Responses are units of data in Capris that represent the stdout, stderr, exit code, environment, history, and exception that running command used/raised. In short, they are encapsulation around several components.
You will get responses if you run non-transactional runnables using their run method. For the rest of this documentation I will use run instead of “call the run method” where suitable. An example of a response:
>>> res = git.log(n=10).run()
>>> res.ok()
True
To check the exit code of a command you need to use the status_code attribute. For example:
>>> res = r.run()
>>> res.status_code
0
However, keep in mind that if the runnable is either a pipe or an iostream object the exit code of the response object will only give you the exit code of the last command ran. For example, if you did the following:
>>> r = git.log(n=10, pretty="format:%m") | grep('message')
>>> res = r.run()
The status_code of your res variable will be that of grep, not git. To access that of git you will have to look into the history of the response object.
The history of a response stores the previously ran commands prior to the last command (the one which the current response object represents). This attribute is only set in the last command, prehaps best illustrated with an example:
>>> res = pipe.run()
>>> res.history
[<Response [git]>, <Response [grep]>]
>>> res.history[-1].history
[]
If you run pipes or iostreams for pipes, you should get the above if the pipe contains more than one command. The validity of the history is not enforced by the response object, i.e. you can append to it and modify it as you like.
To access the standard output of the command, as well as the contents of stderr streams you can use the std_out and std_err attribtues respectively. For a simple example:
>>> res = echo('message').run()
>>> res.std_out
'message\n'
>>> res.std_err
''