207202315
 Check for new language extensions added in Cabal-1.2 and also 1.6.
 Simplify the checking of -X ghc flags. Now always suggest using
 the extensions field, as we separately warn about new extenssons.
[Tweak docs for VersionRange and VersionIntervals
Duncan Coutts <duncan at>**20081207184749] 
[Correct and simplify checkVersion
Duncan Coutts <duncan at>**20081205232845] 
[Make users of VersionIntervals use the new view function
Duncan Coutts <duncan at>**20081205232707] 
[Make VersionIntervals an abstract type
Duncan Coutts <duncan at>**20081205232041
 Provide asVersionIntervals as the view function for a VersionRange
 This will let us enforce the internal data invariant
[Slight clarity improvement in compiler language extension handling
Duncan Coutts <duncan at>**20081205210747] 
[Slightly simplify the maintenance burden of adding new language extensions
Duncan Coutts <duncan at>**20081205210543] 
[Distributing a package with no synopsis and no description is inexcusable
Duncan Coutts <duncan at>**20081205160719
 Previously if one or the other or both were missing we only warned.
 Now if neither are given it's an error. We still warn about either
[Add Test.Laws module for checking class laws
Duncan Coutts <duncan at>**20081204144238
 For Functor, Monoid and Traversable.
[Add QC Arbitrary instances for Version and VersionRange
Duncan Coutts <duncan at>**20081204144204] 
[Remove accidentally added bianry file
Duncan Coutts <duncan at>**20081203000824] 
[Fix #396 and add let .Haddock find autogen modules
Andrea Vezzosi <sanzhiyan at>**20081201114853] 
[Add checks for new and unknown licenses
Duncan Coutts <duncan at>**20081202172742] 
[Add MIT and versioned GPL and LGPL licenses
Duncan Coutts <duncan at>**20081202171033
 Since Cabal-1.4 we've been able to parse versioned licenses
 and unknown licenses without the parser falling over.
[Don't nub lists of dependencies
Duncan Coutts <duncan at>**20081202162259
 It's pretty meaningless since it's only a syntactic check.
 The proper thing is to maintain a dependency set or to
 simplify dependencies before printing them.
[Fix the date in the LICENSE file
Duncan Coutts <duncan at>**20081202161457] 
[Fix the version number in the makefile
Duncan Coutts <duncan at>**20081202161441] 
[Use VersionRange abstractly
Duncan Coutts <duncan at>**20081202160321] 
[Do the cabal version check properly.
Duncan Coutts <duncan at>**20081202155410
 Instead of matching on the actual expression ">= x.y" we use the
 sematic view of the version range so we can do it precisely.
 Also use foldVersionRange to simplify a couple functions.
[Drop support for ghc-6.4 era OPTIONS pragmas
Duncan Coutts <duncan at>**20081202154744
 It's still possible to build with ghc-6.4 but you have to pass
 extra flags like "ghc --make -cpp -fffi Setup.hs" We could not
 keep those OPTIONS pragmas and make it warning-free with ghc-6.10.
 See for details.
[Almost make the VersionRange type abstract
Duncan Coutts <duncan at>**20081202154307
 Export constructor functions and deprecate all the real constructors
 We should not be pattern matching on this type because it's just
 syntax. For meaningful questions we should be matching on the
 VersionIntervals type which represents the semantics.
[Change isAnyVersion to be a semantic rather than syntactic test
Duncan Coutts <duncan at>**20081202142123
 Also add simplify and isNoVersion.
[Add VersionIntervals, a view of VersionRange
Duncan Coutts <duncan at>**20081202141040
 as a sequence of non-overlapping intervals. This provides a canonical
 representation for the semantics of a VersionRange. This makes several
 operations easier.
