7
(The current version of this document is available in the file ``HACKING``
8
in the source tree, or at http://doc.bazaar-vcs.org/bzr.dev/hacking.html)
14
Exploring the Bazaar Platform
15
=============================
17
Before making changes, it's a good idea to explore the work already
18
done by others. Perhaps the new feature or improvement you're looking
19
for is available in another plug-in already? If you find a bug,
20
perhaps someone else has already fixed it?
22
To answer these questions and more, take a moment to explore the
23
overall Bazaar Platform. Here are some links to browse:
25
* The Plugins page on the Wiki - http://bazaar-vcs.org/BzrPlugins
27
* The Bazaar product family on Launchpad - https://launchpad.net/bazaar
29
* Bug Tracker for the core product - https://bugs.launchpad.net/bzr/
31
* Blueprint Tracker for the core product - https://blueprints.launchpad.net/bzr/
33
If nothing else, perhaps you'll find inspiration in how other developers
34
have solved their challenges.
37
Planning and Discussing Changes
38
===============================
40
There is a very active community around Bazaar. Mostly we meet on IRC
41
(#bzr on irc.freenode.net) and on the mailing list. To join the Bazaar
42
community, see http://bazaar-vcs.org/BzrSupport.
44
If you are planning to make a change, it's a very good idea to mention it
45
on the IRC channel and/or on the mailing list. There are many advantages
46
to involving the community before you spend much time on a change.
49
* you get to build on the wisdom on others, saving time
51
* if others can direct you to similar code, it minimises the work to be done
53
* it assists everyone in coordinating direction, priorities and effort.
55
In summary, maximising the input from others typically minimises the
56
total effort required to get your changes merged. The community is
57
friendly, helpful and always keen to welcome newcomers.
60
Bazaar Development in a Nutshell
61
================================
63
Looking for a 10 minute introduction to submitting a change?
64
See http://bazaar-vcs.org/BzrGivingBack.
66
TODO: Merge that Wiki page into this document.
69
Understanding the Development Process
70
=====================================
72
The development team follows many best-practices including:
74
* a public roadmap and planning process in which anyone can participate
76
* time based milestones everyone can work towards and plan around
78
* extensive code review and feedback to contributors
80
* complete and rigorous test coverage on any code contributed
82
* automated validation that all tests still pass before code is merged
83
into the main code branch.
85
The key tools we use to enable these practices are:
87
* Launchpad - https://launchpad.net/
89
* Bazaar - http://bazaar-vcs.org/
91
* Bundle Buggy - http://bundlebuggy.aaronbentley.com/
93
* Patch Queue Manager - https://launchpad.net/pqm/
95
For further information, see http://bazaar-vcs.org/BzrDevelopment.
98
A Closer Look at the Merge & Review Process
99
===========================================
101
If you'd like to propose a change, please post to the
102
bazaar@lists.canonical.com list with a bundle, patch, or link to a
103
branch. Put '[PATCH]' or '[MERGE]' in the subject so Bundle Buggy
104
can pick it out, and explain the change in the email message text.
105
Remember to update the NEWS file as part of your change if it makes any
106
changes visible to users or plugin developers. Please include a diff
107
against mainline if you're giving a link to a branch.
109
You can generate a bundle like this::
111
bzr bundle > mybundle.patch
113
A .patch extension is recommended instead of .bundle as many mail clients
114
will send the latter as a binary file. If a bundle would be too long or your
115
mailer mangles whitespace (e.g. implicitly converts Unix newlines to DOS
116
newlines), use the merge-directive command instead like this::
118
bzr merge-directive http://bazaar-vcs.org http://example.org/my_branch > my_directive.patch
120
See the help for details on the arguments to merge-directive.
122
Please do **NOT** put [PATCH] or [MERGE] in the subject line if you don't
123
want it to be merged. If you want comments from developers rather than
124
to be merged, you can put '[RFC]' in the subject line.
126
Anyone is welcome to review code. There are broadly three gates for
129
* Doesn't reduce test coverage: if it adds new methods or commands,
130
there should be tests for them. There is a good test framework
131
and plenty of examples to crib from, but if you are having trouble
132
working out how to test something feel free to post a draft patch
135
* Doesn't reduce design clarity, such as by entangling objects
136
we're trying to separate. This is mostly something the more
137
experienced reviewers need to help check.
139
* Improves bugs, features, speed, or code simplicity.
141
Code that goes in should pass all three. The core developers take care
142
to keep the code quality high and understandable while recognising that
143
perfect is sometimes the enemy of good. (It is easy for reviews to make
144
people notice other things which should be fixed but those things should
145
not hold up the original fix being accepted. New things can easily be
146
recorded in the Bug Tracker instead.)
