def command_snuffle(args):
from pyrocko.gui import snuffler
parser, options, args = cl_parse('map', args)
if len(args) == 0:
args.append('.')
fn = get_scenario_yml(args[0])
if not fn:
parser.print_help()
sys.exit(1)
engine = get_engine()
scenario = guts.load(filename=fn)
scenario.init_modelling(engine)
return snuffler.snuffle(scenario.get_pile(),
stations=scenario.get_stations(),
events=scenario.get_events())
def command_snuffle(args):
from pyrocko.gui import snuffler
parser, options, args = cl_parse('map', args)
if len(args) == 0:
args.append('.')
fn = get_scenario_yml(args[0])
if not fn:
parser.print_help()
sys.exit(1)
engine = get_engine()
scenario = guts.load(filename=fn)
scenario.init_modelling(engine)
return snuffler.snuffle(
scenario.get_pile(),
stations=scenario.get_stations(),
events=scenario.get_events())
def process(self):
t = time.time()
C = config.Config(self.eventpath)
Config = C.parseConfig('config')
cfg = ConfigObj(dict=Config)
Origin = C.parseConfig('origin')
if cfg.pyrocko_download() is True:
if cfg.quantity() == 'displacement':
disp = True
else:
disp = False
Meta = readpyrockostations(self.eventpath, disp)
elif cfg.colesseo_input() is True:
scenario = guts.load(filename=cfg.colosseo_scenario_yml())
scenario_path = cfg.colosseo_scenario_yml()[:-12]
Meta = readcolosseostations(scenario_path)
events = scenario.get_events()
ev = events[0]
Origin['strike'] = str(ev.moment_tensor.strike1)
Origin['rake'] = str(ev.moment_tensor.rake1)
Origin['dip'] = str(ev.moment_tensor.dip1)
Origin['lat'] = str(ev.lat)
Origin['lon'] = str(ev.lon)
Origin['depth'] = str(ev.depth / 1000.)
else:
Meta = readMetaInfoFile(self.eventpath)
Folder = createFolder(self.eventpath)
FilterMeta = filterStations(Meta, Config, Origin)
try:
km(Config, FilterMeta, Folder, Origin, t)
except Exception:
pass
return True
def processLoop():
#==================================get meta info==========================================
C = config.Config(evpath)
Origin = C.parseConfig('origin')
try:
Syn_in = C.parseConfig('syn')
syn_in = SynthCfg(Syn_in)
except:
pass
Config = C.parseConfig('config')
cfg = ConfigObj(dict=Config)
if cfg.pyrocko_download() == True:
Meta = C.readpyrockostations() #
elif cfg.colesseo_input() == True:
scenario = guts.load(filename=cfg.colosseo_scenario_yml())
scenario_path = cfg.colosseo_scenario_yml()[:-12]
Meta = C.readcolosseostations(scenario_path)
else:
Meta = C.readMetaInfoFile()
#==================================get meta info==========================================
#==================================do prerequiries========================================
Folder = C.createFolder()
#C.cpSkeleton(Folder,Config)
C.writeConfig(Config, Origin, Folder)
filter = FilterCfg(Config)
ntimes = int(
(cfg.UInt('forerun') + cfg.UInt('duration')) / cfg.UInt('step'))
origin = OriginCfg(Origin)
if cfg.colesseo_input() == True:
from pyrocko import util
events = scenario.get_events()
ev = events[0]
origin.strike = str(ev.moment_tensor.strike1)
origin.rake = str(ev.moment_tensor.rake1)
origin.dip = str(ev.moment_tensor.dip1)
strike = ev.moment_tensor.strike1
origin.lat = str(ev.lat)
origin.lon = str(ev.lon)
origin.depth = str(ev.depth / 1000.)
