5
This document describes the Bazaar internals and the development process.
6
It's meant for people interested in developing Bazaar, and some parts will
7
also be useful to people developing Bazaar plugins.
9
If you have any questions or something seems to be incorrect, unclear or
10
missing, please talk to us in ``irc://irc.freenode.net/#bzr``, or write to
11
the Bazaar mailing list. To propose a correction or addition to this
12
document, send a merge request or new text to the mailing list.
14
The current version of this document is available in the file
15
``doc/developers/HACKING.txt`` in the source tree, or at
16
http://doc.bazaar-vcs.org/bzr.dev/en/developer-guide/HACKING.html
19
`Bazaar Developer Documentation Catalog <../../developers/index.html>`_.
27
Exploring the Bazaar Platform
28
=============================
30
Before making changes, it's a good idea to explore the work already
31
done by others. Perhaps the new feature or improvement you're looking
32
for is available in another plug-in already? If you find a bug,
33
perhaps someone else has already fixed it?
35
To answer these questions and more, take a moment to explore the
36
overall Bazaar Platform. Here are some links to browse:
38
* The Plugins page on the Wiki - http://bazaar-vcs.org/BzrPlugins
40
* The Bazaar product family on Launchpad - https://launchpad.net/bazaar
42
* Bug Tracker for the core product - https://bugs.launchpad.net/bzr/
44
* Blueprint Tracker for the core product - https://blueprints.launchpad.net/bzr/
46
If nothing else, perhaps you'll find inspiration in how other developers
47
have solved their challenges.
50
Planning and Discussing Changes
51
===============================
53
There is a very active community around Bazaar. Mostly we meet on IRC
54
(#bzr on irc.freenode.net) and on the mailing list. To join the Bazaar
55
community, see http://bazaar-vcs.org/BzrSupport.
57
If you are planning to make a change, it's a very good idea to mention it
58
on the IRC channel and/or on the mailing list. There are many advantages
59
to involving the community before you spend much time on a change.
62
* you get to build on the wisdom on others, saving time
64
* if others can direct you to similar code, it minimises the work to be done
66
* it assists everyone in coordinating direction, priorities and effort.
68
In summary, maximising the input from others typically minimises the
69
total effort required to get your changes merged. The community is
70
friendly, helpful and always keen to welcome newcomers.
73
Bazaar Development in a Nutshell
74
================================
76
Looking for a 10 minute introduction to submitting a change?
77
See http://bazaar-vcs.org/BzrGivingBack.
79
TODO: Merge that Wiki page into this document.
82
Understanding the Development Process
83
=====================================
85
The development team follows many practices including:
87
* a public roadmap and planning process in which anyone can participate
89
* time based milestones everyone can work towards and plan around
91
* extensive code review and feedback to contributors
93
* complete and rigorous test coverage on any code contributed
95
* automated validation that all tests still pass before code is merged
96
into the main code branch.
98
The key tools we use to enable these practices are:
100
* Launchpad - https://launchpad.net/
102
* Bazaar - http://bazaar-vcs.org/
104
* Bundle Buggy - http://bundlebuggy.aaronbentley.com/
106
* Patch Queue Manager - https://launchpad.net/pqm/
108
For further information, see http://bazaar-vcs.org/BzrDevelopment.
113
Preparing a Sandbox for Making Changes to Bazaar
114
================================================
116
Bazaar supports many ways of organising your work. See
117
http://bazaar-vcs.org/SharedRepositoryLayouts for a summary of the
118
popular alternatives.
120
Of course, the best choice for you will depend on numerous factors:
121
the number of changes you may be making, the complexity of the changes, etc.
122
As a starting suggestion though:
124
* create a local copy of the main development branch (bzr.dev) by using
127
bzr branch http://bazaar-vcs.org/bzr/bzr.dev/ bzr.dev
129
* keep your copy of bzr.dev prestine (by not developing in it) and keep
130
it up to date (by using bzr pull)
132
* create a new branch off your local bzr.dev copy for each issue
133
(bug or feature) you are working on.
135
This approach makes it easy to go back and make any required changes
136
after a code review. Resubmitting the change is then simple with no
137
risk of accidentially including edits related to other issues you may
138
be working on. After the changes for an issue are accepted and merged,
139
the associated branch can be deleted or archived as you wish.
142
Navigating the Code Base
143
========================
145
.. Was at <http://bazaar-vcs.org/NewDeveloperIntroduction>
147
Some of the key files in this directory are:
150
The command you run to start Bazaar itself. This script is pretty
151
short and just does some checks then jumps into bzrlib.
154
This file covers a brief introduction to Bazaar and lists some of its
158
Summary of changes in each Bazaar release that can affect users or
162
Installs Bazaar system-wide or to your home directory. To perform
163
development work on Bazaar it is not required to run this file - you
164
can simply run the bzr command from the top level directory of your
165
development copy. Note: That if you run setup.py this will create a
166
'build' directory in your development branch. There's nothing wrong
167
with this but don't be confused by it. The build process puts a copy
168
of the main code base into this build directory, along with some other
169
files. You don't need to go in here for anything discussed in this
173
Possibly the most exciting folder of all, bzrlib holds the main code
174
base. This is where you will go to edit python files and contribute to
178
Holds documentation on a whole range of things on Bazaar from the
179
origination of ideas within the project to information on Bazaar
180
features and use cases. Within this directory there is a subdirectory
181
for each translation into a human language. All the documentation
182
is in the ReStructuredText markup language.
185
Documentation specifically targetted at Bazaar and plugin developers.
186
(Including this document.)
190
Automatically-generated API reference information is available at
191
<http://starship.python.net/crew/mwh/bzrlibapi/>.
192
(There is an experimental editable version at
193
<http://starship.python.net/crew/mwh/bzrlibapi-oe/>.)
