def test_adding_arbitrary_files(tmpdir):
"""
Tests that adding arbitrary files works.
"""
test_filename = os.path.join(tmpdir.strpath, "temp.json")
test_dict = {"a": 1, "b": 2}
with open(test_filename, "wt") as fh:
json.dump(test_dict, fh, sort_keys=True)
asdf_filename = os.path.join(tmpdir.strpath, "test.h5")
data_set = ASDFDataSet(asdf_filename)
data_set.add_auxiliary_data_file(
test_filename, tag="test_file", parameters={"1": 1})
data_set.__del__()
del data_set
new_data_set = ASDFDataSet(asdf_filename)
# Extraction works the same as always, but now has a special attribute,
# that returns the data as a BytesIO.
aux_data = new_data_set.auxiliary_data.File.test_file
assert aux_data.parameters == {"1": 1}
assert aux_data.tag == "test_file"
new_test_dict = json.loads(aux_data.file.read().decode())
assert test_dict == new_test_dict
aux_data.file.seek(0, 0)
with open(test_filename, "rb") as fh:
assert fh.read() == aux_data.file.read()
def test_get_provenance_document_for_id(tmpdir):
asdf_filename = os.path.join(tmpdir.strpath, "test.h5")
data_set = ASDFDataSet(asdf_filename)
filename = os.path.join(data_dir,
"example_schematic_processing_chain.xml")
doc = prov.read(filename)
data_set.provenance["test_provenance"] = doc
assert data_set.provenance.get_provenance_document_for_id(
'{http://seisprov.org/seis_prov/0.1/#}sp002_dt_f87sf7sf78') == \
{"name": "test_provenance", "document": doc}
assert data_set.provenance.get_provenance_document_for_id(
'{http://seisprov.org/seis_prov/0.1/#}sp004_lp_f87sf7sf78') == \
{"name": "test_provenance", "document": doc}
# Id not found.
with pytest.raises(ASDFValueError) as err:
data_set.provenance.get_provenance_document_for_id(
'{http://seisprov.org/seis_prov/0.1/#}bogus_id')
assert err.value.args[0] == (
"Document containing id "
"'{http://seisprov.org/seis_prov/0.1/#}bogus_id'"
" not found in the data set.")
# Not a qualified id.
with pytest.raises(ASDFValueError) as err:
data_set.provenance.get_provenance_document_for_id("bla")
assert err.value.args[0] == ("Not a valid qualified name.")
data_set.__del__()
def test_waveform_accessor_printing(example_data_set):
"""
Pretty printing of the waveform accessor proxy objects.
"""
data_set = ASDFDataSet(example_data_set.filename)
assert data_set.waveforms.AE_113A.__str__() == (
"Contents of the data set for station AE.113A:\n"
" - Has a StationXML file\n"
" - 1 Waveform Tag(s):\n"
" raw_recording")
data_set.__del__()
del data_set
def test_accessing_non_existent_tag_raises(example_data_set):
"""
Accessing a non-existing station should raise.
"""
data_set = ASDFDataSet(example_data_set.filename)
try:
with pytest.raises(WaveformNotInFileException) as excinfo:
data_set.waveforms.AE_113A.asdfasdf
assert excinfo.value.args[0] == ("Tag 'asdfasdf' not part of the data "
"set for station 'AE.113A'.")
finally:
data_set.__del__()
def test_empty_asdf_file_has_no_quakeml_dataset(tmpdir):
"""
There is no reason an empty ASDF file should have a QuakeML group.
"""
asdf_filename = os.path.join(tmpdir.strpath, "test.h5")
data_set = ASDFDataSet(asdf_filename)
data_set.__del__()
f = h5py.File(asdf_filename)