depth = ev.depth
origin.time = util.time_to_str(ev.time)
time_ev = util.time_to_str(ev.time)
lat = ev.lat
lon = ev.lon
rake = ev.moment_tensor.rake1
dip = ev.moment_tensor.dip1
Origin['strike'] = str(ev.moment_tensor.strike1)
Origin['rake'] = str(ev.moment_tensor.rake1)
Origin['dip'] = str(ev.moment_tensor.dip1)
Origin['lat'] = str(ev.lat)
Origin['lon'] = str(ev.lon)
Origin['time'] = util.time_to_str(ev.time)
Origin['depth'] = str(ev.depth / 1000.)
ev = Event(lat, lon, depth, time_ev, strike=strike, dip=dip, rake=rake)
else:
default = 0
strike = origin.strike(default) # Origin.get ('strike', default)
dip = origin.dip(default) # Origin.get ('dip', default)
rake = origin.rake(default) # Origin.get ('rake', default)
ev = Event(origin.lat(),
origin.lon(),
origin.depth(),
origin.time(),
strike=strike,
dip=dip,
rake=rake)
filtername = filter.filterName()
Logfile.add('filtername = ' + filtername)
#todo crosscorrelation for all arrays before processing
XDict = {}
RefDict = {}
SL = {}
if cfg.Int('xcorr') == 1:
newFreq = str(filter.newFrequency())
fobjreferenceshiftname = newFreq + '_' + filtername + '.refpkl'
rp = os.path.join(Folder['semb'], fobjreferenceshiftname)
fobjpickleshiftname = newFreq + '_' + filtername + '.xcorrpkl'
ps = os.path.join(Folder['semb'], fobjpickleshiftname)
if (os.path.isfile(rp) and os.path.getsize(rp) != 0
and os.path.isfile(ps) and os.path.getsize(ps) != 0):
Logfile.add('file exits : ' + rp)
Logfile.add('load refshifts')
f = open(rp)
RefDict = pickle.load(f)
x = open(ps)
XDict = pickle.load(x)
xcorrnetworks = cfg.String('networks').split(',')
for i in xcorrnetworks:
SL[i] = len(Config[i].split('|'))
else:
SL = {}
xcorrnetworks = cfg.String('networks').split(',')
for i in xcorrnetworks:
W = {}
refshift = 0
network = cfg.String(i).split('|')
FilterMeta = ttt.filterStations(Meta, Config, Origin, network)
arrayfolder = os.path.join(Folder['semb'], i)
if os.access(arrayfolder, os.F_OK) == False:
os.makedirs(arrayfolder)
if cfg.pyrocko_download() == True:
A = Xcorr(ev, FilterMeta, evpath, Config, Syn_in,
arrayfolder)
else:
A = Xcorr(ev, FilterMeta, evpath, Config, Syn_in,
arrayfolder)
print "run Xcorr"
W, triggerobject = A.runXcorr()
XDict[i] = W
RefDict[i] = triggerobject.tdiff
SL[i] = len(network)
#endfor
fobjrefshift = open(rp, 'w')
pickle.dump(RefDict, fobjrefshift)
fobjrefshift.close()
output = open(ps, 'w')
pickle.dump(XDict, output)
output.close()
for i in sorted(XDict.iterkeys()):
Logfile.red('Array %s has %3d of %3d Stations left' %
(i, len(XDict[i]), SL[i]))
logger.info(
'\033[31mFor proceeding without changes press enter or give new comma seperatet network list or quit for exit\033[0m'
)
while True:
nnl = raw_input("please enter your choice: ")
#Logfile.add ('Choise = ' + nnl)
if len(nnl) == 0:
if not Basic.question('Process all networks ?'): continue
Logfile.red('This networks will be used for processing: %s' %
(Config['networks']))
break
elif str(nnl) == 'quit':
sys.exit()
elif str(nnl) == 'rerun':
event = os.path.join(*evpath.split('/')[-1:])
try:
os.remove(rp)
os.remove(ps)
except:
pass
