bzr add: exceptions.AttributeError: children on add
Affects | Status | Importance | Assigned to | Milestone | |
---|---|---|---|---|---|
Bazaar |
Fix Released
|
Medium
|
Martin Packman |
Bug Description
during a bzr add, bzr complains
------8<--------
bash-3.2$ bzr add information/
bzr: ERROR: exceptions.
Traceback (most recent call last):
File "commands.py", line 846, in run_bzr_
File "commands.py", line 802, in run_bzr
File "commands.py", line 504, in run_argv_aliases
File "builtins.py", line 387, in run
File "mutabletree.py", line 52, in tree_write_locked
File "mutabletree.py", line 413, in smart_add
AttributeError: children
bzr 1.4 on python 2.5.1 (cygwin)
arguments: ['/usr/bin/bzr', 'add', 'information/
encoding: 'US-ASCII', fsenc: 'US-ASCII', lang: None
plugins:
launchpad /usr/lib/
*** Bazaar has encountered an internal error.
Please report a bug at https:/
including this traceback, and a description of what you
were doing when the error occurred.
------8<--------
Before, a bzr status delivered:
------8<--------
bash-3.2$ bzr status
kind changed:
information/
modified:
.gnupg/
.gnupg/
unknown:
.bzr.log
.gnupg/
.gnupg/
information/
information/
information/
information/
information/
information/
information/
information/
information/
information/
information/
information/
information/
------8<--------
The exotic situation seems to have arisen from the fact, that some time before a
mv information/
mkdir information/
mv information/
had been done "without informing" bzr.
Undoing the described manual "mv" activities and doing the move in a bzr-aware style {bzr mv ...} let the problem disappear.
At least would the innocent benefit from a more transparent error message than above.
Related branches
- Martin Pool: Approve
-
Diff: 78 lines (+19/-16) (has conflicts)3 files modifiedbzrlib/mutabletree.py (+7/-0)
bzrlib/tests/per_workingtree/test_smart_add.py (+5/-16)
doc/en/release-notes/bzr-2.3.txt (+7/-0)
Changed in bzr: | |
status: | In Progress → Fix Released |
milestone: | none → 2.3b3 |
Hi,
Thanks for taking the time to report this bug.
It looks like bug 192859, but that is about a symlink, whereas
it appears your tree changes a file to a directory. Is that correct?
Thanks,
James