195
See also the `Bazaar Architectural Overview <../../developers/overview.html>`_.
198
The Code Review Process
199
#######################
201
All code changes coming in to Bazaar are reviewed by someone else.
202
Normally changes by core contributors are reviewed by one other core
203
developer, and changes from other people are reviewed by two core
204
developers. Use intelligent discretion if the patch is trivial.
206
Good reviews do take time. They also regularly require a solid
207
understanding of the overall code base. In practice, this means a small
208
number of people often have a large review burden - with knowledge comes
209
responsibility. No one like their merge requests sitting in a queue going
210
nowhere, so reviewing sooner rather than later is strongly encouraged.
214
Sending patches for review
215
==========================
217
If you'd like to propose a change, please post to the
218
bazaar@lists.canonical.com list with a bundle, patch, or link to a
219
branch. Put ``[PATCH]`` or ``[MERGE]`` in the subject so Bundle Buggy
220
can pick it out, and explain the change in the email message text.
221
Remember to update the NEWS file as part of your change if it makes any
222
changes visible to users or plugin developers. Please include a diff
223
against mainline if you're giving a link to a branch.
225
You can generate a merge request like this::
227
bzr send -o bug-1234.patch
229
A ``.patch`` extension is recommended instead of .bundle as many mail clients
230
will send the latter as a binary file.
232
``bzr send`` can also send mail directly if you prefer; see the help.
234
Please do **NOT** put [PATCH] or [MERGE] in the subject line if you don't
235
want it to be merged. If you want comments from developers rather than
236
to be merged, you can put ``[RFC]`` in the subject line.
238
If this change addresses a bug, please put the bug number in the subject
239
line too, in the form ``[#1]`` so that Bundle Buggy can recognize it.
241
If the change is intended for a particular release mark that in the
242
subject too, e.g. ``[1.6]``.
248
Please put a "cover letter" on your merge request explaining:
250
* the reason **why** you're making this change
252
* **how** this change achieves this purpose
254
* anything else you may have fixed in passing
256
* anything significant that you thought of doing, such as a more
257
extensive fix or a different approach, but didn't or couldn't do now
259
A good cover letter makes reviewers' lives easier because they can decide
260
from the letter whether they agree with the purpose and approach, and then
261
assess whether the patch actually does what the cover letter says.
262
Explaining any "drive-by fixes" or roads not taken may also avoid queries
263
from the reviewer. All in all this should give faster and better reviews.
264
Sometimes writing the cover letter helps the submitter realize something
265
else they need to do. The size of the cover letter should be proportional
266
to the size and complexity of the patch.
269
Reviewing proposed changes
270
==========================
272
Anyone is welcome to review code, and reply to the thread with their
275
The simplest way to review a proposed change is to just read the patch on
276
the list or in Bundle Buggy. For more complex changes it may be useful
277
to make a new working tree or branch from trunk, and merge the proposed
278
change into it, so you can experiment with the code or look at a wider
281
There are three main requirements for code to get in:
283
* Doesn't reduce test coverage: if it adds new methods or commands,
284
there should be tests for them. There is a good test framework
285
and plenty of examples to crib from, but if you are having trouble
286
working out how to test something feel free to post a draft patch
289
* Doesn't reduce design clarity, such as by entangling objects
290
we're trying to separate. This is mostly something the more
291
experienced reviewers need to help check.
293
* Improves bugs, features, speed, or code simplicity.
295
Code that goes in should not degrade any of these aspects. Patches are
296
welcome that only cleanup the code without changing the external
297
behaviour. The core developers take care to keep the code quality high
298
and understandable while recognising that perfect is sometimes the enemy
301
It is easy for reviews to make people notice other things which should be
302
fixed but those things should not hold up the original fix being accepted.
303
New things can easily be recorded in the Bug Tracker instead.
305
It's normally much easier to review several smaller patches than one large
306
one. You might want to use ``bzr-loom`` to maintain threads of related
307
work, or submit a preparatory patch that will make your "real" change
311
Checklist for reviewers
312
=======================
314
* Do you understand what the code's doing and why?
316
* Will it perform reasonably for large inputs, both in memory size and
317
run time? Are there some scenarios where performance should be
320
* Is it tested, and are the tests at the right level? Are there both
321
blackbox (command-line level) and API-oriented tests?
323
* If this change will be visible to end users or API users, is it
324
appropriately documented in NEWS?
326
* Does it meet the coding standards below?
328
* If it changes the user-visible behaviour, does it update the help
329
strings and user documentation?
331
* If it adds a new major concept or standard practice, does it update the
332
developer documentation?
334
* (your ideas here...)
337
Bundle Buggy and review outcomes
338
================================
340
Anyone can "vote" on the mailing list by expressing an opinion. Core
341
developers can also vote using Bundle Buggy. Here are the voting codes and
344
:approve: Reviewer wants this submission merged.
345
:tweak: Reviewer wants this submission merged with small changes. (No
347
:abstain: Reviewer does not intend to vote on this patch.
348
:resubmit: Please make changes and resubmit for review.
349
:reject: Reviewer doesn't want this kind of change merged.
350
:comment: Not really a vote. Reviewer just wants to comment, for now.
352
If a change gets two approvals from core reviewers, and no rejections,
353
then it's OK to come in. Any of the core developers can bring it into the
354
bzr.dev trunk and backport it to maintenance branches if required. The
355
Release Manager will merge the change into the branch for a pending
356
release, if any. As a guideline, core developers usually merge their own
357
changes and volunteer to merge other contributions if they were the second
358
reviewer to agree to a change.
360
To track the progress of proposed changes, use Bundle Buggy. See
361
http://bundlebuggy.aaronbentley.com/help for a link to all the
362
outstanding merge requests together with an explanation of the columns.
