I recently ran into a similar problem while developing a native
python DBus implementation. Initially I implemented only the
client-side portion and pointed all my tests at the OS-provided
session bus. However, once I got around to implementing the
server-side portion of DBus, simply pointing al
On Mon, Sep 26, 2011 at 01:37:16PM +0200, Laurens Van Houtven wrote:
> The obvious solution is to have a makeClient function defined in module
> scope for the given test module. I could then have a bin/liveTest.py that
> takes a username, password and cert, and monkeypatches that module. I'm
> assu
On Sep 26, 2011, at 7:37 AM, Laurens Van Houtven wrote:
> Are there any better ways to pass test data to trial?
Environment variables.
This isn't a great option, but I'd say it's definitely a better one than
modules which monkeypatch stuff.
You can also have your test cases synthesize stuff o
Hey,
In order to properly test txievery (a twisted library for dealing with
Paypal's APIs), I'm building a mock Paypal server. The idea is to run the
same high-level tests (which basically use the API and eventually make it
make real HTTPS requests) against both the mock server and the Paypal
San
