1
============================
2
Guidelines for modifying bzr
3
============================
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/current/hacking.html)
13
* New functionality should have test cases. Preferably write the
14
test before writing the code.
16
In general, you can test at either the command-line level or the
17
internal API level. See Writing Tests below for more detail.
19
* Try to practice Test-Driven Development. before fixing a bug, write a
20
test case so that it does not regress. Similarly for adding a new
21
feature: write a test case for a small version of the new feature before
22
starting on the code itself. Check the test fails on the old code, then
23
add the feature or fix and check it passes.
25
* Exceptions should be defined inside bzrlib.errors, so that we can
26
see the whole tree at a glance.
28
* Imports should be done at the top-level of the file, unless there is
29
a strong reason to have them lazily loaded when a particular
30
function runs. Import statements have a cost, so try to make sure
31
they don't run inside hot functions.
33
* Module names should always be given fully-qualified,
34
i.e. ``bzrlib.hashcache`` not just ``hashcache``.
36
* Commands should return non-zero when they encounter circumstances that
37
the user should really pay attention to - which includes trivial shell
40
Recommended values are
42
1. Conflicts in merge-like operations, or changes are present in
44
2. Unrepresentable diff changes (i.e. binary files that we cannot show
46
3. An error or exception has occurred.
51
We have a commitment to 6 months API stability - any supported symbol in a
52
release of bzr MUST NOT be altered in any way that would result in
53
breaking existing code that uses it. That means that method names,
54
parameter ordering, parameter names, variable and attribute names etc must
55
not be changed without leaving a 'deprecated forwarder' behind. This even
56
applies to modules and classes.
58
If you wish to change the behaviour of a supported API in an incompatible
59
way, you need to change its name as well. For instance, if I add an optional keyword
60
parameter to branch.commit - that's fine. On the other hand, if I add a
61
keyword parameter to branch.commit which is a *required* transaction
62
object, I should rename the API - i.e. to 'branch.commit_transaction'.
64
When renaming such supported API's, be sure to leave a deprecated_method (or
65
_function or ...) behind which forwards to the new API. See the
66
bzrlib.symbol_versioning module for decorators that take care of the
67
details for you - such as updating the docstring, and issuing a warning
68
when the old api is used.
70
For unsupported API's, it does not hurt to follow this discipline, but it's
71
not required. Minimally though, please try to rename things so that
72
callers will at least get an AttributeError rather than weird results.
75
Standard parameter types
76
------------------------
78
There are some common requirements in the library: some parameters need to be
79
unicode safe, some need byte strings, and so on. At the moment we have
80
only codified one specific pattern: Parameters that need to be unicode
81
should be checked via 'bzrlib.osutils.safe_unicode'. This will coerce the
82
input into unicode in a consistent fashion, allowing trivial strings to be
83
used for programmer convenience, but not performing unpredictably in the
84
presence of different locales.
89
If you change the behaviour of a command, please update its docstring
90
in bzrlib/commands.py. This is displayed by the 'bzr help' command.
95
If you make a user-visible change, please add a note to the NEWS file.
96
The description should be written to make sense to someone who's just
97
a user of bzr, not a developer: new functions or classes shouldn't be
98
mentioned, but new commands, changes in behaviour or fixed nontrivial
99
bugs should be listed. See the existing entries for an idea of what
102
Within each release, entries in the news file should have the most
103
user-visible changes first. So the order should be approximately:
105
* changes to existing behaviour - the highest priority because the
106
user's existing knowledge is incorrect
107
* new features - should be brought to their attention
108
* bug fixes - may be of interest if the bug was affecting them, and
109
should include the bug number if any
110
* major documentation changes
111
* changes to internal interfaces
113
People who made significant contributions to each change are listed in
114
parenthesis. This can include reporting bugs (particularly with good
115
details or reproduction recipes), submitting patches, etc.
120
Functions, methods, classes and modules should have docstrings
121
describing how they are used.
123
The first line of the docstring should be a self-contained sentence.
125
For the special case of Command classes, this acts as the user-visible
126
documentation shown by the help command.
128
The docstrings should be formatted as reStructuredText_ (like this
129
document), suitable for processing using the epydoc_ tool into HTML
132
.. _reStructuredText: http://docutils.sourceforge.net/rst.html
133
.. _epydoc: http://epydoc.sourceforge.net/
140
Please write PEP-8__ compliant code.
142
One often-missed requirement is that the first line of docstrings
143
should be a self-contained one-sentence summary.
