5
* Knit files now wait to create their contents until the first data is
6
added. The old code used to create an empty .knit and a .kndx with just
7
the header. However, this caused a lot of extra round trips over sftp.
8
This can change the time for ``bzr push`` to create a new remote branch
9
from 160s down to 100s. This also affects ``bzr commit`` performance when
10
adding new files, ``bzr commit`` on a new kernel-like tree drops from 50s
11
down to 40s (John Arbash Meinel, #44692)
13
* When an entire subtree has been deleted, commit will now report that
14
just the top of the subtree has been deleted, rather than reporting
15
all the individual items. (Robert Collins)
17
* Commit performs one less XML parse. (Robert Collins)
19
* ``bzr checkout`` now operates on readonly branches as well
20
as readwrite branches. This fixes bug #39542. (Robert Collins)
22
* ``bzr bind`` no longer synchronises history with the master branch.
23
Binding should be followed by an update or push to synchronise the
24
two branches. This is closely related to the fix for bug #39542.
27
* ``bzrlib.lazy_import.lazy_import`` function to create on-demand
28
objects. This allows all imports to stay at the global scope, but
29
modules will not actually be imported if they are not used.
32
* Support bzr:// and bzr+ssh:// urls to work with the new RPC-based
33
transport which will be used with the upcoming high-performance smart
34
server. The new command ``bzr serve`` will invoke bzr in server mode,
35
which processes these requests. (Andrew Bennetts, Robert Collins, Martin
38
* New command ``bzr version-info`` which can be used to get a summary
39
of the current state of the tree. This is especially useful as part
40
of a build commands. See ``doc/version_info.txt`` for more information
45
* Transform._set_mode() needs to stat the right file.
46
(John Arbash Meinel, #56549)
48
* Raise WeaveFormatError rather than StopIteration when trying to read
49
an empty Weave file. (John Arbash Meinel, #46871)
51
* Don't access e.code for generic URLErrors, only HTTPErrors have .code.
52
(Vincent Ladeuil, #59835)
54
* Handle boundary="" lines properly to allow access through a Squid proxy.
55
(John Arbash Meinel, #57723)
57
* revert now removes newly-added directories (Aaron Bentley, #54172)
59
* ``bzr upgrade sftp://`` shouldn't fail to upgrade v6 branches if there
60
isn't a working tree. (David Allouche, #40679)
62
* Give nicer error messages when a user supplies an invalid --revision
63
parameter. (John Arbash Meinel, #55420)
65
* Handle when LANG is not recognized by python. Emit a warning, but
66
just revert to using 'ascii'. (John Arbash Meinel, #35392)
68
* Don't use preexec_fn on win32, as it is not supported by subprocess.
71
* Skip specific tests when the dependencies aren't met. This includes
72
some ``setup.py`` tests when ``python-dev`` is not available, and
73
some tests that depend on paramiko. (John Arbash Meinel, Mattheiu Moy)
75
* Fallback to Paramiko properly, if no ``ssh`` executable exists on
76
the system. (Andrew Bennetts, John Arbash Meinel)
78
* ``Branch.bind(other_branch)`` no longer takes a write lock on the
79
other branch, and will not push or pull between the two branches.
80
API users will need to perform a push or pull or update operation if they
81
require branch synchronisation to take place. (Robert Collins, #47344)
83
* When creating a tarball or zipfile export, export unicode names as utf-8
84
paths. This may not work perfectly on all platforms, but has the best
85
chance of working in the common case. (John Arbash Meinel, #56816)
89
* TestCaseInTempDir now creates a separate directory for HOME, rather
90
than having HOME set to the same location as the working directory.
93
* run_bzr_subprocess() can take an optional 'env_changes={}' parameter,
94
which will update os.environ inside the spawned child. It also can
95
take a 'universal_newlines=True', which helps when checking the output
96
of the command. (John Arbash Meinel)
98
* Refactor SFTP vendors to allow easier re-use when ssh is used.
101
* Transport.list_dir() and Transport.iter_files_recursive() should always
102
return urlescaped paths. This is now tested (there were bugs in a few
103
of the transports) (Andrew Bennetts, David Allouche, John Arbash Meinel)
105
* New utility function symbol_versioning.deprecation_string. Returns the
106
formatted string for a callable, deprecation format pair. (Robert Collins)
108
* New TestCase helper applyDeprecated. This allows you to call a callable
109
which is deprecated without it spewing to the screen, just by supplying
110
the deprecation format string issued for it. (Robert Collins)
112
* Transport.append and Transport.put have been deprecated in favor of
113
.append_bytes, .append_file, .put_bytes, and .put_file. This removes the
114
ambiguity in what type of object the functions take.