363
Bundle Buggy will also mail you a link to track just your change.
365
Coding Style Guidelines
366
#######################
371
``hasattr`` should not be used because it swallows exceptions including
372
``KeyboardInterrupt``. Instead, say something like ::
374
if getattr(thing, 'name', None) is None
380
Please write PEP-8__ compliant code.
382
__ http://www.python.org/peps/pep-0008.html
384
One often-missed requirement is that the first line of docstrings
385
should be a self-contained one-sentence summary.
387
We use 4 space indents for blocks, and never use tab characters. (In vim,
390
Lines should be no more than 79 characters if at all possible.
391
Lines that continue a long statement may be indented in either of
394
within the parenthesis or other character that opens the block, e.g.::
400
or indented by four spaces::
406
The first is considered clearer by some people; however it can be a bit
407
harder to maintain (e.g. when the method name changes), and it does not
408
work well if the relevant parenthesis is already far to the right. Avoid
411
self.legbone.kneebone.shinbone.toebone.shake_it(one,
417
self.legbone.kneebone.shinbone.toebone.shake_it(one,
423
self.legbone.kneebone.shinbone.toebone.shake_it(
426
For long lists, we like to add a trailing comma and put the closing
427
character on the following line. This makes it easier to add new items in
430
from bzrlib.goo import (
436
There should be spaces between function paramaters, but not between the
437
keyword name and the value::
439
call(1, 3, cheese=quark)
443
;(defface my-invalid-face
444
; '((t (:background "Red" :underline t)))
445
; "Face used to highlight invalid constructs or other uglyties"
448
(defun my-python-mode-hook ()
449
;; setup preferred indentation style.
450
(setq fill-column 79)
451
(setq indent-tabs-mode nil) ; no tabs, never, I will not repeat
452
; (font-lock-add-keywords 'python-mode
453
; '(("^\\s *\t" . 'my-invalid-face) ; Leading tabs
454
; ("[ \t]+$" . 'my-invalid-face) ; Trailing spaces
455
; ("^[ \t]+$" . 'my-invalid-face)); Spaces only
459
(add-hook 'python-mode-hook 'my-python-mode-hook)
461
The lines beginning with ';' are comments. They can be activated
462
if one want to have a strong notice of some tab/space usage
469
* Imports should be done at the top-level of the file, unless there is
470
a strong reason to have them lazily loaded when a particular
471
function runs. Import statements have a cost, so try to make sure
472
they don't run inside hot functions.
474
* Module names should always be given fully-qualified,
475
i.e. ``bzrlib.hashcache`` not just ``hashcache``.
481
Functions, methods or members that are "private" to bzrlib are given
482
a leading underscore prefix. Names without a leading underscore are
483
public not just across modules but to programmers using bzrlib as an
484
API. As a consequence, a leading underscore is appropriate for names
485
exposed across modules but that are not to be exposed to bzrlib API
488
We prefer class names to be concatenated capital words (``TestCase``)
489
and variables, methods and functions to be lowercase words joined by
490
underscores (``revision_id``, ``get_revision``).
492
For the purposes of naming some names are treated as single compound
493
words: "filename", "revno".
495
Consider naming classes as nouns and functions/methods as verbs.
497
Try to avoid using abbreviations in names, because there can be
498
inconsistency if other people use the full name.
504
``revision_id`` not ``rev_id`` or ``revid``
506
Functions that transform one thing to another should be named ``x_to_y``
507
(not ``x2y`` as occurs in some old code.)
513
Python destructors (``__del__``) work differently to those of other
514
languages. In particular, bear in mind that destructors may be called
515
immediately when the object apparently becomes unreferenced, or at some
516
later time, or possibly never at all. Therefore we have restrictions on
517
what can be done inside them.
519
0. If you think you need to use a ``__del__`` method ask another
520
developer for alternatives. If you do need to use one, explain
523
1. Never rely on a ``__del__`` method running. If there is code that
524
must run, do it from a ``finally`` block instead.
526
2. Never ``import`` from inside a ``__del__`` method, or you may crash the
529
3. In some places we raise a warning from the destructor if the object
530
has not been cleaned up or closed. This is considered OK: the warning
531
may not catch every case but it's still useful sometimes.
537
In some places we have variables which point to callables that construct
538
new instances. That is to say, they can be used a lot like class objects,
539
but they shouldn't be *named* like classes:
541
> I think that things named FooBar should create instances of FooBar when
542
> called. Its plain confusing for them to do otherwise. When we have
543
> something that is going to be used as a class - that is, checked for via
544
> isinstance or other such idioms, them I would call it foo_class, so that
545
> it is clear that a callable is not sufficient. If it is only used as a
546
> factory, then yes, foo_factory is what I would use.
552
Several places in Bazaar use (or will use) a registry, which is a
553
mapping from names to objects or classes. The registry allows for
554
loading in registered code only when it's needed, and keeping
555
associated information such as a help string or description.
558
InterObject and multiple dispatch
559
=================================
561
The ``InterObject`` provides for two-way `multiple dispatch`__: matching
562
up for example a source and destination repository to find the right way
563
to transfer data between them.
565
.. __: http://en.wikipedia.org/wiki/Multiple_dispatch
567
There is a subclass ``InterObject`` classes for each type of object that is
568
dispatched this way, e.g. ``InterRepository``. Calling ``.get()`` on this
569
class will return an ``InterObject`` instance providing the best match for
570
those parameters, and this instance then has methods for operations
573
inter = InterRepository.get(source_repo, target_repo)
574
inter.fetch(revision_id)
576
``InterRepository`` also acts as a registry-like object for its
577
subclasses, and they can be added through ``.register_optimizer``. The
578
right one to run is selected by asking each class, in reverse order of
579
registration, whether it ``.is_compatible`` with the relevant objects.
