plus a whole bunch of changes listed at the end of this message.
New in v0.6.21 (2013/01/23)
---------------------------
Enhancements:
- fix python 2.4 vs 2.5 syntax error
* Merged in lp:~mterry/duplicity/u1-oauthlib
- As the Ubuntu packager for duplicity, I would prefer u1backend.py
used oauthlib instead of oauth. oauthlib is well maintained upstream
(unlike oauth), has a python3 port (for the future), and is in Ubuntu
main (so is oauth right now, but hopefully in the future we can drop
it to universe, in which case duplicity can't use it anymore).
* Merged in lp:~mterry/duplicity/delete-new-sig-in-cache
- In duplicity 0.6.20, we fixed bug 1031269. This means that we no longer
leave sig files on the remote location. Leaving sig files on the remote
location also caused a bug with deleting cache files. Code used to leave
remote new-sig but delete the locale cache new-sig; this meant that we would
keep downloadoing the new-sig all the time from remote. We had worked around
that by just not deleting the new-sig in the cache, which was sort of the
wrong side of that problem to tackle. Now that we handle the remote
new-sigs better (by deleting them), I don't think we need this code anymore.
* Merged in lp:~mterry/duplicity/u1-ascii-error
- Fix for u1backend unicode error. Patch by Paul Barker.
* Merged in lp:~satwell/duplicity/caching
- Add a cache for password and group lookups. This significantly improves
runtime with very large password and group configurations.
- more formatting fixes, clarifications in sections EXAMPLES, FILE SELECTION
- Allow .netrc auth for lftp backend
* Merged in lp:~mterry/duplicity/946988
- This fixes bug 946988 by not duplicating the checks for when we should ask
for the password (those same checks are done more correctly inside
get_passphrase). And add a test to reproduce the bug.
* Merged in lp:~lenharo-h/duplicity/duplicity
- Generate encrypted backups without revealing the user's key id
via option --hidden-encrypt-key
* Merged in lp:~mterry/duplicity/u1-utf8
- Make sure u1backend returns filenames as utf8
* Merged in lp:~carlos-abalde/duplicity/gdocs-backend-gdata-2.0.16.-upgrade
- Upgrade of GoogleDocs backend to python gdata lib >= 2.0.15:
Stop using get_everything method.
- bugfix: webdav retrying broke on ERRORS like "error: [Errno 32] Broken pipe" in
added a more generalized 'retry_fatal' decorator which makes retrying backend
methods even easier
- Clear up PASSPHRASE reusage as sign passphrase. Minor fixes.
- Manpage
- document Ubuntu One required python libs
- added continuous contributors and backend author notes
- U1backend
- lazily import non standard python libs, fixes
- fix "not bytearray" prevents PUT with python 2.6
- don't hang after putting in credentials (cause it silently retries in background)
but go through with backup
* Fixed 1091269 Data corruption when resuming with --no-encryption
- Patches from Pascual Abellan that make block size consistent and
that add -n (no-encryption) option to manual-ctrl-c-test.sh.
- Modified gpg.py patch to use 64k block size so unit test passes.
* Merged in lp:~mterry/duplicity/static-corruption
- This branch fixes three possible ways a backup could get data-corrupted.
Inspired by bug 1091269.
A) If resuming after a volume that ended in a one-block file, we would
skip the first block of the next file.
B) If resuming after a volume that ended in a multi-block file, we would
skip the first block of the next file.
C) If resuming after a volume that spanned a multi-block file, we would
skip some data inside the file.
- A and B are because when finding the right place in the source files to
restart the backup, the iteration loop didn't handle None block numbers
very well (which are used to indicate the end of a file).
- C is what bug 1091269 talks about. This was because data block sizes would
get smaller as the difftar file got closer and closer to the volsize.
Standard block sizes were 64 * 1024. But say we were close to the end of
the difftar... When resuming, duplicity doesn't know the custom block sizes
used by the previous run, so it uses standard block sizes. And it doesn't
always match up, as you can imagine. So we would leave chunks of data out
of the backed up file.
- Tests added for these cases.
- This branch is called 'static-corruption' because all these issues occur
even when the source data doesn't change. I still think there are some
corruption issues when a file changes in between duplicity runs. I haven't
started looking into that yet, but that's next on my list.
- C only happened without encryption (because the gpg writer function already
happened to force a constant data block size). A and B happened with or
without encryption.
- added ssl certificate verification (see man page)
- more robust retry routine to survive ssl errors, broken pipe errors
- added http redirect support
- explanation of webdav changes above
* Merged in lp:~mterry/duplicity/pygi
- Python bindings for the gobject stack (used in the gio backend) have changed
from static to dynamically-generated bindings. The old static bindings are
deprecated. So here's a branch to change the gio backend from old to new ones.
* Merged in lp:~mterry/duplicity/py3rsync
- This branch lets one build the _librsync module with Python 3. You can't
really do anything useful with it, but it's a nicely-isolated piece to add
Python 3 support for.
- The changes are a mix of modernization and #ifdef logic.
- All tests still pass in Python 2.7 and 2.4. I tested manually that the module
worked as expected in Python 3.
* Merged in lp:~duplicity-team/duplicity/po-updates
- Updated translations