#!/usr/bin/env python # -*- coding: utf-8 -*- purpose = "Take in values and spit out a JSON (JavaScript Object Notation) object...something that looks like a python dictionary." """ program: %s purpose: %s author: Joycelyn Thomson Jones date: 2019.02.04 //=== File Prolog ============================================================ // // This code was developed by NASA, Goddard Space Flight Center, Code 580 // for the Community Coordinated Modeling Center (CCMC) project. // //--- Notes ------------------------------------------------------------------ // Development history follows this notice. // //--- Warning ---------------------------------------------------------------- // This software is property of the National Aeronautics and Space // Administration. Unauthorized use or duplication of this software is // strictly prohibited. Authorized users are subject to the following // restrictions: // * Neither the author, their corporation, nor NASA is responsible for // any consequence of the use of this software. // * The origin of this software must not be misrepresented either by // explicit claim or by omission. // * Altered versions of this software must be plainly marked as such. // * This notice may not be removed or altered. // //=== End File Prolog ======================================================== modifications are listed at the end of the file. """ % (__file__, purpose) #### NOTES ################################### #### IMPORTS ################################## import argparse import collections import datetime import logging import json import os import string import sys #### CLASSES ################################## def InitLogger(log_dir, log_starter): """ Initialize logger for logging messages. """ logger = logging.getLogger(__name__) first_msg = '' if not os.path.exists(log_dir): try: os.mkdir(log_dir) except: first_msg = 'WARNING: can\'t make logs directory. Using current directory.' print(first_msg) log_dir= './' n = datetime.datetime.now() now_ts = '{}{:02d}{:02d}{:02d}{:02d}{:02d}'.format(n.year, n.month, n.day, n.hour, n.minute, n.second) log_file = os.path.join(log_dir, '{}.{}.log'.format(log_starter, now_ts)) hdlr = logging.FileHandler(log_file) formatter = logging.Formatter('%(asctime)s %(levelname)s %(message)s') hdlr.setFormatter(formatter) logger.addHandler(hdlr) logger.setLevel(logging.INFO) # replace INFO with DEBUG, WARNING, ERROR, or CRITICAL, as desired if first_msg != '': logger.warning(first_msg) return logger # end InitLogger def InitParser(desc): """ """ parser = argparse.ArgumentParser(description=desc, formatter_class=argparse.RawTextHelpFormatter) # Make output JSON filename an arg, otherwise give it a good default parser.add_argument("-o", "--output", dest="output_file_basename", help="Beginning of JSON output filename. Default is the ..") parser.add_argument("-d", "--output-dir", default='./', help="Full path to output directory. Default is current directory.") # Make logging an option parser.add_argument("-n", "--no-logging", dest="log_msgs", action='store_false', default=True, help="Turn off logging. It is turned on by default.") # Make log directory an option, otherwise make it the current directory parser.add_argument("-l", "--log-dir", default='./', help="Full path to log directory. Default is current directory.") # Make log filename an option, otherwise give it a good default parser.add_argument("-b", "--log-basename", dest="log_starter", default='isep_model_run', help="Beginning of the log filename (date and time will be added automatically). Default is \'isep_model_run\'.") parser.add_argument('--import-data-dictionary', action='store_true', default=False, help='import the data dictionary (in the \'sep_forecast_submission_dataDict\' variable from a file named \'named input_sep.py\'.') parser.add_argument('--contact-name', nargs='*', action='append', help='Model developer/administrator POCs, in case of datastream issues or model questions. (At least one is required, more than one is allowed)') parser.add_argument('--contact-email', action='append', help='Contact email (One per contact-name required).') parser.add_argument('--model-short-name', nargs='*', help='Short name (e.g. acronym) of model to appear on scoreboard. Consider including version number with acronym if distinction needed. 30 character limit.(Required)') parser.add_argument('--spase-id', help='Link to URL of full model description metadata in CCMC metadata registry in SPASE format (contact CCMC to register your model).(Required)') parser.add_argument('--issue-time', help='Forecast issue time (e.g. model run is complete and forecast is created)(Required)') parser.add_argument('--mode', default='forecast', help='Allowed values: forecast, historical. Default is \'forecast\'. (Optional)') # optional Triggers/cme (>1 allowed) parser.add_argument('--cme-start-time', action='append', help='Provide if forecast is issued based on a CME trigger. Timestamp of 1st coronagraph image CME is visible in. (Optional)') parser.add_argument('--cme-liftoff-time', action='append', help='Timestamp of coronagraph image with 1st indication of CME liftoff (used by CACTUS). (Optional)') parser.add_argument('--cme-lat', action='append', help='CME latitude (deg). (Optional)') parser.add_argument('--cme-lon', action='append', help='CME latitude (deg). (Optional)') parser.add_argument('--cme-pa', action='append', help='CME plane-of-sky position angle (measured from solar north in degrees counter-clockwise ). (Optional)') parser.add_argument('--cme-half-width', action='append', help='CME half-width (deg). (Optional)') parser.add_argument('--cme-speed', action='append', help='CME speed (km/s). (Optional)') parser.add_argument('--cme-acceleration', action='append', help='CME acceleration (km/s^2). (Optional)') parser.add_argument('--cme-height', action='append', help='CME height at which the above parameters were derived (solar radii from Sun center). (Optional)') parser.add_argument('--cme-time-at-height-time', action='append', help='CME time at specificied height. (Optional, required with cme_time_at_height_height)') parser.add_argument('--cme-time-at-height-height', action='append', help='Specificied height in solar radii. (Optional, required with cme_time_at_height_time)') parser.add_argument('--cme-coordinates', action='append', help='Coordinate system for CME lat/lon parameters (e.g. HEEQ or Carrington) (Optional, required with lat or lon)') parser.add_argument('--cme-catalog', action='append', help='Name of catalog where CME information was pulled from. Allowed values: ARTEMIS, DONKI, HELCATS, JHU APL, CACTUS_NRL, CACTUS_SIDC, CORIMP, SEEDS, SOHO_CDAW, STEREO_COR1 (contact us to add a new catalog name) (Optional)') parser.add_argument('--cme-urls', nargs='*', action='append', help='List of urls where CME information can be found, or information was pulled from. (Optional, more than one is allowed)') # optional Triggers/flare (>1 allowed) parser.add_argument('--flare-last-data-time', action='append', help='Last time data timestamp that was used to create forecast (relevant for forecasts issued before flare end times) (Optional)') parser.add_argument('--flare-start-time', action='append', help='Flare start time (Optional, required it flare-last-data-time used)') parser.add_argument('--flare-peak-time', action='append', help='Flare peak time (Optional)') parser.add_argument('--flare-end-time', action='append', help='Flare end time (Optional)') parser.add_argument('--flare-location', action='append', help='Flare location in Stonyhurst coordinates (i.e., N00W00/S00E00 format). (Optional)') parser.add_argument('--flare-intensity', action='append', help='Flare intensity (W/m^2) (Optional)') parser.add_argument('--flare-integrated-intensity', action='append', help='Flare integrated intensity (J/m^2) (Optional)') parser.add_argument('--flare-noaa-region', action='append', help='Associated NOAA active region number (including the preceding 1) (Optional)') parser.add_argument('--flare-urls', nargs='*', action='append', help='List of urls where flare information can be found, or information was pulled from. (Optional, more than one is allowed)') # optional Triggers/cme_simulation (>1 allowed) parser.add_argument('--cme-sim-model', action='append', help='Model name (Optional)') parser.add_argument('--cme-sim-completion-time', action='append', help='Simulation completion time (Optional, required if cme-sim-model is used)') parser.add_argument('--cme-sim-urls', nargs='*', action='append', help='List of urls where simulation information can be found, or information was pulled from. (Optional, more than one is allowed)') # optional Triggers/particle_intensity (>1 allowed) parser.add_argument('--pi-observatory', action='append', help='Name of observatory/spacecraft data are from. (Optional)') parser.add_argument('--pi-instrument', action='append', help='Name of instrument data are from. (Optional, required if pi-observatory used)') parser.add_argument('--pi-last-data-time', action='append', help='Last time data timestamp used to create forecast. (Optional, required if pi-observatory used)') #parser.add_argument('--pi-ongoing-events', nargs='*', action='append', help='If an ongoing event triggers your forecast, list the properties you used') # start_time, threshold, energy_min, energy_max (all are required, if any) parser.add_argument('--pi-ongoing-events-start-time', nargs='*', action='append', help='If an ongoing event triggers your forecast, this is the start time. (Optional)') parser.add_argument('--pi-ongoing-events-threshold', nargs='*', action='append', help='If an ongoing event triggers your forecast, this is the threshold used to define the event in pfu. (Optional, required if pi-ongoing-events-start-time used)') parser.add_argument('--pi-ongoing-events-energy-min', nargs='*', action='append', help='If an ongoing event triggers your forecast, this is the min of energy channel range in MeV. (Optional, required if pi-ongoing-events-start-time used)') parser.add_argument('--pi-ongoing-events-energy-max', nargs='*', action='append', help='If an ongoing event triggers your forecast, this is the max of energy channel range in MeV. -1 represents an unbounded integral channel. (Optional, required if pi-ongoing-events-start-time used)') # required Forecast # required Forecast/energy_channel parser.add_argument('--energy-min', action='append', help='Min of energy channel range. (Required)') parser.add_argument('--energy-max', action='append', help='Max of energy channel range. -1 represented an unbounded integral channel. (Required)') parser.add_argument('--energy-units', action='append', help='Energy channel units (Required)') parser.add_argument('--species', action='append', help='Allowed