Sandpiper usage¶

Sandpiper is a WPS server that resolves rules for climatological impacts. To get started, first instatiate the client. Here, the client will try to connect to a remote sandpiper instance using the url parameter.

In [ ]:
from birdy import WPSClient
import os
from wps_tools.testing import get_target_url
from importlib.resources import files

# Ensure we are in the working directory with access to the data
while os.path.basename(os.getcwd()) != "sandpiper":
    os.chdir('../')
In [14]:
# NBVAL_IGNORE_OUTPUT
url = get_target_url("sandpiper")
print(f"Using sandpiper on {url}")

Using sandpiper on https://docker-dev03.pcic.uvic.ca/twitcher/ows/proxy/sandpiper/wps
In [15]:
sandpiper = WPSClient(url)

The list of available processes can be diplayed using the ? command (ex. bird?). Furthermore, help for individual processes can be diplayed using the same command (ex. bird.process?).

In [16]:
# NBVAL_IGNORE_OUTPUT
sandpiper?
sandpiper.resolve_rules?

Signature:
sandpiper.resolve_rules(
    csv=None,
    date_range='2080',
    region='bc',
    geoserver='https://docker-dev03.pcic.uvic.ca/geoserver/bc_regions/ows',
    ensemble='p2a_rules',
    loglevel='INFO',
    connection_string=None,
    thredds=True,
    output_formats=None,
)
Docstring:
Resolve climatological impacts rules

Parameters
----------
csv : ComplexData:mimetype:`text/csv`, :mimetype:`text/plain`
    A CSV document
date_range : {'2020', '2050', '2080'}string
    30 year period for data
region : {'bc', 'alberni_clayoquot', 'boreal_plains', 'bulkley_nechako', 'capital', 'cariboo', 'central_coast', 'central_kootenay', 'central_okanagan', 'columbia_shuswap', ...}string
    Impacted region
geoserver : string
    Geoserver URL
connection_string : string
    Database connection string
ensemble : string
    Ensemble name filter for data files
thredds : boolean
    Data from thredds server. It is not recommended to change from the default (True)
loglevel : {'CRITICAL', 'ERROR', 'WARNING', 'INFO', 'DEBUG', 'NOTSET'}string
    Logging level

Returns
-------
json : ComplexData:mimetype:`application/json`
    JSON file
File:      ~/github/pcic/sandpiper/</tmp/sandpiper-venv/lib/python3.8/site-packages/birdy/client/base.py-3>
Type:      method

Now that we know which process we wish to run, we can use the docstring to ensure we provide the appropriate parameters. The main input is the csv file containing all the rules. You can either pass it raw output from github, or pass a local file. The contents are read as a stream meaning you will have to do a little processing first with a local file.

In [17]:
output_online = sandpiper.resolve_rules(
    csv = 'https://raw.githubusercontent.com/pacificclimate/sandpiper/master/tests/data/rules_small.csv'   
)
In [ ]:
output_local = sandpiper.resolve_rules(
    csv = (files("tests") / "data/rules_small.csv").read_text(),
)
In [19]:
# Access the json contents as a dictionary by setting asobj=True
rules_online = output_online.get(asobj=True)[0]
rules_local = output_local.get(asobj=True)[0]

Once the process has completed we can extract the results and ensure it is what we expected.

In [20]:
expected_rules = {
    'rule_snow': True,
    'rule_hybrid': True,
    'rule_rain': True
}

assert rules_online == expected_rules
assert rules_local == expected_rules
In [ ]: