[darcs-devel] buildbot failure in darcs on ertai Linux-2.6.18 i686

Wed Oct 15 18:21:15 UTC 2008

[cc'ed David and Eric, since they are the guys to decide on it]

On Wed, Oct 15, 2008 at 10:22:07AM -0600, zooko wrote:
> >Is it possible to let the buildbot start with a *clean* environment.
> >Or could somebody name the minimum set environment variables needed
> >for the tests? Then we could scrub the environment with env(1).
> 
> Sure.  I'm not familiar with how to use env(1) for this purpose, but  
> you can either change the master.cfg:
> 
> http://buildbot.darcs.net/master.cfg
> 
> To run tests in a different way than this:
> 
>     f.addStep(Test, command=["sh", "-c", make + " test"],  
> description=["testing"], descriptionDone=["tested"], timeout=None)
> 
> Or else change the GNUmakefile to do something different when "make  
> test" is invoked.

I had a look at the shell_harness script and attempted fixed things
there.  This may be the completely wrong place, so feel free to
delete my patches. On the other hand, fixing it in the buildbot
configuration may also be wrong (I even don't have a buildbot
installed and always run the tests right from my shell or from the
OpenBSD ports tree).

There're two patches attached: the first one that repairs the
functionality of overriding the darcs binary used in the tests with
the environment variable DARCS, whereas the second patch just zaps
the complete environment *except* PATH and DARCS.

It would be a good idea to tests the patches on several environments
before someone pushes them (it works for me, but it may break on
other systems).

Ciao,
	Kili
-------------- next part --------------
Wed Oct 15 19:38:05 CEST 2008  Matthias Kilian <kili at outback.escape.de>
  * Fix DARCS override.
  Even if there was a DARCS override in the environment, its path
  never got prepended to PATH. While here, chomp() the output of which
  and dirname to avoid whitespace in the PATH.

New patches:

[Fix DARCS override.
Matthias Kilian <kili at outback.escape.de>**20081015173805
 Even if there was a DARCS override in the environment, its path
 never got prepended to PATH. While here, chomp() the output of which
 and dirname to avoid whitespace in the PATH.
] hunk ./tests/shell_harness 21

 # Put the right darcs first in PATH
 my $darcspath="$ENV{HOME}/..";
+my $actualdarcs;
 if ($ENV{DARCS}) {
   # User has asked for a particular darcs...
hunk ./tests/shell_harness 24
-  my $actualdarcs=`which $ENV{DARCS}`;
-  my $darcspath=`dirname "$actualdarcs"`;
+  chomp($actualdarcs=`which $ENV{DARCS}`);
+  chomp($darcspath=`dirname "$actualdarcs"`);
 }
 $ENV{PATH} = "$darcspath:$ENV{PATH}";

Context:

[add haskeline test to setup.hs.
David Roundy <droundy at darcs.net>**20081014201513
 Ignore-this: 1472b07696d83f70bf8ad8f7b3d5cf1
] 
[finished cleanup of default_boring entries
benjamin.franksen at bessy.de**20081014100840
 Ignore-this: e5ad0567cabe64142bf3876c2d73535e
] 
[clean up setup.hs by using lookForModule instead of catching exceptions.
David Roundy <droundy at darcs.net>**20081014160159
 Ignore-this: 8f5c7bd5be2ca099fda5bd52d35dc429
] 
[simplify pull_many_files.sh.
David Roundy <droundy at darcs.net>**20081014165155
 Ignore-this: 9f7a7cc5b02a9d51810149a8875e987f
] 
[add test for siginfo.h (which shows up on SunOS maybe?)
David Roundy <droundy at darcs.net>**20081014154814
 Ignore-this: 72d32ff1b4b57dbf75eab937be5a0e22
] 
[Use fgrep in printer.sh test (more portable than grep -F)
Eric Kow <eric.kow at gmail.com>**20081014135138] 
[have makefile make hspwd rather than shell_harness.
David Roundy <droundy at darcs.net>**20081013192943
 Ignore-this: 31008f1993c5a2281b229b9c5cdca66c
 I think there was a problem that make -j check could cause ghc --make
 to be run simultaneously.  Now it's run just once, before running any
 tests.
] 
[slightly hokey approach to solving pwd problem.
David Roundy <droundy at darcs.net>**20081013184209
 Ignore-this: 762f0339c99f1d4a529ab031ad353cd0
 I think this may help the Solaris/BSD test portability issue, and it's
 also (I think) a slightly cleaner alternative to the portable_pwd
 hack.
] 
[clean up test-franchise-build.sh a bit more...
David Roundy <droundy at darcs.net>**20081013175711
 Ignore-this: 619fff6a0be4bc8172cd8c39a5591079
] 
[clean up test-franchise-build.sh
David Roundy <droundy at darcs.net>**20081013162935
 Ignore-this: a1bd36f8bb8d3d75519abcc8d8388a99
 Among other infelicities, it wasn't using a clean franchise build
 except under wine.
] 
[add test that timestamps are properly set.
David Roundy <droundy at darcs.net>**20081013152933
 Ignore-this: caf36b9fff70f9a44b0f35078f3a5aee
] 
[on windows, try renameFile before deleting target file.
David Roundy <droundy at darcs.net>**20081010204009
 Ignore-this: 6946caec86d1c7177510baf62e2ed275
 The trouble was that if we were renaming a file to itself, we ended up
 deleting the file we were renaming (oops!).  It's a stupid hack, but
 that's what we're forced to do when running on windows.  (Note that
 it's potentially hard to determine if two paths describe the same
 file.)
] 
[clean up setup.hs configure a bit.
David Roundy <droundy at darcs.net>**20081010154959
 Ignore-this: 10855edc3e6fda315e03b6ea1c64b05e
 This removes redundant output, and warns when fast http libraries
 aren't found (or no http library at all).
] 
[simplify write_default_prefs.
David Roundy <droundy at darcs.net>**20081013121512
 Ignore-this: 4f72699d2406802e771a0ceadc913550
 It's easier to read this way, and shorter, too.
] 
[documented provenance of .tmp_versions and .DS_Store files in default boring
ben.franksen at online.de**20081012220207] 
[Fix curl version in setup.hs.
Dmitry Kurochkin <dmitry.kurochkin at gmail.com>**20081012094214
 Ignore-this: 5c44008cb7105279c37a2297e668a070
] 
[cleanup default boring
ben.franksen at online.de**20081011234148

 Also turn default_boring and default_binaries into (pure) CAFs.
] 
[Remove bashisms and GNUisms
Matthias Kilian <kili at outback.escape.de>**20081012123017

 Replace some bashisms and GNUisms by more standards compliant command
 lines, i.e.:

 - Don't use `\|' in BREs, since it's undefined according to POSIX.
   Instead just use grep -E with an ERE.

 - Don't create "funny" strings with the $'\x01...' bashism but use
   printf(1) instead. Note that this isn't strictly necessary to fix
   printer.sh on non-GNUish systems, but it makes it much simpler to test
   what's going on.

 - In test_line(), just use grep -F, and unescape all second arguments
   passed to test_line(). This makes the script much more readable and
   actually fixes it for non-GNUish systems.
] 
[avoid re-checking for windows.
David Roundy <droundy at darcs.net>**20081011223905
 Ignore-this: 93c42ffc6b67c6cca803e1dcb18a9497
] 
[setup.hs: don't fail if darcs doesn't yet exist.
David Roundy <droundy at darcs.net>**20081011223824
 Ignore-this: 6d318961490a9343f5e792f2d740fd27
] 
[fix spelling error in setup.hs.
David Roundy <droundy at darcs.net>**20081011205353
 Ignore-this: 6d521f633d74aae8705391eff9489b2c
] 
[make setup.hs look for libcurl
David Roundy <droundy at darcs.net>**20081011195923
 Ignore-this: c9a267184e89b188f074171f8887b3b4
] 
[resolve issue1139: make special cases handle no arguments gracefully.
David Roundy <droundy at darcs.net>**20081011161059
 Ignore-this: 12b17a3940558860dd4caf7bb2619ec7
 I accidentally broke this when refactoring Darcs.Match.  This change
 restores some efficiency when no arguments are given and thus we are
 creating copies of all the files in the repository.
] 
[add tests for issue1139.
David Roundy <droundy at darcs.net>**20081011155324
 Ignore-this: ff50784059114f969fb4657ea5e92025
] 
[Copyright and licensing notes and rationales.
Trent W. Buck <trentbuck at gmail.com>**20081011035925] 
[Document DARCS_KEEP_TMPDIR in manual.
Dmitry Kurochkin <dmitry.kurochkin at gmail.com>**20081011115026
 Ignore-this: bd77d5a85fa44094dcd1a867c2a63d2e
] 
[Documentation for DARCS_KEEP_TMPDIR.
Dmitry Kurochkin <dmitry.kurochkin at gmail.com>**20081011113042
 Ignore-this: 5e3d6598aed4b681a099341c3ec5017a
] 
[Simplify Darcs.Lock.withDir.
Dmitry Kurochkin <dmitry.kurochkin at gmail.com>**20081011110639
 Ignore-this: c01257f9e0fc40c415562448cb52d67d
] 
[Do not remove temporary directories if DARCS_KEEP_TMPDIR environment variable is set.
Dmitry Kurochkin <dmitry.kurochkin at gmail.com>**20081011105808
 Ignore-this: 5ea0e206e45dc9be73c29c3bf89643b9
] 
[add new test that optimize --reorder behaves itself on lazy repos.
David Roundy <droundy at darcs.net>**20081010194834
 Ignore-this: 5246d20700c153490da1997caede4369
] 
[give more debug output in test.
David Roundy <droundy at darcs.net>**20081010191209
 Ignore-this: a13ec79ae69f4fe9621ea3c16c9b41d6
] 
[remove reimplementation of zipWith.
David Roundy <droundy at darcs.net>**20081010191051
 Ignore-this: 277f49de513ec84bc6db0b4da6c633b1
] 
[Fix spurious diff and missing cd in diff test.
Eric Kow <eric.kow at gmail.com>**20081008192142

