def __init__(self, date, directory = '~/Data/', extract='CHANNEL4'):
"""Routine to acquire the relevant data for a particular day for the
PROBA2 GI"""
# Which date?
self.date = date
# Acquire the HEK data
self.hek = fevent(self.date, directory=os.path.join(directory,'HEK/'), verbose=True)
# Acquire the LYRA data
self.lyra = LYRALightCurve.create(self.date)
# Which channel?
self.extract = extract
# No event times
self.no_event_times = self.hek.onoff(frm_name='combine').complement().times()
# Event times
self.event_times = self.hek.onoff(frm_name='combine').times()
# Spikes
self.all_spike_times =[]
for i,tr in enumerate(self.no_event_times):
lc = self.lyra.truncate(tr).extract(self.extract)
if len(lc.data) > 2:
# Look for spikes in the time series: uses time-series, not lightcurves
spike_times = find_spike(lc.data)
if spike_times is not None:
for st in spike_times:
self.all_spike_times.append(st)
def main():
hek_directory='~/Data/HEK'
tstart = parse_time('2012/06/08')
tend = parse_time('2012/07/08')
channel='CHANNEL4'
while tstart <= tend:
# Get the HEK data
hek = fevent(tstart, directory=hek_directory, verbose=True)
# Get the time-ranges of when there is NO event
non_event_times = hek.onoff(frm_name='combine').complement().times()
# Get the PROBA2 data
lyra = LYRALightCurve.create(tstart)
# Get the spike times in the non flaring data
spike_times=[]
for tr in non_event_times:
lyra_bit = lyra.data[channel][tr.start():tr.end()]
spike_times = proba2gi2.find_spike(lyra_bit)
if spike_times != []:
spike_times.append(spike_times)
# Combine the spike times and the non-flaring data
times_list =
# Identify the time series that begin with a time which is the start of a non-event
# time series, and end in a spike
after_flare = []
for i in range(len(times_list)):
if (times_list[i] is EventTimeStart) and (times_list[i+1] is SpikeTimeStart):
after_flare.append(TimeRange(times_list[i],times_list[i+1]))
# Identify the time series that end with an event_start
before_flare = []
for i in range(len(times_list)):
if (times_list[i] is SpikeTimeEnd) and (times_list[i+1] is EventTimeEnd):
before_flare.append(TimeRange(times_list[i],times_list[i+1]))
# Save these time-series.
# Advance
tstart = tstart + timedelta(days=1)