Hi all,
Here's an amended version of the patches I recently sent, which turned
out to still be buggy (as the test suite revealed). These are now
well and truly pushed, with all tests passing.
David
Tue Sep 16 11:20:28 EDT 2008 David Roundy <[EMAIL PROTECTED]>
* simplify issue965 test (which took quite a while for me to figure out).
Tue Sep 16 13:03:33 EDT 2008 David Roundy <[EMAIL PROTECTED]>
* first-stage fix for issue709.
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).
Tue Sep 16 13:30:02 EDT 2008 David Roundy <[EMAIL PROTECTED]>
* resolve issue709: avoid adding changes to pending in rmpend when possible.
Tue Sep 16 13:30:30 EDT 2008 David Roundy <[EMAIL PROTECTED]>
* resolve issue885: fix patchSetToRepository to work with hashed.
Tue Sep 16 13:31:16 EDT 2008 David Roundy <[EMAIL PROTECTED]>
* resolve issue1012: it seems to be fixed by the fix for issue709.
Tue Sep 16 13:31:36 EDT 2008 David Roundy <[EMAIL PROTECTED]>
* remove now-unused is_addrmfile and is_addrmdir.
New patches:
[simplify issue965 test (which took quite a while for me to figure out).
David Roundy <[EMAIL PROTECTED]>**20080916152028] hunk ./bugs/issue865_replace_in_moved.sh 28
darcs add file1
darcs record -a -m "0" --author X
-ed file1 <<EOF
-1s/aa/aaaa/
-w
-q
+cat <<EOF > file1
+aaaa
+bb
+cc
+aa
+bb
+cc
EOF
hunk ./bugs/issue865_replace_in_moved.sh 37
-darcs replace aa aaaa file1
+darcs replace aa aaaa file1 > out
+cat out
+grep 'Skipping file' out && exit 1
darcs mv file1 file2
hunk ./bugs/issue865_replace_in_moved.sh 43
-ed file2 <<EOF
-2s/bb/beebee/
-w
-q
+cat <<EOF > file2
+aaaa
+beebee
+cc
+aa
+bb
+cc
EOF
hunk ./bugs/issue865_replace_in_moved.sh 52
-! darcs replace bb beebee file2 2>&1 |
-tee /dev/tty |
-grep -q 'Skipping file'
-exit=$?
+darcs replace bb beebee file2 > out
+cat out
+grep 'Skipping file' out && exit 1
hunk ./bugs/issue865_replace_in_moved.sh 56
-cd .. && rm -rf temp
-
-exit $exit
+cd ..
+rm -rf temp
[first-stage fix for issue709.
David Roundy <[EMAIL PROTECTED]>**20080916170333
Here I fix the bug which leads to a corrupt pending being left, with a
rmfile change but no hunk removing the contents. This doesn't fix
issue709, since an incorrect pending is left, it's just no longer a
corrupt pending (i.e. it's still got the rmfile, but if you record it
there's no problem).
] hunk ./src/Darcs/Repository/Internal.lhs 113
patch2patchinfo, readPatch,
writePatch, effect, invert,
is_addfile, is_adddir,
- is_addrmfile, is_addrmdir, is_setpref,
+ is_setpref,
apply, apply_to_slurpy,
empty_markedup_file, MarkedUpFile
)
hunk ./src/Darcs/Repository/Internal.lhs 474
fromPrims_ = fromPrims
is_simple :: Prim C(x y) -> Bool
-is_simple x = is_hunk x || is_binary x || is_setpref x ||
- is_addrmfile x || is_addrmdir x
+is_simple x = is_hunk x || is_binary x || is_setpref x
crude_sift :: FL Prim C(x y) -> FL Prim C(x y)
crude_sift xs = if allFL is_simple xs then filterFL ishunkbinary xs else xs
hunk ./bugs/issue709_pending_look-for-adds.sh 3
#!/bin/sh
set -ve
+
+## I would use the builtin !, but that has the wrong semantics.