115
Transport.non_atomic_put_{bytes,file} has also been added. Which works
116
similarly to Transport.append() except for SFTP, it doesn't have a round
117
trip when opening the file. Also, it provides functionality for creating
118
a parent directory when trying to create a file, rather than raise
119
NoSuchFile and forcing the caller to repeat their request.
122
* WorkingTree has a new api ``unversion`` which allow the unversioning of
123
entries by their file id. (Robert Collins)
125
* WorkingTree.pending_merges is deprecated. Please use the get_parent_ids
126
(introduced in 0.10) method instead. (Robert Collins)
128
* WorkingTree has a new lock_tree_write method which locks the branch for
129
read rather than write. This is appropriate for actions which only need
130
the branch data for reference rather than mutation. A new decorator
131
needs_tree_write_lock is provided in the workingtree module. Like the
132
needs_read_lock and needs_write_lock decorators this allows static
133
declaration of the locking requirements of a function to ensure that
134
a lock is taken out for casual scripts. (Robert Collins)
136
* All WorkingTree methods which write to the tree, but not to the branch
137
have been converted to use ``needs_tree_write_lock`` rather than
138
``needs_write_lock``. Also converted is the revert, conflicts and tree
139
transform modules. This provides a modest performance improvement on
140
metadir style trees, due to the reduce lock-acquisition, and a more
141
significant performance improvement on lightweight checkouts from
142
remote branches, where trivial operations used to pay a significant
143
penalty. It also provides the basis for allowing readonly checkouts.
146
* Special case importing the standard library 'copy' module. This shaves
147
off 40ms of startup time, while retaining compatibility. See:
148
``bzrlib/inspect_for_copy.py`` for more details. (John Arbash Meinel)
150
* WorkingTree has a new parent class MutableTree which represents the
151
specialisations of Tree which are able to be altered. (Robert Collins)
153
* New methods mkdir and put_file_bytes_non_atomic on MutableTree that
154
mutate the tree and its contents. (Robert Collins)
156
* Transport behaviour at the root of the URL is now defined and tested.
157
(Andrew Bennetts, Robert Collins)
161
* New test helper classs MemoryTree. This is typically accessed via
162
``self.make_branch_and_memory_tree()`` in test cases. (Robert Collins)
164
* Add start_bzr_subprocess and stop_bzr_subprocess to allow test code to
165
continue running concurrently with a subprocess of bzr. (Andrew Bennetts,
168
* Add a new method ``Transport.get_smart_client()``. This is provided to
169
allow upgrades to a richer interface than the VFS one provided by
170
Transport. (Andrew Bennetts, Martin Pool)
175
* 'merge' now takes --uncommitted, to apply uncommitted changes from a
176
tree. (Aaron Bentley)
178
* 'bzr add --file-ids-from' can be used to specify another path to use
179
for creating file ids, rather than generating all new ones. Internally,
180
the 'action' passed to smart_add_tree() can return file_ids that
181
will be used, rather than having bzrlib generate new ones.
182
(John Arbash Meinel, #55781)
184
* ``bzr selftest --benchmark`` now allows a ``--cache-dir`` parameter.
185
This will cache some of the intermediate trees, and decrease the
186
setup time for benchmark tests. (John Arbash Meinel)
188
* Inverse forms are provided for all boolean options. For example,
189
--strict has --no-strict, --no-recurse has --recurse (Aaron Bentley)
191
* Serialize out Inventories directly, rather than using ElementTree.
192
Writing out a kernel sized inventory drops from 2s down to ~350ms.
193
(Robert Collins, John Arbash Meinel)
197
* Help diffutils 2.8.4 get along with binary tests (Marien Zwart: #57614)
199
* Change LockDir so that if the lock directory doesn't exist when
200
lock_write() is called, an attempt will be made to create it.
201
(John Arbash Meinel, #56974)
203
* ``bzr uncommit`` preserves pending merges. (John Arbash Meinel, #57660)
205
* Active FTP transport now works as intended. (ghozzy, #56472)
207
* Really fix mutter() so that it won't ever raise a UnicodeError.
208
It means it is possible for ~/.bzr.log to contain non UTF-8 characters.
209
But it is a debugging log, not a real user file.
210
(John Arbash Meinel, #56947, #53880)
212
* Change Command handle to allow Unicode command and options.
213
At present we cannot register Unicode command names, so we will get
214
BzrCommandError('unknown command'), or BzrCommandError('unknown option')
215
But that is better than a UnicodeError + a traceback.