145
__ http://www.python.org/peps/pep-0008.html
152
Functions, methods or members that are in some sense "private" are given
153
a leading underscore prefix. This is just a hint that code outside the
154
implementation should probably not use that interface.
156
We prefer class names to be concatenated capital words (``TestCase``)
157
and variables, methods and functions to be lowercase words joined by
158
underscores (``revision_id``, ``get_revision``).
160
For the purposes of naming some names are treated as single compound
161
words: "filename", "revno".
163
Consider naming classes as nouns and functions/methods as verbs.
169
``revision_id`` not ``rev_id`` or ``revid``
171
Functions that transform one thing to another should be named ``x_to_y``
172
(not ``x2y`` as occurs in some old code.)
178
Python destructors (``__del__``) work differently to those of other
179
languages. In particular, bear in mind that destructors may be called
180
immediately when the object apparently becomes unreferenced, or at some
181
later time, or possibly never at all. Therefore we have restrictions on
182
what can be done inside them.
184
0. Never use a __del__ method without asking Martin/Robert first.
186
1. Never rely on a ``__del__`` method running. If there is code that
187
must run, do it from a ``finally`` block instead.
189
2. Never ``import`` from inside a ``__del__`` method, or you may crash the
192
3. In some places we raise a warning from the destructor if the object
193
has not been cleaned up or closed. This is considered OK: the warning
194
may not catch every case but it's still useful sometimes.
200
In some places we have variables which point to callables that construct
201
new instances. That is to say, they can be used a lot like class objects,
202
but they shouldn't be *named* like classes:
204
> I think that things named FooBar should create instances of FooBar when
205
> called. Its plain confusing for them to do otherwise. When we have
206
> something that is going to be used as a class - that is, checked for via
207
> isinstance or other such idioms, them I would call it foo_class, so that
208
> it is clear that a callable is not sufficient. If it is only used as a
209
> factory, then yes, foo_factory is what I would use.
215
Several places in Bazaar use (or will use) a registry, which is a
216
mapping from names to objects or classes. The registry allows for
217
loading in registered code only when it's needed, and keeping
218
associated information such as a help string or description.
224
To make startup time faster, we use the ``bzrlib.lazy_import`` module to
225
delay importing modules until they are actually used. ``lazy_import`` uses
226
the same syntax as regular python imports. So to import a few modules in a
229
from bzrlib.lazy_import import lazy_import
230
lazy_import(globals(), """
239
revision as _mod_revision,
241
import bzrlib.transport
245
At this point, all of these exist as a ``ImportReplacer`` object, ready to
246
be imported once a member is accessed. Also, when importing a module into
247
the local namespace, which is likely to clash with variable names, it is
248
recommended to prefix it as ``_mod_<module>``. This makes it clean that
249
the variable is a module, and these object should be hidden anyway, since
250
they shouldn't be imported into other namespaces.
253
Modules versus Members
254
~~~~~~~~~~~~~~~~~~~~~~
256
While it is possible for ``lazy_import()`` to import members of a module
257
when using the ``from module import member`` syntax, it is recommended to
258
only use that syntax to load sub modules ``from module import submodule``.
259
This is because variables and classes can frequently be used without
260
needing a sub-member for example::
262
lazy_import(globals(), """
263
from module import MyClass
267
return isinstance(x, MyClass)
269
This will incorrectly fail, because ``MyClass`` is a ``ImportReplacer``
270
object, rather than the real class.
273
Passing to other variables
274
~~~~~~~~~~~~~~~~~~~~~~~~~~
276
It also is incorrect to assign ``ImportReplacer`` objects to other variables.
277
Because the replacer only knows about the original name, it is unable to
278
replace other variables. The ``ImportReplacer`` class will raise an
279
``IllegalUseOfScopeReplacer`` exception if it can figure out that this
280
happened. But it requires accessing a member more than once from the new
281
variable, so some bugs are not detected right away.
287
(The strategy described here is what we want to get to, but it's not
288
consistently followed in the code at the moment.)
290
bzrlib is intended to be a generically reusable library. It shouldn't
291
write messages to stdout or stderr, because some programs that use it
292
might want to display that information through a GUI or some other
295
We can distinguish two types of output from the library:
297
1. Structured data representing the progress or result of an
298
operation. For example, for a commit command this will be a list
299
of the modified files and the finally committed revision number
302
These should be exposed either through the return code or by calls
303
to a callback parameter.
305
A special case of this is progress indicators for long-lived
306
operations, where the caller should pass a ProgressBar object.
308
2. Unstructured log/debug messages, mostly for the benefit of the
309
developers or users trying to debug problems. This should always
310
be sent through ``bzrlib.trace`` and Python ``logging``, so that
311
it can be redirected by the client.