values: electron, proton, helium, helium3, helium4, oxygen, iron, ion. (Required)') parser.add_argument('--location', action='append', help='Allowed values: mercury, venus, earth, mars, psp, stereoa, stereob, dawn, juno, L1, L2, L4, L5. (Required)') parser.add_argument('--prediction-window', nargs=2, action='append', help='Start time and end time (in that order) of the prediction window that is relevant to the given data. Start of forecast prediction window must be within one hour of forecast issue time when in \'forecast\' mode. (Required)') # start_time, end_time (both are required) # optional Forecast/peak_intensity parser.add_argument('--peak-intensity', action='append', help='Forecast peak intensity value. (Optional)') parser.add_argument('--peak-intensity-units', action='append', help='Forecast peak intensity value units. (Optional, required if peak-intensity used)') parser.add_argument('--peak-intensity-uncertainty', action='append', help='Forecast peak intensity uncertainty value (same units as peak intensity). (Optional)') parser.add_argument('--peak-intensity-time', action='append', help='Forecast time for reaching peak intensity value. (Optional)') # optional Forecast/peak_intensity_esp parser.add_argument('--peak-intensity-esp', action='append', help='Forecast peak intensity value in the vicinity of shock passage. (Optional)') parser.add_argument('--peak-intensity-esp-units', action='append', help='Forecast peak intensity units in the vicinity of shock passage. (Optional, required if peak-intensity-esp used)') parser.add_argument('--peak-intensity-esp-time', action='append', help='Forecast time for reaching peak intensity value in the vicinity of shock passage. (Optional)') # optional Forecast/fluence parser.add_argument('--fluence', action='append', help='Forecast fluence value (corresponds to event length). (Optional)') parser.add_argument('--fluence-units', action='append', help='Forecast fluence units. (Optional, required if fluence used)') # optional Forecast/event_length parser.add_argument('--event-length-start-time', action='append', help='Event length must fall within prediction window. Forecast energetic particle event start time (\'onset\' time). (Optional)') parser.add_argument('--event-length-end-time', action='append', help='Forecast energetic particle event end time. (Optional)') parser.add_argument('--event-length-threshold', action='append', help='Threshold used to extract start and end times. (Optional, required if event-length-start-time used)') parser.add_argument('--event-length-threshold-units', action='append', help='Units of threshold. (Optional, required if event-length-start-time used)') # optional Forecast/threshold_crossings (>1 allowed) parser.add_argument('--thresh-crossing-times', nargs='*', action='append', help='Multiple threshold_corssings can be provided for the same forecast energy channel. Forecast threshold crossing time. (Optional, more than one is allowed)') parser.add_argument('--thresh-uncertainties', nargs='*', action='append', help='Forecast crossing time uncertainty in hours. (Optional, more than one is allowed)') parser.add_argument('--crossing-thresholds', nargs='*', action='append', help='Particle intensity threshold value crossing time refers to. (Optional, required if thresh-crossing-times used, more than one is allowed)') parser.add_argument('--crossing-threshold-units', nargs='*', action='append', help='Units of threshold. (Optional, required if thresh-crossing-times used, more than one is allowed)') # optional Forecast/probabilities (>1 allowed) parser.add_argument('--probabilities', nargs='*', action='append', help='Multiple probabilities can be provided for the same forecast energy channel. forecast probability value (range 0 to 1). (Optional, more than one is allowed)') parser.add_argument('--prob-uncertainties', nargs='*', action='append', help='Plus/minus error bar for probabilty_value (in probability_value units). (Optional, more than one is allowed)') parser.add_argument('--prob-thresholds', nargs='*', action='append', help='Particle intensity threshold value probability forecast refers to. (Optional, required if probabilities is used, more than one is allowed)') parser.add_argument('--prob-threshold-units', nargs='*', action='append', help='Units of threshold. (Optional, required if probabilities is used, more than one is allowed)') # optional Forecast/all_clear parser.add_argument('--all-clear', action='append', help="""There are three situations for setting all_clear_boolean=false: (1) for >10MeV energy channel, your forecast of peak intensity OR threshold crossing exceeds 10 pfu OR your probabilty forecast for a threshold of 10 pfu exceeds your custom probability_threshold; (2) for the >100MeV energy channel, your forecast of peak intensity OR threshold crossing exceeds 1 pfu OR your probabilty forecast for a threshold of 1 pfu exceeds your custom probability_threshold; (3) for your custom (non-integral) energy channel, your forecast peak intensity OR threshold crossing exceeds your custom threshold. Custom cases (3) are being stored but will not be used in the all-clear scoreboard display. (Optional)""") parser.add_argument('--all-clear-threshold', action='append', help='Particle intensity threshold value all_clear_boolean refers to. Can be 10 pfu for >10MeV channel, 1 pfu for >100MeV channel, or a custom threshold value. (Optional, required if all-clear is used)') parser.add_argument('--all-clear-threshold-units', action='append', help='Units of threshold. (Optional, required if all-clear is used)') parser.add_argument('--all-clear-probability-threshold', action='append', help='Probability threshold value all_clear_boolean refers to. Must specify this threshold if setting all_clear_boolean based on probability forecast. (Optional)') parser.add_argument('--sep-profile', action='append', help='Text file with 2 columns: datetime string and predicted SEP intensity for this energy channel. (Optional)') parser.add_argument('--native-id', action='append', help='Specify only if forecast has a native id from your model run. (Optional)') return parser # end InitParser def LogMsg(msg, d): """ """ # initialize log message (from parameters), if logging is selected logger = InitLogger(d['log_dir'], d['log_starter']) # log the message. It if is being called from here, it's an error message. logger.error(msg) return # end LogMsg def ThrowArgError(msg, d): """ """ print(msg) if d['log_msgs']: LogMsg(msg, d) sys.exit() # end ThrowArgError def CheckForRequiredArgs(d, req_argsL=None, err_msg=None): """Check the given (or default) list and see if we got all the required args. """ if req_argsL == None: # this means it is the top level required args. req_argsL = ['contact_name', 'contact_email', 'model_short_name', 'spase_id', 'issue_time', 'mode', 'energy_min', 'energy_max', 'energy_units', 'species', 'location', 'prediction_window'] if err_msg == None: err_msg = 'Missing required arg: \'{}\'' keyL = d.keys() for a in req_argsL: if a not in keyL: ThrowArgError(err_msg.format(a.replace('_', '-')), d) DontAllowNoneValues(d[a], a.replace('_', '-'), d) return # end CheckForRequiredArgs def DontAllowNoneValues(v, field_name, d): """ """ if isinstance(v, list): print('{} is a list? {}'.format(field_name, v)) if v == []: msg = 'A \'None\' value was given for {}, but it isn\'t allowed.'.format(field_name) ThrowArgError(msg, d) else: for v_ in v: DontAllowNoneValues(v_, field_name, d) else: print('{} is not a list {}'.format(field_name, v)) if v in [None, 'None', 'none', 0, '0']: msg = 'A \'None\' value was given for {}, but it isn\'t allowed.'.format(field_name) ThrowArgError(msg, d) return # end DontAllowNoneValues def ParseArguments(parser): """ Build the data dictionary based on the arguments the user included on the command line. """ #print('====Starting to parse arguments.===========================================') args = parser.parse_args() if args.import_data_dictionary: try: from input_sep import sep_forecast_submission_dataDict return (args.output_file_basename, args.output_dir, args.log_msgs, args.log_dir, args.log_starter, sep_forecast_submission_dataDict) except ImportError: print """ ################################################################################### #!!!! Error while trying to import the sep_forecast_submission_dataDict !!!! # # # # You MUST name the dictionary variable sep_forecast_submission_dataDict. # # You MUST save it to a file named input_sep.py. # # Make sure the user can read the input_sep.py file. # # input_sep.py MUST be in the same directory as this (sep_json_writer.py) script. # # # # Then run the ./sep_json_writer.py with the --import-data-dictionary option. # # you can use these options as well # # --output # # --output-dir # # --no-logging # # --log-basename # # --log-dir # ################################################################################### """ sys.exit() d = vars(args) argsL = ['contact_name', 'contact_email', 'model_short_name', 'spase_id', 'issue_time', 'mode', 'cme_start_time', 'cme_liftoff_time', 'cme_lat', 'cme_lon', 'cme_pa', 'cme_half_width', 'cme_speed', 'cme_acceleration', 'cme_height', 'cme_time_at_height_time', 'cme_time_at_height_height', 'cme_coordinates', 'cme_catalog', 'cme_urls', 'flare_last_data_time', 'flare_start_time', 'flare_peak_time', 'flare_end_time', 'flare_location', 'flare_intensity', 'flare_integrated_intensity', 'flare_noaa_region', 'flare_urls', 'cme_sim_model', 'cme_sim_completion_time', 'cme_sim_urls', 'pi_observatory', 'pi_instrument', 'pi_last_data_time', 'pi_ongoing_events_start_time', 'pi_ongoing_events_threshold', 'pi_ongoing_events_energy_min', 'pi_ongoing_events_energy_max', 'energy_min', 'energy_max', 'energy_units', 'species', 'location', 'prediction_window', 'peak_intensity', 'peak_intensity', 'peak_intensity_units', 'peak_intensity_uncertainty', 'peak_intensity_time', 'peak_intensity_esp', 'peak_intensity_esp_units', 'peak_intensity_esp_time', 'fluence', 'fluence_units', 'event_length_start_time', 'event_length_end_time', 'event_length_threshold', 'event_length_threshold_units', 'thresh_crossing_times', 'thresh_uncertainties', 'crossing_thresholds', 'crossing_threshold_units', 'probabilities', 'prob_uncertainties', 'prob_thresholds', 'prob_threshold_units', 'all_clear', 'all_clear_threshold', 'all_clear_threshold_units', 'all_clear_probability_threshold', 'sep_profile', 'native_id'] noneList = [None, 'None', 'none', ['none'], ['None'], [None]] # NOTE: 0 is not included because it is valid/needed in many fields dataDict = collections.OrderedDict() checkedL = [] CheckForRequiredArgs(d) # if this returns, we have the required arg values #print d for a in argsL: if d[a] not in noneList: # look at each value and determine how to handle it. #print('working on {}.'.format(d[a])) if a == 'contact_name': if len(d[a]) != len(d['contact_email']): ThrowArgError('mismatch on the number of values given for \'contact_name\' and \'contact_email\'', d) else: tL = [] # temp list for i in range(0, len(d[a])): v = ' '.join(d[a][i]) DontAllowNoneValues(v, 'contact-name', d) # check for 'None', if it is None, throw an error DontAllowNoneValues(d['contact_email'][i], 'contact-email', d) tL.append({'name':v, 'email':d['contact_email'][i]}) dataDict['contacts'] = tL elif a == 'model_short_name': DontAllowNoneValues(d['model_short_name'], 'model-short-name', d) DontAllowNoneValues(d['spase_id'], 'spase-id', d) msn = ' '.join(d[a]) # concatenating model short name values, if needed msn = msn.strip() dataDict['model'] = {'short_name':msn, 'spase_id':d['spase_id']} elif a in ['issue_time', 'mode']: DontAllowNoneValues(d[a], a.replace('_', '-'), d) dataDict[a] = d[a] elif a == 'cme_start_time' and a in d.keys(): # this is not a required arg # there can be more than one cme set. # each arg defined with nargs and append is a list of lists co = len('cme_') # chop_off (how much of the arg name to chop off to convert it to a key in dataDict got_argL = [] for aa in ('cme_start_time', 'cme_liftoff_time', 'cme_lat', 'cme_lon', 'cme_pa', 'cme_half_width', 'cme_speed', 'cme_acceleration', 'cme_height', 'cme_time_at_height_time', 'cme_time_at_height_height', 'cme_coordinates', 'cme_catalog', 'cme_urls', ): if aa in d.keys() and d[aa] not in noneList: got_argL.append(aa) # handle 'required' args, which there aren't many, except coordinates if lat or lon are given. if 'cme_lat' in got_argL and d['cme_lat'] not in noneList: CheckForRequiredArgs(d, ['cme_coordinates'], 'Missing \'cmd-coordinates\', which is required if you give a \'cme-lat\' value.') elif 'cme_lon' in got_argL and d['cme_lon'] not in noneList: CheckForRequiredArgs(d, ['cme_coordinates'], 'Missing \'cmd-coordinates\', which is required if you give a \'cme-lon\' value.') # build the dictionary to go into the triggers list. list_length = len(d['cme_start_time']) #print('initial length of cme_start_time arg is {}'.format(list_length)) for aa in got_argL: if d[aa] not in noneList and len(d[aa]) != list_length: ThrowArgError('mismatch on the number of values given for \'cme_start_time\' and \'{}\''.format(aa), d) for i in range(0, len(d['cme_start_time'])): tD = collections.OrderedDict() # temp dictionary for aa in got_argL: if aa in ['cme_time_at_height_time', 'cme_time_at_height_height']: # if you get cme_time_at_height_time, cme_time_at_height_height is required and vice versa. if aa == 'cme_time_at_height_time': # we only want to do this once if 'cme_time_at_height_height' not in got_argL: msg = 'missing \'cme_time_at_height_height\' arg. Required because you used \'cme_time_at_height_time\' arg.' ThrowArgError(msg, d) if d[aa][i] not in noneList and d['cme_time_at_height_height'][i] not in noneList: tD['time_at_height'] = {'time':d[aa][i], 'height':d['cme_time_at_height_height'][i]} elif aa == 'cme_time_at_height_height': # Both args should have been given. Thus, 'cme_time_at_height_height' should have already been processed when # 'cme_time_at_height_time' came up. So when you get here, all you need to do is verify if 'cme_time_at_height_time' # was included and throw an error if it is missing. if 'cme_time_at_height_time' not in got_argL: msg = 'missing \'cme_time_at_height_time\' arg. Required because you used \'cme_time_at_height_height\' arg.' ThrowArgError(msg, d) elif d[aa][i] not in noneList: tD[aa[co:]] = d[aa][i] if 'triggers' in dataDict.keys(): dataDict['triggers'].append({'cme':tD}) else: dataDict['triggers'] = [{'cme':tD}] elif a == 'flare_last_data_time' and a in d.keys(): # this is not a required arg # there can be more than one cme set. # each arg defined with nargs and append is a list of lists co = len('flare_') # chop_off (how much of the arg name to chop off to convert it to a key in dataDict got_argL = [] for aa in ('flare_last_data_time', 'flare_start_time', 'flare_peak_time', 'flare_end_time', 'flare_location', 'flare_intensity', 'flare_integrated_intensity', 'flare_noaa_region', 'flare_urls'): if aa in d.keys() and d[aa] not in noneList: got_argL.append(aa) # handle 'required' args, which are 'flare_last_data_time' and 'flare_start_time' for aa in ('flare_last_data_time', 'flare_start_time'): DontAllowNoneValues(d[aa], aa, d) if aa == 'flare_start_time': err_msg = 'Missing \'flare-start-time\', which is required if you give a \'flare-last-data-time\' value.' CheckForRequiredArgs(d, ['flare_start_time'], err_msg) list_length = len(d['flare_last_data_time']) #print('initial length of flare_last_data_time arg is {}'.format(list_length)) for aa in got_argL: if d[aa] not in noneList and len(d[aa]) != list_length: ThrowArgError('mismatch on the number of values given for \'flare_last_data_time\' and \'{}\''.format(aa), d) for i in range(0, len(d['flare_last_data_time'])): tD = collections.OrderedDict() # temp dictionary for aa in got_argL: if d[aa][i] not in noneList: tD[aa[co:]] = d[aa][i] if 'triggers' in dataDict.keys(): dataDict['triggers'].append({'flare':tD}) else: dataDict['triggers'] = [{'flare':tD}] elif a == 'cme_sim_model' and a in d.keys(): # this is not a required arg # there can be more than one cme set. # each arg defined with nargs and append is a list of lists co = len('cme_sim_') # chop_off (how much of the arg name to chop off to convert it to a key in dataDict got_argL = [] for aa in ('cme_sim_model', 'cme_sim_completion_time', 'cme_sim_urls'): if aa in d.keys() and d[aa] not in noneList: got_argL.append(aa) list_length = len(d['cme_sim_model']) #print('initial length of cme_sim_model arg is {}'.format(list_length)) for aa in got_argL: if d[aa] not in noneList and len(d[aa]) != list_length: ThrowArgError('mismatch on the number of values given for \'cme_sim_model\' and \'{}\''.format(aa), d) for i in range(0, len(d['cme_sim_model'])): tD = collections.OrderedDict() # temp dictionary for aa in got_argL: if aa == 'cme_sim_completion_time': if d[aa][i] not in noneList: tD['simulation_completion_time'] = d[aa][i] elif d[aa][i] not in noneList: tD[aa[co:]] = d[aa][i] if 'triggers' in dataDict.keys(): dataDict['triggers'].append({'cme_simulation':tD}) else: dataDict['triggers'] = [{'cme_simulation':tD}] elif a == 'pi_observatory' and a in d.keys() and d[a] not in noneList: # this is not a required arg # there can be more than one particle_intensity set. # each arg defined with nargs and append is a list of lists got_argL = ['pi_observatory'] possible_fieldsL = ['pi_instrument', 'pi_last_data_time', 'pi_ongoing_events_start_time', 'pi_ongoing_events_threshold', 'pi_ongoing_events_energy_min', 'pi_ongoing_events_energy_max'] for aa in possible_fieldsL: if aa in d.keys() and d[aa] not in noneList: got_argL.append(aa) # handle 'required' (if you got here) args, which are 'pi_instrument', 'pi_last_data_time' req_argsL = ['pi_observatory', 'pi_instrument', 'pi_last_data_time'] err_msg = 'Missing \'{}\', which is required if you give a \'pi-observatory\' value.' CheckForRequiredArgs(d, req_argsL, err_msg) list_length = len(d['pi_observatory']) for aa in got_argL: if d[aa] not in noneList and len(d[aa]) != list_length: ThrowArgError('mismatch on the number of values given for \'pi_observatory\' and \'{}\''.format(aa), d) for i in range(0, len(d['pi_observatory'])): co = len('pi_') # chop_off (how much of the arg name to chop off to convert it to a key in dataDict tD = collections.OrderedDict() # temp dictionary for aa in got_argL: if aa == 'pi_ongoing_events_start_time' and d[aa][i] not in noneList: # more than one is allowed # make sure that you have all four values because all are required, if any are given inner_req_argsL = ['pi_ongoing_events_threshold', 'pi_ongoing_events_energy_min', 'pi_ongoing_events_energy_max'] for ra in inner_req_argsL: if ra not in got_argL: msg = 'Got pi-ongoing-events-start-time, but did not get required {}.'.format(ra) ThrowArgError(msg, d) else: DontAllowNoneValues(d[ra][i], ra, d) # checks for an empty list if var is of type list, o/w None or 'None' co2 = len('pi_ongoing_events_') t2L = [] for j in range(0, len(d['pi_ongoing_events_start_time'][i])): t2D = collections.OrderedDict() for f in ['pi_ongoing_events_start_time'] + inner_req_argsL: if f in d.keys() and d[f][i][j] not in noneList: t2D[f[co2:]] = d[f][i][j] else: # throw error msg = 'Got pi-ongoing-events-start-time, but did not get required {}.'.format(f) ThrowArgError(msg, d) t2L.append(t2D) tD['ongoing_events'] = t2L elif aa in req_argsL: if d[aa][i] not in noneList: tD[aa[co:]] = d[aa][i] # this makes it so that we don't get duplicate oe values if 'triggers' in dataDict.keys(): dataDict['triggers'].append({'particle_intensity':tD}) else: dataDict['triggers'] = [{'particle_intensity':tD}] elif a == 'energy_min': # it's required # Here's the start of the forecast forecastLOD = [] # temp list to hold all the forecasts #print('Got to forecasts/energy_min.') threshold_crossingsT = ('thresh_crossing_times', 'thresh_uncertainties', 'crossing_thresholds', 'crossing_threshold_units', ) probabilitiesT = ('probabilities', 'prob_uncertainties', 'prob_thresholds', 'prob_threshold_units', ) fT = ('energy_min', 'energy_max', 'energy_units', 'species', 'location', 'prediction_window', 'peak_intensity', 'peak_intensity', 'peak_intensity_units', 'peak_intensity_uncertainty', 'peak_intensity_time', 'peak_intensity_esp', 'peak_intensity_esp_units', 'peak_intensity_esp_time', 'fluence', 'fluence_units', 'event_length_start_time', 'event_length_end_time', 'event_length_threshold', 'event_length_threshold_units', threshold_crossingsT, probabilitiesT, 'all_clear', 'all_clear_threshold', 'all_clear_threshold_units', 'all_clear_probability_threshold', 'sep_profile', 'native_id') req_argsT = ('energy_min', 'energy_max', 'energy_units', 'species', 'location', 'prediction_window') CheckForRequiredArgs(d, req_argsT) got_argL = [] for aa in fT: if aa in [threshold_crossingsT, probabilitiesT]: # you've got the threshold crossing tuple or probabilties tuple for aaa in aa: if aaa in d.keys() and d[aaa] not in noneList: got_argL.append(aaa) else: if aa in d.keys() and d[aa] not in noneList: got_argL.append(aa) #print('Got these args: {}'.format(got_argL)) list_length = len(d['energy_min']) #print('initial length of energy_min arg is {}'.format(list_length)) for aa in got_argL: if d[aa] not in noneList and len(d[aa]) != list_length: ThrowArgError('mismatch on the number of values given for \'energy_min\' and \'{}\''.format(aa), d) for i in range(0, len(d['energy_min'])): tD = collections.OrderedDict() # temp dictionary for aa in got_argL: if aa == 'energy_min': # required, as are energy_max and energy_units t2D = collections.OrderedDict() for k in ['energy_min', 'energy_max', 'energy_units']: if d[k][i] in noneList: # throw error msg = 'Got invalid {} value of {} and it is a required value.'