342 / 478
We also need a way of resetting the server database in between each test. I’ll
explain the logic of the session-creation code, which should also explain how
this works.
An Additional Hop via subprocess
Because our tests are Python 3, we can’t directly call our Fabric functions, which are
Python 2. Instead, we have to do an extra hop and call the
fab
command as a new
process, like we do from the command line when we do server deploys. Here’s how that
looks, in a module called
server_tools
:
functional_tests/server_tools.py.
from
os
import
path
import
subprocess
THIS_FOLDER
=
path
.
abspath
(
path
.
dirname
(
__file__
))
def
create_session_on_server
(
host
,
):
return
subprocess
.
check_output
(
[
'fab'
,
'create_session_on
_server:email={}'
.
format
(
),
#
'--host={}'
.
format
(
host
),
'--hide=everything,status'
,
#
],
cwd
=
THIS_FOLDER
)
.
decode
()
.
strip
()
#
def
reset_database
(
host
):
subprocess
.
check_call
(
[
'fab'
,
'reset_database'
,
'--host={}'
.
format
(
host
)],
cwd
=
THIS_FOLDER
)
Here we use the
subprocess
module to call some Fabric functions using the
fab
command.
As you can see, the command-line syntax for arguments to
fab
functions is quite
simple, a colon and then a variable=argument syntax.
Incidentally, this is also the first time I’ve shown the “new-style” string
formatting syntax. As you can see it uses curly brackets
{}
instead of
%s
. I slightly
prefer it to the old style, but you’re bound to come across both if you spend any
time with Python. Take a look at some of the examples in the
Python docsto
learn more.
Because of all the hopping around via Fabric and subprocesses, we’re forced to
be quite careful about extracting the session key as a string from the output of
the command as it gets run on the server.
314
|
Chapter 17: Test Fixtures, Logging, and Server-Side Debugging