mainfolder = os.path.join(os.path.sep, *evpath.split('/')[:-2])
os.chdir(mainfolder)
cmd = ('%s arraytool.py process %s') % (sys.executable, event)
Logfile.add('cmd = ' + cmd)
os.system(cmd)
sys.exit()
else:
# Check if selected array(s) exists
names = nnl.split(',')
isOk = True
for array in names:
arrayfolder = os.path.join(Folder['semb'], array)
if not os.path.isdir(arrayfolder):
Logfile.error('Illegal network name ' + str(array))
isOk = False
break
#endfor
if not isOk: continue # Illegal network : input again
# use these networks
Logfile.add('This networks will be used for processing: %s' %
(nnl))
Config['networks'] = nnl
break
for i in range(3, 0, -1):
time.sleep(1)
Logfile.red('Start processing in %d seconds ' % (i))
wd = Origin['depth']
start, stop, step = cfg.String('depths').split(',')
start = int(start)
stop = int(stop) + 1
step = int(step)
filters = cfg.String('filters')
filters = int(filters)
Logfile.add('working on ' + Config['networks'])
#==================================loop over depth======================
for filterindex in xrange(0, filters):
for depthindex in xrange(start, stop, step):
workdepth = float(wd) + depthindex
Origin['depth'] = workdepth
ev = Event(Origin['lat'],
Origin['lon'],
Origin['depth'],
Origin['time'],
strike=strike,
dip=dip,
rake=rake)
Logfile.add('WORKDEPTH: ' + str(Origin['depth']))
#==================================do prerequiries===============
#==================================loop over arrays================
ASL = []
weights = []
array_centers = []
networks = Config['networks'].split(',')
counter = 1
TriggerOnset = []
Wdfs = []
for i in networks:
arrayname = i
arrayfolder = os.path.join(Folder['semb'], arrayname)
network = Config[i].split('|')
Logfile.add('network: ' + str(network))
FilterMeta = ttt.filterStations(Meta, Config, Origin, network)
#if len(FilterMeta) < 3: continue #hs : wieder rein
if len(FilterMeta) < 3: continue
W = XDict[i]
refshift = RefDict[i]
FilterMeta = cmpFilterMetavsXCORR(W, FilterMeta)
Logfile.add(
'BOUNDING BOX DIMX: %s DIMY: %s GRIDSPACING: %s \n' %
(Config['dimx'], Config['dimy'], Config['gridspacing']))
##############=======================PARALLEL===========================================
Logfile.red('Calculating Traveltime Grid')
t1 = time.time()
isParallel = False #10.12.2015
TTTGridMap = []
mint = []
maxt = []
try:
f = open(
'../tttgrid/tttgrid_%s_%s_%s.pkl' %
(ev.time, arrayname, workdepth), 'rb')
print "loading travel time grid_%s_%s_%s.pkl" % (
ev.time, arrayname, workdepth)
TTTGridMap, mint, maxt = pickle.load(f)
f.close()
print "loading of travel time grid sucessful"
except:
print "loading of travel time grid unsucessful, will now calculate the grid:"
if isParallel: #hs
# maxp = int (Config['ncore'])
maxp = 6 #hs
po = multiprocessing.Pool(maxp)
for i in xrange(len(FilterMeta)):
po.apply_async(ttt.calcTTTAdv,
(Config, FilterMeta[i], Origin, i,
arrayname, W, refshift))
po.close()
po.join()
else: #hs+
for i in xrange(len(FilterMeta)):
t1 = time.time()
ttt.calcTTTAdv(Config, FilterMeta[i], Origin, i,
arrayname, W, refshift)
Logfile.add('ttt.calcTTTAdv : ' +
str(time.time() - t1) + ' sec.')