.format(k, d[k][i]) ThrowArgError(msg, d) t2D['min'] = d['energy_min'][i] t2D['max'] = d['energy_max'][i] t2D['units'] = d['energy_units'][i] tD['energy_channel'] = t2D elif aa in ['species', 'location']: # required tD[aa] = d[aa][i] elif aa == 'prediction_window': # required try: tD['prediction_window'] = {'start_time':d['prediction_window'][i][0], 'end_time': d['prediction_window'][i][1]} except: msg = 'Didn\'t get enough values in prediction_window. There should be two values: start_time and then end_time.' ThrowArgError(msg, d) elif aa == 'peak_intensity' and d['peak_intensity'][i] not in noneList: # optional if 'peak_intensity_units' not in got_argL: msg = 'Got peak_intensity, but did not get required peak_intensity_units.' ThrowArgError(msg, d) else: DontAllowNoneValues(d['peak_intensity_units'][i], 'peak_intensity_units', d) t2D = collections.OrderedDict() # temp dictionary t2D = {'intensity':d['peak_intensity'][i], 'units': d['peak_intensity_units'][i]} for aaa in ['peak_intensity_uncertainty', 'peak_intensity_time']: # optional co = len('peak_intensity_') if aaa in got_argL and d[aaa][i] not in noneList: if d[aaa][i] not in noneList: t2D[aaa[co:]] = d[aaa][i] tD['peak_intensity'] = t2D elif aa == 'peak_intensity_esp' and d['peak_intensity_esp'][i] not in noneList: # optional if 'peak_intensity_esp_units' not in got_argL: msg = 'Got peak_intensity_esp, but did not get required peak_intensity_esp_units.' ThrowArgError(msg, d) else: DontAllowNoneValues(d['peak_intensity_esp_units'][i], 'peak_intensity_esp_units', d) t2D = collections.OrderedDict() # temp dictionary t2D['intensity'] = d['peak_intensity_esp'][i] t2D['units'] = d['peak_intensity_esp_units'][i] if 'peak_intensity_esp_time' in got_argL and d['peak_intensity_esp_time'][i] not in noneList: # optional co = len('peak_intensity_esp_') if aaa in got_argL: if d[aaa][i] not in noneList: t2D[aaa[co:]] = d[aaa][i] tD['peak_intensity_esp'] = t2D elif aa == 'fluence' and d['fluence'][i] not in noneList: # optional if 'fluence_units' not in got_argL: # required if 'fluence' is given msg = 'Got fluence, but did not get required fluence_units.' ThrowArgError(msg, d) else: DontAllowNoneValues(d['fluence_units'][i], 'fluence_units', d) for k in ['fluence', 'fluence_units']: if d[k][i] in noneList: # throw error msg = 'Got invalid {} value of {} and it is a required value.'.format(k, d[k][i]) ThrowArgError(msg, d) t2D = collections.OrderedDict() # temp dictionary t2D['fluence_value'] = d['fluence'][i] t2D['units'] = d['fluence_units'][i] tD['fluence'] = t2D elif aa == 'event_length_start_time' and d['event_length_start_time'][i] not in noneList: # optional # checking for 'required' (if event_length given) args rL = ['event_length_threshold', 'event_length_threshold_units'] for ra in rL: if ra not in got_argL: msg = 'Got event_length_start_time, but did not get required {}.'.format(ra) ThrowArgError(msg, d) else: DontAllowNoneValues(d[ra][i], ra, d) co = len('event_length_') for k in ['event_length_start_time', 'event_length_threshold', 'event_length_threshold_units']: if d[k][i] in noneList: # throw error msg = 'Got invalid {} value of {} and it is a required value.'.format(k, d[k][i]) ThrowArgError(msg, d) t2D = collections.OrderedDict() # temp dictionary t2D['start_time'] = d['event_length_start_time'][i] if 'event_length_end_time' in got_argL and d['event_length_end_time'][i] not in noneList: # optional t2D['end_time'] = d['event_length_end_time'][i] t2D['threshold'] = d['event_length_threshold'][i] t2D['threshold_units'] = d['event_length_threshold_units'][i] tD['event_length'] = t2D elif aa == 'thresh_crossing_times' and d['thresh_crossing_times'][i] not in noneList: # optional # going to get a list of lists for these args because they are nargs=* and append req_argsL = ['crossing_thresholds', 'crossing_threshold_units'] for ra in req_argsL: if ra not in got_argL: msg = 'Got thresh_crossing_times, but did not get required {}.'.format(ra) ThrowArgError(msg, d) else: DontAllowNoneValues(d[ra][i], ra, d) translate_arg2keyD = { 'thresh_crossing_times':'crossing_time', 'thresh_uncertainties':'uncertainty', 'crossing_thresholds':'threshold', 'crossing_threshold_units':'threshold_units', } t2L = [] for j in range(0, len(d['thresh_crossing_times'][i])): t2D = collections.OrderedDict() updated_dict = False for f in ['thresh_crossing_times', 'thresh_uncertainties', 'crossing_thresholds', 'crossing_threshold_units']: if f in got_argL and d[f] != []: try: #print('[L525] in {} d[f] is {}'.format(f, d[f])) #print('[L526] in {} d[f][i] is {}'.format(f, d[f][i])) #print('[L527] in {} d[f][i][j] is {}'.format(f, d[f][i][j])) if d[f][i][j] not in noneList: t2D[translate_arg2keyD[f]] = d[f][i][j] updated_dict = True except: print('[L512] some error with {}. Exiting.'.format(f)) sys.exit() if updated_dict: t2L.append(t2D) if t2L != []: tD['threshold_crossings'] = t2L elif aa == 'probabilities' and d['probabilities'][i] not in noneList: # going to get a list of lists for these because they are nargs=* and append req_argsL = ['prob_thresholds', 'prob_threshold_units'] for ra in req_argsL: #print('{}: {} testing for NULL synonyms.'.format(ra, d[ra][i])) if ra not in got_argL: msg = 'Got probabilities, but did not get required {}.'.format(ra) ThrowArgError(msg, d) else: DontAllowNoneValues(d[ra][i], ra, d) # checks for an empty list if var is of type list, o/w None or 'None' translate_arg2keyD = { 'probabilities': 'probability_value', 'prob_uncertainties': 'uncertainty', 'prob_thresholds': 'threshold', 'prob_threshold_units': 'threshold_units', } t2L = [] for j in range(0, len(d['probabilities'][i])): t2D = collections.OrderedDict() for f in ['probabilities', 'prob_uncertainties', 'prob_thresholds', 'prob_threshold_units']: if f == 'prob_uncertainties': if f in d.keys() and d[f][i][j] not in noneList: # optional t2D[translate_arg2keyD[f]] = d[f][i][j] else: # required (if probabiltities given) if f in d.keys() and d[f][i][j] not in noneList: t2D[translate_arg2keyD[f]] = d[f][i][j] else: # throw error msg = 'Got probabilties, but did not get required {}.'.format(f) ThrowArgError(msg, d) t2L.append(t2D) tD['probabilities'] = t2L elif aa == 'all_clear' and d['all_clear'][i] not in noneList: # checking for 'required' (if event_length given) args rL = ['all_clear_threshold', 'all_clear_threshold_units'] for ra in rL: print('is {} in got_argL? {}'.format(ra, ra in got_argL)) if ra not in got_argL: msg = 'Got all_clear, but did not get required {}.'.format(ra) ThrowArgError(msg, d) co = len('all_clear_') t2D = collections.OrderedDict() # temp dictionary for j in ['all_clear', 'all_clear_threshold', 'all_clear_threshold_units', 'all_clear_probability_threshold']: if j in got_argL: if j == 'all_clear': k = 'all_clear_boolean' v = d[j][i] if v in ['True', 'true', True]: v = True elif v in ['False', 'false', False]: v = False else: msg = 'Invalid \'all_clear\' value given. It need to be \'True\' or \'False\'.' ThrowArgError(msg, d) t2D[k] = v else: k = j[co:] if d[j][i] not in noneList: t2D[k] = d[j][i] tD['all_clear'] = t2D elif aa in ['sep_profile', 'native_id']: if aa in d.keys() and d[aa][i] not in noneList: # optional tD[aa] = d[aa][i] #print('forcast tD going into LOD: {}'.format(tD)) forecastLOD.append(tD) dataDict['forecasts'] = forecastLOD #print('====Finished parsing arguments.===========================================') #print(dataDict) #print('====Finished parsing arguments.===========================================') return (args.output_file_basename, args.output_dir, args.log_msgs, args.log_dir, args.log_starter, dataDict) # end ParseArguments class ConvertToJSON: def __init__(self, dataD, output_file_basename, output_dir, log_msgs, log_dir, log_starter): self.dataDict = dataD self.forecast_or_historical_mode = 'forecast' # default data mode self.orderedDict = collections.OrderedDict() self.noneList = [None, 'None', '0', 0] self.now = n = datetime.datetime.now() # datetime obj self.now_ts = '{}{:02d}{:02d}{:02d}{:02d}{:02d}'.format(n.year, n.month, n.day, n.hour, n.minute, n.second) # string created from datetime obj if output_file_basename in self.noneList: # prep model short_name to be used in default output_file_basename value msn = self.dataDict['model']['short_name'] for (a, b) in [(' ', '_'), ('-', '_')]: msn = msn.replace(a, b) # prep issue_time to be used in default output_file_basename value t = self.dataDict['issue_time'] it = '{}{}{}{}{}'.format(t[:4], t[5:7], t[8:10], t[11:13], t[14:16]) self.output_file_basename = '{}.{}'.format(msn, it) # the current time and .json extension are added later else: self.output_file_basename = output_file_basename self.output_dir = output_dir self.log_msgs = log_msgs self.log_dir = log_dir self.log_starter = log_starter self.WriteJSON() def ConvertDTString2DTO(self, dts): """ Convert incoming string in 'YYYY-MM-DDTHH:MMZ' format to a DateTime Object. This assumes that the DateTimeString validation step has already been run. """ y = int(dts[0:4]) m = int(dts[5:7]) d = int(dts[8:10]) h = int(dts[11:13]) n = int(dts[14:16]) # n --> thinking m for minute, but it's already used so bumped it to 'n' return datetime.datetime(y, m, d, h, n) # end ConvertDTString2DTO def IJWError(self, m, log=False, exit=False): """ """ print "{0}: {1}".format(os.path.basename(__file__), m) if log and exit: logging.critical("{0}: {1}".format(os.path.basename(__file__), m)) logging.critical("{0}: Exiting.".format(os.path.basename(__file__))) sys.exit() elif log: logging.error("{0}: {1}".format(os.path.basename(__file__), m)) elif exit: logging.critical("{0}: Exiting".format(os.path.basename(__file__))) sys.exit() return def IJWWarning(self, m, log=False): # in the style of argparse.error output print "{0}: {1}".format(os.path.basename(__file__), m) if log: logging.warning("{0}: {1}".format(os.path.basename(__file__), m)) return def __del__(self): pass #self.cx.close() def InitLogger(self): """ Initialize logger for logging messages. """ logger = logging.getLogger(__name__) # need to add a temp aspect to this... n = datetime.datetime.now() if os.path.exists(self.log_dir): pass else: try: os.mkdir(self.log_dir) except: print('ERROR: can\'t make logs directory. Using current directory.') self.log_dir= './' log_file = os.path.join(self.log_dir, '{}.