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
|
Testing output to streams
=========================
.. currentmodule:: testfixtures
In many situations, it's perfectly legitimate for output to be printed
to one of the standard streams. To aid with testing this kind of
output, testfixtures provides the :class:`OutputCapture` helper.
This helper is a context manager that captures output sent to
``sys.stdout`` and ``sys.stderr`` and provides a
:meth:`~OutputCapture.compare` method to check that the output was as
expected.
Here's a simple example:
.. code-block:: python
from testfixtures import OutputCapture
import sys
with OutputCapture() as output:
# code under test
print("Hello!")
print("Something bad happened!", file=sys.stderr)
output.compare('\n'.join([
"Hello!",
"Something bad happened!",
]))
To make life easier, both the actual and expected output are stripped
of leading and trailing whitespace before the comparison is done:
>>> with OutputCapture() as o:
... print(' Bar! ')
... o.compare(' Foo! ')
Traceback (most recent call last):
...
AssertionError: 'Foo!' (expected) != 'Bar!' (actual)
However, if you need to make very explicit assertions about what has
been written to the stream then you can do so using the `captured`
property of the :class:`OutputCapture`:
>>> with OutputCapture() as o:
... print(' Bar! ')
>>> print(repr(o.captured))
' Bar! \n'
If you need to explicitly check whether output went to ``stdout`` or ``stderr``,
`separate` mode can be used:
.. code-block:: python
from testfixtures import OutputCapture
import sys
with OutputCapture(separate=True) as output:
print("Hello!")
print("Something bad happened!", file=sys.stderr)
output.compare(
stdout="Hello!",
stderr="Something bad happened!",
)
Finally, you may sometimes want to disable an :class:`OutputCapture`
without removing it from your code. This often happens when you want
to insert a :any:`breakpoint` call while an :class:`OutputCapture` is active;
if it remains enabled, all debugger output will be captured making the
debugger very difficult to use!
To deal with this problem, the :class:`OutputCapture` may be disabled
and then re-enabled as follows:
>>> with OutputCapture() as o:
... print('Foo')
... o.disable()
... print('Bar')
... o.enable()
... print('Baz')
Bar
>>> print(o.captured)
Foo
Baz
<BLANKLINE>
.. note::
Some debuggers, notably :mod:`pdb`, do interesting things with streams
such that calling :meth:`~OutputCapture.disable` from within the debugger
will have no effect. A good fallback is to type the following, which will
almost always restore output to where you want it:
.. code-block:: python
import sys; sys.stdout=sys.__stdout__
|