+not () { "$@" && exit 1 || :; }
+
rm -rf temp1
mkdir temp1
cd temp1
hunk ./bugs/issue709_pending_look-for-adds.sh 11
darcs init
+
+# Here we check whether recording just one of two --look-for-add
+# addfiles causes any trouble (which it doesn't)
+
+date > f1
+date > f2
+echo yyd | darcs record -l -m ff
+
+cat _darcs/patches/pending
+
+not darcs wh
+rm f2
+
+# Try recording a file add without --look-for-adds, with a setpref
+# patch present that we don't record.
+
darcs setpref boringfile .boring
hunk ./bugs/issue709_pending_look-for-adds.sh 28
-echo bar > foo
-echo yyd | darcs record --look-for-adds -mP1
+
+echo bar > bar
+darcs add bar
+echo yyd | darcs record -mbar
+
+cat _darcs/patches/pending
+
darcs whatsnew -s
hunk ./bugs/issue709_pending_look-for-adds.sh 36
+
test -z "`darcs whatsnew -s`"
hunk ./bugs/issue709_pending_look-for-adds.sh 38
+
+# Now try the same thing using --look-for-adds
+
+echo foo > foo
+
+darcs wh -l
+
+echo yyd | darcs record --look-for-adds -mfoo
+
+cat _darcs/patches/pending
+
+darcs whatsnew -s
+
+test -z "`darcs whatsnew -s`"
+
+cd ..
+rm -rf temp1
[resolve issue709: avoid adding changes to pending in rmpend when possible.
David Roundy <[EMAIL PROTECTED]>**20080916173002] move ./bugs/issue709_pending_look-for-adds.sh ./tests/issue709_pending_look-for-adds.sh
hunk ./src/Darcs/Repository/Internal.lhs 75
import Workaround ( createDirectoryIfMissing, getCurrentDirectory, renameFile, setExecutable )
import FastPackedString ( PackedString, readFilePS, gzReadFilePS, nilPS, packString, takePS )
import Darcs.Patch ( Patch, RealPatch, Effect, is_hunk, is_binary, description,
- try_to_shrink, commuteFL )
+ try_to_shrink, commuteFL, commute )
import Darcs.Patch.Prim ( try_shrinking_inverse, Conflict )
import Darcs.Patch.Bundle ( scan_bundle, make_bundle )
import Darcs.SlurpDirectory ( Slurpy, slurp_unboring, mmap_slurp, co_slurp,
hunk ./src/Darcs/Repository/Internal.lhs 133
getCaches )
import Darcs.Lock ( withLock, writeDocBinFile, withDelayedDir, removeFileMayNotExist,
withTempDir, withPermDir )
-import Darcs.Sealed ( Sealed(Sealed), mapSeal, seal, FlippedSeal(FlippedSeal), flipSeal )
+import Darcs.Sealed ( Sealed(Sealed), seal, FlippedSeal(FlippedSeal), flipSeal )
import Darcs.Repository.InternalTypes( Repository(..), RepoType(..) )
import Darcs.Global ( darcsdir )
#include "impossible.h"
hunk ./src/Darcs/Repository/Internal.lhs 468
rmpend NilFL x = Sealed x
rmpend _ NilFL = Sealed NilFL
rmpend (x:>:xs) xys | Just ys <- removeFL x xys = rmpend xs ys
- rmpend (x:>:xs) ys = case commuteWhatWeCanFL (x:>xs) of
- a:>x':>b -> (invert (x':>:b) +>+) `mapSeal` rmpend a ys
+ rmpend (x:>:xs) ys =
+ case commuteWhatWeCanFL (x:>xs) of
+ a:>x':>b -> case rmpend a ys of
+ Sealed ys' -> case commute (invert (x':>:b) :> ys') of
+ Just (ys'' :> _) -> seal ys''
+ Nothing -> seal $ invert (x':>:b)+>+ys'
+ -- DJR: I don't think this
+ -- last case should be
+ -- reached, but it also
+ -- shouldn't lead to
+ -- corruption.
fromPrims_ :: FL Prim C(a b) -> Patch C(a b)
fromPrims_ = fromPrims
[resolve issue885: fix patchSetToRepository to work with hashed.
