4
* ``bzr help commands`` output is now shorter (Aaron Bentley)
6
* ``bzr`` now uses lazy importing to reduce the startup time. This has
7
a moderate effect on lots of actions, especially ones that have
8
little to do. For example ``bzr rocks`` time is down to 116ms from
9
283ms. (John Arbash Meinel)
13
* ``TestCase.run_bzr``, ``run_bzr_captured``, and ``run_bzr_subprocess``
14
can take a ``working_dir='foo'`` parameter, which will change directory
15
for the command. (John Arbash Meinel)
17
* ``bzrlib.lazy_regex.lazy_compile`` can be used to create a proxy
18
around a regex, which defers compilation until first use.
23
* Don't require ``Content-Type`` in range responses. Assume they are a
24
single range if ``Content-Type`` does not exist.
25
(John Arbash Meinel, #62473)
27
* bzr branch/pull no longer complain about progress bar cleanup when
28
interrupted during fetch. (Aaron Bentley, #54000)
30
* ``WorkingTree.set_parent_trees()`` uses the trees to directly write
31
the basis inventory, rather than going through the repository. This
32
allows us to have 1 inventory read, and 2 inventory writes when
33
committing a new tree. (John Arbash Meinel)
35
* When reverting, files that are not locally modified that do not exist
36
in the target are deleted, not just unversioned (Aaron Bentley)
38
* When trying to acquire a lock, don't fail immediately. Instead, try
39
a few times (up to 1 hour) before timing out. Also, report why the
40
lock is unavailable (John Arbash Meinel, #43521, #49556)
42
* Leave HttpTransportBase daughter classes decides how they
43
implement cloning. (Vincent Ladeuil, #61606)
45
* diff3 does not indicate conflicts on clean merge. (Aaron Bentley)
49
* New test base class TestCaseWithMemoryTransport offers memory-only
50
testing facilities: its not suitable for tests that need to mutate disk
51
state, but most tests should not need that and should be converted to
52
TestCaseWithMemoryTransport. (Robert Collins)
56
* Smart server transport test failures on windows fixed. (Lukáš Lalinský).
58
bzr 0.11rc2 2006-09-27
62
* Test suite hangs on windows fixed. (Andrew Bennets, Alexander Belchenko).
64
* Commit performance regression fixed. (Aaron Bentley, Robert Collins, John
67
bzr 0.11rc1 2006-09-25
71
* Knit files now wait to create their contents until the first data is
72
added. The old code used to create an empty .knit and a .kndx with just
73
the header. However, this caused a lot of extra round trips over sftp.
74
This can change the time for ``bzr push`` to create a new remote branch
75
from 160s down to 100s. This also affects ``bzr commit`` performance when
76
adding new files, ``bzr commit`` on a new kernel-like tree drops from 50s
77
down to 40s (John Arbash Meinel, #44692)
79
* When an entire subtree has been deleted, commit will now report that
80
just the top of the subtree has been deleted, rather than reporting
81
all the individual items. (Robert Collins)
83
* Commit performs one less XML parse. (Robert Collins)
85
* ``bzr checkout`` now operates on readonly branches as well
86
as readwrite branches. This fixes bug #39542. (Robert Collins)
88
* ``bzr bind`` no longer synchronises history with the master branch.
89
Binding should be followed by an update or push to synchronise the
90
two branches. This is closely related to the fix for bug #39542.
93
* ``bzrlib.lazy_import.lazy_import`` function to create on-demand
94
objects. This allows all imports to stay at the global scope, but
95
modules will not actually be imported if they are not used.
98
* Support bzr:// and bzr+ssh:// urls to work with the new RPC-based
99
transport which will be used with the upcoming high-performance smart
100
server. The new command ``bzr serve`` will invoke bzr in server mode,
101
which processes these requests. (Andrew Bennetts, Robert Collins, Martin
104
* New command ``bzr version-info`` which can be used to get a summary
105
of the current state of the tree. This is especially useful as part
106
of a build commands. See ``doc/version_info.txt`` for more information
111
* 'bzr inventory [FILE...]' allows restricting the file list to a
112
specific set of files. (John Arbash Meinel, #3631)
114
* Don't abort when annotating empty files (John Arbash Meinel, #56814)
116
* Add ``Stanza.to_unicode()`` which can be passed to another Stanza
117
when nesting stanzas. Also, add ``read_stanza_unicode`` to handle when
118
reading a nested Stanza. (John Arbash Meinel)
120
* Transform._set_mode() needs to stat the right file.
