Groups
Groups
Sign in
Groups
Groups
bup-list
Conversations
About
Send feedback
Help
bup-list
1–30 of 3242
Mark all as read
Report group
0 selected
Mark Hewitt
, …Â
Rob Browning
10
Apr 1
Possible race condition in midx
Mark J Hewitt <m...@idnet.com> writes: > for idx in $(find -name "*.idx"); do git
unread,
Possible race condition in midx
Mark J Hewitt <m...@idnet.com> writes: > for idx in $(find -name "*.idx"); do git
Apr 1
Martha David
2
Mar 31
Duct and dryer vent cleaning cleaning in just $199 Only!
2 slots left let me know if anyone is interested? On Fri, Mar 6, 2026, 00:28 Martha David <
unread,
Duct and dryer vent cleaning cleaning in just $199 Only!
2 slots left let me know if anyone is interested? On Fri, Mar 6, 2026, 00:28 Martha David <
Mar 31
Rob Browning
Mar 28
Fix prune-older --pretend format; test more get invocations
A few more minor fixes and test improvements. Pushed to main. -- Rob Browning rlb @defaultvalue.org
unread,
Fix prune-older --pretend format; test more get invocations
A few more minor fixes and test improvements. Pushed to main. -- Rob Browning rlb @defaultvalue.org
Mar 28
Rob Browning
Mar 28
[PATCH 3/4] get: move check for missing targets to parse_args()
Signed-off-by: Rob Browning <r...@defaultvalue.org> Tested-by: Rob Browning <rlb@defaultvalue
unread,
[PATCH 3/4] get: move check for missing targets to parse_args()
Signed-off-by: Rob Browning <r...@defaultvalue.org> Tested-by: Rob Browning <rlb@defaultvalue
Mar 28
Rob Browning
Mar 28
[PATCH 2/4] prune-older: add missing "keeping all KIND" newline
Signed-off-by: Rob Browning <r...@defaultvalue.org> Tested-by: Rob Browning <rlb@defaultvalue
unread,
[PATCH 2/4] prune-older: add missing "keeping all KIND" newline
Signed-off-by: Rob Browning <r...@defaultvalue.org> Tested-by: Rob Browning <rlb@defaultvalue
Mar 28
Rob Browning
Mar 28
[PATCH 4/4] get: test additional invalid invocations
Signed-off-by: Rob Browning <r...@defaultvalue.org> Tested-by: Rob Browning <rlb@defaultvalue
unread,
[PATCH 4/4] get: test additional invalid invocations
Signed-off-by: Rob Browning <r...@defaultvalue.org> Tested-by: Rob Browning <rlb@defaultvalue
Mar 28
Rob Browning
Mar 28
[PATCH 1/4] prune-older: remove extra space in --pretend output
Match bup-prune-older(1). Signed-off-by: Rob Browning <r...@defaultvalue.org> Tested-by: Rob
unread,
[PATCH 1/4] prune-older: remove extra space in --pretend output
Match bup-prune-older(1). Signed-off-by: Rob Browning <r...@defaultvalue.org> Tested-by: Rob
Mar 28
Rob Browning
3
Mar 27
[PATCH 7/8] rm: test --unsafe arg
Cover some additional error cases. Signed-off-by: Rob Browning <r...@defaultvalue.org> Tested-by
unread,
[PATCH 7/8] rm: test --unsafe arg
Cover some additional error cases. Signed-off-by: Rob Browning <r...@defaultvalue.org> Tested-by
Mar 27
Rob Browning
Mar 27
[PATCH 5/8] test_render_url: add some missing cases
Signed-off-by: Rob Browning <r...@defaultvalue.org> Tested-by: Rob Browning <rlb@defaultvalue
unread,
[PATCH 5/8] test_render_url: add some missing cases
Signed-off-by: Rob Browning <r...@defaultvalue.org> Tested-by: Rob Browning <rlb@defaultvalue
Mar 27
Rob Browning
Mar 27
[PATCH 1/8] Revert "gc: refuse to proceed when there are no refs"
Revert this for now because if we want it, it will require further accommodations, eg it breaks the
unread,
[PATCH 1/8] Revert "gc: refuse to proceed when there are no refs"
Revert this for now because if we want it, it will require further accommodations, eg it breaks the
Mar 27
Rob Browning
Mar 27