148
Anyone can "vote" on the mailing list. Core developers can also vote using
149
Bundle Buggy. Here are the voting codes and their explanations.
151
-1 really don't want it in current form
152
-0 somewhat uncomfortable
153
+0 comfortable but resubmission after changes requested
154
+1 conditional good to go after some minor changes
157
+1 conditional is used as a way to avoid another submit/review cycle for
158
patches that need small changes.
160
If a change gets two +1 votes from core reviewers, and no
161
vetos, then it's OK to come in. Any of the core developers can bring it
162
into the bzr.dev trunk and backport it to maintenance branches if required.
163
The Release Manager will merge the change into the branch for a pending
164
release, if any. As a guideline, core developers usually merge their own
165
changes and volunteer to merge other contributions if they were the second
166
reviewer to agree to a change.
168
To track the progress of proposed changes, use Bundle Buggy. See
169
http://bundlebuggy.aaronbentley.com/help for a link to all the
170
outstanding merge requests together with an explanation of the columns.
171
Bundle Buggy will also mail you a link to track just your change.
174
Preparing a Sandbox for Making Changes to Bazaar
175
================================================
177
Bazaar supports many ways of organising your work. See
178
http://bazaar-vcs.org/SharedRepositoryLayouts for a summary of the
179
popular alternatives.
181
Of course, the best choice for you will depend on numerous factors:
182
the number of changes you may be making, the complexity of the changes, etc.
183
As a starting suggestion though:
185
* create a local copy of the main development branch (bzr.dev) by using
188
bzr branch http://bazaar-vcs.org/bzr/bzr.dev/ bzr.dev
190
* keep your copy of bzr.dev prestine (by not developing in it) and keep
191
it up to date (by using bzr pull)
193
* create a new branch off your local bzr.dev copy for each issue
194
(bug or feature) you are working on.
196
This approach makes it easy to go back and make any required changes
197
after a code review. Resubmitting the change is then simple with no
198
risk of accidentially including edits related to other issues you may
199
be working on. After the changes for an issue are accepted and merged,
200
the associated branch can be deleted or archived as you wish.
203
Navigating the Code Base
204
========================
206
TODO: List and describe in one line the purpose of each directory
207
inside an installation of bzr.
209
TODO: Refer to a central location holding an up to date copy of the API
210
documentation generated by epydoc, e.g. something like
211
http://starship.python.net/crew/mwh/bzrlibapi/bzrlib.html.
217
The Importance of Testing
218
=========================
220
Reliability is a critical success factor for any Version Control System.
221
We want Bazaar to be highly reliable across multiple platforms while
222
evolving over time to meet the needs of its community.
224
In a nutshell, this is want we expect and encourage:
226
* New functionality should have test cases. Preferably write the
227
test before writing the code.
229
In general, you can test at either the command-line level or the
230
internal API level. See Writing tests below for more detail.
232
* Try to practice Test-Driven Development: before fixing a bug, write a
233
test case so that it does not regress. Similarly for adding a new
234
feature: write a test case for a small version of the new feature before
235
starting on the code itself. Check the test fails on the old code, then
236
add the feature or fix and check it passes.
238
By doing these things, the Bazaar team gets increased confidence that
239
changes do what they claim to do, whether provided by the core team or
240
by community members. Equally importantly, we can be surer that changes
241
down the track do not break new features or bug fixes that you are
244
As of May 2007, Bazaar ships with a test suite containing over 6000 tests
245
and growing. We are proud of it and want to remain so. As community
246
members, we all benefit from it. Would you trust version control on
247
your project to a product *without* a test suite like Bazaar has?
250
Running the Test Suite
251
======================
253
Currently, bzr selftest is used to invoke tests.
254
You can provide a pattern argument to run a subset. For example,
255
to run just the blackbox tests, run::
257
./bzr selftest -v blackbox
259
To skip a particular test (or set of tests), use the --exclude option
260
(shorthand -x) like so::
262
./bzr selftest -v -x blackbox
264
To list tests without running them, use the --list-only option like so::
266
./bzr selftest --list-only
268
This option can be combined with other selftest options (like -x) and
269
filter patterns to understand their effect.
275
In general tests should be placed in a file named test_FOO.py where
276
FOO is the logical thing under test. That file should be placed in the
277
tests subdirectory under the package being tested.
279
For example, tests for merge3 in bzrlib belong in bzrlib/tests/test_merge3.py.
280
See bzrlib/tests/test_sampler.py for a template test script.
282
Tests can be written for the UI or for individual areas of the library.
283
Choose whichever is appropriate: if adding a new command, or a new command
284
option, then you should be writing a UI test. If you are both adding UI
285
functionality and library functionality, you will want to write tests for
286
both the UI and the core behaviours. We call UI tests 'blackbox' tests
287
and they are found in ``bzrlib/tests/blackbox/*.py``.