121
(John Arbash Meinel, #56549)
123
* Raise WeaveFormatError rather than StopIteration when trying to read
124
an empty Weave file. (John Arbash Meinel, #46871)
126
* Don't access e.code for generic URLErrors, only HTTPErrors have .code.
127
(Vincent Ladeuil, #59835)
129
* Handle boundary="" lines properly to allow access through a Squid proxy.
130
(John Arbash Meinel, #57723)
132
* revert now removes newly-added directories (Aaron Bentley, #54172)
134
* ``bzr upgrade sftp://`` shouldn't fail to upgrade v6 branches if there
135
isn't a working tree. (David Allouche, #40679)
137
* Give nicer error messages when a user supplies an invalid --revision
138
parameter. (John Arbash Meinel, #55420)
140
* Handle when LANG is not recognized by python. Emit a warning, but
141
just revert to using 'ascii'. (John Arbash Meinel, #35392)
143
* Don't use preexec_fn on win32, as it is not supported by subprocess.
146
* Skip specific tests when the dependencies aren't met. This includes
147
some ``setup.py`` tests when ``python-dev`` is not available, and
148
some tests that depend on paramiko. (John Arbash Meinel, Mattheiu Moy)
150
* Fallback to Paramiko properly, if no ``ssh`` executable exists on
151
the system. (Andrew Bennetts, John Arbash Meinel)
153
* ``Branch.bind(other_branch)`` no longer takes a write lock on the
154
other branch, and will not push or pull between the two branches.
155
API users will need to perform a push or pull or update operation if they
156
require branch synchronisation to take place. (Robert Collins, #47344)
158
* When creating a tarball or zipfile export, export unicode names as utf-8
159
paths. This may not work perfectly on all platforms, but has the best
160
chance of working in the common case. (John Arbash Meinel, #56816)
162
* When committing, only files that exist in working tree or basis tree
163
may be specified (Aaron Bentley, #50793)
167
* Fixes to run on Python 2.5 (Brian M. Carlson, Martin Pool, Marien Zwart)
171
* TestCaseInTempDir now creates a separate directory for HOME, rather
172
than having HOME set to the same location as the working directory.
175
* run_bzr_subprocess() can take an optional 'env_changes={}' parameter,
176
which will update os.environ inside the spawned child. It also can
177
take a 'universal_newlines=True', which helps when checking the output
178
of the command. (John Arbash Meinel)
180
* Refactor SFTP vendors to allow easier re-use when ssh is used.
183
* Transport.list_dir() and Transport.iter_files_recursive() should always
184
return urlescaped paths. This is now tested (there were bugs in a few
185
of the transports) (Andrew Bennetts, David Allouche, John Arbash Meinel)
187
* New utility function symbol_versioning.deprecation_string. Returns the
188
formatted string for a callable, deprecation format pair. (Robert Collins)
190
* New TestCase helper applyDeprecated. This allows you to call a callable
191
which is deprecated without it spewing to the screen, just by supplying
192
the deprecation format string issued for it. (Robert Collins)
194
* Transport.append and Transport.put have been deprecated in favor of
195
.append_bytes, .append_file, .put_bytes, and .put_file. This removes the
196
ambiguity in what type of object the functions take.
197
Transport.non_atomic_put_{bytes,file} has also been added. Which works
198
similarly to Transport.append() except for SFTP, it doesn't have a round
199
trip when opening the file. Also, it provides functionality for creating
200
a parent directory when trying to create a file, rather than raise
201
NoSuchFile and forcing the caller to repeat their request.
204
* WorkingTree has a new api ``unversion`` which allow the unversioning of
205
entries by their file id. (Robert Collins)
207
* WorkingTree.pending_merges is deprecated. Please use the get_parent_ids
208
(introduced in 0.10) method instead. (Robert Collins)
210
* WorkingTree has a new lock_tree_write method which locks the branch for
211
read rather than write. This is appropriate for actions which only need
212
the branch data for reference rather than mutation. A new decorator
213
needs_tree_write_lock is provided in the workingtree module. Like the
214
needs_read_lock and needs_write_lock decorators this allows static
215
declaration of the locking requirements of a function to ensure that
216
a lock is taken out for casual scripts. (Robert Collins, #54107)
218
* All WorkingTree methods which write to the tree, but not to the branch
219
have been converted to use ``needs_tree_write_lock`` rather than
220
``needs_write_lock``. Also converted is the revert, conflicts and tree
221
transform modules. This provides a modest performance improvement on
222
metadir style trees, due to the reduce lock-acquisition, and a more
223
significant performance improvement on lightweight checkouts from
224
remote branches, where trivial operations used to pay a significant
225
penalty. It also provides the basis for allowing readonly checkouts.
