Source code for tawhiri.api

# Copyright 2014 (C) Priyesh Patel
#
# This file is part of Tawhiri.
#
# Tawhiri is free software: you can redistribute it and/or modify
# it under the terms of the GNU General Public License as published by
# the Free Software Foundation, either version 3 of the License, or
# (at your option) any later version.
#
# Tawhiri is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
# GNU General Public License for more details.
#
# You should have received a copy of the GNU General Public License
# along with Tawhiri.  If not, see <http://www.gnu.org/licenses/>.

"""
Provide the HTTP API for Tawhiri.
"""

from flask import Flask, jsonify, request, g
from datetime import datetime
import time
import strict_rfc3339

from tawhiri import solver, models
from tawhiri.dataset import Dataset as WindDataset
from ruaumoko import Dataset as ElevationDataset

app = Flask(__name__)

API_VERSION = 1
LATEST_DATASET_KEYWORD = "latest"
PROFILE_STANDARD = "standard_profile"
PROFILE_FLOAT = "float_profile"


# Util functions ##############################################################
[docs]def ruaumoko_ds(): if not hasattr("ruaumoko_ds", "once"): ds_loc = app.config.get('ELEVATION_DATASET', ElevationDataset.default_location) ruaumoko_ds.once = ElevationDataset(ds_loc) return ruaumoko_ds.once
def _rfc3339_to_timestamp(dt): """ Convert from a RFC3339 timestamp to a UNIX timestamp. """ return strict_rfc3339.rfc3339_to_timestamp(dt) def _timestamp_to_rfc3339(dt): """ Convert from a UNIX timestamp to a RFC3339 timestamp. """ return strict_rfc3339.timestamp_to_rfc3339_utcoffset(dt) # Exceptions ##################################################################
[docs]class APIException(Exception): """ Base API exception. """ status_code = 500
[docs]class RequestException(APIException): """ Raised if request is invalid. """ status_code = 400
[docs]class InvalidDatasetException(APIException): """ Raised if the dataset specified in the request is invalid. """ status_code = 404
[docs]class PredictionException(APIException): """ Raised if the solver raises an exception. """ status_code = 500
[docs]class InternalException(APIException): """ Raised when an internal error occurs. """ status_code = 500
[docs]class NotYetImplementedException(APIException): """ Raised when the functionality has not yet been implemented. """ status_code = 501
# Request #####################################################################
[docs]def parse_request(data): """ Parse the request. """ req = {"version": API_VERSION} # Generic fields req['launch_latitude'] = \ _extract_parameter(data, "launch_latitude", float, validator=lambda x: -90 <= x <= 90) req['launch_longitude'] = \ _extract_parameter(data, "launch_longitude", float, validator=lambda x: 0 <= x < 360) req['launch_datetime'] = \ _extract_parameter(data, "launch_datetime", _rfc3339_to_timestamp) req['launch_altitude'] = \ _extract_parameter(data, "launch_altitude", float, ignore=True) # If no launch altitude provided, use Ruaumoko to look it up if req['launch_altitude'] is None: try: req['launch_altitude'] = ruaumoko_ds().get(req['launch_latitude'], req['launch_longitude']) except Exception: raise InternalException("Internal exception experienced whilst " + "looking up 'launch_altitude'.") # Prediction profile req['profile'] = _extract_parameter(data, "profile", str, PROFILE_STANDARD) launch_alt = req["launch_altitude"] if req['profile'] == PROFILE_STANDARD: req['ascent_rate'] = _extract_parameter(data, "ascent_rate", float, validator=lambda x: x > 0) req['burst_altitude'] = \ _extract_parameter(data, "burst_altitude", float, validator=lambda x: x > launch_alt) req['descent_rate'] = _extract_parameter(data, "descent_rate", float, validator=lambda x: x > 0) elif req['profile'] == PROFILE_FLOAT: req['ascent_rate'] = _extract_parameter(data, "ascent_rate", float, validator=lambda x: x > 0) req['float_altitude'] = \ _extract_parameter(data, "float_altitude", float, validator=lambda x: x > launch_alt) req['stop_datetime'] = \ _extract_parameter(data, "stop_datetime", _rfc3339_to_timestamp, validator=lambda x: x > req['launch_datetime']) else: raise RequestException("Unknown profile '%s'." % req['profile']) # Dataset req['dataset'] = _extract_parameter(data, "dataset", _rfc3339_to_timestamp, LATEST_DATASET_KEYWORD) return req
