1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
|
Nova Style Commandments
=======================
- Step 1: Read http://www.python.org/dev/peps/pep-0008/
- Step 2: Read http://www.python.org/dev/peps/pep-0008/ again
- Step 3: Read on
General
-------
- Put two newlines between top-level code (funcs, classes, etc)
- Put one newline between methods in classes and anywhere else
- Long lines should be wrapped in parentheses
in preference to using a backslash for line continuation.
- Do not write "except:", use "except Exception:" at the very least
- Include your name with TODOs as in "#TODO(termie)"
- Do not shadow a built-in or reserved word. Example::
def list():
return [1, 2, 3]
mylist = list() # BAD, shadows `list` built-in
class Foo(object):
def list(self):
return [1, 2, 3]
mylist = Foo().list() # OKAY, does not shadow built-in
Imports
-------
- Do not import objects, only modules (*)
- Do not import more than one module per line (*)
- Do not make relative imports
- Do not make new nova.db imports in nova/virt/*
- Order your imports by the full module path
- Organize your imports according to the following template
(*) exceptions are:
- imports from ``migrate`` package
- imports from ``sqlalchemy`` package
- imports from ``nova.db.sqlalchemy.session`` module
Example::
# vim: tabstop=4 shiftwidth=4 softtabstop=4
{{stdlib imports in human alphabetical order}}
\n
{{third-party lib imports in human alphabetical order}}
\n
{{nova imports in human alphabetical order}}
\n
\n
{{begin your code}}
Human Alphabetical Order Examples
---------------------------------
Example::
import httplib
import logging
import random
import StringIO
import time
import unittest
import eventlet
import webob.exc
import nova.api.ec2
from nova.api import openstack
from nova.auth import users
from nova.endpoint import cloud
import nova.flags
from nova import test
Docstrings
----------
Example::
"""A one line docstring looks like this and ends in a period."""
"""A multi line docstring has a one-line summary, less than 80 characters.
Then a new paragraph after a newline that explains in more detail any
general information about the function, class or method. Example usages
are also great to have here if it is a complex class for function.
When writing the docstring for a class, an extra line should be placed
after the closing quotations. For more in-depth explanations for these
decisions see http://www.python.org/dev/peps/pep-0257/
If you are going to describe parameters and return values, use Sphinx, the
appropriate syntax is as follows.
:param foo: the foo parameter
:param bar: the bar parameter
:returns: return_type -- description of the return value
:returns: description of the return value
:raises: AttributeError, KeyError
"""
Dictionaries/Lists
------------------
If a dictionary (dict) or list object is longer than 80 characters, its items
should be split with newlines. Embedded iterables should have their items
indented. Additionally, the last item in the dictionary should have a trailing
comma. This increases readability and simplifies future diffs.
Example::
my_dictionary = {
"image": {
"name": "Just a Snapshot",
"size": 2749573,
"properties": {
"user_id": 12,
"arch": "x86_64",
},
"things": [
"thing_one",
"thing_two",
],
"status": "ACTIVE",
},
}
Calling Methods
---------------
Calls to methods 80 characters or longer should format each argument with
newlines. This is not a requirement, but a guideline::
unnecessarily_long_function_name('string one',
'string two',
kwarg1=constants.ACTIVE,
kwarg2=['a', 'b', 'c'])
Rather than constructing parameters inline, it is better to break things up::
list_of_strings = [
'what_a_long_string',
'not as long',
]
dict_of_numbers = {
'one': 1,
'two': 2,
'twenty four': 24,
}
object_one.call_a_method('string three',
'string four',
kwarg1=list_of_strings,
kwarg2=dict_of_numbers)
Internationalization (i18n) Strings
-----------------------------------
In order to support multiple languages, we have a mechanism to support
automatic translations of exception and log strings.
Example::
msg = _("An error occurred")
raise HTTPBadRequest(explanation=msg)
If you have a variable to place within the string, first internationalize the
template string then do the replacement.
Example::
msg = _("Missing parameter: %s") % ("flavor",)
LOG.error(msg)
If you have multiple variables to place in the string, use keyword parameters.
This helps our translators reorder parameters when needed.
Example::
msg = _("The server with id %(s_id)s has no key %(m_key)s")
LOG.error(msg % {"s_id": "1234", "m_key": "imageId"})
Creating Unit Tests
-------------------
For every new feature, unit tests should be created that both test and
(implicitly) document the usage of said feature. If submitting a patch for a
bug that had no unit test, a new passing unit test should be added. If a
submitted bug fix does have a unit test, be sure to add a new one that fails
without the patch and passes with the patch.
For more information on creating unit tests and utilizing the testing
infrastructure in OpenStack Nova, please read nova/testing/README.rst.
openstack-common
----------------
A number of modules from openstack-common are imported into the project.
These modules are "incubating" in openstack-common and are kept in sync
with the help of openstack-common's update.py script. See:
http://wiki.openstack.org/CommonLibrary#Incubation
The copy of the code should never be directly modified here. Please
always update openstack-common first and then run the script to copy
the changes across.
OpenStack Trademark
-------------------
OpenStack is a registered trademark of OpenStack, LLC, and uses the
following capitalization:
OpenStack
Commit Messages
---------------
Using a common format for commit messages will help keep our git history
readable. Follow these guidelines:
First, provide a brief summary (it is recommended to keep the commit title
under 50 chars).
The first line of the commit message should provide an accurate
description of the change, not just a reference to a bug or
blueprint. It must be followed by a single blank line.
If the change relates to a specific driver (libvirt, xenapi, qpid, etc...),
begin the first line of the commit message with the driver name, lowercased,
followed by a colon.
Following your brief summary, provide a more detailed description of
the patch, manually wrapping the text at 72 characters. This
description should provide enough detail that one does not have to
refer to external resources to determine its high-level functionality.
Once you use 'git review', two lines will be appended to the commit
message: a blank line followed by a 'Change-Id'. This is important
to correlate this commit with a specific review in Gerrit, and it
should not be modified.
For further information on constructing high quality commit messages,
and how to split up commits into a series of changes, consult the
project wiki:
http://wiki.openstack.org/GitCommitMessages
|