289
When writing blackbox tests please honour the following conventions:
291
1. Place the tests for the command 'name' in
292
bzrlib/tests/blackbox/test_name.py. This makes it easy for developers
293
to locate the test script for a faulty command.
295
2. Use the 'self.run_bzr("name")' utility function to invoke the command
296
rather than running bzr in a subprocess or invoking the
297
cmd_object.run() method directly. This is a lot faster than
298
subprocesses and generates the same logging output as running it in a
299
subprocess (which invoking the method directly does not).
301
3. Only test the one command in a single test script. Use the bzrlib
302
library when setting up tests and when evaluating the side-effects of
303
the command. We do this so that the library api has continual pressure
304
on it to be as functional as the command line in a simple manner, and
305
to isolate knock-on effects throughout the blackbox test suite when a
306
command changes its name or signature. Ideally only the tests for a
307
given command are affected when a given command is changed.
309
4. If you have a test which does actually require running bzr in a
310
subprocess you can use ``run_bzr_subprocess``. By default the spawned
311
process will not load plugins unless ``--allow-plugins`` is supplied.
317
We make selective use of doctests__. In general they should provide
318
*examples* within the API documentation which can incidentally be tested. We
319
don't try to test every important case using doctests -- regular Python
320
tests are generally a better solution.
322
Most of these are in ``bzrlib/doc/api``. More additions are welcome.
324
__ http://docs.python.org/lib/module-doctest.html
327
Skipping tests and test requirements
328
------------------------------------
330
In our enhancements to unittest we allow for some addition results beyond
331
just success or failure.
333
If a test can't be run, it can say that it's skipped. This is typically
334
used in parameterized tests - for example if a transport doesn't support
335
setting permissions, we'll skip the tests that relating to that. ::
338
return self.branch_format.initialize(repo.bzrdir)
339
except errors.UninitializableFormat:
340
raise tests.TestSkipped('Uninitializable branch format')
342
Raising TestSkipped is a good idea when you want to make it clear that the
343
test was not run, rather than just returning which makes it look as if it
346
A subtly different case is a test that should run, but can't run in the
347
current environment. This covers tests that can only run in particular
348
operating systems or locales, or that depend on external libraries. Here
349
we want to inform the user that they didn't get full test coverage, but
350
they possibly could if they installed more libraries. These are expressed
351
as a dependency on a feature so we can summarise them, and so that the
352
test for the feature is done only once. (For historical reasons, as of
353
May 2007 many cases that should depend on features currently raise
354
TestSkipped.) The typical use is::
356
class TestStrace(TestCaseWithTransport):
358
_test_needs_features = [StraceFeature]
360
which means all tests in this class need the feature. The feature itself
361
should provide a ``_probe`` method which is called once to determine if
368
Known failures are when a test exists but we know it currently doesn't
369
work, allowing the test suite to still pass. These should be used with
370
care, we don't want a proliferation of quietly broken tests. It might be
371
appropriate to use them if you've committed a test for a bug but not the
372
fix for it, or if something works on Unix but not on Windows.
375
Testing exceptions and errors
376
-----------------------------
378
It's important to test handling of errors and exceptions. Because this
379
code is often not hit in ad-hoc testing it can often have hidden bugs --
380
it's particularly common to get NameError because the exception code
381
references a variable that has since been renamed.
383
.. TODO: Something about how to provoke errors in the right way?
385
In general we want to test errors at two levels:
387
1. A test in ``test_errors.py`` checking that when the exception object is
388
constructed with known parameters it produces an expected string form.
389
This guards against mistakes in writing the format string, or in the
390
``str`` representations of its parameters. There should be one for
391
each exception class.
393
2. Tests that when an api is called in a particular situation, it raises
394
an error of the expected class. You should typically use
395
``assertRaises``, which in the Bazaar test suite returns the exception
396
object to allow you to examine its parameters.
398
In some cases blackbox tests will also want to check error reporting. But
399
it can be difficult to provoke every error through the commandline
400
interface, so those tests are only done as needed -- eg in response to a
401
particular bug or if the error is reported in an unusual way(?) Blackbox
402
tests should mostly be testing how the command-line interface works, so
403
should only test errors if there is something particular to the cli in how
404
they're displayed or handled.
407
Essential Domain Classes
408
########################
410
Introducing the Object Model
411
============================
413
The core domain objects within the bazaar model are:
423
Transports are explained below. See http://bazaar-vcs.org/Classes/
424
for an introduction to the other key classes.
429
The ``Transport`` layer handles access to local or remote directories.
430
Each Transport object acts like a logical connection to a particular
431
directory, and it allows various operations on files within it. You can
432
*clone* a transport to get a new Transport connected to a subdirectory or
435
Transports are not used for access to the working tree. At present
436
working trees are always local and they are accessed through the regular
437
Python file io mechanisms.