313
The distinction between the two is a bit subjective, but in general if
314
there is any chance that a library would want to see something as
315
structured data, we should make it so.
317
The policy about how output is presented in the text-mode client
318
should be only in the command-line tool.
324
In general tests should be placed in a file named test_FOO.py where
325
FOO is the logical thing under test. That file should be placed in the
326
tests subdirectory under the package being tested.
328
For example, tests for merge3 in bzrlib belong in bzrlib/tests/test_merge3.py.
329
See bzrlib/selftest/test_sampler.py for a template test script.
331
Tests can be written for the UI or for individual areas of the library.
332
Choose whichever is appropriate: if adding a new command, or a new command
333
option, then you should be writing a UI test. If you are both adding UI
334
functionality and library functionality, you will want to write tests for
335
both the UI and the core behaviours. We call UI tests 'blackbox' tests
336
and they are found in ``bzrlib/tests/blackbox/*.py``.
338
When writing blackbox tests please honour the following conventions:
340
1. Place the tests for the command 'name' in
341
bzrlib/tests/blackbox/test_name.py. This makes it easy for developers
342
to locate the test script for a faulty command.
344
2. Use the 'self.run_bzr("name")' utility function to invoke the command
345
rather than running bzr in a subprocess or invoking the
346
cmd_object.run() method directly. This is a lot faster than
347
subprocesses and generates the same logging output as running it in a
348
subprocess (which invoking the method directly does not).
350
3. Only test the one command in a single test script. Use the bzrlib
351
library when setting up tests and when evaluating the side-effects of
352
the command. We do this so that the library api has continual pressure
353
on it to be as functional as the command line in a simple manner, and
354
to isolate knock-on effects throughout the blackbox test suite when a
355
command changes its name or signature. Ideally only the tests for a
356
given command are affected when a given command is changed.
358
4. If you have a test which does actually require running bzr in a
359
subprocess you can use ``run_bzr_subprocess``. By default the spawned
360
process will not load plugins unless ``--allow-plugins`` is supplied.
366
We make selective use of doctests__. In general they should provide
367
*examples* within the API documentation which can incidentally be tested. We
368
don't try to test every important case using doctests -- regular Python
369
tests are generally a better solution.
371
Most of these are in ``bzrlib/doc/api``. More additions are welcome.
373
__ http://docs.python.org/lib/module-doctest.html
378
Currently, bzr selftest is used to invoke tests.
379
You can provide a pattern argument to run a subset. For example,
380
to run just the blackbox tests, run::
382
./bzr selftest -v blackbox
384
To skip a particular test (or set of tests), you need to use a negative
387
./bzr selftest '^(?!.*blackbox)'
390
Errors and exceptions
391
=====================
393
Errors are handled through Python exceptions. They can represent user
394
errors, environmental errors or program bugs. Sometimes we can't be sure
395
at the time it's raised which case applies. See bzrlib/errors.py for
396
details on the error-handling practices.
403
Integer identifier for a revision on the main line of a branch.
404
Revision 0 is always the null revision; others are 1-based
405
indexes into the branch's revision history.
411
The ``Transport`` layer handles access to local or remote directories.
412
Each Transport object acts like a logical connection to a particular
413
directory, and it allows various operations on files within it. You can
414
*clone* a transport to get a new Transport connected to a subdirectory or
417
Transports are not used for access to the working tree. At present
418
working trees are always local and they are accessed through the regular
419
Python file io mechanisms.
424
Transports work in URLs. Take note that URLs are by definition only
425
ASCII - the decision of how to encode a Unicode string into a URL must be
426
taken at a higher level, typically in the Store. (Note that Stores also
427
escape filenames which cannot be safely stored on all filesystems, but
428
this is a different level.)
430
The main reason for this is that it's not possible to safely roundtrip a
431
URL into Unicode and then back into the same URL. The URL standard
432
gives a way to represent non-ASCII bytes in ASCII (as %-escapes), but
433
doesn't say how those bytes represent non-ASCII characters. (They're not
434
guaranteed to be UTF-8 -- that is common but doesn't happen everywhere.)
436
For example if the user enters the url ``http://example/%e0`` there's no
437
way to tell whether that character represents "latin small letter a with
438
grave" in iso-8859-1, or "latin small letter r with acute" in iso-8859-2
439
or malformed UTF-8. So we can't convert their URL to Unicode reliably.
