12
* Fetching via bzr+ssh will no longer fill ghosts by default (this is
13
consistent with pack-0.92 fetching over SFTP). (Robert Collins)
17
* New smart method, ``Repository.stream_revisions_chunked``, for fetching
18
revision data that streams revision data via a chunked encoding. This
19
avoids buffering large amounts of revision data on the server and on the
20
client. (Andrew Bennetts, #178353)
24
* Fetching between different repository formats with compatible models now
25
takes advantage of the smart method to stream revisions. (Andrew Bennetts)
29
* Calculate remote path relative to the shared medium in _SmartClient. This
30
is related to the problem in bug #124089. (Andrew Bennetts)
32
* ``reconfigure`` can safely be interrupted while fetching.
33
(Aaron Bentley, #179316)
35
* Unknown hostnames when connecting to a ``bzr://`` URL no longer cause
36
tracebacks. (Andrew Bennetts, #182849)
40
* Add -Dtimes debug flag, which records a timestamp against each mutter to
41
the trace file, relative to the first mutter. (Andrew Bennetts)
43
* Fetching now passes the find_ghosts flag through to the
44
``InterRepository.missing_revision_ids`` call consistently for all
45
repository types. This will enable faster missing revision discovery with
46
bzr+ssh. (Robert Collins)
48
* New method ``iter_inventories`` on Repository for access to many
49
inventories. This is primarily used by the ``revision_trees`` method, as
50
direct access to inventories is discouraged. (Robert Collins)
52
* New method ``next_with_ghosts`` on the Graph breadth-first-search objects
53
which will split out ghosts and present parents into two separate sets,
54
useful for code which needs to be aware of ghosts (e.g. fetching data
55
cares about ghosts during revision selection). (Robert Collins)
57
* Repository has a new method ``has_revisions`` which signals the presence
58
of many revisions by returning a set of the revisions listed which are
59
present. This can be done by index queries without reading data for parent
60
revision names etc. (Robert Collins)
66
(no changes from 1.1rc1)
12
73
* The syntax ``bzr diff branch1 branch2`` is no longer supported.
13
74
Use ``bzr diff branch1 --new branch2`` instead. This change has
14
75
been made to remove the ambiguity where ``branch2`` is in fact a
15
76
specific file to diff within ``branch1``.
17
* Fetching via bzr+ssh will no longer fill ghosts by default (this is
18
consistent with pack-0.92 fetching over SFTP). (Robert Collins)
22
80
* New option to use custom template-based formats in ``bzr version-info``.
25
* New smart method, ``Repository.stream_revisions_chunked``, for fetching
26
revision data that streams revision data via a chunked encoding. This
27
avoids buffering large amounts of revision data on the server and on the
28
client. (Andrew Bennetts, #178353)
30
83
* diff '--using' allows an external diff tool to be used for files.
138
182
removed use characters not supported in the terminal encoding.
141
* Unknown hostnames when connecting to a ``bzr://`` URL no longer cause
142
tracebacks. (Andrew Bennetts, #182849)
144
185
* When dumb http servers return whole files instead of the requested ranges,
145
186
read the remaining bytes by chunks to avoid overflowing network buffers.
146
187
(Vincent Ladeuil, #175886)
164
* Add -Dtimes debug flag, which records a timestamp against each mutter to
165
the trace file, relative to the first mutter. (Andrew Bennetts)
167
* Fetching now passes the find_ghosts flag through to the
168
``InterRepository.missing_revision_ids`` call consistently for all
169
repository types. This will enable faster missing revision discovery with
170
bzr+ssh. (Robert Collins)
172
* find_* methods available for BzrDirs, Branches and WorkingTrees.
175
* Help topics can now be loaded from files.
176
(Ian Clatworthy, Alexander Belchenko)
178
* get_parent_map now always provides tuples as its output. (Aaron Bentley)
180
* New method ``iter_inventories`` on Repository for access to many
181
inventories. This is primarily used by the ``revision_trees`` method, as
182
direct access to inventories is discouraged. (Robert Collins)
184
* New method ``next_with_ghosts`` on the Graph breadth-first-search objects
185
which will split out ghosts and present parents into two separate sets,
186
useful for code which needs to be aware of ghosts (e.g. fetching data
187
cares about ghosts during revision selection). (Robert Collins)
189
* Parent Providers should now implement ``get_parent_map`` returning a
190
dictionary instead of ``get_parents`` returning a list.
191
``Graph.get_parents`` is now deprecated. (John Arbash Meinel,
194
* Patience Diff now supports arbitrary python objects, as long as they
195
support ``hash()``. (John Arbash Meinel)
197
* Reduce selftest overhead to establish test names by memoization.
200
* Repository has a new method ``has_revisions`` which signals the presence
201
of many revisions by returning a set of the revisions listed which are
202
present. This can be done by index queries without reading data for parent
203
revision names etc. (Robert Collins)
205
* find_* methods available for BzrDirs, Branches and WorkingTrees.
208
* Help topics can now be loaded from files.
209
(Ian Clatworthy, Alexander Belchenko)
211
* get_parent_map now always provides tuples as its output. (Aaron Bentley)
213
* Parent Providers should now implement ``get_parent_map`` returning a
214
dictionary instead of ``get_parents`` returning a list.
215
``Graph.get_parents`` is now deprecated. (John Arbash Meinel,
218
* Patience Diff now supports arbitrary python objects, as long as they
219
support ``hash()``. (John Arbash Meinel)
221
* Reduce selftest overhead to establish test names by memoization.