David Roundy <[EMAIL PROTECTED]>**20080916173030] move ./bugs/issue885_get_to_match.sh ./tests/issue885_get_to_match.sh
hunk ./src/Darcs/Commands/Get.lhs 159
debugMessage "Using economical get --to-match handling"
fromrepo <- identifyRepositoryFor repository repodir
Sealed patches_to_get <- get_one_patchset fromrepo opts
- patchSetToRepository patches_to_get opts
+ patchSetToRepository fromrepo patches_to_get opts
debugMessage "Finished converting selected patch set to new repository"
else copy_repo_and_go_to_chosen_version opts repodir rfsource rf putInfo
where am_informative = not $ Quiet `elem` opts
hunk ./src/Darcs/Repository.lhs 237
-- repository with the --to-match flag and the new repository is not in hashed format.
-- This function does not (yet) work for hashed repositories. If the passed @[EMAIL PROTECTED] tell
-- darcs to create a hashed repository, this function will call @[EMAIL PROTECTED]
-patchSetToRepository :: RepoPatch p => PatchSet p C(x) -> [DarcsFlag] -> IO (Repository p C(r u t))
-patchSetToRepository patchset opts = do
+patchSetToRepository :: RepoPatch p => Repository p C(r1 u1 r1) -> PatchSet p C(x)
+ -> [DarcsFlag] -> IO (Repository p C(r u t))
+patchSetToRepository (Repo fromrepo _ rf _) patchset opts = do
+ when (format_has HashedInventory rf) $ -- set up sources and all that
+ do writeFile "_darcs/tentative_pristine" "" -- this is hokey
+ repox <- writePatchSet patchset opts
+ HashedRepo.copy_repo repox opts fromrepo
repo <- writePatchSet patchset opts
read_repo repo >>= (apply_patches opts . reverseRL . concatRL)
debugMessage "Writing the pristine"
[resolve issue1012: it seems to be fixed by the fix for issue709.
David Roundy <[EMAIL PROTECTED]>**20080916173116] move ./bugs/issue1012_unrecord_remove.sh ./tests/issue1012_unrecord_remove.sh
[remove now-unused is_addrmfile and is_addrmdir.
David Roundy <[EMAIL PROTECTED]>**20080916173136] hunk ./src/Darcs/Patch.lhs 56
showContextPatch, showPatch, showNicely,
infopatch, changepref,
thing, things,
- is_similar, is_addfile, is_addrmfile, is_hunk, is_setpref,
+ is_similar, is_addfile, is_hunk, is_setpref,
#ifndef GADT_WITNESSES
merger, is_merger, merge,
commute, commutex, list_touched_files,
hunk ./src/Darcs/Patch.lhs 67
#endif
resolve_conflicts,
Effect, effect,
- is_binary, gzWritePatch, writePatch, is_adddir, is_addrmdir,
+ is_binary, gzWritePatch, writePatch, is_adddir,
invert, invertFL, invertRL, identity,
commuteFL, commuteRL,
readPatch,
hunk ./src/Darcs/Patch.lhs 126
sort_coalesceFL,
rmdir, rmfile, tokreplace, adddir, addfile,
binary, changepref, hunk, move,
- is_adddir, is_addrmdir, is_addfile, is_addrmfile,
+ is_adddir, is_addfile,
is_hunk, is_binary, is_setpref,
is_similar,
try_to_shrink )
hunk ./src/Darcs/Patch/Prim.lhs 33
formatFileName, FileNameFormat(..),
adddir, addfile, binary, changepref,
hunk, move, rmdir, rmfile, tokreplace,
- is_addrmfile, is_addfile, is_hunk, is_binary, is_setpref,
- is_similar, is_adddir, is_addrmdir, is_filepatch,
+ is_addfile, is_hunk, is_binary, is_setpref,
+ is_similar, is_adddir, is_filepatch,
canonize, try_to_shrink, modernizePrim,
subcommutes, sort_coalesceFL, join,
applyBinary, try_tok_internal,
hunk ./src/Darcs/Patch/Prim.lhs 136
is_adddir (DP _ AddDir) = True
is_adddir _ = False
-is_addrmfile :: Prim C(x y) -> Bool
-is_addrmfile (FP _ AddFile) = True
-is_addrmfile (FP _ RmFile) = True
-is_addrmfile _ = False
-
-is_addrmdir :: Prim C(x y) -> Bool
-is_addrmdir (DP _ AddDir) = True
-is_addrmdir (DP _ RmDir) = True
-is_addrmdir _ = False
-
is_hunk :: Prim C(x y) -> Bool
is_hunk (FP _ (Hunk _ _ _)) = True
is_hunk _ = False
Context:
[Make match.pl test understand Windows dates.