442
Transports work in URLs. Take note that URLs are by definition only
443
ASCII - the decision of how to encode a Unicode string into a URL must be
444
taken at a higher level, typically in the Store. (Note that Stores also
445
escape filenames which cannot be safely stored on all filesystems, but
446
this is a different level.)
448
The main reason for this is that it's not possible to safely roundtrip a
449
URL into Unicode and then back into the same URL. The URL standard
450
gives a way to represent non-ASCII bytes in ASCII (as %-escapes), but
451
doesn't say how those bytes represent non-ASCII characters. (They're not
452
guaranteed to be UTF-8 -- that is common but doesn't happen everywhere.)
454
For example if the user enters the url ``http://example/%e0`` there's no
455
way to tell whether that character represents "latin small letter a with
456
grave" in iso-8859-1, or "latin small letter r with acute" in iso-8859-2
457
or malformed UTF-8. So we can't convert their URL to Unicode reliably.
459
Equally problematic if we're given a url-like string containing non-ascii
460
characters (such as the accented a) we can't be sure how to convert that
461
to the correct URL, because we don't know what encoding the server expects
462
for those characters. (Although this is not totally reliable we might still
463
accept these and assume they should be put into UTF-8.)
465
A similar edge case is that the url ``http://foo/sweet%2Fsour`` contains
466
one directory component whose name is "sweet/sour". The escaped slash is
467
not a directory separator. If we try to convert URLs to regular Unicode
468
paths this information will be lost.
470
This implies that Transports must natively deal with URLs; for simplicity
471
they *only* deal with URLs and conversion of other strings to URLs is done
472
elsewhere. Information they return, such as from ``list_dir``, is also in
473
the form of URL components.
482
We have a commitment to 6 months API stability - any supported symbol in a
483
release of bzr MUST NOT be altered in any way that would result in
484
breaking existing code that uses it. That means that method names,
485
parameter ordering, parameter names, variable and attribute names etc must
486
not be changed without leaving a 'deprecated forwarder' behind. This even
487
applies to modules and classes.
489
If you wish to change the behaviour of a supported API in an incompatible
490
way, you need to change its name as well. For instance, if I add an optional keyword
491
parameter to branch.commit - that's fine. On the other hand, if I add a
492
keyword parameter to branch.commit which is a *required* transaction
493
object, I should rename the API - i.e. to 'branch.commit_transaction'.
495
When renaming such supported API's, be sure to leave a deprecated_method (or
496
_function or ...) behind which forwards to the new API. See the
497
bzrlib.symbol_versioning module for decorators that take care of the
498
details for you - such as updating the docstring, and issuing a warning
499
when the old api is used.
501
For unsupported API's, it does not hurt to follow this discipline, but it's
502
not required. Minimally though, please try to rename things so that
503
callers will at least get an AttributeError rather than weird results.
506
Coding Style Guidelines
507
=======================
509
Please write PEP-8__ compliant code.
511
One often-missed requirement is that the first line of docstrings
512
should be a self-contained one-sentence summary.
514
__ http://www.python.org/peps/pep-0008.html
520
* Imports should be done at the top-level of the file, unless there is
521
a strong reason to have them lazily loaded when a particular
522
function runs. Import statements have a cost, so try to make sure
523
they don't run inside hot functions.
525
* Module names should always be given fully-qualified,
526
i.e. ``bzrlib.hashcache`` not just ``hashcache``.
532
Functions, methods or members that are in some sense "private" are given
533
a leading underscore prefix. This is just a hint that code outside the
534
implementation should probably not use that interface.
536
We prefer class names to be concatenated capital words (``TestCase``)
537
and variables, methods and functions to be lowercase words joined by
538
underscores (``revision_id``, ``get_revision``).
540
For the purposes of naming some names are treated as single compound
541
words: "filename", "revno".
543
Consider naming classes as nouns and functions/methods as verbs.
545
Try to avoid using abbreviations in names, because there can be
546
inconsistency if other people use the full name.
552
``revision_id`` not ``rev_id`` or ``revid``
554
Functions that transform one thing to another should be named ``x_to_y``
555
(not ``x2y`` as occurs in some old code.)
561
Python destructors (``__del__``) work differently to those of other
562
languages. In particular, bear in mind that destructors may be called
563
immediately when the object apparently becomes unreferenced, or at some
564
later time, or possibly never at all. Therefore we have restrictions on
565
what can be done inside them.
567
0. Never use a __del__ method without asking Martin/Robert first.
569
1. Never rely on a ``__del__`` method running. If there is code that
570
must run, do it from a ``finally`` block instead.
