def example_data_set(tmpdir):
"""
Fixture creating a small example file.
"""
asdf_filename = os.path.join(tmpdir.strpath, "test.h5")
data_path = os.path.join(data_dir, "small_sample_data_set")
data_set = ASDFDataSet(asdf_filename)
for filename in glob.glob(os.path.join(data_path, "*.xml")):
if "quake.xml" in filename:
data_set.add_quakeml(filename)
else:
data_set.add_stationxml(filename)
for filename in glob.glob(os.path.join(data_path, "*.mseed")):
data_set.add_waveforms(filename, tag="raw_recording",
event_id=data_set.events[0])
# Flush and finish writing.
del data_set
# Return filename and path to tempdir, no need to always create a
# new one.
return Namespace(filename=asdf_filename, tmpdir=tmpdir.strpath)
def combine_asdf(base_asdf_path, append_asdf_path, output_asdf_path):
"""
combine_asdf: merge the waveforms in append_asdf to base_asdf, and generate a new asdf.
"""
base_asdf = ASDFDataSet(base_asdf_path, mode="r", mpi=False)
append_asdf = ASDFDataSet(append_asdf_path, mode="r", mpi=False)
output_asdf = ASDFDataSet(output_asdf_path, mpi=False)
# * add events
events = base_asdf.events
event = events[0]
output_asdf.add_quakeml(events)
# * add waveforms and stationxml
# firstly we add base asdf
rep_net_sta = base_asdf.waveforms.list()[0]
tag_default = base_asdf.waveforms[rep_net_sta].get_waveform_tags()[0]
for each_net_sta in base_asdf.waveforms.list():
tag = base_asdf.waveforms[each_net_sta].get_waveform_tags()[0]
assert tag == tag_default
st = base_asdf.waveforms[each_net_sta][tag]
inv = base_asdf.waveforms[each_net_sta]["StationXML"]
output_asdf.add_waveforms(st, tag=tag, event_id=event)
output_asdf.add_stationxml(inv)
# secondly we add append asdf
for each_net_sta in append_asdf.waveforms.list():
tag = append_asdf.waveforms[each_net_sta].get_waveform_tags()[0]
assert tag == tag_default
st = append_asdf.waveforms[each_net_sta][tag]
inv = append_asdf.waveforms[each_net_sta]["StationXML"]
output_asdf.add_waveforms(st, tag=tag, event_id=event)
output_asdf.add_stationxml(inv)
del base_asdf
del append_asdf
del output_asdf
def test_data_set_creation(tmpdir):
"""
Test data set creation with a small test dataset.
It tests that the the stuff that goes in is correctly saved and
can be retrieved again.
"""
asdf_filename = os.path.join(tmpdir.strpath, "test.h5")
data_path = os.path.join(data_dir, "small_sample_data_set")
data_set = ASDFDataSet(asdf_filename)
for filename in glob.glob(os.path.join(data_path, "*.mseed")):
data_set.add_waveforms(filename, tag="raw_recording")
for filename in glob.glob(os.path.join(data_path, "*.xml")):
if "quake.xml" in filename:
data_set.add_quakeml(filename)
else:
data_set.add_stationxml(filename)
# Flush and finish writing.
del data_set
# Open once again
data_set = ASDFDataSet(asdf_filename)
# ObsPy is tested enough to make this comparison meaningful.
for station in (("AE", "113A"), ("TA", "POKR")):
# Test the waveforms
stream_asdf = \
getattr(data_set.waveforms, "%s_%s" % station).raw_recording
stream_file = obspy.read(os.path.join(
data_path, "%s.%s.*.mseed" % station))
# Delete the file format specific stats attributes. These are
# meaningless inside ASDF data sets.
for trace in stream_file:
del trace.stats.mseed
del trace.stats._format
for trace in stream_asdf:
del trace.stats.asdf
del trace.stats._format
assert stream_asdf == stream_file
# Test the inventory data.
inv_asdf = \
getattr(data_set.waveforms, "%s_%s" % station).StationXML
inv_file = obspy.read_inventory(
os.path.join(data_path, "%s.%s..BH*.xml" % station))
assert inv_file == inv_asdf
# Test the event.
cat_file = obspy.readEvents(os.path.join(data_path, "quake.xml"))
cat_asdf = data_set.events
# from IPython.core.debugger import Tracer; Tracer(colors="Linux")()
assert cat_file == cat_asdf
def convert_to_asdf(filelist,
asdf_fn,
quakemlfile,
staxml_filelist=None,
tag=None):
"""
Convert files(sac or mseed) to asdf
"""
nfiles = len(filelist)
if nfiles == 0:
print "No file specified. Return..."