228
* Special case importing the standard library 'copy' module. This shaves
229
off 40ms of startup time, while retaining compatibility. See:
230
``bzrlib/inspect_for_copy.py`` for more details. (John Arbash Meinel)
232
* WorkingTree has a new parent class MutableTree which represents the
233
specialisations of Tree which are able to be altered. (Robert Collins)
235
* New methods mkdir and put_file_bytes_non_atomic on MutableTree that
236
mutate the tree and its contents. (Robert Collins)
238
* Transport behaviour at the root of the URL is now defined and tested.
239
(Andrew Bennetts, Robert Collins)
243
* New test helper classs MemoryTree. This is typically accessed via
244
``self.make_branch_and_memory_tree()`` in test cases. (Robert Collins)
246
* Add start_bzr_subprocess and stop_bzr_subprocess to allow test code to
247
continue running concurrently with a subprocess of bzr. (Andrew Bennetts,
250
* Add a new method ``Transport.get_smart_client()``. This is provided to
251
allow upgrades to a richer interface than the VFS one provided by
252
Transport. (Andrew Bennetts, Martin Pool)
257
* 'merge' now takes --uncommitted, to apply uncommitted changes from a
258
tree. (Aaron Bentley)
260
* 'bzr add --file-ids-from' can be used to specify another path to use
261
for creating file ids, rather than generating all new ones. Internally,
262
the 'action' passed to smart_add_tree() can return file_ids that
263
will be used, rather than having bzrlib generate new ones.
264
(John Arbash Meinel, #55781)
266
* ``bzr selftest --benchmark`` now allows a ``--cache-dir`` parameter.
267
This will cache some of the intermediate trees, and decrease the
268
setup time for benchmark tests. (John Arbash Meinel)
270
* Inverse forms are provided for all boolean options. For example,
271
--strict has --no-strict, --no-recurse has --recurse (Aaron Bentley)
273
* Serialize out Inventories directly, rather than using ElementTree.
274
Writing out a kernel sized inventory drops from 2s down to ~350ms.
275
(Robert Collins, John Arbash Meinel)
279
* Help diffutils 2.8.4 get along with binary tests (Marien Zwart: #57614)
281
* Change LockDir so that if the lock directory doesn't exist when
282
lock_write() is called, an attempt will be made to create it.
283
(John Arbash Meinel, #56974)
285
* ``bzr uncommit`` preserves pending merges. (John Arbash Meinel, #57660)
287
* Active FTP transport now works as intended. (ghozzy, #56472)
289
* Really fix mutter() so that it won't ever raise a UnicodeError.
290
It means it is possible for ~/.bzr.log to contain non UTF-8 characters.
291
But it is a debugging log, not a real user file.
292
(John Arbash Meinel, #56947, #53880)
294
* Change Command handle to allow Unicode command and options.
295
At present we cannot register Unicode command names, so we will get
296
BzrCommandError('unknown command'), or BzrCommandError('unknown option')
297
But that is better than a UnicodeError + a traceback.
298
(John Arbash Meinel, #57123)
300
* Handle TZ=UTC properly when reading/writing revisions.
301
(John Arbash Meinel, #55783, #56290)
303
* Use GPG_TTY to allow gpg --cl to work with gpg-agent in a pipeline,
304
(passing text to sign in on stdin). (John Arbash Meinel, #54468)
306
* External diff does the right thing for binaries even in foreign
307
languages. (John Arbash Meinel, #56307)
309
* Testament handles more cases when content is unicode. Specific bug was
310
in handling of revision properties. (John Arbash Meinel, Holger Krekel,
313
* The bzr selftest was failing on installed versions due to a bug in a new
314
test helper. (John Arbash Meinel, Robert Collins, #58057)
318
* ``bzrlib.cache_utf8`` contains ``encode()`` and ``decode()`` functions
319
which can be used to cache the conversion between utf8 and Unicode.
320
Especially helpful for some of the knit annotation code, which has to
321
convert revision ids to utf8 to annotate lines in storage.
324
* ``setup.py`` now searches the filesystem to find all packages which
325
need to be installed. This should help make the life of packagers
326
easier. (John Arbash Meinel)
9
328
bzr 0.9.0 2006-08-11
added
removed
NEWS
