mirror/borg
1
0
Fork 0
mirror of https://github.com/borgbackup/borg.git synced 2024-12-25 09:19:31 +00:00
Code Issues Releases Wiki Activity
3408 commits 9 branches 139 tags 60 MiB
Commit graph

3408 commits

This branch
This branch
All branches
Author SHA1 Message Date
TW
b10426397f Merge pull request #1964 from enkore/issue/1759 
Fix processing of remote ~/ and ~user/ paths
 2016-12-18 21:23:06 +01:00
enkore
0c959cb67a pytest 2 compat 2016-12-17 15:58:51 +01:00
Marian Beermann
84d13751d5 Fix processing of remote ~/ and ~user/ paths 2016-12-17 14:33:40 +01:00
enkore
4225009e60 Merge pull request #1955 from ThomasWaldmann/merge-1.0-maint 
merge 1.0-maint
 2016-12-17 14:11:11 +01:00
Marian Beermann
63ce627a35 fix in-file checkpoints when clock jumps 2016-12-17 13:59:37 +01:00
Marian Beermann
a9db2a2e55 Merge branch '1.0-maint' into master 
# Conflicts:
#	src/borg/archive.py
#	src/borg/archiver.py
#	src/borg/helpers.py
 2016-12-17 13:26:28 +01:00
Marian Beermann
87f8b2bb64 Merge branch 'merge-1.0-maint' of https://github.com/ThomasWaldmann/borg into ThomasWaldmann-merge-1.0-maint 2016-12-17 13:21:58 +01:00
enkore
49c05719d8 Merge pull request #1941 from enkore/issue/1936 
Fix some clock problems
 2016-12-17 13:20:56 +01:00
Marian Beermann
baa8baafdb create: fix duration if --timestamp is given 2016-12-17 12:55:16 +01:00
Marian Beermann
445365b3ff update changes 2016-12-17 12:00:25 +01:00
Marian Beermann
f5d6093ccc fix checkpoints when clock jumps 2016-12-17 11:58:26 +01:00
Marian Beermann
a8d921a54c base archive timestamps on end time 
The assumption is that if the clock jumps during the Borg run that it
was jump-corrected and is now correct, while the start timestamp would
be wrong.
 2016-12-17 11:58:26 +01:00
Marian Beermann
2dc558a02e fix create progress not updating if clock jumps 2016-12-17 11:58:25 +01:00
Marian Beermann
420c984f05 fix wrong duration if clock jumps during create 2016-12-17 11:58:25 +01:00
enkore
9396aacb21 Merge pull request #1962 from ThomasWaldmann/win10-linux-subsys 
document windows 10 linux subsystem install
 2016-12-17 11:34:25 +01:00
enkore
20d3eff8c4 Merge pull request #1963 from ThomasWaldmann/fix-win10-lxsys-tests 
catch errno.ENOSYS for mknod (win 10 lxsys)
 2016-12-17 11:33:52 +01:00
Thomas Waldmann
61370082d6 catch errno.ENOSYS for mknod (win 10 lxsys) 
mknod raises this when running as non-root under Windows 10's Linux Subsystem.
 2016-12-17 00:37:00 +01:00
Thomas Waldmann
6b6ddecd93 document windows 10 linux subsystem install 
also add note about remote repos being broken on cygwin.
 2016-12-17 00:25:49 +01:00
TW
4c768624dc Merge pull request #1957 from enkore/f/bio-failop 
mention failed operation in per-file warnings
 2016-12-16 21:10:13 +01:00
TW
4f0c2ab68c Merge pull request #1959 from ThomasWaldmann/pretty-test-fails2 
pytest: only rewrite the testsuite, fixes #1938
 2016-12-16 00:25:24 +01:00
Thomas Waldmann
04340ae8b1 pytest: only rewrite the testsuite, fixes #1938 
do not rewrite the borg application code, just the test code,
so the bytecode tested is identical / very close to the bytecode used in practice.
 2016-12-15 03:02:06 +01:00
Marian Beermann
34e19ccb6a mention failed operation in per-file warnings 
on the one hand one can say it's ugly global state, on the other it's
totally handy!

just have to keep that in mind for MT, but it's rather obvious.
 2016-12-14 15:20:08 +01:00
Thomas Waldmann
82cd1fd392 run setup.py build_usage 2016-12-14 03:42:38 +01:00
Thomas Waldmann
c855c687eb Merge branch '1.0-maint' 2016-12-14 03:33:53 +01:00
TW
7f63ca8922 Merge pull request #1953 from ThomasWaldmann/fix-1932 
add a borg debug/key dummy command, fixes #1932
 2016-12-14 02:50:08 +01:00
Thomas Waldmann
5a40870416 add a borg debug/key dummy command, fixes #1932 
the problem was that there neither was a do_debug implementation for the case
someone just enters "borg debug", nor did the parser inherit from common_parser
(so accessing .umask triggered an exception before setup_logging() was called,
which triggered another exception when log output should have been emitted).