584
To make startup time faster, we use the ``bzrlib.lazy_import`` module to
585
delay importing modules until they are actually used. ``lazy_import`` uses
586
the same syntax as regular python imports. So to import a few modules in a
589
from bzrlib.lazy_import import lazy_import
590
lazy_import(globals(), """
599
revision as _mod_revision,
601
import bzrlib.transport
605
At this point, all of these exist as a ``ImportReplacer`` object, ready to
606
be imported once a member is accessed. Also, when importing a module into
607
the local namespace, which is likely to clash with variable names, it is
608
recommended to prefix it as ``_mod_<module>``. This makes it clearer that
609
the variable is a module, and these object should be hidden anyway, since
610
they shouldn't be imported into other namespaces.
612
While it is possible for ``lazy_import()`` to import members of a module
613
when using the ``from module import member`` syntax, it is recommended to
614
only use that syntax to load sub modules ``from module import submodule``.
615
This is because variables and classes can frequently be used without
616
needing a sub-member for example::
618
lazy_import(globals(), """
619
from module import MyClass
623
return isinstance(x, MyClass)
625
This will incorrectly fail, because ``MyClass`` is a ``ImportReplacer``
626
object, rather than the real class.
628
It also is incorrect to assign ``ImportReplacer`` objects to other variables.
629
Because the replacer only knows about the original name, it is unable to
630
replace other variables. The ``ImportReplacer`` class will raise an
631
``IllegalUseOfScopeReplacer`` exception if it can figure out that this
632
happened. But it requires accessing a member more than once from the new
633
variable, so some bugs are not detected right away.
639
The null revision is the ancestor of all revisions. Its revno is 0, its
640
revision-id is ``null:``, and its tree is the empty tree. When referring
641
to the null revision, please use ``bzrlib.revision.NULL_REVISION``. Old
642
code sometimes uses ``None`` for the null revision, but this practice is
646
Object string representations
647
=============================
649
Python prints objects using their ``__repr__`` method when they are
650
written to logs, exception tracebacks, or the debugger. We want
651
objects to have useful representations to help in determining what went
654
If you add a new class you should generally add a ``__repr__`` method
655
unless there is an adequate method in a parent class. There should be a
658
Representations should typically look like Python constructor syntax, but
659
they don't need to include every value in the object and they don't need
660
to be able to actually execute. They're to be read by humans, not
661
machines. Don't hardcode the classname in the format, so that we get the
662
correct value if the method is inherited by a subclass. If you're
663
printing attributes of the object, including strings, you should normally
664
use ``%r`` syntax (to call their repr in turn).
666
Try to avoid the representation becoming more than one or two lines long.
667
(But balance this against including useful information, and simplicity of
670
Because repr methods are often called when something has already gone
671
wrong, they should be written somewhat more defensively than most code.
672
The object may be half-initialized or in some other way in an illegal
673
state. The repr method shouldn't raise an exception, or it may hide the
674
(probably more useful) underlying exception.
679
return '%s(%r)' % (self.__class__.__name__,
686
A bare ``except`` statement will catch all exceptions, including ones that
687
really should terminate the program such as ``MemoryError`` and
688
``KeyboardInterrupt``. They should rarely be used unless the exception is
689
later re-raised. Even then, think about whether catching just
690
``Exception`` (which excludes system errors in Python2.5 and later) would
697
All code should be exercised by the test suite. See `Guide to Testing
698
Bazaar <testing.html>`_ for detailed information about writing tests.
707
We have a commitment to 6 months API stability - any supported symbol in a
708
release of bzr MUST NOT be altered in any way that would result in
709
breaking existing code that uses it. That means that method names,
710
parameter ordering, parameter names, variable and attribute names etc must
711
not be changed without leaving a 'deprecated forwarder' behind. This even
712
applies to modules and classes.
714
If you wish to change the behaviour of a supported API in an incompatible
715
way, you need to change its name as well. For instance, if I add an optional keyword
716
parameter to branch.commit - that's fine. On the other hand, if I add a
717
keyword parameter to branch.commit which is a *required* transaction
718
object, I should rename the API - i.e. to 'branch.commit_transaction'.
720
When renaming such supported API's, be sure to leave a deprecated_method (or
721
_function or ...) behind which forwards to the new API. See the
722
bzrlib.symbol_versioning module for decorators that take care of the
723
details for you - such as updating the docstring, and issuing a warning
724
when the old api is used.
726
For unsupported API's, it does not hurt to follow this discipline, but it's
727
not required. Minimally though, please try to rename things so that
728
callers will at least get an AttributeError rather than weird results.
731
Deprecation decorators
732
----------------------
734
``bzrlib.symbol_versioning`` provides decorators that can be attached to
735
methods, functions, and other interfaces to indicate that they should no
736
longer be used. For example::
738
@deprecated_method(deprecated_in((0, 1, 4)))
740
return self._new_foo()
742
To deprecate a static method you must call ``deprecated_function``
743
(**not** method), after the staticmethod call::
746
@deprecated_function(deprecated_in((0, 1, 4)))
747
def create_repository(base, shared=False, format=None):
749
When you deprecate an API, you should not just delete its tests, because
750
then we might introduce bugs in them. If the API is still present at all,
751
it should still work. The basic approach is to use
752
``TestCase.applyDeprecated`` which in one step checks that the API gives
753
the expected deprecation message, and also returns the real result from
754
the method, so that tests can keep running.
756
Deprecation warnings will be suppressed for final releases, but not for
757
development versions or release candidates, or when running ``bzr
758
selftest``. This gives developers information about whether their code is
759
using deprecated functions, but avoids confusing users about things they
766
Processing Command Lines
767
------------------------
769
bzrlib has a standard framework for parsing command lines and calling
770
processing routines associated with various commands. See builtins.py
771
for numerous examples.
