Make pipelines able to run and load data and check errors from dynamic context

[lisad]

I started to write documentation for install.md, and found a couple things I really wanted to do for an example, but the library can't do yet.

• Loading data: it should be possible to build the pipeline into HTTP views, so that the request.body is treated as data fort he pipeline.
• Checking errors: it should be possible to check the whole pipeline for errors e.g. for the use case below.
• Output: I didn't even get to this part... in the example of the pipeline running in response to an HTTP request, how does the pipeline's output get into the db.

from phaser import Pipeline, Phase, row_step

@row_step
def my_step(row):
    row['pudding'] = 'proof'

my_pipeline = Pipeline(
    phases = [
        Phase(steps=[my_step])
    ]
)

def my_import_data_view(request):
    data = request.body
    my_pipeline.load(data)
    my_pipeline.run()
    if my_pipeline.context.has_errors():
        return HttpResponse(status=401)
    else:
        return HttpResponse(status=200)