[PATCH 8/8] Treat all cli parse_num ValueErrors as misuses
Previously, we'd just crash with a backtrace. Catch the anticipated ValueErrors and treat them as
unread,
[PATCH 8/8] Treat all cli parse_num ValueErrors as misuses
Previously, we'd just crash with a backtrace. Catch the anticipated ValueErrors and treat them as
Mar 27
Rob Browning
Mar 27
[PATCH 6/8] test_split_trees: remove executable bit
Thanks to Johannes Berg for mentioning it. Signed-off-by: Rob Browning <r...@defaultvalue.org>
unread,
[PATCH 6/8] test_split_trees: remove executable bit
Thanks to Johannes Berg for mentioning it. Signed-off-by: Rob Browning <r...@defaultvalue.org>
Mar 27
Rob Browning
Mar 27
[PATCH 3/8] test-cat-file: test more invalid invocations
Signed-off-by: Rob Browning <r...@defaultvalue.org> Tested-by: Rob Browning <rlb@defaultvalue
unread,
[PATCH 3/8] test-cat-file: test more invalid invocations
Signed-off-by: Rob Browning <r...@defaultvalue.org> Tested-by: Rob Browning <rlb@defaultvalue
Mar 27
Rob Browning
Mar 27
[PATCH 4/8] parse_bytes_path_url: remove some unreachable "no authority" code
The _userinfo_host_port_rx will always match when auth isn't empty, and we've already handled
unread,
[PATCH 4/8] parse_bytes_path_url: remove some unreachable "no authority" code
The _userinfo_host_port_rx will always match when auth isn't empty, and we've already handled
Mar 27
Sarah Tylor0160
Mar 19
$180 For Each Duct, Dryer and Chimney cleaning Group rate
Just wanted to let you know that me and a few neighbors had our ducts,dryer and Chimney cleaned by
unread,
$180 For Each Duct, Dryer and Chimney cleaning Group rate
Just wanted to let you know that me and a few neighbors had our ducts,dryer and Chimney cleaned by
Mar 19
Rob Browning
5
Mar 15
Simplfy drecurse and don't chdir during traversal
Switch to open_noatime so that we'll avoid changing the atime when possible, and replace
unread,
Simplfy drecurse and don't chdir during traversal
Switch to open_noatime so that we'll avoid changing the atime when possible, and replace
Mar 15
Rob Browning
Mar 15
[PATCH 2/7] recursive_dirlist: remove redundant top-level path stat
We already have an lstat of the path, and from that we know it's a directory, so relying on the
unread,
[PATCH 2/7] recursive_dirlist: remove redundant top-level path stat
We already have an lstat of the path, and from that we know it's a directory, so relying on the
Mar 15
Rob Browning
Mar 15
[PATCH 3/7] drecurse: simplify the logic and only normpath() once
Flatten the logic via more loop continue short-circuting, and remember the normalized path if we
unread,
[PATCH 3/7] drecurse: simplify the logic and only normpath() once
Flatten the logic via more loop continue short-circuting, and remember the normalized path if we
Mar 15
Rob Browning
Mar 15
[PATCH 1/7] Add openat_noatime wrapping openat(2) and matching open_noatime
This will be used by a forthcoming file descriptor based drecurse that doesn't chdir. Signed-off-
unread,
[PATCH 1/7] Add openat_noatime wrapping openat(2) and matching open_noatime
This will be used by a forthcoming file descriptor based drecurse that doesn't chdir. Signed-off-
Mar 15
Rob Browning
Mar 15
Fix O_NOATIME failures introduced by recent hashsplit change
The O_NOATIME EPERM rejections that our wrapper suggested were long lost behaviors are alive and well
unread,
Fix O_NOATIME failures introduced by recent hashsplit change
The O_NOATIME EPERM rejections that our wrapper suggested were long lost behaviors are alive and well
Mar 15
Rob Browning
Mar 15
