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
|
<?xml version="1.0" encoding="utf-8" ?>
<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd">
<html xmlns="http://www.w3.org/1999/xhtml" xml:lang="en" lang="en">
<head>
<meta http-equiv="Content-Type" content="text/html; charset=utf-8" />
<meta name="generator" content="Docutils 0.4: http://docutils.sourceforge.net/" />
<title>Programming guidelines</title>
<link rel="stylesheet" href="html4css1.css" type="text/css" />
</head>
<body>
<div class="header">
<a class="reference" href="connection-ref.html">Prev</a> <a class="reference" href="index.html">Up</a> <a class="reference" href="tests.html">Next</a>
<hr class="header"/>
</div>
<div class="document" id="programming-guidelines">
<h1 class="title">Programming guidelines</h1>
<p>There are certain guidelines and idioms which should be adhered to
when using the <tt class="docutils literal"><span class="pre">processing</span></tt> package.</p>
<div class="section">
<h1><a id="all-platforms" name="all-platforms">All platforms</a></h1>
<dl class="docutils">
<dt><em>Avoid shared state</em></dt>
<dd><p class="first">As far as possible one should try to avoid shifting large amounts
of data between processes.</p>
<p class="last">It is probably best to stick to using queues or pipes for
communication between processes rather than using the lower level
synchronization primitives from the <tt class="docutils literal"><span class="pre">threading</span></tt> module.</p>
</dd>
<dt><em>Picklability</em>:</dt>
<dd>Ensure that the arguments to the methods of proxies are
picklable.</dd>
<dt><em>Thread safety of proxies</em>:</dt>
<dd><p class="first">Do not use a proxy object from more than one thread unless you
protect it with a lock.</p>
<p class="last">(There is never a problem with different processes using the
'same' proxy.)</p>
</dd>
<dt><em>Joining zombie processes</em></dt>
<dd>On Unix when a process finishes but has not been joined it becomes
a zombie. There should never be very many because each time a new
process starts (or <tt class="docutils literal"><span class="pre">activeChildren()</span></tt> is called) all completed
processes which have not yet been joined will be joined. Also
calling a finished process's <tt class="docutils literal"><span class="pre">isAlive()</span></tt> will join the process.
Even so it is probably good practice to explicitly join all the
processes that you start.</dd>
<dt><em>Better to inherit than pickle/unpickle</em></dt>
<dd>On Windows many of types from the <tt class="docutils literal"><span class="pre">processing</span></tt> package need to be
picklable so that child processes can use them. However, one
should generally avoid sending shared objects to other processes
using pipes or queues. Instead you should arrange the program so
that a process which need access to a shared resource created
elsewhere can inherit it from an ancestor process.</dd>
<dt><em>Avoid terminating processes</em></dt>
<dd><p class="first">Using the <tt class="docutils literal"><span class="pre">terminate()</span></tt> method to stop a process is liable to
cause any shared resources (such as locks, semaphores, pipes and
queues) currently being used by the process to become broken or
unavailable to other processes.</p>
<p class="last">Therefore it is probably best to only consider using <tt class="docutils literal"><span class="pre">terminate()</span></tt>
on processes which never use any shared resources.</p>
</dd>
<dt><em>Joining processes that use queues</em></dt>
<dd><p class="first">Bear in mind that a process that has put items in a queue will
wait before terminating until all the buffered items are fed by
the "feeder" thread to the underlying pipe. (The child process
can call the <tt class="docutils literal"><span class="pre">cancelJoin()</span></tt> method of the queue to avoid this
behaviour.)</p>
<p>This means that whenever you use a queue you need to make sure
that all items which have been put on the queue will eventually be
removed before the process is joined. Otherwise you cannot be
sure that processes which have put items on the queue will
terminate. Remember also that non-daemonic processes will be
automatically be joined.</p>
<p>An example which will deadlock is the following:</p>
<pre class="literal-block">
from processing import Process, Queue
def f(q):
q.put('X' * 1000000)
if __name__ == '__main__':
queue = Queue()
p = Process(target=f, args=(queue,))
p.start()
p.join() # this deadlocks
obj = queue.get()
</pre>
<p class="last">A fix here would be to swap the last two lines round (or simply
remove the <tt class="docutils literal"><span class="pre">p.join()</span></tt> line).</p>
</dd>
<dt><em>Explicity pass resources to child processes</em></dt>
<dd><p class="first">On Unix a child process can make use of a shared resource created
in a parent process using a global resource. However, it is
better to pass the object as an argument to the constructor for
the child process.</p>
<p>Apart from making the code (potentially) compatible with Windows
this also ensures that as long as the child process is still alive
the object will not be garbage collected in the parent process.