same for do_key ("borg key").

added a generic handler that just prints the subcommand help.
 2016-12-14 01:57:48 +01:00
TW
60bbd7a944 update CHANGES (1.0-maint) (#1954) 2016-12-14 01:29:43 +01:00
TW
d13854853c Merge pull request #1937 from ThomasWaldmann/location-parser 
fix the user part parsing in the Location parser
 2016-12-13 23:12:53 +01:00
Thomas Waldmann
292ff42655 refactor common regex part into optional_user_re 2016-12-13 22:47:12 +01:00
TW
ccf882a885 Merge pull request #1945 from enkore/ThomasWaldmann-update-1.1-changes 
Update 1.1 changes
 2016-12-13 22:32:06 +01:00
enkore
60b6f5a30e Merge pull request #1951 from ThomasWaldmann/pretty-test-fails 
get back pytest's pretty assertion failures, fixes #1938
 2016-12-13 20:58:58 +01:00
Thomas Waldmann
c6017abfb7 get back pytest's pretty assertion failures, fixes #1938 2016-12-13 19:27:01 +01:00
enkore
796ea9edde Merge pull request #1939 from ThomasWaldmann/update-1.0-changes 
update CHANGES (1.0-maint)
 2016-12-11 15:20:22 +01:00
Marian Beermann
9147c7038b update 1.1 beta CHANGES 2016-12-11 15:18:52 +01:00
Thomas Waldmann
09b1079b18 update CHANGES (master / 1.1 beta) 2016-12-11 06:20:10 +01:00
enkore
9d315384b4 Merge pull request #1940 from enkore/f/cache-sync 
Cache sync stuff
 2016-12-09 21:25:37 +01:00
Thomas Waldmann
bd8b4a4489 update CHANGES (1.0-maint) 2016-12-09 04:42:23 +01:00
Thomas Waldmann
335d599db4 fix location parser for archives with @ char, add test, fixes #1930 
we must exclude colon and slash chars from the username, otherwise the term for the user part
will match everything up to a @ char in the archive name.

a slash can't be in a username as the home directory would contain a illegal slash (slash is path sep),

a colon likely also should not be in a username because chown user:group ... syntax.
 2016-12-09 03:37:13 +01:00
Marian Beermann
be18418b74 cache: no archive caches => work directly on master cache (no merges) 2016-12-08 22:45:59 +01:00
Marian Beermann
e169510116 cache: don't create Item in fetch_and_build_idx 2016-12-08 22:39:04 +01:00
enkore
4affad73d1 Merge pull request #1924 from ThomasWaldmann/pr-template 
add a PR template pointing to guidelines
 2016-12-03 20:40:33 +01:00
Thomas Waldmann
dfd37d09d5 add a PR template pointing to guidelines 2016-12-03 18:34:34 +01:00
enkore
7e5ed40e2f Merge pull request #1911 from enkore/f/itemnt 
Work on metadata handling speed
 2016-12-03 18:06:27 +01:00
Marian Beermann
b7eaeee266 clean imports, remove unused code 2016-12-03 17:50:50 +01:00
Marian Beermann
8b2e7ec680 don't do "bigint" conversion for nanosecond mtime 
2**63 nanoseconds are 292 years, so this change is good until 2262.
See also https://en.wikipedia.org/wiki/Time_formatting_and_storage_bugs#Year_2262

I expect that we will have plenty of time to revert this commit in time
for 2262.

timespec := time_t + long, so it's probably only 64 bits on some platforms
anyway.
 2016-12-03 11:52:49 +01:00
Marian Beermann
b3707f7175 Replace backup_io with a singleton 
This is some 15 times faster than @contextmanager, because no instance
creation is involved and no generator has to be maintained. Overall
difference is low, but still nice for a very simple change.
 2016-12-03 11:52:48 +01:00
Marian Beermann
b885841c39 make item native code 
This makes an surprisingly large difference. Test case: ~70000 empty files.
(Ie. little data shoveling, lots of metadata shoveling). Before: 9.1 seconds
+- 0.1 seconds. After: 8.4 seconds +- 0.1 seconds.). That's a huge
win for changing a few lines.

I'd expect that this improves performance in almost all areas that touch
the items (list, delete, prune).
 2016-12-03 00:11:36 +01:00
enkore
e9d7f928e2 Merge pull request #1921 from enkore/f/recreate-fixes 
recreate fixes
 2016-12-03 00:02:01 +01:00
Marian Beermann
a9395dd8b1 recreate: don't rechunkify unless explicitly told so 2016-12-02 20:19:59 +01:00
Marian Beermann
30df63c509 recreate: remove special-cased --dry-run 2016-12-02 18:15:11 +01:00