774
Standard Parameter Types
775
------------------------
777
There are some common requirements in the library: some parameters need to be
778
unicode safe, some need byte strings, and so on. At the moment we have
779
only codified one specific pattern: Parameters that need to be unicode
780
should be checked via ``bzrlib.osutils.safe_unicode``. This will coerce the
781
input into unicode in a consistent fashion, allowing trivial strings to be
782
used for programmer convenience, but not performing unpredictably in the
783
presence of different locales.
789
(The strategy described here is what we want to get to, but it's not
790
consistently followed in the code at the moment.)
792
bzrlib is intended to be a generically reusable library. It shouldn't
793
write messages to stdout or stderr, because some programs that use it
794
might want to display that information through a GUI or some other
797
We can distinguish two types of output from the library:
799
1. Structured data representing the progress or result of an
800
operation. For example, for a commit command this will be a list
801
of the modified files and the finally committed revision number
804
These should be exposed either through the return code or by calls
805
to a callback parameter.
807
A special case of this is progress indicators for long-lived
808
operations, where the caller should pass a ProgressBar object.
810
2. Unstructured log/debug messages, mostly for the benefit of the
811
developers or users trying to debug problems. This should always
812
be sent through ``bzrlib.trace`` and Python ``logging``, so that
813
it can be redirected by the client.
815
The distinction between the two is a bit subjective, but in general if
816
there is any chance that a library would want to see something as
817
structured data, we should make it so.
819
The policy about how output is presented in the text-mode client
820
should be only in the command-line tool.
827
Bazaar has online help for various topics through ``bzr help COMMAND`` or
828
equivalently ``bzr command -h``. We also have help on command options,
829
and on other help topics. (See ``help_topics.py``.)
831
As for python docstrings, the first paragraph should be a single-sentence
832
synopsis of the command.
834
The help for options should be one or more proper sentences, starting with
835
a capital letter and finishing with a full stop (period).
837
All help messages and documentation should have two spaces between
841
Handling Errors and Exceptions
842
==============================
844
Commands should return non-zero when they encounter circumstances that
845
the user should really pay attention to - which includes trivial shell
848
Recommended values are:
851
1. Conflicts in merge-like operations, or changes are present in
852
diff-like operations.
853
2. Unrepresentable diff changes (i.e. binary files that we cannot show
855
3. An error or exception has occurred.
856
4. An internal error occurred (one that shows a traceback.)
858
Errors are handled through Python exceptions. Exceptions should be defined
859
inside bzrlib.errors, so that we can see the whole tree at a glance.
861
We broadly classify errors as either being either internal or not,
862
depending on whether ``internal_error`` is set or not. If we think it's our
863
fault, we show a backtrace, an invitation to report the bug, and possibly
864
other details. This is the default for errors that aren't specifically
865
recognized as being caused by a user error. Otherwise we show a briefer
866
message, unless -Derror was given.
868
Many errors originate as "environmental errors" which are raised by Python
869
or builtin libraries -- for example IOError. These are treated as being
870
our fault, unless they're caught in a particular tight scope where we know
871
that they indicate a user errors. For example if the repository format
872
is not found, the user probably gave the wrong path or URL. But if one of
873
the files inside the repository is not found, then it's our fault --
874
either there's a bug in bzr, or something complicated has gone wrong in
875
the environment that means one internal file was deleted.
877
Many errors are defined in ``bzrlib/errors.py`` but it's OK for new errors
878
to be added near the place where they are used.
880
Exceptions are formatted for the user by conversion to a string
881
(eventually calling their ``__str__`` method.) As a convenience the
882
``._fmt`` member can be used as a template which will be mapped to the
883
error's instance dict.
885
New exception classes should be defined when callers might want to catch
886
that exception specifically, or when it needs a substantially different
889
#. If it is something that a caller can recover from, a custom exception
892
#. If it is a data consistency issue, using a builtin like
893
``ValueError``/``TypeError`` is reasonable.
895
#. If it is a programmer error (using an api incorrectly)
896
``AssertionError`` is reasonable.
898
#. Otherwise, use ``BzrError`` or ``InternalBzrError``.
900
Exception strings should start with a capital letter and should not have a
901
final fullstop. If long, they may contain newlines to break the text.
907
Do not use the Python ``assert`` statement, either in tests or elsewhere.
908
A source test checks that it is not used. It is ok to explicitly raise
913
* It makes the behaviour vary depending on whether bzr is run with -O
914
or not, therefore giving a chance for bugs that occur in one case or
915
the other, several of which have already occurred: assertions with
916
side effects, code which can't continue unless the assertion passes,
917
cases where we should give the user a proper message rather than an
919
* It's not that much shorter than an explicit if/raise.
920
* It tends to lead to fuzzy thinking about whether the check is
921
actually needed or not, and whether it's an internal error or not
922
* It tends to cause look-before-you-leap patterns.
923
* It's unsafe if the check is needed to protect the integrity of the
925
* It tends to give poor messages since the developer can get by with
926
no explanatory text at all.
927
* We can't rely on people always running with -O in normal use, so we
928
can't use it for tests that are actually expensive.
929
* Expensive checks that help developers are better turned on from the
930
test suite or a -D flag.
931
* If used instead of ``self.assert*()`` in tests it makes them falsely pass with -O.
937
When you change bzrlib, please update the relevant documentation for the
938
change you made: Changes to commands should update their help, and
939
possibly end user tutorials; changes to the core library should be
940
reflected in API documentation.
945
If you make a user-visible change, please add a note to the NEWS file.