572
2. Never ``import`` from inside a ``__del__`` method, or you may crash the
575
3. In some places we raise a warning from the destructor if the object
576
has not been cleaned up or closed. This is considered OK: the warning
577
may not catch every case but it's still useful sometimes.
583
In some places we have variables which point to callables that construct
584
new instances. That is to say, they can be used a lot like class objects,
585
but they shouldn't be *named* like classes:
587
> I think that things named FooBar should create instances of FooBar when
588
> called. Its plain confusing for them to do otherwise. When we have
589
> something that is going to be used as a class - that is, checked for via
590
> isinstance or other such idioms, them I would call it foo_class, so that
591
> it is clear that a callable is not sufficient. If it is only used as a
592
> factory, then yes, foo_factory is what I would use.
598
Several places in Bazaar use (or will use) a registry, which is a
599
mapping from names to objects or classes. The registry allows for
600
loading in registered code only when it's needed, and keeping
601
associated information such as a help string or description.
607
To make startup time faster, we use the ``bzrlib.lazy_import`` module to
608
delay importing modules until they are actually used. ``lazy_import`` uses
609
the same syntax as regular python imports. So to import a few modules in a
612
from bzrlib.lazy_import import lazy_import
613
lazy_import(globals(), """
622
revision as _mod_revision,
624
import bzrlib.transport
628
At this point, all of these exist as a ``ImportReplacer`` object, ready to
629
be imported once a member is accessed. Also, when importing a module into
630
the local namespace, which is likely to clash with variable names, it is
631
recommended to prefix it as ``_mod_<module>``. This makes it clearer that
632
the variable is a module, and these object should be hidden anyway, since
633
they shouldn't be imported into other namespaces.
636
Modules versus Members
637
~~~~~~~~~~~~~~~~~~~~~~
639
While it is possible for ``lazy_import()`` to import members of a module
640
when using the ``from module import member`` syntax, it is recommended to
641
only use that syntax to load sub modules ``from module import submodule``.
642
This is because variables and classes can frequently be used without
643
needing a sub-member for example::
645
lazy_import(globals(), """
646
from module import MyClass
650
return isinstance(x, MyClass)
652
This will incorrectly fail, because ``MyClass`` is a ``ImportReplacer``
653
object, rather than the real class.
656
Passing to Other Variables
657
~~~~~~~~~~~~~~~~~~~~~~~~~~
659
It also is incorrect to assign ``ImportReplacer`` objects to other variables.
660
Because the replacer only knows about the original name, it is unable to
661
replace other variables. The ``ImportReplacer`` class will raise an
662
``IllegalUseOfScopeReplacer`` exception if it can figure out that this
663
happened. But it requires accessing a member more than once from the new
664
variable, so some bugs are not detected right away.
670
The null revision is the ancestor of all revisions. Its revno is 0, its
671
revision-id is ``null:``, and its tree is the empty tree. When referring
672
to the null revision, please use ``bzrlib.revision.NULL_REVISION``. Old
673
code sometimes uses ``None`` for the null revision, but this practice is
680
Processing Command Lines
681
------------------------
683
bzrlib has a standard framework for parsing command lines and calling
684
processing routines associated with various commands. See builtins.py
685
for numerous examples.
688
Standard Parameter Types
689
------------------------
691
There are some common requirements in the library: some parameters need to be
692
unicode safe, some need byte strings, and so on. At the moment we have
693
only codified one specific pattern: Parameters that need to be unicode
694
should be checked via ``bzrlib.osutils.safe_unicode``. This will coerce the
695
input into unicode in a consistent fashion, allowing trivial strings to be
696
used for programmer convenience, but not performing unpredictably in the
697
presence of different locales.
703
(The strategy described here is what we want to get to, but it's not
704
consistently followed in the code at the moment.)
706
bzrlib is intended to be a generically reusable library. It shouldn't
707
write messages to stdout or stderr, because some programs that use it
708
might want to display that information through a GUI or some other
711
We can distinguish two types of output from the library:
713
1. Structured data representing the progress or result of an
714
operation. For example, for a commit command this will be a list
715
of the modified files and the finally committed revision number
718
These should be exposed either through the return code or by calls
719
to a callback parameter.
721
A special case of this is progress indicators for long-lived
722
operations, where the caller should pass a ProgressBar object.
724
2. Unstructured log/debug messages, mostly for the benefit of the
725
developers or users trying to debug problems. This should always
726
be sent through ``bzrlib.trace`` and Python ``logging``, so that
727
it can be redirected by the client.
729
The distinction between the two is a bit subjective, but in general if
730
there is any chance that a library would want to see something as
731
structured data, we should make it so.
733
The policy about how output is presented in the text-mode client
734
should be only in the command-line tool.
741
Bazaar has online help for various topics through ``bzr help COMMAND`` or
742
equivalently ``bzr command -h``. We also have help on command options,
743
and on other help topics. (See ``help_topics.py``.)