{}.log'.format(self.log_starter, self.now_ts)) hdlr = logging.FileHandler(log_file) formatter = logging.Formatter('%(asctime)s %(levelname)s %(message)s') hdlr.setFormatter(formatter) logger.addHandler(hdlr) logger.setLevel(logging.INFO) # replace INFO with DEBUG, WARNING, ERROR, or CRITICAL, as desired return logger def PrepContacts(self): """ Prep the Contacts data for the ordered dictionary, from which the JSON is written. """ stub_found = False if self.VerifyKeyInDict('contacts', required=True): pfn = 'contacts' # parent field name newL = [] valueL = self.dataDict['contacts'] # if valueL is not a list, throw an error if not isinstance(valueL, list): m = 'ERROR: \'{}\' is supposed to be a list, but it isn\'t. Exiting.'.format(pfn) self.IJWError(m, self.log_msgs, True) # (msg, log, exit) for d in valueL: L = [ ('name', 'Model Developer Name', True, None, None, None, False), ('email', 'developer@email.com', True, None, None, None, False), ] for (k, s, r, allow_neg, allow_neg_one, must_be_in_past, allow_stub_value) in L: if self.VerifyKeyInDict(k, d, required=r): value = d[k] if not allow_stub_value: stub_found = self.VerifyNonStubValue(value, s, '{}/{}'.format(pfn, k), required=r) # check for non-stub value if allow_stub_value or not stub_found: if k == 'name': self.ValidateAlphaNumeric(value, '{}/{}'.format(pfn, k), allow_addtl_chars=',. ') elif k == 'email': self.ValidateEmail(value) if d['name'] not in self.noneList and d['email'] not in self.noneList: newL.append({'name':d['name'], 'email':d['email']}) self.orderedDict['contacts'] = newL return # end of PrepContacts def PrepForecastAllClear(self, d): """ Prep the Forecast's all_clear data for the ordered dictionary, from which the JSON is written. """ stub_found = False pfn = 'forecasts/all_clear' # parent field name toD = collections.OrderedDict() L = [ ('all_clear_boolean', False, True, None, None, None, True), ('threshold', 0, True, False, False, None, False), ('threshold_units', 'pfu', True, None, None, None, True), ('probability_threshold', 0, False, False, False, None, True), ] for (k, s, r, allow_neg, allow_neg_one, must_be_in_past, allow_stub_value) in L: if self.VerifyKeyInDict(k, d, required=r): value = d[k] if not allow_stub_value: stub_found = self.VerifyNonStubValue(value, s, '{}/{}'.format(pfn, k), required=r) # check for non-stub value if allow_stub_value or not stub_found: if k == 'all_clear_boolean': self.ValidateBoolean(value, '{}/{}'.format(pfn, k)) elif s == 0: # 'threshold' Can be 10 pfu, 1 pfu, or custom self.ValidateFloat(value, '{}/{}'.format(pfn, k), neg_allowed=allow_neg, neg_one_allowed=allow_neg_one) elif k == 'threshold_units': self.ValidateAlphaNumeric(value, '{}/{}'.format(pfn, k)) toD[k] = value return toD # end PrepForecastAllClear def PrepForecastEnergyChannel(self, d): """Prep the Forecast's energy_channel data for the ordered dictionary, from which the JSON is written. """ stub_found = False pfn = 'forecasts/energy_channel' # parent field name toD = collections.OrderedDict() L = [ ('min', 0, True, False, False, None, False), # is allow_stub_value correct? ('max', 0, True, False, True, None, False), # is allow_stub_value correct? ('units', 'pfu', True, None, None, None, True), ] for (k, s, r, allow_neg, allow_neg_one, must_be_in_past, allow_stub_value) in L: if self.VerifyKeyInDict(k, d, required=r): value = d[k] if not allow_stub_value: stub_found = self.VerifyNonStubValue(value, s, '{}/{}'.format(pfn, k), required=r) # check for non-stub value if allow_stub_value or not stub_found: if k in ['min', 'max']: self.ValidateFloat(value, '{}/{}'.format(pfn, k), neg_allowed=allow_neg, neg_one_allowed=allow_neg_one) elif k == 'units': self.ValidateAlphaNumeric(value, '{}/{}'.format(pfn, k)) toD[k] = value return toD # end PrepForecastEnergyChannel def PrepForecastEventLength(self, d, prediction_window_d): """ """ stub_found = False pfn = 'forecasts/event_length' # parent field name toD = collections.OrderedDict() L = [ ('start_time', 'YYYY-MM-DDTHH:MMZ', True, None, None, False, False), ('end_time', 'YYYY-MM-DDTHH:MMZ', False, None, None, False, False), ('threshold', 0, True, False, False, None, False), ('threshold_units', 'pfu', True, None, None, None, True), ] for (k, s, r, allow_neg, allow_neg_one, must_be_in_past, allow_stub_value) in L: if self.VerifyKeyInDict(k, d, required=r): value = d[k] if not allow_stub_value: stub_found = self.VerifyNonStubValue(value, s, '{}/{}'.format(pfn, k), required=r) # check for non-stub value if allow_stub_value or not stub_found: if s == 'YYYY-MM-DDTHH:MMZ': self.ValidateDateTimeStamp(value, '{}/{}'.format(pfn, k), ensure_in_past=must_be_in_past) elif k == 'threshold': self.ValidateFloat(value, '{}/{}'.format(pfn, k), neg_allowed=allow_neg, neg_one_allowed=allow_neg_one) elif k == 'threshold_units': self.ValidateAlphaNumeric(value, '{}/{}'.format(pfn, k)) toD[k] = value s = d['start_time'] e = None if 'end_time' in d.keys(): e = d['end_time'] self.ValidateForecastEventLength(s, e, prediction_window_d) return toD # end PrepForecastEventLength def PrepForecastFluence(self, d): """ """ stub_found = False pfn = 'forecasts/fluence' # parent field name toD = collections.OrderedDict() L = [ ('fluence_value', '0', True, False, False, None, True), ('units', 'pfu', True, None, None, None, True), ] for (k, s, r, allow_neg, allow_neg_one, must_be_in_past, allow_stub_value) in L: if self.VerifyKeyInDict(k, d, required=r): value = d[k] if not allow_stub_value: stub_found = self.VerifyNonStubValue(value, s, '{}/{}'.format(pfn, k), required=r) # check for non-stub value if allow_stub_value or not stub_found: if k == 'units': self.ValidateAlphaNumeric(value, '{}/{}'.format(pfn, k), allow_addtl_chars=' -_+*^()/') # should I allow other characters? else: self.ValidateFloat(value, '{}/{}'.format(pfn, k), neg_allowed=allow_neg, neg_one_allowed=allow_neg_one) toD[k] = value return toD # end PrepForecastFluence def PrepForecastPeakIntensity(self, d): """ """ stub_found = False pfn = 'forecasts/peak_intensity' # parent field name toD = collections.OrderedDict() L = [ ('intensity', '0', True, False, False, None, False), ('units', 'pfu', True, None, None, None, True), ('uncertainty', 0, False, False, False, None, True), ('time', 'YYYY-MM-DDTHH:MMZ', False, None, None, False, False), ] for (k, s, r, allow_neg, allow_neg_one, must_be_in_past, allow_stub_value) in L: if self.VerifyKeyInDict(k, d, required=r): value = d[k] if not allow_stub_value: stub_found = self.VerifyNonStubValue(value, s, '{}/{}'.format(pfn, k), required=r) # check for non-stub value if allow_stub_value or not stub_found: if k == 'intensity': self.ValidateFloat(value, '{}/{}'.format(pfn, k), neg_allowed=allow_neg, neg_one_allowed=allow_neg_one) elif k == 'units': self.ValidateAlphaNumeric(value, '{}/{}'.format(pfn, k), allow_addtl_chars=' -_+*^()/') # should I allow other characters? elif k == 'time': self.ValidateDateTimeStamp(value, '{}/{}'.format(pfn, k), ensure_in_past=must_be_in_past) toD[k] = value return toD # end PrepForecastPeakIntensity def PrepForecastPeakIntensityESP(self, d): """ESP = Electron? Storm Particle """ stub_found = False pfn = 'forecasts/peak_intensity_esp' # parent field name toD = collections.OrderedDict() L = [ ('intensity', '0', True, False, False, None, False), ('units', 'pfu', True, None, None, None, True), ('time', 'YYYY-MM-DDTHH:MMZ', False, None, None, False, False), ] for (k, s, r, allow_neg, allow_neg_one, must_be_in_past, allow_stub_value) in L: if self.VerifyKeyInDict(k, d, required=r): value = d[k] if not allow_stub_value: stub_found = self.VerifyNonStubValue(value, s, '{}/{}'.format(pfn, k), required=r) # check for non-stub value if allow_stub_value or not stub_found: if k == 'intensity': self.ValidateFloat(value, '{}/{}'.format(pfn, k), neg_allowed=allow_neg, neg_one_allowed=allow_neg_one) elif k == 'units': self.ValidateAlphaNumeric(value, '{}/{}'.format(pfn, k), allow_addtl_chars=' -_+*^()/') # should I allow other characters? elif k == 'time': self.ValidateDateTimeStamp(value, '{}/{}'.format(pfn, k), ensure_in_past=must_be_in_past) toD[k] = value return toD # end PrepForecastPeakIntensityESP def PrepForecastPredictionWindow(self, d): """ start of forecast prediction window (must be within one hour of forecast issue time for "forecast" mode) don't worry about it being in the past if in 'historical' mode """ stub_found = False pfn = 'forecasts/prediction_window' # parent field name toD = collections.OrderedDict() L = [ ('start_time', 'YYYY-MM-DDTHH:MMZ', True, None, None, False, False), ('end_time', 'YYYY-MM-DDTHH:MMZ', True, None, None, False, False), ] for (k, s, r, allow_neg, allow_neg_one, must_be_in_past, allow_stub_value) in L: if self.VerifyKeyInDict(k, d, required=r): value = d[k] if not allow_stub_value: stub_found = self.VerifyNonStubValue(value, s, '{}/{}'.format(pfn, k), required=r) # check for non-stub value if allow_stub_value or not stub_found: self.ValidateDateTimeStamp(value, '{}/{}'.format(pfn, k), ensure_in_past=must_be_in_past) if k == 'start_time': pass # TODO start of forecast prediction window (must be within one hour of forecast issue time for "forecast" mode) toD[k] = value return toD # end PrepForecastPredictionWindow def PrepForecastProbabilities(self, lod): """ """ stub_found = False pfn = 'forecasts/probabilities' # parent field name L = [ ('probability_value', 0, True, False, False, None, False), ('uncertainty', 0, False, False, False, None, True), ('threshold', 0, True, False, False, None, False), ('threshold_units', 'pfu', True, None, None, None, True), ] newL = [] for d in lod: toD = collections.OrderedDict() for (k, s, r, allow_neg, allow_neg_one, must_be_in_past, allow_stub_value) in L: if self.VerifyKeyInDict(k, d, required=r): value = d[k] if not allow_stub_value: stub_found = self.VerifyNonStubValue(value, s, '{}/{}'.format(pfn, k), required=r) # check for non-stub value if allow_stub_value or not stub_found: if s == 0: self.ValidateFloat(value, '{}/{}'.format(pfn, k), neg_allowed=allow_neg, neg_one_allowed=allow_neg_one) if k == 'probability_value': self.ValidateForecastProbabilityValue(value) elif k == 'units': self.ValidateAlphaNumeric(value, '{}/{}'.format(pfn, k), allow_addtl_chars=' -_+*^()/') # should I allow other characters? toD[k] = value newL.append(toD) return newL # end PrepForecastProbabilities def PrepForecastThresholdCrossings(self, lod): """What is """ stub_found = False pfn = 'forecasts/threshold_crossings' # parent field name L = [ ('crossing_time', 'YYYY-MM-DDTHH:MMZ', True, None, None, False, False), ('uncertainty', 0, False, False, False, None, True), ('threshold', 0, True, False, False, None, False), ('threshold_units', 'pfu', True, None, None, None, True), ] newL = [] for d in lod: toD = collections.OrderedDict() for (k, s, r, allow_neg, allow_neg_one, must_be_in_past, allow_stub_value) in L: if self.VerifyKeyInDict(k, d, required=r): value = d[k] if not allow_stub_value: stub_found = self.VerifyNonStubValue(value, s, '{}/{}'.format(pfn, k), required=r) # check for non-stub value if allow_stub_value or not stub_found: if s == 'YYYY-MM-DDTHH:MMZ': self.ValidateDateTimeStamp(value, '{}/{}'.format(pfn, k), ensure_in_past=must_be_in_past) elif s == 0: self.ValidateFloat(value, '{}/{}'.format(pfn, k), neg_allowed=allow_neg, neg_one_allowed=allow_neg_one) elif k == 'units': self.ValidateAlphaNumeric(value, '{}/{}'.format(pfn, k), allow_addtl_chars=' -_+*^()/') # should I allow other characters? toD[k] = value newL.append(toD) return newL # end PrepForecastThresholdCrossings def PrepForecast(self, d): """ """ stub_found = False pfn = 'forecasts' # parent field name toD = collections.OrderedDict() # temp ordered dictionary leafD = { 'species': (self.noneList, True, self.ValidateForecastSpecies), 'location': (self.noneList, True, self.ValidateForecastLocation), 'sep_profile': ("filename_energychannel.txt", False, self.ValidateForecastSEPProfile), 'native_id': (self.noneList, False, self.ValidateForecastNativeID), } L = [ ('energy_channel', True, self.PrepForecastEnergyChannel), ('species', True, None), ('location', True, None), ('prediction_window', True, self.PrepForecastPredictionWindow), ('peak_intensity', False, self.PrepForecastPeakIntensity), ('peak_intensity_esp', False, self.PrepForecastPeakIntensityESP), ('fluence', False, self.PrepForecastFluence), ('event_length', False, self.PrepForecastEventLength), ('threshold_crossings', False, self.PrepForecastThresholdCrossings), ('probabilities', False, self.PrepForecastProbabilities), ('all_clear', False, self.PrepForecastAllClear), ('sep_profile', False, None), ('native_id', False, None), ] for (k, r, meth) in L: if self.VerifyKeyInDict(k, d, required=r): if meth is None: if k in leafD.keys(): value = d[k] (s, r, val_meth) = leafD[k] stub_found = self.VerifyNonStubValue(value, s, '{}/{}'.format(pfn, k), required=r) if not stub_found: val_meth(value) toD[k] = value else: #print('about to call {} with arg *{}*.'.format(str(meth), d[k])) if k == 'event_length': toD[k] = meth(d[k], d['prediction_window']) else: toD[k] = meth(d[k]) return toD # end PrepForecast (NOT PrepForecasts !!!) def PrepForecasts(self): """ """ pfn = 'forecasts' if self.VerifyKeyInDict(pfn, required=True): L = self.dataDict[pfn] # if L is not a list, throw an error if not isinstance(L, list): m = 'ERROR: \'{}\' is supposed to be a list, but it isn\'t. Exiting.'.format(pfn) self.IJWError(m, self.log_msgs, True) # (msg, log, exit) newL = [] for d in L: toD = self.PrepForecast(d) newL.append(toD) if newL != []: self.orderedDict[pfn] = newL else: m = 'ERROR: \'{}\' did not get any data. Exiting.'.format(pfn) self.IJWError(m, self.log_msgs, True) # (msg, log, exit) return # end PrepForecasts def PrepMode(self): """ Prep the mode data for the ordered dictionary, from which the JSON is written. """ pfn = 'mode' if self.VerifyKeyInDict(pfn, required=True): value = self.dataDict[pfn] # the stub value is 'forecast' and that's a legit (and likely) value, so skip the VerifyNonStubValue step valid_valuesL = ['forecast', 'historical'] self.ValidateEnum(value, valid_valuesL, pfn) self.forecast_or_historical_mode = value self.orderedDict[pfn] = value return # end PrepMode def PrepModel(self): """ Prep the model data for the ordered dictionary, from which the JSON is written. """ stub_found = False k = pfn = 'model' # parent field name if self.VerifyKeyInDict(k, required=True): d = self.dataDict[k] toD = collections.OrderedDict() L = [ ('short_name', 'Short name for your model', True, None, None, None, False), ('spase_id', 'spase://CCMC/SimulationModel/MODEL_NAME/VERSION', True, None, None, None, False), ] for (k, s, r, allow_neg, allow_neg_one, must_be_in_past, allow_stub_value) in L: if self.VerifyKeyInDict(k, d, required=r): value = d[k] if not allow_stub_value: stub_found = self.VerifyNonStubValue(value, s, '{}/{}'.format(pfn, k), required=r) # check for non-stub value if allow_stub_value or not stub_found: if k == 'short_name': self.ValidateAlphaNumeric(value, '{}/{}'.format(pfn, k), allow_addtl_chars=' -_+') # should I allow other characters? elif k == 'spase_id': self.ValidateURL(value, '{}/{}'.format(pfn, k), spase_id=True) # Validate single url value toD[k] = value self.orderedDict[pfn] = toD return # end PrepModel def PrepIssueTime(self): """ Prep the issue_time data for the ordered dictionary, from which the JSON is written. """ stub_found = False k = pfn = 'issue_time' # parent field name if self.VerifyKeyInDict(k, required=True): # make sure the key exists L = [ ('issue_time', 'YYYY-MM-DDTHH:MMZ', True, None, None, True, False), ] for (k, s, r, allow_neg, allow_neg_one, must_be_in_past, allow_stub_value) in L: value = self.dataDict[k] if not allow_stub_value: stub_found = self.VerifyNonStubValue(value, s, '{}'.format(k), required=r) # check for non-stub value if allow_stub_value or not stub_found: self.ValidateDateTimeStamp(value, '{}'.format(k), ensure_in_past=must_be_in_past) # ensure proper format 'YYYY-MM-DDTHH:MMZ' self.orderedDict[k] = value return # end PrepIssueTime def PrepTriggers(self): """ Prep the triggers data (if it exists) for the ordered dictionary, from which the JSON is written. """ newL = [] if self.VerifyKeyInDict('triggers', required=False): L = self.dataDict['triggers'] for d in L: newD = {} if self.VerifyKeyInDict('cme', d, required=False): newD = {'cme': self.PrepTriggersCME(d['cme'])} if self.VerifyKeyInDict('flare', d, required=False): newD = {'flare': self.PrepTriggersFlare(d['flare'])} if self.VerifyKeyInDict('cme_simulation', d, required=False): newD = {'cme_simulation': self.PrepTriggersCMESimulation(d['cme_simulation'])} if self.VerifyKeyInDict('particle_intensity', d, required=False): newD = {'particle_intensity': self.PrepTriggersParticleIntensity(d['particle_intensity'])} if newD != {}: newL.append(newD) if newL != []: self.orderedDict['triggers'] = newL # put Triggers in ordered dictionary return def PrepTriggersCME(self, d): """Prep the triggers/cme data (if it exists) for the ordered dictionary, from which the JSON is written.""" stub_found = False pfn = 'triggers/cme' # parent field names toD = collections.OrderedDict() # temp ordered dictionary L = [ ('start_time', 'YYYY-MM-DDTHH:MMZ', True, None, None, False, False, None, None), ('liftoff_time', 'YYYY-MM-DDTHH:MMZ', False, None, None, False, False, None, None), ('lat', 0, False, True, True, None, True, -90, 90), ('lon', 0, False, True, True, None, True, -180, 180), ('pa', 0, False, True, True, None, True, 0, 360), ('half_width', 0, False, True, True, None, True, 0, 180), ('speed', 0, False, True, True, None, True, 0, 5000), ('acceleration', 0, False, False, False, None, True, None, None), ('height', 0, False, True, True, None, False, 1, 250), ] for (k, s, r, allow_neg, allow_neg_one, must_be_in_past, allow_stub_value, min_v, max_v) in L: if self.VerifyKeyInDict(k, d, required=r): value = d[k] if not allow_stub_value: stub_found = self.VerifyNonStubValue(value, s, '{}/{}'.format(pfn, k), required=r) # check for non-stub value if allow_stub_value or not stub_found: if s == 'YYYY-MM-DDTHH:MMZ': self.ValidateDateTimeStamp(value, '{}/{}'.format(pfn, k), ensure_in_past=must_be_in_past) elif s == 0: self.ValidateFloat(value, '{}/{}'.format(pfn, k), allow_neg, allow_neg_one, min_v, max_v) toD[k] = value k = 'time_at_height' if self.VerifyKeyInDict(k, d, required=False): toD[k] = self.PrepTriggersCMETimeAtHeight(d[k]) # coordinates and catalog stub_found = False if self.VerifyKeyInDict('lat', d, required=False) or self.VerifyKeyInDict('lon', d, required=False): L = [('coordinates', self.noneList, True, None, None, None, False)] else: L = [('coordinates', self.noneList, False, None, None, None, False)] L.append(('catalog', self.noneList, False, None, None, None, False)) for (k, s, r, allow_neg, allow_neg_one, must_be_in_past, allow_stub_value) in L: if self.VerifyKeyInDict(k, d, required=r): value = d[k] if not allow_stub_value: stub_found = self.VerifyNonStubValue(value, s, '{}/{}'.format(pfn, k), required=r) # check for non-stub value if allow_stub_value or not stub_found: if k == 'coordinates': self.ValidateCoordinates(value) elif k == 'catalog': self.ValidateCatalog(value) toD[k] = value # urls k = 'urls' if self.VerifyKeyInDict(k, d, required=False): value = d[k] toD[k] = self.PrepTriggersURLs(value, '{}/{}'.format(pfn, k)) # verifies list, non-None, validates, returns fresh List return toD # end PrepTriggersCME def PrepTriggersCMESimulation(self, d): """ """ stub_found = False pfn = 'triggers/cme_simulation' # parent field name toD = collections.OrderedDict() # temp ordered dictionary L = [ ('model', self.noneList, True, None, None, None, False), ('simulation_completion_time', 'YYYY-MM-DDTHH:MMZ', False, None, None, True, False), ('urls', [], False, None, None, None, True), ] for (k, s, r, allow_neg, allow_neg_one, must_be_in_past, allow_stub_value) in L: if self.VerifyKeyInDict(k, d, required=r): value = d[k] if not allow_stub_value: stub_found = self.VerifyNonStubValue(value, s, '{}/{}'.format(pfn, k), required=r) # check for non-stub value if allow_stub_value or not stub_found: if k == 'model': self.ValidateAlphaNumeric(value, '{}/{}'.format(pfn, k), allow_addtl_chars=' -_+') toD[k] = value elif s == 'YYYY-MM-DDTHH:MMZ': self.ValidateDateTimeStamp(value, '{}/{}'.format(pfn, k), ensure_in_past=must_be_in_past) toD[k] = value elif k == 'urls': toD[k] = self.PrepTriggersURLs(value, '{}/{}'.format(pfn, k)) return toD # end PrepTriggersCMESimulation def PrepTriggersCMETimeAtHeight(self, d): """ """ stub_found = False pfn = 'triggers/cme/time_at_height' # parent field name toD = collections.OrderedDict() # temp ordered dictionary L = [ ('time', 'YYYY-MM-DDTHH:MMZ', True, None, None, False, False, None, None), ('height', 0, True, False, False, None, False, 1, 9000), ] for (k, s, r, allow_neg, allow_neg_one, must_be_in_past, allow_stub_value, min_v, max_v) in L: if self.VerifyKeyInDict(k, d, required=r): value = d[k] if not allow_stub_value: stub_found = self.VerifyNonStubValue(value, s, '{}/{}'.format(pfn, k), required=r) # check for non-stub value if allow_stub_value or not stub_found: if s == 'YYYY-MM-DDTHH:MMZ': self.ValidateDateTimeStamp(value, '{}/{}'.format(pfn, k), ensure_in_past=must_be_in_past) elif s == 0: self.ValidateFloat(value, '{}/{}'.format(pfn, k), allow_neg, allow_neg_one, min_v, max_v) toD[k] = value return toD # end PrepTriggersCMETimeAtHeight def PrepTriggersFlare(self, d): """ """ stub_found = False pfn = 'triggers/flare' # parent field name toD = collections.OrderedDict() # temp ordered dictionary L = [ ('last_data_time', 'YYYY-MM-DDTHH:MMZ', True, None, None, False, False), ('start_time', 