#endif #hs-
assert len(FilterMeta) > 0
TTTGridMap = deserializer.deserializeTTT(len(FilterMeta))
mint, maxt = deserializer.deserializeMinTMaxT(
len(FilterMeta))
f = open(
'../tttgrid/tttgrid_%s_%s_%s.pkl' %
(ev.time, arrayname, workdepth), 'wb')
print "dumping the traveltime grid for this array"
pickle.dump([TTTGridMap, mint, maxt], f)
f.close()
t2 = time.time()
Logfile.red('%s took %0.3f s' % ('TTT', (t2 - t1)))
switch = filterindex
tw = times.calculateTimeWindows(mint, maxt, Config, ev, switch)
if cfg.pyrocko_download() == True:
if cfg.quantity() == 'displacement':
Wd = waveform.readWaveformsPyrocko_restituted(
FilterMeta, tw, evpath, ev)
else:
Wd = waveform.readWaveformsPyrocko(
FilterMeta, tw, evpath, ev)
# Wdf = waveform.processpyrockoWaveforms(Wd, Config, Folder, arrayname, FilterMeta, ev, switch, W)
elif cfg.colesseo_input() == True:
Wd = waveform.readWaveforms_colesseo(
FilterMeta, tw, evpath, ev, C)
else:
Wd = waveform.readWaveforms(FilterMeta, tw, evpath, ev)
if cfg.Bool('synthetic_test') is True:
Wdf = waveform.processdummyWaveforms(
Wd, Config, Folder, arrayname, FilterMeta, ev, switch,
W)
Wdfs.append(Wdf)
else:
Wdf = waveform.processWaveforms(Wd, Config, Folder,
arrayname, FilterMeta, ev,
switch, W)
Wdfs.append(Wdf)
C.writeStationFile(FilterMeta, Folder, counter)
Logfile.red('%d Streams added for Processing' % (len(Wd)))
t1 = time.time()
f = open(
'../tttgrid/tttgrid_%s_%s_%s.pkl' %
(ev.time, arrayname, workdepth), 'rb')
print "loading travel time grid_%s_%s_%s.pkl" % (
ev.time, arrayname, workdepth)
TTTGridMap, mint, maxt = pickle.load(f)
f.close()
if cfg.optimize() == True:
optim.solve(counter, Config, Wdf, FilterMeta, mint, maxt,
TTTGridMap, Folder, Origin, ntimes, switch, ev,
arrayfolder, syn_in)
else:
arraySemb, weight, array_center = sembCalc.doCalc(
counter, Config, Wdf, FilterMeta, mint, maxt,
TTTGridMap, Folder, Origin, ntimes, switch, ev,
arrayfolder, syn_in)
t2 = time.time()
Logfile.add('CALC took %0.3f sec' % (t2 - t1))
weights.append(weight)
array_centers.append(array_center)
ASL.append(arraySemb)
counter += 1
sembCalc.writeSembMatricesSingleArray(arraySemb, Config,
Origin, arrayfolder,
ntimes, switch)
fileName = os.path.join(arrayfolder, 'stations.txt')
Logfile.add('Write to file ' + fileName)
fobjarraynetwork = open(fileName, 'w')
for i in FilterMeta:
fobjarraynetwork.write(
('%s %s %s\n') % (i.getName(), i.lat, i.lon))
fobjarraynetwork.close()
TTTGridMAP = []
if cfg.optimize_all() == True:
import optim_csemb
from optim_csemb import solve
sembmax = sembCalc.collectSemb(ASL, Config, Origin, Folder,
ntimes, len(networks), switch)
optim_csemb.solve(counter, Config, Wdf, FilterMeta, mint, maxt,
TTTGridMap, Folder, Origin, ntimes, switch,
ev, arrayfolder, syn_in, ASL, sembmax,
evpath, XDict, RefDict, workdepth,
filterindex, Wdfs)
if ASL:
Logfile.red('collect semblance matrices from all arrays')
sembmax = sembCalc.collectSemb(ASL, Config, Origin, Folder,
ntimes, len(networks), switch,
array_centers)
if cfg.Bool('weight_by_noise') == True:
sembCalc.collectSembweighted(ASL, Config, Origin, Folder,
ntimes, len(networks), switch,
weights)
else:
Logfile.red('Nothing to do -> Finish')
print "depth:"
print workdepth