745
As for python docstrings, the first paragraph should be a single-sentence
746
synopsis of the command.
748
The help for options should be one or more proper sentences, starting with
749
a capital letter and finishing with a full stop (period).
751
All help messages and documentation should have two spaces between
758
In general tests should be placed in a file named test_FOO.py where
759
FOO is the logical thing under test. That file should be placed in the
760
tests subdirectory under the package being tested.
762
For example, tests for merge3 in bzrlib belong in bzrlib/tests/test_merge3.py.
763
See bzrlib/tests/test_sampler.py for a template test script.
765
Tests can be written for the UI or for individual areas of the library.
766
Choose whichever is appropriate: if adding a new command, or a new command
767
option, then you should be writing a UI test. If you are both adding UI
768
functionality and library functionality, you will want to write tests for
769
both the UI and the core behaviours. We call UI tests 'blackbox' tests
770
and they are found in ``bzrlib/tests/blackbox/*.py``.
772
When writing blackbox tests please honour the following conventions:
774
1. Place the tests for the command 'name' in
775
bzrlib/tests/blackbox/test_name.py. This makes it easy for developers
776
to locate the test script for a faulty command.
778
2. Use the 'self.run_bzr("name")' utility function to invoke the command
779
rather than running bzr in a subprocess or invoking the
780
cmd_object.run() method directly. This is a lot faster than
781
subprocesses and generates the same logging output as running it in a
782
subprocess (which invoking the method directly does not).
784
3. Only test the one command in a single test script. Use the bzrlib
785
library when setting up tests and when evaluating the side-effects of
786
the command. We do this so that the library api has continual pressure
787
on it to be as functional as the command line in a simple manner, and
788
to isolate knock-on effects throughout the blackbox test suite when a
789
command changes its name or signature. Ideally only the tests for a
790
given command are affected when a given command is changed.
792
4. If you have a test which does actually require running bzr in a
793
subprocess you can use ``run_bzr_subprocess``. By default the spawned
794
process will not load plugins unless ``--allow-plugins`` is supplied.
800
We have a rich collection of tools to support writing tests. Please use
801
them in preference to ad-hoc solutions as they provide portability and
802
performance benefits.
807
The ``TreeBuilder`` interface allows the construction of arbitrary trees
808
with a declarative interface. A sample session might look like::
810
tree = self.make_branch_and_tree('path')
811
builder = TreeBuilder()
812
builder.start_tree(tree)
813
builder.build(['foo', "bar/", "bar/file"])
814
tree.commit('commit the tree')
815
builder.finish_tree()
817
Please see bzrlib.treebuilder for more details.
822
The ``BranchBuilder`` interface allows the creation of test branches in a
823
quick and easy manner. A sample session::
825
builder = BranchBuilder(self.get_transport().clone('relpath'))
826
builder.build_commit()
827
builder.build_commit()
828
builder.build_commit()
829
branch = builder.get_branch()
831
Please see bzrlib.branchbuilder for more details.
836
We make selective use of doctests__. In general they should provide
837
*examples* within the API documentation which can incidentally be tested. We
838
don't try to test every important case using doctests -- regular Python
839
tests are generally a better solution.
841
Most of these are in ``bzrlib/doc/api``. More additions are welcome.
843
__ http://docs.python.org/lib/module-doctest.html
848
Currently, bzr selftest is used to invoke tests.
849
You can provide a pattern argument to run a subset. For example,
850
to run just the blackbox tests, run::
852
./bzr selftest -v blackbox
854
To skip a particular test (or set of tests), use the --exclude option
855
(shorthand -x) like so::
857
./bzr selftest -v -x blackbox
859
To list tests without running them, use the --list-only option like so::
861
./bzr selftest --list-only
863
This option can be combined with other selftest options (like -x) and
864
filter patterns to understand their effect.
867
Handling Errors and Exceptions
868
==============================
870
Commands should return non-zero when they encounter circumstances that
871
the user should really pay attention to - which includes trivial shell
874
Recommended values are:
877
1. Conflicts in merge-like operations, or changes are present in
878
diff-like operations.
879
2. Unrepresentable diff changes (i.e. binary files that we cannot show
881
3. An error or exception has occurred.
883
Errors are handled through Python exceptions. Exceptions should be defined
884
inside bzrlib.errors, so that we can see the whole tree at a glance.
886
We broadly classify errors as either being either internal or not,
887
depending on whether ``internal_error`` is set or not. If we think it's our
888
fault, we show a backtrace, an invitation to report the bug, and possibly
889
other details. This is the default for errors that aren't specifically
890
recognized as being caused by a user error. Otherwise we show a briefer
891
message, unless -Derror was given.