'YYYY-MM-DDTHH:MMZ', True, None, None, False, False), ('peak_time', 'YYYY-MM-DDTHH:MMZ', False, None, None, False, False), ('end_time', 'YYYY-MM-DDTHH:MMZ', False, None, None, False, False), ('location', self.noneList, False, None, None, None, False), ('intensity', 0, False, None, None, None, False), ('integrated_intensity', 0, False, None, None, None, False), ('noaa_region', self.noneList, False, None, None, None, False), ('urls', [], False, None, None, None, True), ] for (k, s, r, allow_neg, allow_neg_one, must_be_in_past, allow_stub_value) in L: if self.VerifyKeyInDict(k, d, required=r): value = d[k] if not allow_stub_value: stub_found = self.VerifyNonStubValue(value, s, '{}/{}'.format(pfn, k), required=r) # check for non-stub value if allow_stub_value or not stub_found: if s == 'YYYY-MM-DDTHH:MMZ': self.ValidateDateTimeStamp(d[k], '{}/{}'.format(pfn, k), ensure_in_past=must_be_in_past) toD[k] = value elif k == 'location': self.ValidateStonyhurstCoordinates(value, '{}/{}'.format(pfn, k)) # Validate Stonyhurst coordinates (N00W00/S00E00 format) toD[k] = value elif s == 0: self.ValidateFloat(value, '{}/{}'.format(pfn, k), neg_allowed=allow_neg, neg_one_allowed=allow_neg_one) toD[k] = value elif k == 'noaa_region': self.ValidateNOAARegion(value) # Validate noaa_region (include the preceding 1) toD[k] = value elif k == 'urls': toD[k] = self.PrepTriggersURLs(value, '{}/{}'.format(pfn, k)) return toD # end PrepTriggersFlare def PrepTriggersParticleIntensity(self, d): """ """ stub_found = False toD = collections.OrderedDict() # temp ordered dictionary pfn = 'triggers/particle_intensity' # parent field name L = [ ('observatory', self.noneList, True, None, None, None, False), ('instrument', self.noneList, True, None, None, None, False), ('last_data_time', 'YYYY-MM-DDTHH:MMZ', True, None, None, False, False), ] for (k, s, r, allow_neg, allow_neg_one, must_be_in_past, allow_stub_value) in L: if self.VerifyKeyInDict(k, d, required=r): value = d[k] if not allow_stub_value: stub_found = self.VerifyNonStubValue(value, s, '{}/{}'.format(pfn, k), required=r) # check for non-stub value if allow_stub_value or not stub_found: if s == 'YYYY-MM-DDTHH:MMZ': self.ValidateDateTimeStamp(value, '{}/{}'.format(pfn, k), ensure_in_past=must_be_in_past) else: # Validate observatory and instrument values (is that even possible? alphanumeric + some special characters?) self.ValidateAlphaNumeric(value, '{}/{}'.format(pfn, k), allow_addtl_chars='-_') toD[k] = value k = 'ongoing_events' if self.VerifyKeyInDict(k, d, required=False): toD[k] = self.PrepTriggersParticleIntensityOngoingEvents(d[k]) return toD # end PrepTriggersParticleIntensity def PrepTriggersParticleIntensityOngoingEvents(self, List): """ """ stub_found = False pfn = 'triggers/particle_intensity/ongoing_events' # parent field name newL = [] L = [ ("start_time", "YYYY-MM-DDTHH:MMZ", True, None, None, False, False), ("threshold", 0, True, False, False, None, True), ("energy_min", 0, True, False, False, None, True), ("energy_max", 0, True, False, True, None, True), # -1 ok. o/w float ] #print('received List: {} which is of type({}).'.format(List, type(List))) for d in List: toD = collections.OrderedDict() # temp ordered dictionary for (k, s, r, allow_neg, allow_neg_one, must_be_in_past, allow_stub_value) in L: if self.VerifyKeyInDict(k, d, required=r): value = d[k] if not allow_stub_value: stub_found = self.VerifyNonStubValue(value, s, '{}/{}'.format(pfn, k), required=r) # check for non-stub value if allow_stub_value or not stub_found: if s == 'YYYY-MM-DDTHH:MMZ': self.ValidateDateTimeStamp(value, '{}/{}'.format(pfn, k), ensure_in_past=must_be_in_past) else: self.ValidateFloat(value, '{}/{}'.format(pfn, k), neg_allowed=allow_neg, neg_one_allowed=allow_neg_one) toD[k] = value newL.append(toD) return newL # end PrepTriggersParticleIntensityOngoingEvents def PrepTriggersURLs(self, value, field_name): """Verify that the value is a list of urls. Don't return any urls that are in the self.noneList. Validate the URL. """ newL = [] self.VerifyIsInstance(value, list, field_name) # make sure it is a list for url in value: if any([self.VerifyNonStubValue(url, n, field_name, required=False) for n in self.noneList]): pass else: self.ValidateURL(url, field_name) # Validate single url value newL.append(url) return newL # end PrepTriggersURLs def PrintLogMessage(self, m, t, logger=None): """Print the message. Log the message.""" if logger == None: self.InitLogger() if mode == 'debug': #print "%s: %s" % (mode, msg) logger.debug(msg) elif mode == 'info': #print "%s: %s" % (mode, msg) logger.info(msg) elif mode == 'warning': print "%s: %s" % (mode, msg) logger.warning(msg) elif mode == 'error': print "%s: %s" % (mode, msg) logger.error(msg) elif mode == 'critical': print "%s: %s" % (mode, msg) logger.critical(msg) return # end PrintLogMessage def VerifyNonStubValue(self, value, stub_value, field_name, required=True): """Compare the value and the stub value. If they are equivalent, throw an error and quit.""" found_stub_value = False if stub_value == []: if value == stub_value: found_stub_value = True elif stub_value == self.noneList: if any([self.VerifyNonStubValue(value, n, field_name, required=False) for n in self.noneList]): found_stub_value = True elif value == stub_value: found_stub_value = True if found_stub_value: m = "You did not replace the default \'{}\' value.".format(field_name) if required: m = "{} Please do so. Exiting.".format(m) self.IJWError(m, self.log_msgs, True) # (msg, log, exit) else: m = "{} Either replace it or remove the key and default/stub value from the dictionary, please.".format(m) self.IJWWarning(m, log=self.log_msgs) return found_stub_value # end VerifyNonStubValue def ValidateAlphaNumeric(self, value, field_name, allow_addtl_chars=''): """ Validate AlphaNumeric Value """ allowed_chars = string.letters + string.digits + allow_addtl_chars for c in value: if c not in allowed_chars: # throw error m = 'ERROR: An invalid character (\'{}\') was found in \'{}\'. It has to be one of these characters: \'{}\'. Exiting.'.format(c, field_name, allowed_chars) self.IJWError(m, self.log_msgs, True) # (msg, log, exit) return # end ValidateAlphaNumeric def ValidateBoolean(self, value, field_name): """ """ if value not in [True, False]: m = 'ERROR: the boolean value given in \'{}\' is not True or False. Exiting.'.format(field_name) self.IJWError(m, self.log_msgs, True) # (msg, log, exit) return # end ValidateBoolean def ValidateCatalog(self, value): """ """ pfn = 'triggers/cme/catalog' valid_valuesL = ['ARTEMIS', 'DONKI', 'HELCATS', 'JHU APL', 'CACTUS_NRL', 'CACTUS_SIDC', 'CORIMP', 'SEEDS', 'SOHO_CDAW', 'STEREO_COR'] self.ValidateEnum(value, valid_valuesL, pfn) return # end ValidateCatalog def ValidateCoordinates(self, value): """ Validate CME Coordinates Value (in triggers/cme)""" self.ValidateAlphaNumeric(value, 'triggers/cme/coordinates', allow_addtl_chars='._-') return # end ValidateCoordinates def ValidateDateTimeStamp(self, dts_value, field_name, ensure_in_past=False): """Check that the date time stamp has the correct format. Check that it is a valid value. """ #print('In ValidateDateTimeStamp. dts_value is {}'.format(dts_value)) c_typeDict = {} for i in [0,1,2,3,5,6,8,9,11,12,14,15]: c_typeDict[i] = string.digits for i in [4,7]: c_typeDict[i] = '-' c_typeDict[10] = 'T' c_typeDict[13] = ':' c_typeDict[16] = 'Z' #for (k, v) in c_typeDict.items(): # print("{} : {}".format(k, v)) if len(dts_value) != 17: # ensure you have exactly the correct length of date time string m = 'ERROR: the date time string given in \'{}\' is the wrong length. It is \'{}\'. Exiting.'.format(field_name, dts_value) self.IJWError(m, self.log_msgs, True) # (msg, log, exit) for i in range(0, len(dts_value)): #print("dts_value[{0}] is {1} and is of type {2}.".format(i, dts_value[i], type(dts_value[i]))) if dts_value[i] not in c_typeDict[i]: # the character I got in this position is not the correct type, throw an error m = 'ERROR: the character in the {0} index of the date time string given in \'{1}\' is not the correct type. It needs to be one of these characters [{2}]. Exiting.'.format(i, field_name, c_typeDict[i]) self.IJWError(m, self.log_msgs, True) # (msg, log, exit) if ensure_in_past: # make sure the date value is in the past y = int(dts_value[0:4]) m = int(dts_value[5:7]) d = int(dts_value[8:10]) h = int(dts_value[11:13]) n = int(dts_value[14:16]) # n --> thinking m for minute, but it's already used so bumped it to 'n' if datetime.datetime(y, m, d, h, n) > datetime.datetime.now(): # throw error m = 'ERROR: the date time string given in \'{}\' is not in the past. Exiting.'.format(field_name) self.IJWError(m, self.log_msgs, True) # (msg, log, exit) return True # end ValidateDateTimeStamp def ValidateEmail(self, value): """ Validation of email address (only characters + numbers + '_-@.' """ # Validate that there are no illegal characters self.ValidateAlphaNumeric(value, 'email', allow_addtl_chars='.@_-') # make sure there is an '@' in the value and that it is not the first character # and make sure a '.' appears AFTER the '@' with something inbetween i = value.find('@') if i <= 0 or value.find('.', i) <= 0: # throw error m = 'ERROR: the email value given (\'{}\') is not valid. Exiting.'.format(value) self.IJWError(m, self.log_msgs, True) # (msg, log, exit) return # end ValidateEmail def ValidateEnum(self, value, valid_valuesL, field_name): """ Make sure the value is in the valid values list, otherwise, throw an error and exit. """ valid_values_loweredL = [v.lower() for v in valid_valuesL] # do this because we want these strings to be case insensitive if value.lower() not in valid_values_loweredL: # allow the value to be any case # throw error m = 'ERROR: the value given (\'{}\') is not valid. It needs to be one of these values: \n{}\nExiting.'.format(value, str(valid_valuesL)) self.IJWError(m, self.log_msgs, True) # (msg, log, exit) return # end ValidateEnum def ValidateFloat(self, value, field_name, neg_allowed=False, neg_one_allowed=False, min_=None, max_=None): """ sample intensity value: "8.2e-4" """ if isinstance(value, str): try: value = float(value) except TypeError as ex: print('TypeError Exception: {}'.format(ex)) #print('type({}) is {}'.format(field_name, type(value))) allowed = '{}e-.x'.format(string.digits) if isinstance(value, str): value = value.strip() if neg_one_allowed and value == -1: return elif min_ is not None and max_ is not None: if value >= min_ and value <= max_: return else: m = 'ERROR: the float value given (\'{}\') in \'{}\' was not within the specified range of \'{}\' and \'{}\'. Exiting.' m = m.format(value, field_name, min_, max_) self.IJWError(m, self.log_msgs, True) # (msg, log, exit) elif not neg_allowed and value < 0.0: m = 'ERROR: the float value given in \'{}\' was negative and that is not allowed in this field. Exiting.'