assert "QuakeML" not in f
# It should still return an empty catalog object if the events are
# requested.
new_data_set = ASDFDataSet(asdf_filename)
assert len(new_data_set.events) == 0
new_data_set.__del__()
def test_adding_waveforms_with_provenance_id(tmpdir):
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",
provenance_id="{http://example.org}test")
data_set.__del__()
del data_set
new_data_set = ASDFDataSet(asdf_filename)
st = new_data_set.waveforms.TA_POKR.raw_recording
for tr in st:
assert tr.stats.asdf.provenance_id == "{http://example.org}test"
new_data_set.__del__()
del new_data_set
def test_reading_and_writing_auxiliary_data_with_provenance_id(tmpdir):
asdf_filename = os.path.join(tmpdir.strpath, "test.h5")
data_set = ASDFDataSet(asdf_filename)
data = np.random.random((10, 10))
# The data must NOT start with a number.
data_type = "RandomArray"
tag = "test_data"
parameters = {"a": 1, "b": 2.0, "e": "hallo"}
provenance_id = "{http://example.org}test"
data_set.add_auxiliary_data(data=data, data_type=data_type,
tag=tag, parameters=parameters,
provenance_id=provenance_id)
data_set.__del__()
del data_set
new_data_set = ASDFDataSet(asdf_filename)
assert new_data_set.auxiliary_data.RandomArray.test_data.provenance_id \
== provenance_id
def test_adding_auxiliary_data_with_invalid_data_type_name_raises(tmpdir):
asdf_filename = os.path.join(tmpdir.strpath, "test.h5")
data_set = ASDFDataSet(asdf_filename)
data = np.random.random((10, 10))
# The data must NOT start with a number.
data_type = "2DRandomArray"
tag = "test_data"
parameters = {"a": 1, "b": 2.0, "e": "hallo"}
try:
with pytest.raises(ASDFValueError) as err:
data_set.add_auxiliary_data(data=data, data_type=data_type,
tag=tag, parameters=parameters)
assert err.value.args[0] == (
"Data type name '2DRandomArray' is invalid. It must validate "
"against the regular expression '^[A-Z][A-Za-z0-9]*$'.")
finally:
data_set.__del__()
def test_trying_to_add_provenance_record_with_invalid_name_fails(tmpdir):
"""
The name must be valid according to a particular regular expression.
"""
asdf_filename = os.path.join(tmpdir.strpath, "test.h5")
data_set = ASDFDataSet(asdf_filename)
filename = os.path.join(data_dir, "example_schematic_processing_chain.xml")
# First try adding it as a prov document.
doc = prov.read(filename, format="xml")
with pytest.raises(ASDFValueError) as err:
data_set.add_provenance_document(doc, name="a-b-c")
assert err.value.args[0] == (
"Name 'a-b-c' is invalid. It must validate against the regular "
"expression '^[0-9a-z][0-9a-z_]*[0-9a-z]$'.")
# Must sometimes be called to get around some bugs.
data_set.__del__()
def test_adding_auxiliary_data_with_wrong_tag_name_raises(tmpdir):
asdf_filename = os.path.join(tmpdir.strpath, "test.h5")
data_set = ASDFDataSet(asdf_filename)
# With provenance id.
data = np.random.random((10, 10))
# The data must NOT start with a number.
data_type = "RandomArray"
tag = "A.B.C"
with pytest.raises(ASDFValueError) as err:
data_set.add_auxiliary_data(
data=data, data_type=data_type,
tag=tag, parameters={})
assert err.value.args[0] == (
"Tag name 'A.B.C' is invalid. It must validate "
"against the regular expression "
"'^[a-zA-Z0-9][a-zA-Z0-9_]*[a-zA-Z0-9]$'.")
data_set.__del__()
def test_provenance_dicionary_behaviour(tmpdir):
asdf_filename = os.path.join(tmpdir.strpath, "test.h5")
data_set = ASDFDataSet(asdf_filename)
filename = os.path.join(data_dir,
"example_schematic_processing_chain.xml")
# Add it as a document.
doc = prov.read(filename, format="xml")
# Setting via setitem.
data_set.provenance["test_provenance"] = doc
data_set.__del__()
del data_set
new_data_set = ASDFDataSet(asdf_filename)
assert new_data_set.provenance.list() == ["test_provenance"]
assert new_data_set.provenance["test_provenance"] == doc
assert getattr(new_data_set.provenance, "test_provenance") == doc
assert list(new_data_set.provenance.keys()) == ["test_provenance"]
assert list(new_data_set.provenance.values()) == [doc]
assert list(new_data_set.provenance.items()) == [("test_provenance", doc)]