893
Many errors originate as "environmental errors" which are raised by Python
894
or builtin libraries -- for example IOError. These are treated as being
895
our fault, unless they're caught in a particular tight scope where we know
896
that they indicate a user errors. For example if the repository format
897
is not found, the user probably gave the wrong path or URL. But if one of
898
the files inside the repository is not found, then it's our fault --
899
either there's a bug in bzr, or something complicated has gone wrong in
900
the environment that means one internal file was deleted.
902
Many errors are defined in ``bzrlib/errors.py`` but it's OK for new errors
903
to be added near the place where they are used.
905
Exceptions are formatted for the user by conversion to a string
906
(eventually calling their ``__str__`` method.) As a convenience the
907
``._fmt`` member can be used as a template which will be mapped to the
908
error's instance dict.
910
New exception classes should be defined when callers might want to catch
911
that exception specifically, or when it needs a substantially different
914
Exception strings should start with a capital letter and should not have a
915
final fullstop. If long, they may contain newlines to break the text.
921
When you change bzrlib, please update the relevant documentation for the
922
change you made: Changes to commands should update their help, and
923
possibly end user tutorials; changes to the core library should be
924
reflected in API documentation.
929
If you make a user-visible change, please add a note to the NEWS file.
930
The description should be written to make sense to someone who's just
931
a user of bzr, not a developer: new functions or classes shouldn't be
932
mentioned, but new commands, changes in behaviour or fixed nontrivial
933
bugs should be listed. See the existing entries for an idea of what
936
Within each release, entries in the news file should have the most
937
user-visible changes first. So the order should be approximately:
939
* changes to existing behaviour - the highest priority because the
940
user's existing knowledge is incorrect
941
* new features - should be brought to their attention
942
* bug fixes - may be of interest if the bug was affecting them, and
943
should include the bug number if any
944
* major documentation changes
945
* changes to internal interfaces
947
People who made significant contributions to each change are listed in
948
parenthesis. This can include reporting bugs (particularly with good
949
details or reproduction recipes), submitting patches, etc.
954
The docstring of a command is used by ``bzr help`` to generate help output
955
for the command. The list 'takes_options' attribute on a command is used by
956
``bzr help`` to document the options for the command - the command
957
docstring does not need to document them. Finally, the '_see_also'
958
attribute on a command can be used to reference other related help topics.
963
Functions, methods, classes and modules should have docstrings
964
describing how they are used.
966
The first line of the docstring should be a self-contained sentence.
968
For the special case of Command classes, this acts as the user-visible
969
documentation shown by the help command.
971
The docstrings should be formatted as reStructuredText_ (like this
972
document), suitable for processing using the epydoc_ tool into HTML
975
.. _reStructuredText: http://docutils.sourceforge.net/rst.html
976
.. _epydoc: http://epydoc.sourceforge.net/
985
The copyright policy for bzr was recently made clear in this email (edited
986
for grammatical correctness)::
988
The attached patch cleans up the copyright and license statements in
989
the bzr source. It also adds tests to help us remember to add them
990
with the correct text.
992
We had the problem that lots of our files were "Copyright Canonical
993
Development Ltd" which is not a real company, and some other variations
994
on this theme. Also, some files were missing the GPL statements.
996
I want to be clear about the intent of this patch, since copyright can
997
be a little controversial.
999
1) The big motivation for this is not to shut out the community, but
1000
just to clean up all of the invalid copyright statements.
1002
2) It has been the general policy for bzr that we want a single
1003
copyright holder for all of the core code. This is following the model
1004
set by the FSF, which makes it easier to update the code to a new
1005
license in case problems are encountered. (For example, if we want to
1006
upgrade the project universally to GPL v3 it is much simpler if there is
1007
a single copyright holder). It also makes it clearer if copyright is
1008
ever debated, there is a single holder, which makes it easier to defend
1009
in court, etc. (I think the FSF position is that if you assign them
1010
copyright, they can defend it in court rather than you needing to, and
1011
I'm sure Canonical would do the same).
1012
As such, Canonical has requested copyright assignments from all of the
1015
3) If someone wants to add code and not attribute it to Canonical, there
1016
is a specific list of files that are excluded from this check. And the
1017
test failure indicates where that is, and how to update it.
1019
4) If anyone feels that I changed a copyright statement incorrectly, just
1020
let me know, and I'll be happy to correct it. Whenever you have large
1021
mechanical changes like this, it is possible to make some mistakes.
1023
Just to reiterate, this is a community project, and it is meant to stay
1024
that way. Core bzr code is copyright Canonical for legal reasons, and
1025
the tests are just there to help us maintain that.