.format(field_name) self.IJWError(m, self.log_msgs, True) # (msg, log, exit) for c in str(value): if c not in allowed: m = 'ERROR: the float value given in \'{}\' has an illegal character in it. Exiting.'.format(field_name) self.IJWError(m, self.log_msgs, True) # (msg, log, exit) return # end ValidateFloat def ValidateForecastEventLength(self, start, end, prediction_window_d): """Event length must fall within the prediction window. end can be None. """ start_dto = self.ConvertDTString2DTO(start) pw_start_dto = self.ConvertDTString2DTO(prediction_window_d['start_time']) pw_end_dto = self.ConvertDTString2DTO(prediction_window_d['end_time']) if start_dto < pw_start_dto or start_dto > pw_end_dto: # throw error m = 'ERROR: the forecast/event_length/start_time is not within the prediction window. Exiting.' self.IJWError(m, self.log_msgs, True) # (msg, log, exit) if end is not None: end_dto = self.ConvertDTString2DTO(end) if end_dto < pw_start_dto or end_dto > pw_end_dto: # throw error m = 'ERROR: the forecast/event_length/end_time ({}) is not within the prediction window ({} - {}). Exiting.'.format(end_dto, pw_start_dto, pw_end_dto) self.IJWError(m, self.log_msgs, True) # (msg, log, exit) return # end ValidateForecastEventLength def ValidateForecastLocation(self, v): """ """ pfn = 'forecast/location' valid_valuesL = ['mercury', 'venus', 'earth', 'mars', 'psp', 'stereoa', 'stereob', 'dawn', 'juno', 'L1', 'L2', 'L4', 'L5'] self.ValidateEnum(v, valid_valuesL, pfn) return True # end ValidateForecastLocation def ValidateForecastNativeID(self, v): """ """ # make sure there are no wacko characters/scripting hacks self.ValidateAlphaNumeric(v, 'native_id', allow_addtl_chars='_-.') return # end ValidateForecastNativeID def ValidateForecastProbabilityValue(self, v): """Probability value is a float, but it has to be between 0 and 1. """ if isinstance(v, str): v = float(v) if v < 0.0 or v > 1.0: m = 'ERROR: the float value given (\'{}\') in \'forecasts/probability_value\' is invalid. It must be between 0 and 1. Exiting.'.format(v) self.IJWError(m, self.log_msgs, True) # (msg, log, exit) return # end ValidateForecastProbabilityValue def ValidateForecastSEPProfile(self, v): """ """ # make sure there are no wacko characters/scripting hacks self.ValidateAlphaNumeric(v, 'sep_profile', allow_addtl_chars='-_.') return # end ValidateForecastSEPProfile def ValidateForecastSpecies(self, v): """ """ pfn = 'forecast/species' valid_valuesL = ['electron', 'proton', 'helium', 'helium3', 'helium4', 'oxygen', 'iron', 'ion'] self.ValidateEnum(v, valid_valuesL, pfn) return True # end ValidateForecastSpecies def ValidateHTTPURL(self, value, field_name): """Validate a regular HTTP URL """ value = value.strip() # chop off any leading or trailing spaces # if there's a space anywhere in the URL, return False success value if ' ' in value: return False # it must start with http or https if value.startswith('http'): if value.startswith('http'): ci = 4 elif value.startswith('https'): ci = 5 else: return False # followed by :// if value[ci:ci+3] == '://': ci = ci + 3 # try: # followed by x number of alphanumeric characters, including - (hostname)... followed by . ei = value.find('.', ci) self.ValidateAlphaNumeric(value[ci:ei], field_name, allow_addtl_chars='-') ci = ei + 1 # followed by x number of alphanumeric characters, including - (domain name)... followed by . ei = value.find('.', ci) self.ValidateAlphaNumeric(value[ci:ei], field_name, allow_addtl_chars='-') ci = ei + 1 # followed by 2+ number of alpha characters (top-level domain)... followed by / or : ei = value.find(':', ci) if ei > -1: # there's a port number # Validate the TLD self.ValidateAlphaNumeric(value[ci:ei], field_name, allow_addtl_chars='') ci = ei + 1 # Validate the port number ei = value.find('/', ci) self.ValidateNumeric(value[ci:ei], field_name) ci = ei + 1 else: # Validate the TLD ei = value.find('/', ci) if ei == -1: ei = len(value) self.ValidateAlphaNumeric(value[ci:ei], field_name, allow_addtl_chars='') ci = ei + 1 ## followed by x number of alphanumeric characters, including -%=&?~/ #if len(value) >= ci: # ei = len(value) # # Check for blacklisted characters??? # #self.ValidateAlphaNumeric(value[ci:ei], '{}/{}'.format(pfn, k), allow_addtl_chars='-%=&?~/') # This is going to be too strong. # except: return False else: return False return True # end ValidateHTTPURL def ValidateNOAARegion(self, value): """ Validate noaa_region. The NOAA active region number is a 5-digit number and current values are > 10,000, although many people present just the last four digits. """ # make sure all digits. for c in str(value): if c not in string.digits: m = 'ERROR: \'noaa_region\' is not all numerical digits. Exiting.' self.IJWError(m, self.log_msgs, True) # (msg, log, exit) if isinstance(value, str): try: value = int(value) except TypeError as ex: print('TypeError Exception: {}'.format(ex)) # make sure value is > 10000. if value < 10000: m = 'ERROR: \'noaa_region\' has to include the leading 1 in the number. Exiting.' self.IJWError(m, self.log_msgs, True) # (msg, log, exit) if value > 100000: # it should be a 5 digit number, so if it goes over to 6 digits, throw an error m = 'ERROR: \'noaa_region\' is too long. Exiting.' self.IJWError(m, self.log_msgs, True) # (msg, log, exit) return # end ValidateNOAARegion def ValidateNumeric(self, value, field_name): """Validate that every character in the given string is a digit.""" for c in value: # value is already a string if c not in string.digits: m = 'ERROR: the numeric value given in \'{}\' has an illegal character in it. Exiting.'.format(field_name) self.IJWError(m, self.log_msgs, True) # (msg, log, exit) return # end ValidateNumeric def ValidateSpaseURL(self, value, field_name): """ validate that URL follows this convention: "spase_id": "spase://CCMC/SimulationModel/MODEL_NAME/VERSION" """ if value.startswith('spase'): ci = 5 # followed by :// if value[ci:ci+3] == '://': ci = ci + 3 for i in range(0, 4): # for ['CCMC', 'SimulationModel', 'MODEL_NAME', 'VERSION'] try: ei = value.find('/', ci) self.ValidateAlphaNumeric(value[ci:ei], field_name, allow_addtl_chars='._-=') ci = ei + 1 except: return False else: return False return True # end ValidateSpaseURL(value, field_name) def ValidateStonyhurstCoordinates(self, value, field_name): """ Validate the use of Stonyhurst coordinates (use N00W00/S00E00 format). """ self.ValidateAlphaNumeric(value, field_name) c_typeDict = {} for i in [1,2,4,5]: c_typeDict[i] = string.digits c_typeDict[0] = 'NS' c_typeDict[3] = 'EW' #for (k, v) in c_typeDict.items(): # print("{} : {}".format(k, v)) if len(value) != 6: # ensure you have exactly the correct length of coordinates m = 'ERROR: the Stonyhurst coordinate string given in \'{}\' is the wrong length. Exiting.'.format(field_name) self.IJWError(m, self.log_msgs, True) # (msg, log, exit) for i in range(0, len(value)): #print("value[{0}] is {1} and is of type {2}.".format(i, value[i], type(value[i]))) if value[i] not in c_typeDict[i]: # the character I got in this position is not the correct type, throw an error m = 'ERROR: the character in the {0} index of the date time string given in \'{1}\' is not the correct type. It needs to be one of these characters [{2}]. Exiting.'.format(i, field_name, c_typeDict[i]) self.IJWError(m, self.log_msgs, True) # (msg, log, exit) return # end ValidateStonyhurstCoordinates def ValidateURL(self, value, field_name, spase_id=False): """ Validate the URL, at least within reason. """ success = False if value.startswith('http'): success = self.ValidateHTTPURL(value, field_name) elif value.startswith('spase'): success = self.ValidateSpaseURL(value, field_name) else: pass # throw an error return # end ValidateURL def VerifyIsInstance(self, v, t, field_name): """Throw an error if the given value, v, is not of type t.""" if not isinstance(v, t): m = 'ERROR: \'{}\' is supposed to be a {}, but it isn\'t. Exiting.'.format(field_name, str(t)) self.IJWError(m, self.log_msgs, True) # (msg, log, exit) return # end VerifyIsInstance def VerifyKeyInDict(self, key, local_dataDict=None, required=False): """See if the key exists in the appropriate dictionary (self.dataDict or the provided local_dataDict). If it does not, throw an error if it is a required field. """ if local_dataDict == None: local_dataDict = self.dataDict #if not isinstance(local_dataDict, dict): # print("ERROR: expecting a dictionary. Got something else in \'{}\'. Exiting.".format(key)) # sys.exit() keyexists = key in local_dataDict.keys() if required and not keyexists: # issue error #print("===========================================================") print("ERROR: missing key \'{}\' in dictionary. Exiting.".format(key)) #print(" dictionary does have the following keys: ".format(key)) #for k in local_dataDict.keys(): # print(' {}: {}'.format(k, local_dataDict[k])) sys.exit() return keyexists # end VerifyKeyInDict def WriteJSON(self): """This is about controlling the order of the data written to the JSON file. Along the way, prep the data for the ordered dictionary, from which the JSON is written. For each field, make sure to: 1) verify that required field(s) are there. 2) verify that the stub/default/sample data has been replaced (but allow stub values in some cases). 3) validate the data that is there. """ self.PrepContacts() self.PrepModel() self.PrepIssueTime() self.PrepMode() self.PrepTriggers() self.PrepForecasts() # now for the actual writing d = {'sep_forecast_submission' : self.orderedDict} ofn = '{}.{}.json'.format(self.output_file_basename, self.now_ts) # ofn = output file name with open(ofn, 'w') as outfile: json.dump(d, outfile) #### MAIN ##################################### if __name__ == '__main__': program_desc = "This program is supposed to help the modeler to provide their model data in to the CCMC in JSON format. Contact Joycelyn Jones at joycelyn.t.jones@nasa.gov for additional assistance." parser = InitParser(program_desc) (output_file_basename, output_dir, log_msgs, log_dir, log_starter, dataDict) = ParseArguments(parser) ConvertToJSON(dataDict, output_file_basename, output_dir, log_msgs, log_dir, log_starter) """ Modifications: 2019.03.05,Joycelyn Thomson Jones: found latest addition to the visual schema, 'mode' didn't make it into the final JSON output. Added it. 2019.03.05,Joycelyn Thomson Jones: found that 'None' or 'none' values were getting through in lists where they shouldn't. Fixed that. 2019.03.06,Joycelyn Thomson Jones: Improved NOAA Region value validation to check that the number is not too large. 2019.03.06,Joycelyn Thomson Jones: added check for key in the dictionary in PrepTriggersCMESimulation. """