 If tests are run in parallel, each instance of the diff test may
 attempt to create temporary directories for use by darcs diff.
 Darcs itself handles this fine; only the test script gets confused
 by the spurious difference that results.
] 
[Help the test harness clean to up after tests.
Eric Kow <eric.kow at gmail.com>**20081008190340
 A typical use case is if a test sets restrictive permissions
 and dies.  Notes:
 * we do this after each test so that one test result doesn't
   contaminate others
 * we explicitly do /not/ remove these temporary directories
   because you may want them around for forensics
] 
[ratify use of readFile in setup.hs
David Roundy <droundy at darcs.net>**20081010150105
 Ignore-this: 92627010ec07e16d271bcc5e7f325c64
] 
[Minor refactor in URL.urlThread.
Dmitry Kurochkin <dmitry.kurochkin at gmail.com>**20081010071649
 Ignore-this: 809d750f10adc8073ebe0dd3d432f2a2
] 
[clean up code formatting in Hopefully.
David Roundy <droundy at darcs.net>**20081010144943
 Ignore-this: 65a4dbb17505adbc0c0c615c445a8926
] 
[clean up unused exports in Hopefully.
David Roundy <droundy at darcs.net>**20081010144918
 Ignore-this: a016b66fd64f756295555255b268e80a
] 
[haddock documentation for Hopefully
florent.becker at ens-lyon.org**20081001081315] 
[resolve issue1062: ignore setCooked and setEcho errors.
Eric Kow <eric.kow at gmail.com>**20081010122047
 Ignore-this: 2f8c8710faefdcdb6b224244995a00af
 The issue in question is only about setCooked.
 I presume setEcho can similarly fail.
] 
[add script for testing the franchise build.
David Roundy <droundy at darcs.net>**20081010142711
 Ignore-this: 5aa40dc14c0827b199b9d23d1c7ef27d
 This includes testing the cross-compile to windows using wine.  It's
 part of my scripts that are run on each push to darcs-unstable.
] 
[add a --prompt-for-dependencies to restore the default behaviour
florent.becker at ens-lyon.org**20081010123240
 Ignore-this: 92480353c19dea1b67aff89b933759a2
 Add a --prompt-for-dependencies flag so that one can put
 --dont-prompt-for-dependencies in their defaults file and
 still get back the default behavior when needed.
] 
[eliminate autogeneration of Workaround.hs.
David Roundy <droundy at darcs.net>**20081010134727
 Ignore-this: 784a4b08bab1e22c93f54c1c5cd710f1
] 
[add franchise setup file.
David Roundy <droundy at darcs.net>**20081009203410
 Ignore-this: f1738c7608e6093aa18e808d3d41d000
] 
[add missing createLink to win32/System/Posix/Files.hsc.
David Roundy <droundy at darcs.net>**20081010133500
 Ignore-this: 39d246ab1db8c31f78e0b8362ad05ca3
] 
[make make_changelog default to quiet output.
David Roundy <droundy at darcs.net>**20081009203322
 Ignore-this: 55eba7a68a3c8a5683d23cc99848bd6f
 You can restore previous behavior by defining the VERBOSE environment
 variable.
] 
[remove unused and unneeded Workarounds.
David Roundy <droundy at darcs.net>**20081009194541
 Ignore-this: bca5a7a2444d4ddffeed673cbbfa19ec
] 
[rename lazy-deps to dont-prompt-for-dependencies.
David Roundy <droundy at darcs.net>**20081009171030
 Ignore-this: 78393e28c184c9339c89cecaa852dbc9
] 
[fix incompatibilities with ghc 6.6
David Roundy <droundy at darcs.net>**20081009165845
 Ignore-this: 18739c880a49b976ec151363f4ff9296
] 
[basic tests for --lazy-deps
Florent Becker <florent.becker at ens-lyon.org>**20081008124956
 Ignore-this: 79d8a312943813e9fe5448bc2ab86aa0
] 
[added a --lazy-deps option
Florent Becker <florent.becker at ens-lyon.org>**20081008122224
 Ignore-this: d934868e56f0084067f5aedf0512a533
 This option means that you don't get asked about patches which do not match
 a --match, but are depended upon by a patch that does. These patches will get
 ilently selected. 

] 
[refactor of SelectChanges
Florent Becker <florent.becker at ens-lyon.org>**20081008100551
 Ignore-this: e620d36d6e52eb62a2205fe5122d7076
] 
[refactor Match to handle --store-in-memory itself.
David Roundy <droundy at darcs.net>**20081008172748
 Ignore-this: 647c75f78d2d4981e67039ae1b6e96df
 This reduces the amount of duplicate code, at the cost of adding a
 class to Darcs.Match--but at least it isn't exported.
] 
[make writeSlurpy work properly when given "."
David Roundy <droundy at darcs.net>**20081008172207
 Ignore-this: 9d5d077bcdf3a9bb7bfa8477ab9d96d5
 This isn't a bugfix per se, but rather a feature addition.
 writeSlurpy has a "safety feature" that it refuses to overwrite a
 directory, but that safety feature doesn't really apply to the
 directory ".", since "." explicitly means that we want to overwrite
 the directory.
] 
[change a bit of code to accept FilePathLike arguments.
David Roundy <droundy at darcs.net>**20081008172047
 Ignore-this: 87032e071bb49117133d99a90bdfd0e1
 These functions really should expect FileName, and this is an
 incremental move in that direction.
] 
[Implement Commands.Check in terms of Repository.Repair functionality.
me at mornfall.net**20080926144006] 
[Implement Commands.Repair in terms of Repository.Repair functionality.
me at mornfall.net**20080926143919] 
[Export testTentative from Darcs.Repository.
me at mornfall.net**20080926143839] 
[Implement Repository.Repair to provide primitives for check and repair commands.
me at mornfall.net**20080926143721] 
[keep changepref patches from breaking the toSimple optimization.
David Roundy <droundy at darcs.net>**20080924162022
 Ignore-this: 404a4386b048b74dd9afcdde47bc1b11
] 
[rewrite push_coalesce_patch to avoid calls to lengthFL.
David Roundy <droundy at darcs.net>**20080924161517
 Ignore-this: cf746702c3b1e06ab2d400878e20e305
] 
[make various autoconf simplifications.
David Roundy <droundy at darcs.net>**20080923150742
 Ignore-this: 213c53eda0b54c01909c8d15b2a5fdd6
] 
[Restore send-external.sh test.
David Roundy <droundy at darcs.net>**20081009140121
 Ignore-this: 700f9c538033108cac397c1d757fbfc4
] 
[TAG 2.1.0
Eric Kow <E.Y.Kow at brighton.ac.uk>**20081009120532] 
[Rollback send-external test.
Eric Kow <E.Y.Kow at brighton.ac.uk>**20081009101224
 I was over-optimistic about pulling this into stable.
 It fails under Windows.

 rolling back:

 Tue Oct  7 23:28:04 BST 2008  David Roundy <droundy at darcs.net>
   * add test for send --sendmail-command.

     A ./tests/send-external.sh
] 
[ChangeLog entries for 2.1.0 (tweaks).
Eric Kow <E.Y.Kow at brighton.ac.uk>**20081009090247] 
[disable tests/issue1078_symlink.sh on Windows
zooko at zooko.com**20081001130248
 (Windows does not have symlinks.)
] 
[use minimum Cachable time in corner case where f/=f' and c/=c'.
David Roundy <droundy at darcs.net>**20081008173804
 Ignore-this: 51f2a68ac4139420bd116e8293cc5cc2
] 
[Bump version number to 2.1.0.
Eric Kow <E.Y.Kow at brighton.ac.uk>**20081008203033] 
[Canonize Benjamin Franksen.
Eric Kow <E.Y.Kow at brighton.ac.uk>**20081008202951] 
[ChangeLog entries for darcs 2.1.0.
Eric Kow <E.Y.Kow at brighton.ac.uk>**20081008202927] 
[add test of Ignore-this functionality.
David Roundy <droundy at darcs.net>**20081008192621
 Ignore-this: 47b7bf225e2997dd78c351763c769738
] 
[warn users if they try to record a change beginning with Ignore-this:.
David Roundy <droundy at darcs.net>**20081008192001
 Ignore-this: 94828726d609f46e384c2ab8091e49db
 Ignore-this: I am testing the functionality of this feature.
] 
[allow users to add their own ignored information to patches.
David Roundy <droundy at darcs.net>**20081008190038
 Ignore-this: fe8294776896b6fdcd489752a29d1069
] 
[haddock documentation for Printer
florent.becker at ens-lyon.org**20081007211959] 
[Use copyFile+renameFile for safe copy in URL.waitNextUrl.
Dmitry Kurochkin <dmitry.kurochkin at gmail.com>**20081008134308
 Ignore-this: a9a8889c2b57a5dca2785344cbd18a2f
] 
[remove excessive indentation that made the code very hard to read.
David Roundy <droundy at darcs.net>**20081008173358
 Ignore-this: d93e312c27f0d9a1ef993855c6872257
] 
[When requesting same URL with different cachability choose least cachable.
Dmitry Kurochkin <dmitry.kurochkin at gmail.com>**20081008145049
 Ignore-this: fddacf3e706a68c55871f81438d283b
] 
[add simple test of interactive record.
David Roundy <droundy at darcs.net>**20081008172036
 Ignore-this: ebb01acdb477174095e1b66aac230629
] 
[Improve test pull.sh cleanup
Thorkil Naur <naur at post11.tele.dk>**20081008115539] 
[resolve issue1124: Test pull.sh failed when run as root
Thorkil Naur <naur at post11.tele.dk>**20081008114134] 
[Allow dist command to take a tag argument (issue 286).
Stephen Hicks <sdh33 at cornell.edu>**20081006075133
 It should probably be a "match_one_context" instead of a "match_one"
 option type, and then instead of using the "nonranged" matcher, we'd
 use the "patchset" ones (that don't exist yet).  But I couldn't figure
 out how to blend "get_nonrange_match" and "get_one_patchset" in any
 sensical way to get immediate pristine output (and all the more so for
 the "_s" variants).  So for now, contexts are not supported, but it
 would be nice if they were someday.
] 
[clean up RepoPath a bit (removing unused methods from classes).
David Roundy <droundy at darcs.net>**20081008161906
 Ignore-this: 4e69882ebe10f20bb67350b106bd10af
] 
[clean up exportation of SlurmMonad private internals.
David Roundy <droundy at darcs.net>**20081008151758
 Ignore-this: 3763cc352b6f28fa83d8fbe5b1077985
] 
[Optimize clean_hashdir's use of cleanCaches.
Petr Rockai <me at mornfall.net>**20081007191237