This might be important if some resource is freed when the object
is garbage collected in the parent process.</p>
<p>So for instance</p>
<pre class="literal-block">
from processing import Process, Lock
def f():
... do something using "lock" ...
if __name__ == '__main__':
lock = Lock()
for i in range(10):
Process(target=f).start()
</pre>
<p>should be rewritten as</p>
<pre class="last literal-block">
from processing import Process, Lock
def f(l):
... do something using "l" ...
if __name__ == '__main__':
lock = Lock()
for i in range(10):
Process(target=f, args=(lock,)).start()
</pre>
</dd>
</dl>
</div>
<div class="section">
<h1><a id="windows" name="windows">Windows</a></h1>
<p>Since Windows lacks <tt class="docutils literal"><span class="pre">os.fork()</span></tt> it has a few extra restrictions:</p>
<p><em>More picklability</em>:</p>
<blockquote>
<p>Ensure that all arguments to <tt class="docutils literal"><span class="pre">Process.__init__()</span></tt> are picklable.
This means, in particular, that bound or unbound methods cannot be
used directly as the <tt class="docutils literal"><span class="pre">target</span></tt> argument on Windows --- just define
a function and use that instead.</p>
<p>Also, if you subclass <tt class="docutils literal"><span class="pre">Process</span></tt> then make sure that instances
will be picklable when the <tt class="docutils literal"><span class="pre">start()</span></tt> method is called.</p>
</blockquote>
<dl class="docutils">
<dt><em>Global variables</em>:</dt>
<dd><p class="first">Bear in mind that if code run in a child process tries to access a
global variable, then the value it sees (if any) may not be the
same as the value in the parent process at the time that
<tt class="docutils literal"><span class="pre">start()</span></tt> was called.</p>
<p class="last">However, global variables which are just module level constants
cause no problems.</p>
</dd>
<dt><em>Safe importing of main module</em>:</dt>
<dd><p class="first">Make sure that the main module can be safely imported by a new
Python interpreter without causing unintended side effects (such a
starting a new process).</p>
<p>For example, under Windows running the following module would
fail with a <tt class="docutils literal"><span class="pre">RuntimeError</span></tt>:</p>
<pre class="literal-block">
from processing import Process
def foo():
print 'hello'
p = Process(target=foo)
p.start()
</pre>
<p>Instead one should protect the "entry point" of the program by
using <tt class="docutils literal"><span class="pre">if</span> <span class="pre">__name__</span> <span class="pre">==</span> <span class="pre">'__main__':</span></tt> as follows:</p>
<pre class="literal-block">
from processing import Process
def foo():
print 'hello'
if __name__ == '__main__':
freezeSupport()
p = Process(target=foo)
p.start()
</pre>
<p>(The <tt class="docutils literal"><span class="pre">freezeSupport()</span></tt> line can be ommitted if the program will
be run normally instead of frozen.)</p>
<p>This allows the newly spawned Python interpreter to safely import
the module and then run the module's <tt class="docutils literal"><span class="pre">foo()</span></tt> function.</p>
<p class="last">Similar restrictions apply if a pool or manager is created in the
main module.</p>
</dd>
</dl>
</div>
</div>
<div class="footer">
<hr class="footer" />
<a class="reference" href="connection-ref.html">Prev</a> <a class="reference" href="index.html">Up</a> <a class="reference" href="tests.html">Next</a>
</div>
</body>
</html>
|