Eric Kow <[EMAIL PROTECTED]>**20080916011339
Windows does not abbreviate its timezones.
]
[Disable some tests that don't work under Windows.
Eric Kow <[EMAIL PROTECTED]>**20080916000912]
[Translate get.pl test into shell.
Eric Kow <[EMAIL PROTECTED]>**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 <[EMAIL PROTECTED]>**20080915224046]
[Move get_default_flag type to definition.
Dmitry Kurochkin <[EMAIL PROTECTED]>**20080915220316]
[reverse overly-invasive haddocking.
David Roundy <[EMAIL PROTECTED]>**20080915110353]
[In darcs send if POST fails try sendmail.
Dmitry Kurochkin <[EMAIL PROTECTED]>**20080914234314]
[Grammar in Darcs.Commands.Send.
Dmitry Kurochkin <[EMAIL PROTECTED]>**20080914232923]
[Print "Successfully sent patch bundle to" only when we really sent something.
Dmitry Kurochkin <[EMAIL PROTECTED]>**20080914232905]
[Fix overzealous escaping in haddock.
Eric Kow <[EMAIL PROTECTED]>**20080915070926]
[Haddock documentation for English.lhs
[EMAIL PROTECTED]
Docstrings by Eric Kow
]
[Haddock documentation for PrintPatch
[EMAIL PROTECTED]
[Haddock documentation for Flags
[EMAIL PROTECTED]
[Haddock documentation for Motd
[EMAIL PROTECTED]
docstring by Eric Kow
]
[Haddock string for TheCommands
[EMAIL PROTECTED]
Docstring by Eric Kow
]
[Run testing in temp directories to avoid collosions, fail if DarcsURL header is not found.
Dmitry Kurochkin <[EMAIL PROTECTED]>**20080914223930]
[Use tempfile() UNLINK option to automatically remove temp files at exit.
Dmitry Kurochkin <[EMAIL PROTECTED]>**20080914223827]
[Coding style in upload.cgi.
Dmitry Kurochkin <[EMAIL PROTECTED]>**20080914223751]
[Stop after we found the first DarcsURL: in patch bundle.
Dmitry Kurochkin <[EMAIL PROTECTED]>**20080914220421]
[Spaces in upload.cgi.
Dmitry Kurochkin <[EMAIL PROTECTED]>**20080914220324]
[Really allow pull.sh test to pass on Windows.
Eric Kow <[EMAIL PROTECTED]>**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.
]
[rewrite upload.cgi so it won't need any customization by default.
David Roundy <[EMAIL PROTECTED]>**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 <[EMAIL PROTECTED]>**20080913171025]
[sketchy documentation of _darcs/prefs/post
David Roundy <[EMAIL PROTECTED]>**20080913115655]
[ChangeLog entries for darcs 2.0.3pre1
Eric Kow <[EMAIL PROTECTED]>**20080914094144]
[Disable amend-cancelling test under Cygwin.