 We now only ask cleanCaches to look at files we have unlinked and therefore
 might have caused their cached equivalents to drop link-count to 1. Limits
 number of stats to O(n), n = number of cleaned out files.
] 
[Resolve issue1131: accept download requests for different files.
Dmitry Kurochkin <dmitry.kurochkin at gmail.com>**20081008113040
 Ignore-this: a45dc4c336f87a61bda3b3c96a8656fa
] 
[add test for send --sendmail-command.
David Roundy <droundy at darcs.net>**20081007222804
 Ignore-this: f84bb43f50a6a07e0cddde739b602e6a
] 
[fixed accidental merge of two lines in default boring; removed pattern for directory "dist"
ben.franksen at online.de**20081007185233] 
[resolve issue1128: must call execSendmail inside body of withOpenTemp
benjamin.franksen at bessy.de**20081007115303] 
[Nicer bug output (it was missing a space).
Eric Kow <eric.kow at gmail.com>**20081007155214] 
[partial haddock documentation of Arguments.lhs
Florent Becker <florent.becker at ens-lyon.org>**20081007131122] 
[haddock documentation for ColorPrinter
florent.becker at ens-lyon.org**20081002203509] 
[disable progress reports when connecting to an ssh server.
David Roundy <droundy at darcs.net>**20081002215119
 Ignore-this: 5088b46719072abcecf0c8c406b8f6d7
] 
[Resolve issue1104: stop progress reports in exec.
Dmitry Kurochkin <dmitry.kurochkin at gmail.com>**20081002130157
 Ignore-this: 40bd4c27fd600f6dbcc3f89a62db9104
] 
[Resolve issue1109: stop progress reports in exec_interactive.
Dmitry Kurochkin <dmitry.kurochkin at gmail.com>**20081002125805
 Ignore-this: 2938508e82322b41fb9e18db4ee4ceba
] 
[fix tab in Replace that broke haskell-policy.sh test.
David Roundy <droundy at darcs.net>**20081006171036
 Ignore-this: c02875410986f3e2e48da0da19333311
] 
[Restore quiet grep in portable_pwd
Eric Kow <eric.kow at gmail.com>**20081005164212
 Only this time, we use > /dev/null.
] 
[Disable a test that triggers a setCooked failure under Windows.
Eric Kow <eric.kow at gmail.com>**20081005145124] 
[hint about replace --force [issue864]
Tommy Pettersson <ptp at lysator.liu.se>**20081005160640] 
[fix issue966 test, use better temp dir name, and start clean
Tommy Pettersson <ptp at lysator.liu.se>**20081005000117] 
[mv nfs-failure test to the bugs/ folder, since it fails.
David Roundy <droundy at darcs.net>**20081005130112
 Ignore-this: 73c6e55558233f1b56d7ff51f327f454
] 
[further makefile cleanups.
David Roundy <droundy at darcs.net>**20081005124353
 Ignore-this: 8c498e4b793b7049e273d6ac98441256
] 
[Simplify testing in makefile.
Eric Kow <eric.kow at gmail.com>**20081005092416] 
[Revise test documentation to reflect consolidation to shell tests.
Eric Kow <eric.kow at gmail.com>**20081005091131] 
[Remove Perl test harness and convenience copies.
Eric Kow <eric.kow at gmail.com>**20081005090921] 
[Do not use Perl test harness.
Eric Kow <eric.kow at gmail.com>**20081005090903] 
[Hopefully fix issue595 test on Solaris.
Eric Kow <eric.kow at gmail.com>**20081005090156

 I'm not entirely clear on what is causing this to fail.
 The basic test is:
   mkdir -p restrictive/liberal
   chmod 111 restrictive
   cd restrictive/liberal
   darcs get

 This is apparantly known to fail under Solaris.  The original Perl test
 claims it's because you can't cd into restrictive/liberal, but the shell
 variant seems to manage fine.  In any case, I think this touch file
 check is simpler.
] 
[Translate merging_newlines bug into shell and restore bug context.
Eric Kow <eric.kow at gmail.com>**20081005085031

 This used to be a TODO item at the end of the merging_newlines
 test, but when it was moved out, the actions required to set up
 a bug were forgotten.  This patch restores those actions.
] 
[Translate some Perl bugs into shell.
Eric Kow <eric.kow at gmail.com>**20081005084157] 
[fix bug in external.sh translation.
David Roundy <droundy at darcs.net>**20081005122814
 Ignore-this: 1291ebf27ea9c4a04d79c877b0f6f82d
] 
[Translate printer test into shell.
Eric Kow <eric.kow at gmail.com>**20081005020140] 
[Translate still more Perl tests into shell.
Eric Kow <eric.kow at gmail.com>**20081005012058] 
[Translate pull_many_files tests into shell and avoid redoing work.
Eric Kow <eric.kow at gmail.com>**20081005004120
 Just use whatever format is in .darcs/defaults
] 
[Remove residual Perlisms in match-date.
Eric Kow <eric.kow at gmail.com>**20081005000324] 
[Translate date matching test to shell.
Eric Kow <eric.kow at gmail.com>**20081005000243] 
[Skip filepath test which does not work on Windows.
Eric Kow <E.Y.Kow at brighton.ac.uk>**20081004234806] 
[Fix non-portable use of grep -q in tests.
Eric Kow <E.Y.Kow at brighton.ac.uk>**20081004234524] 
[Make sed usage more portable in hidden_conflict2 test.
Eric Kow <E.Y.Kow at brighton.ac.uk>**20081004233717] 
[Use example.com in external.pl tests.
Eric Kow <E.Y.Kow at brighton.ac.uk>**20080928082407
 This avoids timing out in some DNS lookup.  When I first wrote this
 test, I did not know that RFC2606 has reserved example.{com,net,org}
 which seems to be useful in precisely this situation.
] 
[fix incorrect merging_newlines.sh test.
David Roundy <droundy at darcs.net>**20081004223547
 Ignore-this: 33ebc10cd54112ec101750dd6c75fdf9
] 
[Move merging_newlines test to bugs and add a simpler version.
Eric Kow <eric.kow at gmail.com>**20081004135724

 There are two issues I noticed when converting this Perl test into
 shell.  This test was passing, but I made some mistakes translating it,
 and in doing so, noticed two potential bugs.

 The first is a potentially hidden conflict.  If we start from
 a file
   from temp1
   <EOF>
 And then it seems that a change which adds an empty line should
 conflict with a change that adds a line with content.

 The second is that if you do omit the newline, as the original
 test does,
   from temp1<EOF>
 you get some strange-looking changes, namely that appending a
 line to this yields two addline changes and not one.
] 
[fix wrong ../path in failing issue1013 test
Tommy Pettersson <ptp at lysator.liu.se>**20081004140200] 
[Add a toolbox.sh to tests
Eric Kow <eric.kow at gmail.com>**20081004202139
 Like template.sh, this is not a real test, but just helper
 code for writing tests.  The toolbox just provides a single
 place you can go to look up how to do things.
] 
[Translate some more Perl tests into shell.
Eric Kow <eric.kow at gmail.com>**20081004201632] 
[Translate some more Perl tests into shell.
Eric Kow <eric.kow at gmail.com>**20081004163850
 Shell equivalents already exist for these particular tests,
 but they may not do the same thing.
] 
[Move non-date-matching code out of match.pl
Eric Kow <eric.kow at gmail.com>**20081004145445
 The goal is to minimise and eventually eliminate our use of Perl
 for regression tests.
] 
[Fix minor bug in conflict-doppelganger test.
Eric Kow <eric.kow at gmail.com>**20081004163746
 It does not account for --old-fashioned-inventory being
 in .darcs/defaults
] 
[Update tests to reflect --darcs-2 default.
Eric Kow <E.Y.Kow at brighton.ac.uk>**20080925143252
 We need to explicitly say --old-fashioned-inventory to test
 accordingly.
] 
[mv issue1111 to tests/
David Roundy <droundy at darcs.net>**20081004143649
 Ignore-this: 51683a80094a4d7733a31b4a40e94016
] 
[resolve issue1111: patchset_intersection used wrong selection for partitionRL
Tommy Pettersson <ptp at lysator.liu.se>**20081004123851
 We want to commute the non-common patches away, so we can stick the
 remaining common patches to the rest of the common patch set.
] 
[use longer patch names in issue1111 test for safer grep result
Tommy Pettersson <ptp at lysator.liu.se>**20081003230323
 The 'not grep C out' found the author--date line of patch A, which
 contained my timezone (CEST), so the test failed even when it should have
 succeeded.
] 
[Translate some more Perl tests into shell.
Eric Kow <eric.kow at gmail.com>**20081004092507] 
[The pager defaults to less(1), not more(1)
Matthias Kilian <kili at outback.escape.de>**20081003212319

 It would be better to change get_viewer in Darcs/Utils.lhs to default
 to more(1), but since this may be too intrusive for the upcoming
 release, just let the manual tell the truth (i.e., we're using
 less(1) by default).