441
Equally problematic if we're given a url-like string containing non-ascii
442
characters (such as the accented a) we can't be sure how to convert that
443
to the correct URL, because we don't know what encoding the server expects
444
for those characters. (Although this is not totally reliable we might still
445
accept these and assume they should be put into UTF-8.)
447
A similar edge case is that the url ``http://foo/sweet%2Fsour`` contains
448
one directory component whose name is "sweet/sour". The escaped slash is
449
not a directory separator. If we try to convert URLs to regular Unicode
450
paths this information will be lost.
452
This implies that Transports must natively deal with URLs; for simplicity
453
they *only* deal with URLs and conversion of other strings to URLs is done
454
elsewhere. Information they return, such as from ``list_dir``, is also in
455
the form of URL components.
458
Unicode and Encoding Support
459
============================
461
This section discusses various techniques that Bazaar uses to handle
462
characters that are outside the ASCII set.
467
When a ``Command`` object is created, it is given a member variable
468
accessible by ``self.outf``. This is a file-like object, which is bound to
469
``sys.stdout``, and should be used to write information to the screen,
470
rather than directly writing to ``sys.stdout`` or calling ``print``.
471
This file has the ability to translate Unicode objects into the correct
472
representation, based on the console encoding. Also, the class attribute
473
``encoding_type`` will effect how unprintable characters will be
474
handled. This parameter can take one of 3 values:
477
Unprintable characters will be represented with a suitable replacement
478
marker (typically '?'), and no exception will be raised. This is for
479
any command which generates text for the user to review, rather than
480
for automated processing.
481
For example: ``bzr log`` should not fail if one of the entries has text
482
that cannot be displayed.
485
Attempting to print an unprintable character will cause a UnicodeError.
486
This is for commands that are intended more as scripting support, rather
487
than plain user review.
488
For exampl: ``bzr ls`` is designed to be used with shell scripting. One
489
use would be ``bzr ls --null --unknows | xargs -0 rm``. If ``bzr``
490
printed a filename with a '?', the wrong file could be deleted. (At the
491
very least, the correct file would not be deleted). An error is used to
492
indicate that the requested action could not be performed.
495
Do not attempt to automatically convert Unicode strings. This is used
496
for commands that must handle conversion themselves.
497
For example: ``bzr diff`` needs to translate Unicode paths, but should
498
not change the exact text of the contents of the files.
501
``bzrlib.urlutils.unescape_for_display``
502
----------------------------------------
504
Because Transports work in URLs (as defined earlier), printing the raw URL
505
to the user is usually less than optimal. Characters outside the standard
506
set are printed as escapes, rather than the real character, and local
507
paths would be printed as ``file://`` urls. The function
508
``unescape_for_display`` attempts to unescape a URL, such that anything
509
that cannot be printed in the current encoding stays an escaped URL, but
510
valid characters are generated where possible.
516
If you'd like to propose a change, please post to the
517
bazaar-ng@lists.canonical.com list with a patch, bzr changeset, or link to a
518
branch. Please put '[patch]' in the subject so we can pick them out, and
519
include some text explaining the change. Remember to put an update to the NEWS
520
file in your diff, if it makes any changes visible to users or plugin
521
developers. Please include a diff against mainline if you're giving a link to
524
Please indicate if you think the code is ready to merge, or if it's just a
525
draft or for discussion. If you want comments from many developers rather than
526
to be merged, you can put '[rfc]' in the subject lines.
528
Anyone is welcome to review code. There are broadly three gates for
531
* Doesn't reduce test coverage: if it adds new methods or commands,
532
there should be tests for them. There is a good test framework
533
and plenty of examples to crib from, but if you are having trouble
534
working out how to test something feel free to post a draft patch
537
* Doesn't reduce design clarity, such as by entangling objects
538
we're trying to separate. This is mostly something the more
539
experienced reviewers need to help check.
541
* Improves bugs, features, speed, or code simplicity.
543
Code that goes in should pass all three.
545
If you read a patch please reply and say so. We can use a numeric scale
546
of -1, -0, +0, +1, meaning respectively "really don't want it in current
547
form", "somewhat uncomfortable", "ok with me", and "please put it in".
548
Anyone can "vote". (It's not really voting, just a terse expression.)
550
If something gets say two +1 votes from core reviewers, and no
551
vetos, then it's OK to come in. Any of the core developers can bring it
552
into their integration branch, which I'll merge regularly. (If you do
553
so, please reply and say so.)
556
Making installers for OS Windows
557
================================
558
To build a win32 installer, see the instructions on the wiki page:
559
http://bazaar-vcs.org/BzrWin32Installer
562
:: vim: ft=rst tw=74 ai
added
removed
HACKING