Eric Kow <[EMAIL PROTECTED]>**20080913213039]
[Make binary.sh test more portable (avoid copying an exe).
Eric Kow <[EMAIL PROTECTED]>**20080913212843
Under Windows, copying the exe file will result in an .exe extension
being appended to the filename. This confuses the test.
]
[set default upload.cgi to work on darcs-unstable.
David Roundy <[EMAIL PROTECTED]>**20080913112227]
[Improve upload.cgi.
Dmitry Kurochkin <[EMAIL PROTECTED]>**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 <[EMAIL PROTECTED]>**20080912230827]
[fix problem in Darcs.URL refactor pointed out by Eric.
David Roundy <[EMAIL PROTECTED]>**20080913104327]
[Disable a pull.sh test under Windows.
Eric Kow <[EMAIL PROTECTED]>**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 <[EMAIL PROTECTED]>**20080912173611
The algorithms here are not tricky, and I find this easier to read.
]
[change is_file to return false on [EMAIL PROTECTED]:
David Roundy <[EMAIL PROTECTED]>**20080912173501]
[clean up whitespace.
David Roundy <[EMAIL PROTECTED]>**20080912150248]
[fix manual for optional arguments.
David Roundy <[EMAIL PROTECTED]>**20080912150231]
[clean up whitespace.
David Roundy <[EMAIL PROTECTED]>**20080912145708]
[add test for new --output-auto-name feature.
David Roundy <[EMAIL PROTECTED]>**20080912145648]
[Spaces in Darcs.Commands.Send module.
Dmitry Kurochkin <[EMAIL PROTECTED]>**20080912122223]
[Make '--output-auto-name' accept optional directory argument.
Dmitry Kurochkin <[EMAIL PROTECTED]>**20080912120516]
[Add DarcsOptAbsPathOption for options with optional path argument.
Dmitry Kurochkin <[EMAIL PROTECTED]>**20080912112458]
[Refactor Darcs.Repository.Prefs.getCaches.
Dmitry Kurochkin <[EMAIL PROTECTED]>**20080912055204]
[Print warning when '--http-pipelining' option is used, but darcs is compiled without HTTP pipelining support.
Dmitry Kurochkin <[EMAIL PROTECTED]>**20080912054253]
[Do not download URL we have speculated before.
Dmitry Kurochkin <[EMAIL PROTECTED]>**20080912053236]
[Spaces and parentheses in URL module.
Dmitry Kurochkin <[EMAIL PROTECTED]>**20080912053000]
[Coding style in Darcs.Arguments.network_options.
Dmitry Kurochkin <[EMAIL PROTECTED]>**20080911140710]
[Resolve issue1054: --no-cache option to ignore patch caches.
Dmitry Kurochkin <[EMAIL PROTECTED]>**20080911140233]
[Remove unused variable from configure.ac.
Dmitry Kurochkin <[EMAIL PROTECTED]>**20080911132107]
[Comment in configure.ac.
Dmitry Kurochkin <[EMAIL PROTECTED]>**20080911115840]
[Indentation fixes in configure.ac.
Dmitry Kurochkin <[EMAIL PROTECTED]>**20080911115117]
[Formating and minor refactoring in URL.urlThread.
Dmitry Kurochkin <[EMAIL PROTECTED]>**20080910061227]
[insecure printfs - fix the two that I can currently hit
Steve Cotton <[EMAIL PROTECTED]>**20080910230659]
[TAG this version works.
David Roundy <[EMAIL PROTECTED]>**20080910212908]
Patch bundle hash:
80573d015d495c0c6fed4517c30f9d6754620194
_______________________________________________
darcs-users mailing list
[email protected]
http://lists.osuosl.org/mailman/listinfo/darcs-users