] 
[remove -q from diff in test, not supported on solaris
Tommy Pettersson <ptp at lysator.liu.se>**20081003205756] 
[Fix cd bugs in conflict-doppleganger test.
Eric Kow <eric.kow at gmail.com>**20081004094407
 We were not always exiting from darcs repositories when we
 meant to.
] 
[fix test issue1110, remove duplicates of cd ..
Tommy Pettersson <ptp at lysator.liu.se>**20081003182126
 They got us out of the test dir, up in the file tree hierarchy, to the
 darcs root dir, and beyond, where the test continued to run test commands
 and cleanups (ooops!!)
] 
[Translate some Perl tests into shell.
Eric Kow <eric.kow at gmail.com>**20081003183746] 
[Convert tag.pl test into shell.
Eric Kow <eric.kow at gmail.com>**20081003095231] 
[Add a shell test template.
Eric Kow <eric.kow at gmail.com>**20081003095005
 This provides helper functions and a basic repository
 setup.  The idea is that when making a new shell test,
 you start by making a copy of the template.
] 
[Reformat Darcs.CommandsAux comments as haddock.
Eric Kow <E.Y.Kow at brighton.ac.uk>**20080927123217] 
[haddock documentation for ColorPrinter
Tommy Pettersson <ptp at lysator.liu.se>**20081003175214] 
[only  show 'diffing dir' when debugging.
David Roundy <droundy at darcs.net>**20081001134124
 Ignore-this: 277810d9083e36b42f27fa7ac4c47386
] 
[TAG 2.1.0pre3
Eric Kow <E.Y.Kow at brighton.ac.uk>**20081002091241] 
[Bump version number to 2.1.0pre3.
Eric Kow <E.Y.Kow at brighton.ac.uk>**20081002091226] 
[ChangeLog entries for darcs 2.1.0pre3
Eric Kow <E.Y.Kow at brighton.ac.uk>**20081002091155] 
[make boring file and tests work with hpc.
David Roundy <droundy at darcs.net>**20081001202849
 Ignore-this: a699af014f672cc32c1a88863db8ffec
] 
[make default_boring and default_binaries NOINLINE
David Roundy <droundy at darcs.net>**20081001202731
 Ignore-this: f74741738a612bb283d1e871d70aa492
 This is primarily so I won't have to recompile all of darcs when these
 change, but also because they shouldn't be inlined.  It would just be
 a waste of space (and would slow down darcs).
] 
[trim exports of IsoDate.
David Roundy <droundy at darcs.net>**20081001194015
 Ignore-this: 6f1ef595002c05c07821c2d1332e054b
] 
[fix bug in 1105 fix that led to improper overrides of defaults
David Roundy <droundy at darcs.net>**20081001175907
 Ignore-this: 8413d2596f97076ccb096900585f0c62
 The problem is illustrated in the tests/override-defaults.sh that I've
 recorded separately.  We want conflicting defaults in
 ~/.darcs/defaults and _darcs/prefs/defaults to be resolved in favor of
 the default residing in _darcs/prefs/defaults.
] 
[add test that overriding of defaults works right.
David Roundy <droundy at darcs.net>**20081001175645
 Ignore-this: aedfd923676b1fa35c25c72abb88c8f3
] 
[Test for issue1105.
Dmitry Kurochkin <dmitry.kurochkin at gmail.com>**20080929142205
 Ignore-this: 5b7c2f7d270ab614eb1294566048a586
] 
[Resolve issue1105: check if default options are valid.
Dmitry Kurochkin <dmitry.kurochkin at gmail.com>**20080929141948
 Ignore-this: 17b4b27e31545689f21496f7fc9c9fe7
] 
[fix bugs in second issue1110 test.
David Roundy <droundy at darcs.net>**20081001171147
 Ignore-this: 2757852f2b599cfd55936cccd1b83b5f
] 
[Add another test case for issue1110.
Eric Kow <E.Y.Kow at brighton.ac.uk>**20080930220818] 
[generate a tidier hoogle frame, add required files, simplify framed doc generation
Simon Michael <simon at joyful.com>**20080930204518
 Ignore-this: e60b456f1fdd001b5ae456f9aae05999
] 
[haddock documenation for DateTester
florent.becker at ens-lyon.org**20080926165323] 
[haddock documentation for DateMatcher
florent.becker at ens-lyon.org**20080926164753] 
[make installdocs should not install TeX intermediaries.
Trent W. Buck <trentbuck at gmail.com>**20080930030315
 I'm unilaterally classing the DVI and PostScript versions as
 "intermediaries" to the PDF version, and only installing the latter.
] 
[Add test cases for issue1043.
Eric Kow <E.Y.Kow at brighton.ac.uk>**20080930105032
 There are two known variants, one of which was fixed by
 resolve issue1043: fix bug in mergeAfterConflicting.
] 
[resolve issue1110: fix get --hashed.
David Roundy <droundy at darcs.net>**20080929175725
 Ignore-this: d0aaaa26583dd3ab37bedfc738fb6117
] 
[add test that show bug works right.
David Roundy <droundy at darcs.net>**20080929152909
 Ignore-this: 4829e300015120adeed108079324e5e2
] 
[Add a test case for issue1110.
Eric Kow <E.Y.Kow at brighton.ac.uk>**20080929151743] 
[Tone down unnecessarily scary language when cancelling
Simon Michael <simon at joyful.com>**20080928004013
 Ignore-this: c5940d14099953b7bc963c70591739d1
] 
[darcshoogle script and emacs integration example
Simon Michael <simon at joyful.com>**20080927210609
 Ignore-this: 93d5a0ff269d314d32f213dfbe4325be
] 
[Fix doubled verb `be' in documentation on apply posthooks in _darcs/prefs/defaults.
Taylor R Campbell <campbell at mumble.net>**20080928200954] 
[the issue864 test was misnamed
Simon Michael <simon at joyful.com>**20080928231942
 Ignore-this: a0b643bf0abc6f4b6237e5683e9f6dad
] 
[add tests for pull --union and --intersection.
David Roundy <droundy at darcs.net>**20080929150711
 Ignore-this: 356f506f79ca89a2d1246d068aaa8b2b
 --intersection fails, which is issue1111.
] 
[add more output to URL bug message.
David Roundy <droundy at darcs.net>**20080929145544
 Ignore-this: 868e0ba5819dba2a4f5b8819080e9407
 I triggered this bug a few times, but once I added the extra output, I
 wasn't able to do it again.  So I'm leaving the extra information in
 the bug message so if someone else runs into it, we can more easily
 track down the issue.
] 
[answer darcs-doc question: no, that would be wrong.
David Roundy <droundy at darcs.net>**20080926220229
 Ignore-this: 419afd7bd9832e06d3dca45880276296
] 
[haddockification of IsoDate
florent.becker at ens-lyon.org**20080926163257
 Docstrings by Eric Kow
] 
[Make UglyFileName.super_name work with "/foo" paths.
Dmitry Kurochkin <dmitry.kurochkin at gmail.com>**20080926223753
 Ignore-this: 6e4abb72087272ed03e5839a5420f82f
] 
[Restore 'simplify patches in rollback'.
Eric Kow <E.Y.Kow at brighton.ac.uk>**20080926225429] 
[Rollback 'simplify patches in rollback'
Eric Kow <E.Y.Kow at brighton.ac.uk>**20080926225411
 I made the mistake of pulling this into stable despite a darcs 2.1
 feature freeze.
 rolling back:

 Tue Sep 23 16:06:19 BST 2008  David Roundy <droundy at darcs.net>
   * simplify patches in rollback.
] 
[Restore issue27 patches.
Eric Kow <E.Y.Kow at brighton.ac.uk>**20080926222522

 rolling back:

 Fri Sep 26 23:01:19 BST 2008  Eric Kow <E.Y.Kow at brighton.ac.uk>
   * Resolve issue1102: recognise but do not generate patch log 'junk'.

] 
[Resolve issue1102: recognise but do not generate patch log 'junk'.
Eric Kow <E.Y.Kow at brighton.ac.uk>**20080926220119

 This is a partial rollback of the two issue27 patches below.  The patches exist
 to (a) generate patch log 'junk' and (b) hide such junk from users' view.
 Because of a feature freeze, we do not want junk generation to be part of darcs
 2.1; however, we do anticipate junk-generation being part of subsequent
 versions of darcs.  To avoid users being confused by this future junk, we only
 rollback the junk-generation part, retaining the junk-hiding part.

 rolling back:

 Wed Sep 17 16:46:57 BST 2008  David Roundy <droundy at darcs.net>
   * resolve issue27: add junk to patch identifiers.

 Wed Sep 17 18:09:13 BST 2008  David Roundy <droundy at darcs.net>
   * hokey fix to allow unit tests to generate random input.

] 
["make api-doc-frames" converts api docs to a frame layout
Simon Michael <simon at joyful.com>**20080927091919
 Ignore-this: 2a600aefe4399d17a4c2f14ea1cec350
] 
[hoogleindex no longer depends on api-doc dir
Simon Michael <simon at joyful.com>**20080927090139
 Ignore-this: 3a29b9bbd1ba1cdbccea03f4c04727c8
] 
[make hoogle targets more robust
Simon Michael <simon at joyful.com>**20080927083125
 Ignore-this: 2aac81020d3855ce8ffc9c4deef3e949
] 
[fix makefile indentation
Simon Michael <simon at joyful.com>**20080927020444
 Ignore-this: 847051698f40dc1fd03a8e1db93ff7a1
] 
["make hoogleweb" configures the hoogle web interface in hoogle/ 
Simon Michael <simon at joyful.com>**20080927015904
 Ignore-this: dcc4e2526b3c92859ccf9a6f1cb57ad2
 The hoogle index target is now "make hoogleindex". Requires the hoogle
 source tree (perhaps patched to configure result link urls).
 Tested on GNU/Linux with hoogle 4.0.0.5.
] 
[make haddock less noisy 
Simon Michael <simon at joyful.com>**20080927012958
 Ignore-this: 84921fbd34b1f7345dec2571c3ec32ae
] 
[resolve issue1043: fix bug in mergeAfterConflicting.
David Roundy <droundy at darcs.net>**20080926211928
 Ignore-this: 1416605539b44b32c18b348f3b4f459d
 This is moderately deep in the internals of the darcs-2 conflict
 handling code.  I had made an assumption that turned out not to be
 correct.  I fix this by switching to use a variant of the commute
 function that doesn't allow conflicting patches to commute, which I
 think should restore correctness here.  It's a scary bug, though, and
 if anyone were to create a moderately small test case, I'd be
 extremely grateful.
] 
[Roll back Dmitry's drop_dotdot change in Darcs.Patch.Filename.  
David Roundy <droundy at darcs.net>**20080926145013
 Ignore-this: d432374bbbe4cc006a26deeb3d15c3ec

 The key here is that he didn't want to change the internal
 patch-handling code, just the use of drop_dotdots in Darcs.RepoPath,
 which now uses UglyFileName for this purpose (which has Dmitry's code
 in it).

 rolling back:

 Thu Sep 25 13:57:11 EDT 2008  Dmitry Kurochkin <dmitry.kurochkin at gmail.com>
   * Make FileName.drop_dotdot work with absolute paths.