216
(John Arbash Meinel, #57123)
218
* Handle TZ=UTC properly when reading/writing revisions.
219
(John Arbash Meinel, #55783, #56290)
221
* Use GPG_TTY to allow gpg --cl to work with gpg-agent in a pipeline,
222
(passing text to sign in on stdin). (John Arbash Meinel, #54468)
224
* External diff does the right thing for binaries even in foreign
225
languages. (John Arbash Meinel, #56307)
227
* Testament handles more cases when content is unicode. Specific bug was
228
in handling of revision properties. (John Arbash Meinel, Holger Krekel,
231
* The bzr selftest was failing on installed versions due to a bug in a new
232
test helper. (John Arbash Meinel, Robert Collins, #58057)
236
* ``bzrlib.cache_utf8`` contains ``encode()`` and ``decode()`` functions
237
which can be used to cache the conversion between utf8 and Unicode.
238
Especially helpful for some of the knit annotation code, which has to
239
convert revision ids to utf8 to annotate lines in storage.
242
* ``setup.py`` now searches the filesystem to find all packages which
243
need to be installed. This should help make the life of packagers
244
easier. (John Arbash Meinel)
250
* The hard-coded built-in ignore rules have been removed. There are
251
now two rulesets which are enforced. A user global one in
252
~/.bazaar/ignore which will apply to every tree, and the tree
253
specific one '.bzrignore'.
254
~/.bazaar/ignore will be created if it does not exist, but with
255
a more conservative list than the old default.
256
This fixes bugs with default rules being enforced no matter what.
257
The old list of ignore rules from bzr is available by
258
running 'bzr ignore --old-default-rules'.
259
(Robert Collins, Martin Pool, John Arbash Meinel)
261
* 'branches.conf' has been changed to 'locations.conf', since it can apply
262
to more locations than just branch locations.
5
267
* The revision specifier "revno:" is extended to accept the syntax
6
268
revno:N:branch. For example,
7
269
revno:42:http://bazaar-vcs.org/bzr/bzr.dev/ means revision 42 in
8
270
bzr.dev. (Matthieu Moy)
10
* No default ignore rules are applied by bzr - only the rules in .bzrignore
11
are considered. This fixes bugs with default rules being enforced no
12
matter what. The old list of ignore rules from bzr is available by
13
running 'bzr ignore --old-default-rules'.
15
272
* Tests updates to ensure proper URL handling, UNICODE support, and
16
273
proper printing when the user's terminal encoding cannot display
17
274
the path of a file that has been versioned.
157
429
* smart-add recurses through all supplied directories
158
430
(John Arbash Meinel, #52578)
432
* Make the bundle reader extra lines before and after the bundle text.
433
This allows you to parse an email with the bundle inline.
434
(John Arbash Meinel, #49182)
436
* Change the file id generator to squash a little bit more. Helps when
437
working with long filenames on windows. (Also helps for unicode filenames
438
not generating hidden files). (John Arbash Meinel, #43801)
440
* Restore terminal mode on C-c while reading sftp password. (#48923,
441
Nicholas Allen, Martin Pool)
443
* Timestamps are rounded to 1ms, and revision entries can be recreated
444
exactly. (John Arbash Meinel, Jamie Wilkinson, #40693)
446
* Branch.base has changed to a URL, but ~/.bazaar/locations.conf should
447
use local paths, since it is user visible (John Arbash Meinel, #53653)
449
* ``bzr status foo`` when foo was unversioned used to cause a full delta
450
to be generated (John Arbash Meinel, #53638)
452
* When reading revision properties, an empty value should be considered
453
the empty string, not None (John Arbash Meinel, #47782)
455
* ``bzr diff --diff-options`` can now handle binary files being changed.
456
Also, the output is consistent when --diff-options is not supplied.
457
(John Arbash Meinel, #54651, #52930)
459
* Use the right suffixes for loading plugins (John Arbash Meinel, #51810)
461
* Fix Branch.get_parent() to handle the case when the parent is not
462
accessible (John Arbash Meinel, #52976)
162
466
* Combine the ignore rules into a single regex rather than looping over
163
467
them to reduce the threshold where N^2 behaviour occurs in operations
164
468
like status. (Jan Hudec, Robert Collins).
470
* Appending to bzrlib.DEFAULT_IGNORE is now deprecated. Instead, use
471
one of the add functions in bzrlib.ignores. (John Arbash Meinel)
166
473
* 'bzr push' should only push the ancestry of the current revision, not
167
474
all of the history in the repository. This is especially important for
168
475
shared repositories. (John Arbash Meinel)
added
removed
NEWS