[PATCH 1/3] restore: always return to parent after child restore attempt
Signed-off-by: Rob Browning <r...@defaultvalue.org> Tested-by: Rob Browning <rlb@defaultvalue
unread,
[PATCH 1/3] restore: always return to parent after child restore attempt
Signed-off-by: Rob Browning <r...@defaultvalue.org> Tested-by: Rob Browning <rlb@defaultvalue
Mar 15
Rob Browning
Mar 15
[PATCH 3/3] save: use open_noatime, not os.open(..., O_NOATIME)
Switch back to our open_noatime because os.open with O_NOATIME may die with EPERM in some situations
unread,
[PATCH 3/3] save: use open_noatime, not os.open(..., O_NOATIME)
Switch back to our open_noatime because os.open with O_NOATIME may die with EPERM in some situations
Mar 15
Rob Browning
Mar 15
[PATCH 2/3] open_noatime: add flags arg, only add O_LARGEFILE
Add a flags argument to the python function, and stop adding any flags other than O_NOATIME and
unread,
[PATCH 2/3] open_noatime: add flags arg, only add O_LARGEFILE
Add a flags argument to the python function, and stop adding any flags other than O_NOATIME and
Mar 15
Rob Browning
6
Mar 14
[PATCH 7/7] bup-on: restrict local server actions based on command
This allows us to create the Server directly, making it easier to control/customize it without being
unread,
[PATCH 7/7] bup-on: restrict local server actions based on command
This allows us to create the Server directly, making it easier to control/customize it without being
Mar 14
Rob Browning
Mar 14
[PATCH 1/7] Client(): only require set-dir and init-dir when needed
Previously we always required both. Thanks to Johannes Berg for proposing the change. Signed-off-by:
unread,
[PATCH 1/7] Client(): only require set-dir and init-dir when needed
Previously we always required both. Thanks to Johannes Berg for proposing the change. Signed-off-by:
Mar 14
Rob Browning
Mar 14
[PATCH 2/7] Client(): only require list-indexes and send-index when needed
Previously we always required both. Thanks to Johannes Berg for proposing the change. Signed-off-by:
unread,
[PATCH 2/7] Client(): only require list-indexes and send-index when needed
Previously we always required both. Thanks to Johannes Berg for proposing the change. Signed-off-by:
Mar 14
Rob Browning
Mar 14
[PATCH 2/2] dataclass_frozen_for_testing: add and use for WalkItem and GetResult
The dataclass docs say that "there is a tiny perforance penalty when using frozen=True".
unread,
[PATCH 2/2] dataclass_frozen_for_testing: add and use for WalkItem and GetResult
The dataclass docs say that "there is a tiny perforance penalty when using frozen=True".
Mar 14
Rob Browning
Mar 14
[PATCH 1/2] get: improve --exclude*/--ignore* help
Signed-off-by: Rob Browning <r...@defaultvalue.org> Tested-by: Rob Browning <rlb@defaultvalue
unread,
[PATCH 1/2] get: improve --exclude*/--ignore* help
Signed-off-by: Rob Browning <r...@defaultvalue.org> Tested-by: Rob Browning <rlb@defaultvalue
Mar 14
Rob Browning
Mar 14
Limit use of frozen dataclasses when not testing
Pushed to main. Looks like they're too expensive for some uses right now. -- Rob Browning rlb @
unread,
Limit use of frozen dataclasses when not testing
Pushed to main. Looks like they're too expensive for some uses right now. -- Rob Browning rlb @
Mar 14
Rob Browning
14
Mar 14
[PATCH 13/13] pylint: examine test/ext too
Signed-off-by: Rob Browning <r...@defaultvalue.org> --- test/ext/conftest.py | 4 +++- test/ext/
unread,
[PATCH 13/13] pylint: examine test/ext too
Signed-off-by: Rob Browning <r...@defaultvalue.org> --- test/ext/conftest.py | 4 +++- test/ext/
Mar 14