     M ./src/Darcs/Patch/FileName.lhs -7 +12
] 
[split FileName into two modules.
David Roundy <droundy at darcs.net>**20080926144501
 Ignore-this: dbb6650c6300745101bacd41bef431f0
 This duplicates some code, and makes a Darcs.Patch.FileName into the
 module (which FileName originally was) for handling paths within a
 darcs repository.  Code that had gotten agglomerated into this module
 is now in UglyFileName.lhs, which I hope to eliminate.  I went through
 the imports of FileName and tried to separate them between those that
 use FileName to deal with paths in a repository and those that use the
 extra functions what were added there, and those that abuse FileName
 to handle other sorts of paths (absolute paths, in particular).
] 
["make api-doc-with-source" generates docs with links to colourised source code
Simon Michael <simon at joyful.com>**20080925213719
 Ignore-this: c16b935c727838c606fa5daa29ccc41f
 This works with current hs-colour on GNU/Linux but may not be
 portable; I made it a separate make target to start with.  Only
 per-module source links are enabled until there is a fix for the
 haddock/hoogle issue noted.
] 
[issue27.sh is still sporadically buggy.
David Roundy <droundy at darcs.net>**20080926134948
 Ignore-this: ec2bf07485a6dc0fda83c786a7982df1
] 
[add test that fails sporadically on nfs under ghc 6.6.
David Roundy <droundy at darcs.net>**20080925202350
 Ignore-this: 8aae9073ea132935f1951f7e187bc2ea
 I haven't time to track this down, but it doesn't look like a bug in
 the test script.  This was the issue27 test.
] 
[Move issue1078 test from bugs to tests.
Dmitry Kurochkin <dmitry.kurochkin at gmail.com>**20080925180103
 Ignore-this: f735ee2e36bdf8f446cab61d1f7ac334
] 
[Resolve issue1078: make ioAbsolute work with symbolic links in file paths.
Dmitry Kurochkin <dmitry.kurochkin at gmail.com>**20080925175726
 Ignore-this: af4cf0bd842b9aae5e2fffe4500a1aa5
] 
[Make FileName.drop_dotdot work with absolute paths.
Dmitry Kurochkin <dmitry.kurochkin at gmail.com>**20080925175711
 Ignore-this: 46c625a35bb11bea19b0749756d1a225
] 
[Force hidden conflicts test to use the darcs-2 format.
Eric Kow <E.Y.Kow at brighton.ac.uk>**20080925175251
 Move it to the tests directory because it passes if we do
 this.  We consider this to be a bug that is solved by using
 the darcs 2 format.
] 
[Use init+pull instead of get in issue27 test.
Dmitry Kurochkin <dmitry.kurochkin at gmail.com>**20080925142606
 Ignore-this: e8be404b0ccbc56d8f547b11b6e58c76
 This would hopefully make it pass on NFS.
] 
[Exceptions to GPL granted as of 2008-09-11.
Eric Kow <E.Y.Kow at brighton.ac.uk>**20080911120758] 
[Update manual to reflect --darcs-2 default.
Eric Kow <E.Y.Kow at brighton.ac.uk>**20080925142538] 
[resolve issue1003: don't sever transfer-mode connection on IO error.
David Roundy <droundy at darcs.net>**20080925145150
 Ignore-this: 3aecb8cffa83170847b0a2452c5763f0
 There was a bug in Ssh, in which unless the very first file we access
 on a given server was present, we severed the connection.  This fixes
 that bug.
] 
[preliminary hoogle indexing
Simon Michael <simon at joyful.com>**20080925084432
 If haddock and hoogle are installed, "make hoogle" generates api-doc/main.hoo.
 Use it at the command-line like so: hoogle --data=api-doc/main.hoo something
] 
[TAG 2.1.0pre2
Eric Kow <E.Y.Kow at brighton.ac.uk>**20080925081049
 Ignore-this: 99b608f2401e8f14358e121e9b95e211
] 
[Bump version number to 2.1.0pre2.
Eric Kow <E.Y.Kow at brighton.ac.uk>**20080925081019
 Ignore-this: 9d1aa701ce0b8cfc87007216207166fe
 It was initially planned that the upcoming release be called 2.0.3, but
 since we are initializing darcs-2 format repositories by default, we are
 bumping the version number higher.
] 
[ChangeLog entries for 2.1.0pre2
Eric Kow <E.Y.Kow at brighton.ac.uk>**20080925080141
 Ignore-this: 1b1e57d425f8528e00e03e7b4a23ad78
] 
[ChangeLog entries: more stuff to ignore
Eric Kow <E.Y.Kow at brighton.ac.uk>**20080925080129
 Ignore-this: 45362ed8bbabdacf222928cba6756aa4
] 
[More readable length comparison.
Eric Kow <E.Y.Kow at brighton.ac.uk>**20080924142304] 
[Haddock some primitive patch functions.
Eric Kow <E.Y.Kow at brighton.ac.uk>**20080924142157] 
[resolve issue805: make darcs-2 format the default for new repositories.
David Roundy <droundy at darcs.net>**20080924141158
 Ignore-this: e7952cb0cdc3124ffa50b0775822000e
] 
[make flagsToPristine obey repository format.
David Roundy <droundy at darcs.net>**20080924135319
 Ignore-this: 6038a7d05126af9e446406022ca608a0
 This reduces the number of places we set the default repository format
 (hopefully to one?).
] 
[move issue27 test to bugs directory, since it fails.
David Roundy <droundy at darcs.net>**20080923215936
 Ignore-this: 4556b273a9f8728de8ac855aae8442d0
] 
[Add test for issue27.
Dmitry Kurochkin <dmitry.kurochkin at gmail.com>**20080918135923] 
[Add a test case for issue1078
Eric Kow <E.Y.Kow at brighton.ac.uk>**20080923081757
 Ignore-this: 33f7f1f63c7b707ff148531f8229ceb0
] 
[give more useful failure message in HTTP for proxy errors.
David Roundy <droundy at darcs.net>**20080923153400
 Ignore-this: 3d6d204da399175eedf68bedfed8e504
] 
[HTTP: detect proxy server (failing if we want one)
Eric Kow <E.Y.Kow at brighton.ac.uk>**20080923123539
 The HTTP package has proxy server support, but putting it to use seems
 to be complicated.  Since fetchUrl seems to be used only in optional
 situations, it seems safe to just return an error message (as opposed
 to waiting around for a timeout).
] 
[simplify patches in rollback.
David Roundy <droundy at darcs.net>**20080923150619
 Ignore-this: fd3d327f800e2f1799ec97bc4524f612
 This makes it nicer to incrementally rollback changes from one large
 change:  you aren't prompted for changes that have already been rolled
 back.
] 
[fix filepath code to work with FilePath package that preceded filepath.
droundy at darcs.net**20080923000405
 Ignore-this: 6aa0d8b357b0f966403ebe5965dcdec4
] 
[fix type witness bug in createRepository.
David Roundy <droundy at darcs.net>**20080922234321
 Ignore-this: 2c50393ca25740ce3e210dd24fe8d8fa
] 
[Resolve conflict between replace patches in Darcs.Arguments
Eric Kow <E.Y.Kow at brighton.ac.uk>**20080922202647] 
[Resolve issue53: check for windows filename validity in darcs add/mv.
Eric Kow <E.Y.Kow at brighton.ac.uk>**20080922172004] 
[Use --reserved-ok to allow a file with ':' in it in tests.
Eric Kow <E.Y.Kow at brighton.ac.uk>**20080922171519
 It's likely that this test will just fail under Windows and
 we will have to disable it.
] 
[Add test for issue53.
Eric Kow <E.Y.Kow at brighton.ac.uk>**20080922152256] 
[Add --reserved-ok flag for darcs add and mv.
Eric Kow <E.Y.Kow at brighton.ac.uk>**20080922141532
 This is just the flag, not the actual effect.
] 
[Check for filepath package in configure.
Eric Kow <E.Y.Kow at brighton.ac.uk>**20080922140520] 
[simplify fix for issue1041.
David Roundy <droundy at darcs.net>**20080922233019
 Ignore-this: a3002e9bba5271790c62ac634e08f472
 It turns out that the simple solution works once a bug in the
 conflict-doppleganger test was fixed!
] 
[translate conflict-doppleganger test to bash.
David Roundy <droundy at darcs.net>**20080922232839
 Ignore-this: de2a050022dea4251cdc2cc5e8b55c8c
] 
[Translate mark-conflicts test into shell.
Eric Kow <E.Y.Kow at brighton.ac.uk>**20080922224858
 It was failing because it expects init to be completely silent.  Since we
 were going to tweak it anyway, we might as well simplify the script.
] 
[Stop calling the darcs-2 format experimental.
Eric Kow <E.Y.Kow at brighton.ac.uk>**20080922221024] 
[Move repository creation to Darcs.Repository.
Eric Kow <E.Y.Kow at brighton.ac.uk>**20080922215913
 This is just to avoid importing the DarcsRepo and HashedRepo code in higher
 level code.
] 
[remove test_unit from disttest to speed things up.
David Roundy <droundy at darcs.net>**20080922225355
 Ignore-this: b3b21bcd3fa72c8d602b5bd0e601021a
 The unit test is only affected by very rarely-modified code, and it's
 overkill to run it absolutely every single time we push code
 (particularly as it sometimes takes exponentially long to generate its
 test cases).
] 
[resolve issue1041: add test for issue1041.
David Roundy <droundy at darcs.net>**20080922183320
 Ignore-this: 5a6330158d16a24d45f58268c0edb823
 Note that this issue was actually resolved by Vlad Dogaru.  I just
 wrote the test.
] 
[Get: if URL is invalid, direcotry is not created (#1041)
Vlad Dogaru <ddvlad at anaconda.cs.pub.ro>**20080922171705] 
[Replace --without-docs with less ambiguous --without-manual (issue1082).
Trent W. Buck <trentbuck at gmail.com>**20080922002602
 It's confusing for ./configure --without-docs to complain about missing haddock.
] 
[Documentation for --allow-unrelated-repos.
Dmitry Kurochkin <dmitry.kurochkin at gmail.com>**20080922121122
 Ignore-this: d2630826313c8aeb00acb6853030c22d
] 
[Rename --ignore-unrelated-repos to --allow-unrelated-repos.
Dmitry Kurochkin <dmitry.kurochkin at gmail.com>**20080922120727
 Ignore-this: a5990f1741c867316a948e6721118651
] 
[fix  bug I introduced into issue1039 test.
David Roundy <droundy at darcs.net>**20080921213043
 Ignore-this: 5b3c6476abae6bb050be014555d05bbe
] 
[Fix hang after a user input error (for example, EOF).
Judah Jacobson <judah.jacobson at gmail.com>**20080918163017] 
[replace consRLSealed with a more  general mapFlipped.
David Roundy <droundy at darcs.net>**20080921185241
 Ignore-this: c28f73f165254582cba6a14ba6ce93
] 
[make issue1039 fix allow small dissimilar  repositories.
David Roundy <droundy at darcs.net>**20080921184515
 Ignore-this: 918a09df18ef48c649c1bfaa866d6176
] 
[revert refactor that breaks type witnesses.
David Roundy <droundy at darcs.net>**20080921182331
 Ignore-this: dd692cffc1a238d6726448bacfe9cacc
] 
[Add '--ignore-unrelated-repos' option to disable unrelated repositories check.
Dmitry Kurochkin <dmitry.kurochkin at gmail.com>**20080919152631] 
[Resolve issue1039: detect seemingly unrelated repositories when doing push, pull and send.
Dmitry Kurochkin <dmitry.kurochkin at gmail.com>**20080919144553] 
[Refactor in pull_cmd.
Dmitry Kurochkin <dmitry.kurochkin at gmail.com>**20080919135349
 Ignore-this: e26a489a7a53aeaba544ae5ad0006700
] 
[Test for issue1039.
Dmitry Kurochkin <dmitry.kurochkin at gmail.com>**20080919153011] 
[manual: add an example of record --pipe prompts similar to tag --pipe docs
Simon Michael <simon at joyful.com>**20080918205353] 
[user manual corrections regarding what record and tag --pipe prompt for
Simon Michael <simon at joyful.com>**20080918204500] 
[clarify the short help for --pipe
Simon Michael <simon at joyful.com>**20080918193717] 
[Spaces in Darcs.Arguments.
Dmitry Kurochkin <dmitry.kurochkin at gmail.com>**20080919150440] 
[Spaces in Darcs.Commands.Send.
Dmitry Kurochkin <dmitry.kurochkin at gmail.com>**20080919150139] 
[Spaces in Darcs.Commands.Pull.
Dmitry Kurochkin <dmitry.kurochkin at gmail.com>**20080919145812] 
[Spaces in Darcs.Commands.Push.
Dmitry Kurochkin <dmitry.kurochkin at gmail.com>**20080919145553] 
[Print "We have the following patches to send:" only when we really have somthing to send.
Dmitry Kurochkin <dmitry.kurochkin at gmail.com>**20080919114229] 
[Use gmakeisms for prettier output.
Trent W. Buck <trentbuck at gmail.com>**20080919071358] 
[TAG 2.0.3pre1
Eric Kow <E.Y.Kow at brighton.ac.uk>**20080918023645] 
[Bump version number to 2.0.3pre1.
Eric Kow <E.Y.Kow at brighton.ac.uk>**20080918023634] 
[Canonize Florent Becker, Matthias Killian, Trent Buck, Thorkil Naur
Eric Kow <E.Y.Kow at brighton.ac.uk>**20080918020453] 
[ChangeLog entries for darcs 2.0.3pre1 again.
Eric Kow <E.Y.Kow at brighton.ac.uk>**20080918020442] 
[Re-enable haskeline by default.  Require haskeline>=0.3.1 which fixes issue1050.
Judah Jacobson <judah.jacobson at gmail.com>**20080917194007] 
[Throw an error when haskeline returns an EOF/EOT.
Judah Jacobson <judah.jacobson at gmail.com>**20080917180124] 
[Disable more record_editor tests under Windows.
Eric Kow <E.Y.Kow at brighton.ac.uk>**20080917192602
 It still appears to be failing under Cygwin.
] 
[fix changes.pl test (translating to bash)
David Roundy <droundy at darcs.net>**20080917182432
 Ignore-this: 5f8bc7e1f9eadc073402a935142281c4
 This test made assumptions such as that darcs wouldn't ever add a long
 comment to its internal representation of changes, which are now
 broken.
] 
[hokey fix to allow unit tests to generate random input.
David Roundy <droundy at darcs.net>**20080917170913
 Ignore-this: 31e847e82eef741f4c6cc857fd79a245
 A nicer fix would be to move namepatch and patchinfo into some sort of
 random-number monad rather than leaving them in IO and using
 unsafePerformIO in the example-generation scripts.
] 
[resolve issue27: add junk to patch identifiers.
David Roundy <droundy at darcs.net>**20080917154657
 Ignore-this: b91ab6f6e05e0fda25488fa51653b741
] 
[add a couple of tests of unrecord.
David Roundy <droundy at darcs.net>**20080917133738] 
[add a few (passing) tests of pending handling in obliterate and unrecord.
David Roundy <droundy at darcs.net>**20080917130423] 
[Use putStrLn for "Cancelled." message.
Dmitry Kurochkin <dmitry.kurochkin at gmail.com>**20080916205014] 
[remove now-unused is_addrmfile and is_addrmdir.
David Roundy <droundy at darcs.net>**20080916173136] 
[resolve issue1012: it seems to be fixed by the fix for issue709.
David Roundy <droundy at darcs.net>**20080916173116] 
[resolve issue885: fix patchSetToRepository to work with hashed.
David Roundy <droundy at darcs.net>**20080916173030] 
[resolve issue709: avoid adding changes to pending in rmpend when possible.
David Roundy <droundy at darcs.net>**20080916173002] 
[first-stage fix for issue709.
David Roundy <droundy at darcs.net>**20080916170333
 Here I fix the bug which leads to a corrupt pending being left, with a
 rmfile change but no hunk removing the contents.  This doesn't fix
 issue709, since an incorrect pending is left, it's just no longer a
 corrupt pending (i.e. it's still got the rmfile, but if you record it
 there's no problem).
] 
[add new test that we don't do anything when repairing a clean repo.
David Roundy <droundy at darcs.net>**20080916165437] 
[whitespace change in prepend.
David Roundy <droundy at darcs.net>**20080916160425] 
[make shell_harness print summary of passing tests when running bugs tests.
David Roundy <droundy at darcs.net>**20080916145820] 
[trailing whitespace cleanup in Repository.Internal.
David Roundy <droundy at darcs.net>**20080916142112] 
[tiny refactor in Internal.lhs.
David Roundy <droundy at darcs.net>**20080916155922] 
[simplify issue965 test (which took quite a while for me to figure out).
David Roundy <droundy at darcs.net>**20080916152028] 
[Test for issue691.
Dmitry Kurochkin <dmitry.kurochkin at gmail.com>**20080916111332] 
[Make match.pl test understand Windows dates.
Eric Kow <E.Y.Kow at brighton.ac.uk>**20080916011339
 Windows does not abbreviate its timezones.
] 
[Disable some tests that don't work under Windows.
Eric Kow <E.Y.Kow at brighton.ac.uk>**20080916000912] 
[Translate get.pl test into shell.
Eric Kow <E.Y.Kow at brighton.ac.uk>**20080916000227
 The original get.pl uses the Perl Cwd library, which seems not
 to work for purposes of this test under MSYS.
] 
[Resolve issue691: distinguish between NoArg and ReqArg in defaults parser.
Dmitry Kurochkin <dmitry.kurochkin at gmail.com>**20080915224046] 
[Move get_default_flag type to definition.
Dmitry Kurochkin <dmitry.kurochkin at gmail.com>**20080915220316] 
[reverse overly-invasive haddocking.
David Roundy <droundy at darcs.net>**20080915110353] 
[In darcs send if POST fails try sendmail.
Dmitry Kurochkin <dmitry.kurochkin at gmail.com>**20080914234314] 
[Grammar in Darcs.Commands.Send.
Dmitry Kurochkin <dmitry.kurochkin at gmail.com>**20080914232923] 
[Print "Successfully sent patch bundle to" only when we really sent something.
Dmitry Kurochkin <dmitry.kurochkin at gmail.com>**20080914232905] 
[Fix overzealous escaping in haddock.
Eric Kow <E.Y.Kow at brighton.ac.uk>**20080915070926] 
[Haddock documentation for English.lhs
florent.becker at ens-lyon.org**20080913143844
 Docstrings by Eric Kow
] 
[Haddock documentation for PrintPatch
florent.becker at ens-lyon.org**20080913143420] 
[Haddock documentation for Flags
florent.becker at ens-lyon.org**20080913143026] 
[Haddock documentation for Motd
florent.becker at ens-lyon.org**20080913141335
 docstring by Eric Kow
] 
[Haddock string for TheCommands
florent.becker at ens-lyon.org**20080913141040
 Docstring by Eric Kow
] 
[Run testing in temp directories to avoid collosions, fail if DarcsURL header is not found.
Dmitry Kurochkin <dmitry.kurochkin at gmail.com>**20080914223930] 
[Use tempfile() UNLINK option to automatically remove temp files at exit.
Dmitry Kurochkin <dmitry.kurochkin at gmail.com>**20080914223827] 
[Coding style in upload.cgi.
Dmitry Kurochkin <dmitry.kurochkin at gmail.com>**20080914223751] 
[Stop after we found the first DarcsURL: in patch bundle.
Dmitry Kurochkin <dmitry.kurochkin at gmail.com>**20080914220421] 
[Spaces in upload.cgi.
Dmitry Kurochkin <dmitry.kurochkin at gmail.com>**20080914220324] 
[Really allow pull.sh test to pass on Windows.
Eric Kow <E.Y.Kow at brighton.ac.uk>**20080914211458