946
The description should be written to make sense to someone who's just
947
a user of bzr, not a developer: new functions or classes shouldn't be
948
mentioned, but new commands, changes in behaviour or fixed nontrivial
949
bugs should be listed. See the existing entries for an idea of what
952
Within each release, entries in the news file should have the most
953
user-visible changes first. So the order should be approximately:
955
* changes to existing behaviour - the highest priority because the
956
user's existing knowledge is incorrect
957
* new features - should be brought to their attention
958
* bug fixes - may be of interest if the bug was affecting them, and
959
should include the bug number if any
960
* major documentation changes
961
* changes to internal interfaces
963
People who made significant contributions to each change are listed in
964
parenthesis. This can include reporting bugs (particularly with good
965
details or reproduction recipes), submitting patches, etc.
970
The docstring of a command is used by ``bzr help`` to generate help output
971
for the command. The list 'takes_options' attribute on a command is used by
972
``bzr help`` to document the options for the command - the command
973
docstring does not need to document them. Finally, the '_see_also'
974
attribute on a command can be used to reference other related help topics.
979
Functions, methods, classes and modules should have docstrings
980
describing how they are used.
982
The first line of the docstring should be a self-contained sentence.
984
For the special case of Command classes, this acts as the user-visible
985
documentation shown by the help command.
987
The docstrings should be formatted as reStructuredText_ (like this
988
document), suitable for processing using the epydoc_ tool into HTML
991
.. _reStructuredText: http://docutils.sourceforge.net/rst.html
992
.. _epydoc: http://epydoc.sourceforge.net/
1001
The copyright policy for bzr was recently made clear in this email (edited
1002
for grammatical correctness)::
1004
The attached patch cleans up the copyright and license statements in
1005
the bzr source. It also adds tests to help us remember to add them
1006
with the correct text.
1008
We had the problem that lots of our files were "Copyright Canonical
1009
Development Ltd" which is not a real company, and some other variations
1010
on this theme. Also, some files were missing the GPL statements.
1012
I want to be clear about the intent of this patch, since copyright can
1013
be a little controversial.
1015
1) The big motivation for this is not to shut out the community, but
1016
just to clean up all of the invalid copyright statements.
1018
2) It has been the general policy for bzr that we want a single
1019
copyright holder for all of the core code. This is following the model
1020
set by the FSF, which makes it easier to update the code to a new
1021
license in case problems are encountered. (For example, if we want to
1022
upgrade the project universally to GPL v3 it is much simpler if there is
1023
a single copyright holder). It also makes it clearer if copyright is
1024
ever debated, there is a single holder, which makes it easier to defend
1025
in court, etc. (I think the FSF position is that if you assign them
1026
copyright, they can defend it in court rather than you needing to, and
1027
I'm sure Canonical would do the same).
1028
As such, Canonical has requested copyright assignments from all of the
1031
3) If someone wants to add code and not attribute it to Canonical, there
1032
is a specific list of files that are excluded from this check. And the
1033
test failure indicates where that is, and how to update it.
1035
4) If anyone feels that I changed a copyright statement incorrectly, just
1036
let me know, and I'll be happy to correct it. Whenever you have large
1037
mechanical changes like this, it is possible to make some mistakes.
1039
Just to reiterate, this is a community project, and it is meant to stay
1040
that way. Core bzr code is copyright Canonical for legal reasons, and
1041
the tests are just there to help us maintain that.
1044
Miscellaneous Topics
1045
####################
1050
Bazaar has a few facilities to help debug problems by going into pdb_, the
1053
.. _pdb: http://docs.python.org/lib/debugger-commands.html
1055
If the ``BZR_PDB`` environment variable is set
1056
then bzr will go into pdb post-mortem mode when an unhandled exception
1059
If you send a SIGQUIT signal to bzr, which can be done by pressing
1060
Ctrl-\\ on Unix, bzr will go into the debugger immediately. You can
1061
continue execution by typing ``c``. This can be disabled if necessary
1062
by setting the environment variable ``BZR_SIGQUIT_PDB=0``.
1069
Integer identifier for a revision on the main line of a branch.
1070
Revision 0 is always the null revision; others are 1-based
1071
indexes into the branch's revision history.
1074
Unicode and Encoding Support
1075
============================
1077
This section discusses various techniques that Bazaar uses to handle
1078
characters that are outside the ASCII set.
1083
When a ``Command`` object is created, it is given a member variable
1084
accessible by ``self.outf``. This is a file-like object, which is bound to
1085
``sys.stdout``, and should be used to write information to the screen,
1086
rather than directly writing to ``sys.stdout`` or calling ``print``.
1087
This file has the ability to translate Unicode objects into the correct
1088
representation, based on the console encoding. Also, the class attribute
1089
``encoding_type`` will effect how unprintable characters will be
1090
handled. This parameter can take one of 3 values:
1093
Unprintable characters will be represented with a suitable replacement
1094
marker (typically '?'), and no exception will be raised. This is for
1095
any command which generates text for the user to review, rather than
1096
for automated processing.
1097
For example: ``bzr log`` should not fail if one of the entries has text
1098
that cannot be displayed.
1101
Attempting to print an unprintable character will cause a UnicodeError.
1102
This is for commands that are intended more as scripting support, rather
1103
than plain user review.
1104
For exampl: ``bzr ls`` is designed to be used with shell scripting. One
1105
use would be ``bzr ls --null --unknows | xargs -0 rm``. If ``bzr``
1106
printed a filename with a '?', the wrong file could be deleted. (At the
1107
very least, the correct file would not be deleted). An error is used to
1108
indicate that the requested action could not be performed.
1111
Do not attempt to automatically convert Unicode strings. This is used
1112
for commands that must handle conversion themselves.
1113
For example: ``bzr diff`` needs to translate Unicode paths, but should
1114
not change the exact text of the contents of the files.
