Coverage for /usr/local/lib/python3.7/dist-packages/pyrocko/gui/snufflings/rms.py : 36%

# http://pyrocko.org - GPLv3 

# 

# The Pyrocko Developers, 21st Century 

# ---|P------/S----------~Lg---------- 

from __future__ import absolute_import 

from pyrocko.gui.snuffling import Snuffling, Param 

from pyrocko.trace import Trace 

import numpy as num 

class RootMeanSquareSnuffling(Snuffling): 

''' 

Create traces with blockwise root mean square values. 

''' 

def setup(self): 

'''Customization of the snuffling.''' 

self.set_name('Block RMS') 

self.add_parameter(Param( 

'Block Length [s]', 'block_length', 100., 0.1, 3600.)) 

self.set_live_update(False) 

def call(self): 

'''Main work routine of the snuffling.''' 

self.cleanup() 

tinc = self.block_length 

tmin, tmax = self.get_selected_time_range(fallback=True) 

n = int((tmax-tmin)/tinc) 

rms_by_nslc = {} 

for traces in self.chopper_selected_traces( 

tinc=tinc, 

want_incomplete=False, 

fallback=True): 

for tr in traces: 

# ignore the block if it extends past the region of interest 

if tr.tmax > tmax: 

continue 

# don't work traces produced by this (and other) snuffling 

if tr.meta and 'tabu' in tr.meta and tr.meta['tabu']: 

continue 

# create a trace of required length if none has been 

# initialized yet 

if tr.nslc_id not in rms_by_nslc: 

rms_by_nslc[tr.nslc_id] = Trace( 

network=tr.network, 

station=tr.station, 

location=tr.location, 

channel=tr.channel+'-RMS', 

tmin=tmin + 0.5*tinc, 

deltat=tinc, 

ydata=num.zeros(n, dtype=num.float), 

meta={'tabu': True}) 

# create and insert the current sample 

i = int(round((tr.tmin - tmin)/tinc)) 

if 0 <= i and i < n: 

tr.ydata = num.asarray(tr.ydata, num.float) 

tr.ydata -= num.mean(tr.ydata) 

value = num.sqrt(num.sum(tr.ydata**2)/tr.ydata.size) 

rms_by_nslc[tr.nslc_id].ydata[i] = value 

# add the newly created traces to the viewer 

if rms_by_nslc.values(): 

self.add_traces(list(rms_by_nslc.values())) 

def __snufflings__(): 

'''Returns a list of snufflings to be exported by this module.''' 

return [RootMeanSquareSnuffling()]