 It seems that these redirections cause an openBinaryFile failure under
 Windows.  I do not yet understand why using different names makes a
 difference, but it could provide an interesting clue.
] 
[ChangeLog entries for darcs 2.0.3pre1
Eric Kow <E.Y.Kow at brighton.ac.uk>**20080914094144] 
[Disable amend-cancelling test under Cygwin.
Eric Kow <E.Y.Kow at brighton.ac.uk>**20080913213039] 
[Make binary.sh test more portable (avoid copying an exe).
Eric Kow <E.Y.Kow at brighton.ac.uk>**20080913212843
 Under Windows, copying the exe file will result in an .exe extension
 being appended to the filename.  This confuses the test.
] 
[rewrite upload.cgi so it won't need any customization by default.
David Roundy <droundy at darcs.net>**20080913171447
 The downside is that it has to do a darcs get --lazy in order to check if
 the patch can be applied.  If you define $target_repo, however, it doesn't
 need to do this (but then can only accept patches to a single
 locally-present repo).
] 
[when _darcs/prefs/post is present, use _darcs/prefs/post for To: header.
David Roundy <droundy at darcs.net>**20080913171025] 
[sketchy documentation of _darcs/prefs/post
David Roundy <droundy at darcs.net>**20080913115655] 
[set default upload.cgi to work on darcs-unstable.
David Roundy <droundy at darcs.net>**20080913112227] 
[Improve upload.cgi.
Dmitry Kurochkin <dmitry.kurochkin at gmail.com>**20080912230953
 - check if patch is valid before sending
 - use sendmail to send patches or drop to maildir
] 
[Spaces and punctuation in upload.cgi.
Dmitry Kurochkin <dmitry.kurochkin at gmail.com>**20080912230827] 
[fix problem in Darcs.URL refactor pointed out by Eric.
David Roundy <droundy at darcs.net>**20080913104327] 
[Disable a pull.sh test under Windows.
Eric Kow <E.Y.Kow at brighton.ac.uk>**20080912224027
 It relies on darcs not working if we chmod u-r a file.
 This seems to have little effect in Windows.
] 
[refactor Darcs.URL to eliminate use of Regexes.
David Roundy <droundy at darcs.net>**20080912173611
 The algorithms here are not tricky, and I find this easier to read.
] 
[change is_file to return false on droundy at darcs.net:
David Roundy <droundy at darcs.net>**20080912173501] 
[clean up whitespace.
David Roundy <droundy at darcs.net>**20080912150248] 
[fix manual for optional arguments.
David Roundy <droundy at darcs.net>**20080912150231] 
[clean up whitespace.
David Roundy <droundy at darcs.net>**20080912145708] 
[add test for new --output-auto-name feature.
David Roundy <droundy at darcs.net>**20080912145648] 
[Spaces in Darcs.Commands.Send module.
Dmitry Kurochkin <dmitry.kurochkin at gmail.com>**20080912122223] 
[Make '--output-auto-name' accept optional directory argument.
Dmitry Kurochkin <dmitry.kurochkin at gmail.com>**20080912120516] 
[Add DarcsOptAbsPathOption for options with optional path argument.
Dmitry Kurochkin <dmitry.kurochkin at gmail.com>**20080912112458] 
[Refactor Darcs.Repository.Prefs.getCaches.
Dmitry Kurochkin <dmitry.kurochkin at gmail.com>**20080912055204] 
[Print warning when '--http-pipelining' option is used, but darcs is compiled without HTTP pipelining support.
Dmitry Kurochkin <dmitry.kurochkin at gmail.com>**20080912054253] 
[Do not download URL we have speculated before.
Dmitry Kurochkin <dmitry.kurochkin at gmail.com>**20080912053236] 
[Spaces and parentheses in URL module.
Dmitry Kurochkin <dmitry.kurochkin at gmail.com>**20080912053000] 
[Coding style in Darcs.Arguments.network_options.
Dmitry Kurochkin <dmitry.kurochkin at gmail.com>**20080911140710] 
[Resolve issue1054: --no-cache option to ignore patch caches.
Dmitry Kurochkin <dmitry.kurochkin at gmail.com>**20080911140233] 
[Remove unused variable from configure.ac.
Dmitry Kurochkin <dmitry.kurochkin at gmail.com>**20080911132107] 
[Comment in configure.ac.
Dmitry Kurochkin <dmitry.kurochkin at gmail.com>**20080911115840] 
[Indentation fixes in configure.ac.
Dmitry Kurochkin <dmitry.kurochkin at gmail.com>**20080911115117] 
[Formating and minor refactoring in URL.urlThread.
Dmitry Kurochkin <dmitry.kurochkin at gmail.com>**20080910061227] 
[insecure printfs - fix the two that I can currently hit
Steve Cotton <steve at s.cotton.clara.co.uk>**20080910230659] 
[TAG this version works.
David Roundy <droundy at darcs.net>**20080910212908] 
Patch bundle hash:
68f1091703a9281494d1a4f1cc0c05ec4e3f13b9
-------------- next part --------------
Wed Oct 15 19:38:50 CEST 2008  Matthias Kilian <kili at outback.escape.de>
  * Run with a minimal test environment.
  This trims down the environment to only PATH and DARCS (if set).