def _extract_parameter(data, parameter, cast, default=None, ignore=False, validator=None): """ Extract a parameter from the POST request and raise an exception if any parameter is missing or invalid. """ if parameter not in data: if default is None and not ignore: raise RequestException("Parameter '%s' not provided in request." % parameter) return default try: result = cast(data[parameter]) except Exception: raise RequestException("Unable to parse parameter '%s': %s." % (parameter, data[parameter])) if validator is not None and not validator(result): raise RequestException("Invalid value for parameter '%s': %s." % (parameter, data[parameter])) return result # Response ####################################################################
[docs]def run_prediction(req): """ Run the prediction. """ # Response dict resp = { "request": req, "prediction": [], } # Find wind data location ds_dir = app.config.get('WIND_DATASET_DIR', WindDataset.DEFAULT_DIRECTORY) # Dataset try: if req['dataset'] == LATEST_DATASET_KEYWORD: tawhiri_ds = WindDataset.open_latest(persistent=True, directory=ds_dir) else: tawhiri_ds = WindDataset(datetime.fromtimestamp(req['dataset']), directory=ds_dir) except IOError: raise InvalidDatasetException("No matching dataset found.") except ValueError as e: raise InvalidDatasetException(*e.args) # Note that hours and minutes are set to 00 as Tawhiri uses hourly datasets resp['request']['dataset'] = tawhiri_ds.ds_time.strftime( "%Y-%m-%dT%H:00:00Z") # Stages if req['profile'] == PROFILE_STANDARD: stages = models.standard_profile(req['ascent_rate'], req['burst_altitude'], req['descent_rate'], tawhiri_ds, ruaumoko_ds()) elif req['profile'] == PROFILE_FLOAT: stages = models.float_profile(req['ascent_rate'], req['float_altitude'], req['stop_datetime'], tawhiri_ds) else: raise InternalException("No implementation for known profile.") # Run solver try: result = solver.solve(req['launch_datetime'], req['launch_latitude'], req['launch_longitude'], req['launch_altitude'], stages) except Exception as e: raise PredictionException("Prediction did not complete: '%s'." % str(e)) # Format trajectory if req['profile'] == PROFILE_STANDARD: resp['prediction'] = _parse_stages(["ascent", "descent"], result) elif req['profile'] == PROFILE_FLOAT: resp['prediction'] = _parse_stages(["ascent", "float"], result) else: raise InternalException("No implementation for known profile.") # Convert request UNIX timestamps to RFC3339 timestamps for key in resp['request']: if "datetime" in key: resp['request'][key] = _timestamp_to_rfc3339(resp['request'][key]) return resp
def _parse_stages(labels, data): """ Parse the predictor output for a set of stages. """ assert len(labels) == len(data) prediction = [] for index, leg in enumerate(data): stage = {} stage['stage'] = labels[index] stage['trajectory'] = [{ 'latitude': lat, 'longitude': lon, 'altitude': alt, 'datetime': _timestamp_to_rfc3339(dt), } for dt, lat, lon, alt in leg] prediction.append(stage) return prediction # Flask App ################################################################### @app.route('/api/v{0}/'.format(API_VERSION), methods=['GET'])
[docs]def main(): """ Single API endpoint which accepts GET requests. """ g.request_start_time = time.time() response = run_prediction(parse_request(request.args)) g.request_complete_time = time.time() response['metadata'] = _format_request_metadata() return jsonify(response)
@app.errorhandler(APIException)
[docs]def handle_exception(error): """ Return correct error message and HTTP status code for API exceptions. """ response = {} response['error'] = { "type": type(error).__name__, "description": str(error) } g.request_complete_time = time.time() response['metadata'] = _format_request_metadata() return jsonify(response), error.status_code
def _format_request_metadata(): """ Format the request metadata for inclusion in the response. """ return { "start_datetime": _timestamp_to_rfc3339(g.request_start_time), "complete_datetime": _timestamp_to_rfc3339(g.request_complete_time), }