1117
``bzrlib.urlutils.unescape_for_display``
1118
----------------------------------------
1120
Because Transports work in URLs (as defined earlier), printing the raw URL
1121
to the user is usually less than optimal. Characters outside the standard
1122
set are printed as escapes, rather than the real character, and local
1123
paths would be printed as ``file://`` urls. The function
1124
``unescape_for_display`` attempts to unescape a URL, such that anything
1125
that cannot be printed in the current encoding stays an escaped URL, but
1126
valid characters are generated where possible.
1132
The ``bzrlib.osutils`` module has many useful helper functions, including
1133
some more portable variants of functions in the standard library.
1135
In particular, don't use ``shutil.rmtree`` unless it's acceptable for it
1136
to fail on Windows if some files are readonly or still open elsewhere.
1137
Use ``bzrlib.osutils.rmtree`` instead.
1143
We write some extensions in C using pyrex. We design these to work in
1146
* User with no C compiler
1147
* User with C compiler
1150
The recommended way to install bzr is to have a C compiler so that the
1151
extensions can be built, but if no C compiler is present, the pure python
1152
versions we supply will work, though more slowly.
1154
For developers we recommend that pyrex be installed, so that the C
1155
extensions can be changed if needed.
1157
For the C extensions, the extension module should always match the
1158
original python one in all respects (modulo speed). This should be
1159
maintained over time.
1161
To create an extension, add rules to setup.py for building it with pyrex,
1162
and with distutils. Now start with an empty .pyx file. At the top add
1163
"include 'yourmodule.py'". This will import the contents of foo.py into this
1164
file at build time - remember that only one module will be loaded at
1165
runtime. Now you can subclass classes, or replace functions, and only your
1166
changes need to be present in the .pyx file.
1168
Note that pyrex does not support all 2.4 programming idioms, so some
1169
syntax changes may be required. I.e.
1171
- 'from foo import (bar, gam)' needs to change to not use the brackets.
1172
- 'import foo.bar as bar' needs to be 'import foo.bar; bar = foo.bar'
1174
If the changes are too dramatic, consider
1175
maintaining the python code twice - once in the .pyx, and once in the .py,
1176
and no longer including the .py file.
1179
Making Installers for OS Windows
1180
================================
1181
To build a win32 installer, see the instructions on the wiki page:
1182
http://bazaar-vcs.org/BzrWin32Installer
1185
Core Developer Tasks
1186
####################
1191
What is a Core Developer?
1192
-------------------------
1194
While everyone in the Bazaar community is welcome and encouraged to
1195
propose and submit changes, a smaller team is reponsible for pulling those
1196
changes together into a cohesive whole. In addition to the general developer
1197
stuff covered above, "core" developers have responsibility for:
1200
* reviewing blueprints
1202
* managing releases (see the `Releasing Bazaar <../../developers/releasing.html>`_)
1205
Removing barriers to community participation is a key reason for adopting
1206
distributed VCS technology. While DVCS removes many technical barriers,
1207
a small number of social barriers are often necessary instead.
1208
By documenting how the above things are done, we hope to
1209
encourage more people to participate in these activities, keeping the
1210
differences between core and non-core contributors to a minimum.
1213
Communicating and Coordinating
1214
------------------------------
1216
While it has many advantages, one of the challenges of distributed
1217
development is keeping everyone else aware of what you're working on.
1218
There are numerous ways to do this:
1220
#. Assign bugs to yourself in Launchpad
1221
#. Mention it on the mailing list
1222
#. Mention it on IRC
1224
As well as the email notifcations that occur when merge requests are sent
1225
and reviewed, you can keep others informed of where you're spending your
1226
energy by emailing the **bazaar-commits** list implicitly. To do this,
1227
install and configure the Email plugin. One way to do this is add these
1228
configuration settings to your central configuration file (e.g.
1229
``~/.bazaar/bazaar.conf`` on Linux)::
1232
email = Joe Smith <joe.smith@internode.on.net>
1233
smtp_server = mail.internode.on.net:25
1235
Then add these lines for the relevant branches in ``locations.conf``::
1237
post_commit_to = bazaar-commits@lists.canonical.com
1238
post_commit_mailer = smtplib
1240
While attending a sprint, RobertCollins' Dbus plugin is useful for the
1241
same reason. See the documentation within the plugin for information on
1242
how to set it up and configure it.
1251
Of the many workflows supported by Bazaar, the one adopted for Bazaar
1252
development itself is known as "Decentralized with automatic gatekeeper".
1253
To repeat the explanation of this given on
1254
http://bazaar-vcs.org/Workflows:
1257
In this workflow, each developer has their own branch or
1258
branches, plus read-only access to the mainline. A software gatekeeper
1259
(e.g. PQM) has commit rights to the main branch. When a developer wants
1260
their work merged, they request the gatekeeper to merge it. The gatekeeper
1261
does a merge, a compile, and runs the test suite. If the code passes, it
1262
is merged into the mainline.
1264
In a nutshell, here's the overall submission process:
1266
#. get your work ready (including review except for trivial changes)
1267
#. push to a public location
1268
#. ask PQM to merge from that location
1271
At present, PQM always takes the changes to merge from a branch
1272
at a URL that can be read by it. For Bazaar, that means a public,
1273
typically http, URL.
1275
As a result, the following things are needed to use PQM for submissions:
1277
#. A publicly available web server
1278
#. Your OpenPGP key registered with PQM (contact RobertCollins for this)
1279
#. The PQM plugin installed and configured (not strictly required but
1280
highly recommended).
1283
Selecting a Public Branch Location
1284
----------------------------------
1286
If you don't have your own web server running, branches can always be
1287
pushed to Launchpad. Here's the process for doing that:
1289
Depending on your location throughout the world and the size of your
1290
repository though, it is often quicker to use an alternative public
1291
location to Launchpad, particularly if you can set up your own repo and
1292
push into that. By using an existing repo, push only needs to send the
1293
changes, instead of the complete repository every time. Note that it is
1294
easy to register branches in other locations with Launchpad so no benefits
1295
are lost by going this way.
1298
For Canonical staff, http://people.ubuntu.com/~<user>/ is one
1299
suggestion for public http branches. Contact your manager for information
1300
on accessing this system if required.
1302
It should also be noted that best practice in this area is subject to
1303
change as things evolve. For example, once the Bazaar smart server on
1304
Launchpad supports server-side branching, the performance situation will
1305
be very different to what it is now (Jun 2007).
1308
Configuring the PQM Plug-In
1309
---------------------------
1311
While not strictly required, the PQM plugin automates a few things and
1312
reduces the chance of error. Before looking at the plugin, it helps to
1313
understand a little more how PQM operates. Basically, PQM requires an
1314
email indicating what you want it to do. The email typically looks like
1317
star-merge source-branch target-branch
1321
star-merge http://bzr.arbash-meinel.com/branches/bzr/jam-integration http://bazaar-vcs.org/bzr/bzr.dev
1323
Note that the command needs to be on one line. The subject of the email
1324
will be used for the commit message. The email also needs to be ``gpg``
1325
signed with a key that PQM accepts.
1327
The advantages of using the PQM plugin are:
1329
#. You can use the config policies to make it easy to set up public
1330
branches, so you don't have to ever type the full paths you want to merge
1333
#. It checks to make sure the public branch last revision matches the
1334
local last revision so you are submitting what you think you are.
1336
#. It uses the same public_branch and smtp sending settings as bzr-email,
1337
so if you have one set up, you have the other mostly set up.
1339
#. Thunderbird refuses to not wrap lines, and request lines are usually
1340
pretty long (you have 2 long URLs in there).
1342
Here are sample configuration settings for the PQM plugin. Here are the
1343
lines in bazaar.conf::
1346
email = Joe Smith <joe.smith@internode.on.net>
1347
smtp_server=mail.internode.on.net:25
1349
And here are the lines in ``locations.conf`` (or ``branch.conf`` for
1350
dirstate-tags branches)::
1352
[/home/joe/bzr/my-integration]
1353
push_location = sftp://joe-smith@bazaar.launchpad.net/%7Ejoe-smith/bzr/my-integration/
1354
push_location:policy = norecurse
1355
public_branch = http://bazaar.launchpad.net/~joe-smith/bzr/my-integration/
1356
public_branch:policy = appendpath
1357
pqm_email = Bazaar PQM <pqm@bazaar-vcs.org>
1358
pqm_branch = http://bazaar-vcs.org/bzr/bzr.dev
1360
Note that the push settings will be added by the first ``push`` on
1361
a branch. Indeed the preferred way to generate the lines above is to use
1362
``push`` with an argument, then copy-and-paste the other lines into
1369
Here is one possible recipe once the above environment is set up:
1371
#. pull bzr.dev => my-integration
1372
#. merge patch => my-integration
1373
#. fix up any final merge conflicts (NEWS being the big killer here).
1379
The ``push`` step is not required if ``my-integration`` is a checkout of
1382
Because of defaults, you can type a single message into commit and
1383
pqm-commit will reuse that.
1386
Tracking Change Acceptance
1387
--------------------------
1389
The web interface to PQM is https://pqm.bazaar-vcs.org/. After submitting
1390
a change, you can visit this URL to confirm it was received and placed in
1393
When PQM completes processing a change, an email is sent to you with the
1397
Reviewing Blueprints
1398
====================
1400
Blueprint Tracking Using Launchpad
1401
----------------------------------
1403
New features typically require a fair amount of discussion, design and
1404
debate. For Bazaar, that information is often captured in a so-called
1405
"blueprint" on our Wiki. Overall tracking of blueprints and their status
1406
is done using Launchpad's relevant tracker,
1407
https://blueprints.launchpad.net/bzr/. Once a blueprint for ready for
1408
review, please announce it on the mailing list.
1410
Alternatively, send an email begining with [RFC] with the proposal to the
1411
list. In some cases, you may wish to attach proposed code or a proposed
1412
developer document if that best communicates the idea. Debate can then
1413
proceed using the normal merge review processes.
1416
Recording Blueprint Review Feedback
1417
-----------------------------------
1419
Unlike its Bug Tracker, Launchpad's Blueprint Tracker doesn't currently
1420
(Jun 2007) support a chronological list of comment responses. Review
1421
feedback can either be recorded on the Wiki hosting the blueprints or by
1422
using Launchpad's whiteboard feature.
1431
As the two senior developers, Martin Pool and Robert Collins coordinate
1432
the overall Bazaar product development roadmap. Core developers provide
1433
input and review into this, particularly during sprints. It's totally
1434
expected that community members ought to be working on things that
1435
interest them the most. The roadmap is valuable though because it provides
1436
context for understanding where the product is going as a whole and why.
1439
Using Releases and Milestones in Launchpad
1440
------------------------------------------
1442
TODO ... (Exact policies still under discussion)
1448
Keeping on top of bugs reported is an important part of ongoing release
1449
planning. Everyone in the community is welcome and encouraged to raise
1450
bugs, confirm bugs raised by others, and nominate a priority. Practically
1451
though, a good percentage of bug triage is often done by the core
1452
developers, partially because of their depth of product knowledge.
1454
With respect to bug triage, core developers are encouraged to play an
1455
active role with particular attention to the following tasks:
1457
* keeping the number of unconfirmed bugs low
1458
* ensuring the priorities are generally right (everything as critical - or
1459
medium - is meaningless)
1460
* looking out for regressions and turning those around sooner rather than later.
1463
As well as prioritizing bugs and nominating them against a
1464
target milestone, Launchpad lets core developers offer to mentor others in
1469
vim: ft=rst tw=74 ai
added
removed
doc/developers/HACKING.txt