New patches:

[Run with a minimal test environment.
Matthias Kilian <kili at outback.escape.de>**20081015173850
 This trims down the environment to only PATH and DARCS (if set).
] hunk ./tests/shell_harness 12
 # Place to put test output to avoid cluttering test/
 mkdir 'test_output', 0750;

+# Clean out the environment, preserve only PATH and DARCS.
+$saved_path = $ENV{PATH};
+$saved_darcs = $ENV{DARCS};
+undef (%ENV);
+$ENV{PATH} = $saved_path;
+if ($saved_darcs) {
+  $ENV{DARCS} = $saved_darcs;
+}
+
 # Override the users default .darcs settings
 $ENV{HOME} = cwd();
 mkdir '.darcs';

Context:

[add haskeline test to setup.hs.
David Roundy <droundy at darcs.net>**20081014201513
 Ignore-this: 1472b07696d83f70bf8ad8f7b3d5cf1
] 
[finished cleanup of default_boring entries
benjamin.franksen at bessy.de**20081014100840
 Ignore-this: e5ad0567cabe64142bf3876c2d73535e
] 
[clean up setup.hs by using lookForModule instead of catching exceptions.
David Roundy <droundy at darcs.net>**20081014160159
 Ignore-this: 8f5c7bd5be2ca099fda5bd52d35dc429
] 
[simplify pull_many_files.sh.
David Roundy <droundy at darcs.net>**20081014165155
 Ignore-this: 9f7a7cc5b02a9d51810149a8875e987f
] 
[add test for siginfo.h (which shows up on SunOS maybe?)
David Roundy <droundy at darcs.net>**20081014154814
 Ignore-this: 72d32ff1b4b57dbf75eab937be5a0e22
] 
[Use fgrep in printer.sh test (more portable than grep -F)
Eric Kow <eric.kow at gmail.com>**20081014135138] 
[have makefile make hspwd rather than shell_harness.
David Roundy <droundy at darcs.net>**20081013192943
 Ignore-this: 31008f1993c5a2281b229b9c5cdca66c
 I think there was a problem that make -j check could cause ghc --make
 to be run simultaneously.  Now it's run just once, before running any
 tests.
] 
[slightly hokey approach to solving pwd problem.
David Roundy <droundy at darcs.net>**20081013184209
 Ignore-this: 762f0339c99f1d4a529ab031ad353cd0
 I think this may help the Solaris/BSD test portability issue, and it's
 also (I think) a slightly cleaner alternative to the portable_pwd
 hack.
] 
[clean up test-franchise-build.sh a bit more...
David Roundy <droundy at darcs.net>**20081013175711
 Ignore-this: 619fff6a0be4bc8172cd8c39a5591079
] 
[clean up test-franchise-build.sh
David Roundy <droundy at darcs.net>**20081013162935
 Ignore-this: a1bd36f8bb8d3d75519abcc8d8388a99
 Among other infelicities, it wasn't using a clean franchise build
 except under wine.
] 
[add test that timestamps are properly set.
David Roundy <droundy at darcs.net>**20081013152933
 Ignore-this: caf36b9fff70f9a44b0f35078f3a5aee
] 
[on windows, try renameFile before deleting target file.
David Roundy <droundy at darcs.net>**20081010204009
 Ignore-this: 6946caec86d1c7177510baf62e2ed275
 The trouble was that if we were renaming a file to itself, we ended up
 deleting the file we were renaming (oops!).  It's a stupid hack, but
 that's what we're forced to do when running on windows.  (Note that
 it's potentially hard to determine if two paths describe the same
 file.)
] 
[clean up setup.hs configure a bit.
David Roundy <droundy at darcs.net>**20081010154959
 Ignore-this: 10855edc3e6fda315e03b6ea1c64b05e
 This removes redundant output, and warns when fast http libraries
 aren't found (or no http library at all).
] 
[simplify write_default_prefs.
David Roundy <droundy at darcs.net>**20081013121512
 Ignore-this: 4f72699d2406802e771a0ceadc913550
 It's easier to read this way, and shorter, too.
] 
[documented provenance of .tmp_versions and .DS_Store files in default boring
ben.franksen at online.de**20081012220207] 
[Fix curl version in setup.hs.
Dmitry Kurochkin <dmitry.kurochkin at gmail.com>**20081012094214
 Ignore-this: 5c44008cb7105279c37a2297e668a070
] 
[cleanup default boring
ben.franksen at online.de**20081011234148

 Also turn default_boring and default_binaries into (pure) CAFs.
] 
[Remove bashisms and GNUisms
Matthias Kilian <kili at outback.escape.de>**20081012123017

 Replace some bashisms and GNUisms by more standards compliant command
 lines, i.e.:

 - Don't use `\|' in BREs, since it's undefined according to POSIX.
   Instead just use grep -E with an ERE.

 - Don't create "funny" strings with the $'\x01...' bashism but use
   printf(1) instead. Note that this isn't strictly necessary to fix
   printer.sh on non-GNUish systems, but it makes it much simpler to test
   what's going on.

 - In test_line(), just use grep -F, and unescape all second arguments
   passed to test_line(). This makes the script much more readable and
   actually fixes it for non-GNUish systems.
] 
[avoid re-checking for windows.
David Roundy <droundy at darcs.net>**20081011223905
 Ignore-this: 93c42ffc6b67c6cca803e1dcb18a9497
] 
[setup.hs: don't fail if darcs doesn't yet exist.
David Roundy <droundy at darcs.net>**20081011223824
 Ignore-this: 6d318961490a9343f5e792f2d740fd27
] 
[fix spelling error in setup.hs.
David Roundy <droundy at darcs.net>**20081011205353
 Ignore-this: 6d521f633d74aae8705391eff9489b2c
] 
[make setup.hs look for libcurl
David Roundy <droundy at darcs.net>**20081011195923
 Ignore-this: c9a267184e89b188f074171f8887b3b4
] 
[resolve issue1139: make special cases handle no arguments gracefully.
David Roundy <droundy at darcs.net>**20081011161059
 Ignore-this: 12b17a3940558860dd4caf7bb2619ec7
 I accidentally broke this when refactoring Darcs.Match.  This change
 restores some efficiency when no arguments are given and thus we are
 creating copies of all the files in the repository.
] 
[add tests for issue1139.
David Roundy <droundy at darcs.net>**20081011155324
 Ignore-this: ff50784059114f969fb4657ea5e92025
] 
[Copyright and licensing notes and rationales.
Trent W. Buck <trentbuck at gmail.com>**20081011035925] 
[Document DARCS_KEEP_TMPDIR in manual.
Dmitry Kurochkin <dmitry.kurochkin at gmail.com>**20081011115026
 Ignore-this: bd77d5a85fa44094dcd1a867c2a63d2e
] 
[Documentation for DARCS_KEEP_TMPDIR.
Dmitry Kurochkin <dmitry.kurochkin at gmail.com>**20081011113042
 Ignore-this: 5e3d6598aed4b681a099341c3ec5017a
] 
[Simplify Darcs.Lock.withDir.
Dmitry Kurochkin <dmitry.kurochkin at gmail.com>**20081011110639
 Ignore-this: c01257f9e0fc40c415562448cb52d67d
] 
[Do not remove temporary directories if DARCS_KEEP_TMPDIR environment variable is set.
Dmitry Kurochkin <dmitry.kurochkin at gmail.com>**20081011105808
 Ignore-this: 5ea0e206e45dc9be73c29c3bf89643b9
] 
[add new test that optimize --reorder behaves itself on lazy repos.
David Roundy <droundy at darcs.net>**20081010194834
 Ignore-this: 5246d20700c153490da1997caede4369
] 
[give more debug output in test.
David Roundy <droundy at darcs.net>**20081010191209
 Ignore-this: a13ec79ae69f4fe9621ea3c16c9b41d6
] 
[remove reimplementation of zipWith.
David Roundy <droundy at darcs.net>**20081010191051
 Ignore-this: 277f49de513ec84bc6db0b4da6c633b1
] 
[Fix spurious diff and missing cd in diff test.
Eric Kow <eric.kow at gmail.com>**20081008192142