[Fix pretty-printing of version wildcards, was missing leading ==
Duncan Coutts <duncan at>**20081202135949] 
[Add a fold function for the VersionRange
Duncan Coutts <duncan at>**20081202135845
 Use it to simplify the eval / withinRange function
[Improve the error on invalid file globs slightly
Duncan Coutts <duncan at>**20081202135335] 
[Use commaSep everywhere in the Check module
Duncan Coutts <duncan at>**20081202135208] 
[Fix message in the extra-source-files field check
Duncan Coutts <duncan at>**20081202135000] 
[Add checks for file glob syntax
Duncan Coutts <duncan at>**20081202133954
 It requires cabal-version: >= 1.6 to be specified
[Add check for use of "build-depends: foo == 1.*" syntax
Duncan Coutts <duncan at>**20081202131459
 It requires Cabal-1.6 or later.
[Distinguish version wild cards in the VersionRange AST
Duncan Coutts <duncan at>**20081128170513
 Rather than encoding them in existing constructors.
 This will enable us to check that uses of the new syntax
 are flagged in .cabal files with cabal-version: >= 1.6
[Fix comment in LHC module
Duncan Coutts <duncan at>**20081123100710
 Yes, LHC really does use ghc-pkg (with a different package.conf)
[Use the new bug-reports and source-repository info in the .cabal file
Duncan Coutts <duncan at>**20081123100041] 
[Simplify build-depends and base3/4 flags
Duncan Coutts <duncan at>**20081123100003] 
[Simplify default global libdir for LHC
Duncan Coutts <duncan at>**20081123095802
 So it uses libdir=$prefix/lib rather than libdir=/usr/local/lib
[Simplify the compat exceptions stuff
Duncan Coutts <duncan at>**20081123095737] 
[Fix warnings in the LHC module
Duncan Coutts <duncan at>**20081122224011] 
[Distribution/Simple/GHC.hs: remove tabs for whitespace to eliminate warnings in cabal-install
gwern0 at**20081122190011
 Ignore-this: 2fd54090af86e67e25e51ade42992b53
[Warn about use of tabs
Duncan Coutts <duncan at>**20081122154134] 
[Bump Cabal HEAD version to 1.7.x development series
Duncan Coutts <duncan at>**20081122145817
 Support for LHC is the first divergence between 1.7
 and the stable 1.6.x series.
[Update changelog for 1.6.0.x fixes
Duncan Coutts <duncan at>**20081122145758] 
[LHC: Don't use --no-user-package-conf. It doesn't work with ghc-6.8.
Lemmih <lemmih at>**20081122012341
 Ignore-this: 88a837b38cf3e897cc5ed4bb22046cee
[Semi-decent lhc support.
Lemmih <lemmih at>**20081121034138] 
[Make auto-generated *_paths.hs module warning-free.
Thomas Schilling <nominolo at>**20081106142734
 On newer GHCs using {-# OPTIONS_GHC -fffi #-} gives a warning which
 can lead to a compile failure when -Werror is activated.  We therefore
 emit this option if we know that the LANGUAGE pragma is supported 
 (ghc >= 6.6.1).
[Escape ld-options with the -optl prefix when passing them to ghc
Duncan Coutts <duncan at>**20081103151931
 Fixes ticket #389
[Simplify previous pkg-config fix
Duncan Coutts <duncan at>**20081101200309] 
[Fix bug where we'd try to configure an empty set of pkg-config packages
Duncan Coutts <duncan at>**20081101195512
 This happened when the lib used pkg-config but the exe did not.
 It cropped up in hsSqlite3-0.0.5.
[Add GHC 6.10.1's extensions to the list in Language.Haskell.Extension
Ian Lynagh <igloo at>**20081019141408] 
[Ensure that the lib target directory is present when installing
Duncan Coutts <duncan at>**20081017004437
 Variant on a patch from Bryan O'Sullivan
[Release kind is now rc
Duncan Coutts <duncan at>**20081011183201] 
Duncan Coutts <duncan at>**20081011182516] 
Patch bundle hash:

More information about the cabal-devel mailing list