return
if os.path.exists(asdf_fn):
raise Exception("File '%s' exists." % asdf_fn)
ds = ASDFDataSet(asdf_fn)
# Add event
if quakemlfile is not None and os.path.exists(quakemlfile):
print "Event info added"
ds.add_quakeml(quakemlfile)
event = ds.events[0]
else:
raise ValueError("No Event file")
# Add waveforms.
print "Adding Waveform data"
for _i, filename in enumerate(filelist):
if os.path.exists(filename):
#print("Adding file %i of %i: %s" % (_i + 1,
# len(filelist), os.path.basename(filename)))
ds.add_waveforms(filename, tag=tag, event_id=event)
else:
print("File not exist %i of %i")
# Add StationXML files.
if staxml_filelist is not None and len(staxml_filelist) > 0:
for _i, filename in enumerate(staxml_filelist):
if os.path.exists(filename):
#print("Adding StationXML file %i of %i..." % (_i + 1, len(filenames)))
ds.add_stationxml(filename)
else:
print("No stationxml added")
def convert_to_asdf(filelist, asdf_fn, quakemlfile, staxml_filelist=None, tag=None):
"""
Convert files(sac or mseed) to asdf
"""
nfiles = len(filelist)
if nfiles == 0:
print "No file specified. Return..."
return
if os.path.exists(asdf_fn):
raise Exception("File '%s' exists." % asdf_fn)
ds = ASDFDataSet(asdf_fn)
# Add event
if quakemlfile is not None and os.path.exists(quakemlfile):
print "Event info added"
ds.add_quakeml(quakemlfile)
event = ds.events[0]
else:
raise ValueError("No Event file")
# Add waveforms.
print "Adding Waveform data"
for _i, filename in enumerate(filelist):
if os.path.exists(filename):
#print("Adding file %i of %i: %s" % (_i + 1,
# len(filelist), os.path.basename(filename)))
ds.add_waveforms(filename, tag=tag, event_id=event)
else:
print("File not exist %i of %i")
# Add StationXML files.
if staxml_filelist is not None and len(staxml_filelist) > 0:
for _i, filename in enumerate(staxml_filelist):
if os.path.exists(filename):
#print("Adding StationXML file %i of %i..." % (_i + 1, len(filenames)))
ds.add_stationxml(filename)
else:
print("No stationxml added")
import glob
import os
from pyasdf import ASDFDataSet
filename = "synthetic.h5"
if os.path.exists(filename):
raise Exception("File '%s' exists." % filename)
ds = ASDFDataSet(filename)
# Add event
ds.add_quakeml("./GCMT_event_SOUTH_SANDWICH_ISLANDS_REGION_Mag_5.6_2010-3-11-6.xml")
event = ds.events[0]
# Add waveforms.
filenames = glob.glob("./SYNTHETIC_SAC/*.sem")
for _i, filename in enumerate(filenames):
print("Adding SAC file %i of %i..." % (_i + 1, len(filenames)))
ds.add_waveforms(filename, tag="synthetic", event_id=event)
# Add StationXML files.
filenames = glob.glob("./StationXML/*.xml")
for _i, filename in enumerate(filenames):
print("Adding StationXML file %i of %i..." % (_i + 1, len(filenames)))
ds.add_stationxml(filename)
code=network,
stations=[
obspy.core.inventory.Station(
code=station,
latitude=group2[0][2],
longitude=group2[0][3],
elevation=group2[0][4],
start_date=starttime,
end_date=endtime,
creation_date=obspy.UTCDateTime(),
site=obspy.core.inventory.Site(name=""),
channels=[
obspy.core.inventory.Channel(
code=channel,
location_code=location,
latitude=latitude,
longitude=longitude,
elevation=elevation,
depth=depth,
start_date=starttime,
end_date=endtime)
])
])
],
source="pyasdf sac2asdf converter")
ds.add_stationxml(inventory)
# add sac_headers as auxiliary data
ds.add_auxiliary_data_file(dumpjson(headers), path='SacHeaders')
del ds