 If tests are run in parallel, each instance of the diff test may
 attempt to create temporary directories for use by darcs diff.
 Darcs itself handles this fine; only the test script gets confused
 by the spurious difference that results.
] 
[Help the test harness clean to up after tests.
Eric Kow <eric.kow at gmail.com>**20081008190340
 A typical use case is if a test sets restrictive permissions
 and dies.  Notes:
 * we do this after each test so that one test result doesn't
   contaminate others
 * we explicitly do /not/ remove these temporary directories
   because you may want them around for forensics
] 
[ratify use of readFile in setup.hs
David Roundy <droundy at darcs.net>**20081010150105
 Ignore-this: 92627010ec07e16d271bcc5e7f325c64
] 
[Minor refactor in URL.urlThread.
Dmitry Kurochkin <dmitry.kurochkin at gmail.com>**20081010071649
 Ignore-this: 809d750f10adc8073ebe0dd3d432f2a2
] 
[clean up code formatting in Hopefully.
David Roundy <droundy at darcs.net>**20081010144943
 Ignore-this: 65a4dbb17505adbc0c0c615c445a8926
] 
[clean up unused exports in Hopefully.
David Roundy <droundy at darcs.net>**20081010144918
 Ignore-this: a016b66fd64f756295555255b268e80a
] 
[haddock documentation for Hopefully
florent.becker at ens-lyon.org**20081001081315] 
[resolve issue1062: ignore setCooked and setEcho errors.
Eric Kow <eric.kow at gmail.com>**20081010122047
 Ignore-this: 2f8c8710faefdcdb6b224244995a00af
 The issue in question is only about setCooked.
 I presume setEcho can similarly fail.
] 
[add script for testing the franchise build.
David Roundy <droundy at darcs.net>**20081010142711
 Ignore-this: 5aa40dc14c0827b199b9d23d1c7ef27d
 This includes testing the cross-compile to windows using wine.  It's
 part of my scripts that are run on each push to darcs-unstable.
] 
[add a --prompt-for-dependencies to restore the default behaviour
florent.becker at ens-lyon.org**20081010123240
 Ignore-this: 92480353c19dea1b67aff89b933759a2
 Add a --prompt-for-dependencies flag so that one can put
 --dont-prompt-for-dependencies in their defaults file and
 still get back the default behavior when needed.
] 
[eliminate autogeneration of Workaround.hs.
David Roundy <droundy at darcs.net>**20081010134727
 Ignore-this: 784a4b08bab1e22c93f54c1c5cd710f1
] 
[add franchise setup file.
David Roundy <droundy at darcs.net>**20081009203410
 Ignore-this: f1738c7608e6093aa18e808d3d41d000
] 
[add missing createLink to win32/System/Posix/Files.hsc.
David Roundy <droundy at darcs.net>**20081010133500
 Ignore-this: 39d246ab1db8c31f78e0b8362ad05ca3
] 
[make make_changelog default to quiet output.
David Roundy <droundy at darcs.net>**20081009203322
 Ignore-this: 55eba7a68a3c8a5683d23cc99848bd6f
 You can restore previous behavior by defining the VERBOSE environment
 variable.
] 
[remove unused and unneeded Workarounds.
David Roundy <droundy at darcs.net>**20081009194541
 Ignore-this: bca5a7a2444d4ddffeed673cbbfa19ec
] 
[rename lazy-deps to dont-prompt-for-dependencies.
David Roundy <droundy at darcs.net>**20081009171030
 Ignore-this: 78393e28c184c9339c89cecaa852dbc9
] 
[fix incompatibilities with ghc 6.6
David Roundy <droundy at darcs.net>**20081009165845
 Ignore-this: 18739c880a49b976ec151363f4ff9296
] 
[basic tests for --lazy-deps
Florent Becker <florent.becker at ens-lyon.org>**20081008124956
 Ignore-this: 79d8a312943813e9fe5448bc2ab86aa0
] 
[added a --lazy-deps option
Florent Becker <florent.becker at ens-lyon.org>**20081008122224
 Ignore-this: d934868e56f0084067f5aedf0512a533
 This option means that you don't get asked about patches which do not match
 a --match, but are depended upon by a patch that does. These patches will get
 ilently selected. 

] 
[refactor of SelectChanges
Florent Becker <florent.becker at ens-lyon.org>**20081008100551
 Ignore-this: e620d36d6e52eb62a2205fe5122d7076
] 
[refactor Match to handle --store-in-memory itself.
David Roundy <droundy at darcs.net>**20081008172748
 Ignore-this: 647c75f78d2d4981e67039ae1b6e96df
 This reduces the amount of duplicate code, at the cost of adding a
 class to Darcs.Match--but at least it isn't exported.
] 
[make writeSlurpy work properly when given "."
David Roundy <droundy at darcs.net>**20081008172207
 Ignore-this: 9d5d077bcdf3a9bb7bfa8477ab9d96d5
 This isn't a bugfix per se, but rather a feature addition.
 writeSlurpy has a "safety feature" that it refuses to overwrite a
 directory, but that safety feature doesn't really apply to the
 directory ".", since "." explicitly means that we want to overwrite
 the directory.
] 
[change a bit of code to accept FilePathLike arguments.
David Roundy <droundy at darcs.net>**20081008172047
 Ignore-this: 87032e071bb49117133d99a90bdfd0e1
 These functions really should expect FileName, and this is an
 incremental move in that direction.
] 
[Implement Commands.Check in terms of Repository.Repair functionality.
me at mornfall.net**20080926144006] 
[Implement Commands.Repair in terms of Repository.Repair functionality.
me at mornfall.net**20080926143919] 
[Export testTentative from Darcs.Repository.
me at mornfall.net**20080926143839] 
[Implement Repository.Repair to provide primitives for check and repair commands.
me at mornfall.net**20080926143721] 
[keep changepref patches from breaking the toSimple optimization.
David Roundy <droundy at darcs.net>**20080924162022
 Ignore-this: 404a4386b048b74dd9afcdde47bc1b11
] 
[rewrite push_coalesce_patch to avoid calls to lengthFL.
David Roundy <droundy at darcs.net>**20080924161517
 Ignore-this: cf746702c3b1e06ab2d400878e20e305
] 
[make various autoconf simplifications.
David Roundy <droundy at darcs.net>**20080923150742
 Ignore-this: 213c53eda0b54c01909c8d15b2a5fdd6
] 
[Restore send-external.sh test.
David Roundy <droundy at darcs.net>**20081009140121
 Ignore-this: 700f9c538033108cac397c1d757fbfc4
] 
[TAG 2.1.0
Eric Kow <E.Y.Kow at brighton.ac.uk>**20081009120532] 
[Rollback send-external test.
Eric Kow <E.Y.Kow at brighton.ac.uk>**20081009101224
 I was over-optimistic about pulling this into stable.
 It fails under Windows.

 rolling back:

 Tue Oct  7 23:28:04 BST 2008  David Roundy <droundy at darcs.net>
   * add test for send --sendmail-command.

     A ./tests/send-external.sh
] 
[ChangeLog entries for 2.1.0 (tweaks).
Eric Kow <E.Y.Kow at brighton.ac.uk>**20081009090247] 
[disable tests/issue1078_symlink.sh on Windows
zooko at zooko.com**20081001130248
 (Windows does not have symlinks.)
] 
[use minimum Cachable time in corner case where f/=f' and c/=c'.
David Roundy <droundy at darcs.net>**20081008173804
 Ignore-this: 51f2a68ac4139420bd116e8293cc5cc2
] 
[Bump version number to 2.1.0.
Eric Kow <E.Y.Kow at brighton.ac.uk>**20081008203033] 
[Canonize Benjamin Franksen.
Eric Kow <E.Y.Kow at brighton.ac.uk>**20081008202951] 
[ChangeLog entries for darcs 2.1.0.
Eric Kow <E.Y.Kow at brighton.ac.uk>**20081008202927] 
[add test of Ignore-this functionality.
David Roundy <droundy at darcs.net>**20081008192621
 Ignore-this: 47b7bf225e2997dd78c351763c769738
] 
[warn users if they try to record a change beginning with Ignore-this:.
David Roundy <droundy at darcs.net>**20081008192001
 Ignore-this: 94828726d609f46e384c2ab8091e49db
 Ignore-this: I am testing the functionality of this feature.
] 
[allow users to add their own ignored information to patches.
David Roundy <droundy at darcs.net>**20081008190038
 Ignore-this: fe8294776896b6fdcd489752a29d1069
] 
[haddock documentation for Printer
florent.becker at ens-lyon.org**20081007211959] 
[Use copyFile+renameFile for safe copy in URL.waitNextUrl.
Dmitry Kurochkin <dmitry.kurochkin at gmail.com>**20081008134308
 Ignore-this: a9a8889c2b57a5dca2785344cbd18a2f
] 
[remove excessive indentation that made the code very hard to read.
David Roundy <droundy at darcs.net>**20081008173358
 Ignore-this: d93e312c27f0d9a1ef993855c6872257
] 
[When requesting same URL with different cachability choose least cachable.
Dmitry Kurochkin <dmitry.kurochkin at gmail.com>**20081008145049
 Ignore-this: fddacf3e706a68c55871f81438d283b
] 
[add simple test of interactive record.
David Roundy <droundy at darcs.net>**20081008172036
 Ignore-this: ebb01acdb477174095e1b66aac230629
] 
[Improve test pull.sh cleanup
Thorkil Naur <naur at post11.tele.dk>**20081008115539] 
[resolve issue1124: Test pull.sh failed when run as root
Thorkil Naur <naur at post11.tele.dk>**20081008114134] 
[Allow dist command to take a tag argument (issue 286).
Stephen Hicks <sdh33 at cornell.edu>**20081006075133
 It should probably be a "match_one_context" instead of a "match_one"
 option type, and then instead of using the "nonranged" matcher, we'd
 use the "patchset" ones (that don't exist yet).  But I couldn't figure
 out how to blend "get_nonrange_match" and "get_one_patchset" in any
 sensical way to get immediate pristine output (and all the more so for
 the "_s" variants).  So for now, contexts are not supported, but it
 would be nice if they were someday.
] 
[clean up RepoPath a bit (removing unused methods from classes).
David Roundy <droundy at darcs.net>**20081008161906
 Ignore-this: 4e69882ebe10f20bb67350b106bd10af
] 
[clean up exportation of SlurmMonad private internals.
David Roundy <droundy at darcs.net>**20081008151758
 Ignore-this: 3763cc352b6f28fa83d8fbe5b1077985
] 
[Optimize clean_hashdir's use of cleanCaches.
Petr Rockai <me at mornfall.net>**20081007191237