1028
Miscellaneous Topics
1029
####################
1034
Bazaar has a few facilities to help debug problems by going into pdb_, the
1037
.. _pdb: http://docs.python.org/lib/debugger-commands.html
1039
If the ``BZR_PDB`` environment variable is set
1040
then bzr will go into pdb post-mortem mode when an unhandled exception
1043
If you send a SIGQUIT signal to bzr, which can be done by pressing
1044
Ctrl-\\ on Unix, bzr will go into the debugger immediately. You can
1045
continue execution by typing ``c``. This can be disabled if necessary
1046
by setting the environment variable ``BZR_SIGQUIT_PDB=0``.
1053
Integer identifier for a revision on the main line of a branch.
1054
Revision 0 is always the null revision; others are 1-based
1055
indexes into the branch's revision history.
1058
Unicode and Encoding Support
1059
============================
1061
This section discusses various techniques that Bazaar uses to handle
1062
characters that are outside the ASCII set.
1067
When a ``Command`` object is created, it is given a member variable
1068
accessible by ``self.outf``. This is a file-like object, which is bound to
1069
``sys.stdout``, and should be used to write information to the screen,
1070
rather than directly writing to ``sys.stdout`` or calling ``print``.
1071
This file has the ability to translate Unicode objects into the correct
1072
representation, based on the console encoding. Also, the class attribute
1073
``encoding_type`` will effect how unprintable characters will be
1074
handled. This parameter can take one of 3 values:
1077
Unprintable characters will be represented with a suitable replacement
1078
marker (typically '?'), and no exception will be raised. This is for
1079
any command which generates text for the user to review, rather than
1080
for automated processing.
1081
For example: ``bzr log`` should not fail if one of the entries has text
1082
that cannot be displayed.
1085
Attempting to print an unprintable character will cause a UnicodeError.
1086
This is for commands that are intended more as scripting support, rather
1087
than plain user review.
1088
For exampl: ``bzr ls`` is designed to be used with shell scripting. One
1089
use would be ``bzr ls --null --unknows | xargs -0 rm``. If ``bzr``
1090
printed a filename with a '?', the wrong file could be deleted. (At the
1091
very least, the correct file would not be deleted). An error is used to
1092
indicate that the requested action could not be performed.
1095
Do not attempt to automatically convert Unicode strings. This is used
1096
for commands that must handle conversion themselves.
1097
For example: ``bzr diff`` needs to translate Unicode paths, but should
1098
not change the exact text of the contents of the files.
1101
``bzrlib.urlutils.unescape_for_display``
1102
----------------------------------------
1104
Because Transports work in URLs (as defined earlier), printing the raw URL
1105
to the user is usually less than optimal. Characters outside the standard
1106
set are printed as escapes, rather than the real character, and local
1107
paths would be printed as ``file://`` urls. The function
1108
``unescape_for_display`` attempts to unescape a URL, such that anything
1109
that cannot be printed in the current encoding stays an escaped URL, but
1110
valid characters are generated where possible.
1116
The ``bzrlib.osutils`` module has many useful helper functions, including
1117
some more portable variants of functions in the standard library.
1119
In particular, don't use ``shutil.rmtree`` unless it's acceptable for it
1120
to fail on Windows if some files are readonly or still open elsewhere.
1121
Use ``bzrlib.osutils.rmtree`` instead.
1127
We write some extensions in C using pyrex. We design these to work in
1130
* User with no C compiler
1131
* User with C compiler
1134
The recommended way to install bzr is to have a C compiler so that the
1135
extensions can be built, but if no C compiler is present, the pure python
1136
versions we supply will work, though more slowly.
1138
For developers we recommend that pyrex be installed, so that the C
1139
extensions can be changed if needed.
1141
For the C extensions, the extension module should always match the
1142
original python one in all respects (modulo speed). This should be
1143
maintained over time.
1145
To create an extension, add rules to setup.py for building it with pyrex,
1146
and with distutils. Now start with an empty .pyx file. At the top add
1147
"include 'yourmodule.py'". This will import the contents of foo.py into this
1148
file at build time - remember that only one module will be loaded at
1149
runtime. Now you can subclass classes, or replace functions, and only your
1150
changes need to be present in the .pyx file.
1152
Note that pyrex does not support all 2.4 programming idioms, so some
1153
syntax changes may be required. I.e.
1155
- 'from foo import (bar, gam)' needs to change to not use the brackets.
1156
- 'import foo.bar as bar' needs to be 'import foo.bar; bar = foo.bar'
1158
If the changes are too dramatic, consider
1159
maintaining the python code twice - once in the .pyx, and once in the .py,
1160
and no longer including the .py file.
1163
Making Installers for OS Windows
1164
================================
1165
To build a win32 installer, see the instructions on the wiki page:
1166
http://bazaar-vcs.org/BzrWin32Installer
1170
vim: ft=rst tw=74 ai
added
removed
doc/developers/HACKING
