def test_data_key():
rs1_uid = insert_run_start(time=100.0, scan_id=1, owner="nedbrainard", beamline_id="example", uid=str(uuid.uuid4()))
rs2_uid = insert_run_start(time=200.0, scan_id=2, owner="nedbrainard", beamline_id="example", uid=str(uuid.uuid4()))
rs1, = find_run_starts(uid=rs1_uid)
rs2, = find_run_starts(uid=rs2_uid)
data_keys = {"fork": {"source": "_", "dtype": "number"}, "spoon": {"source": "_", "dtype": "number"}}
insert_descriptor(run_start=rs1_uid, data_keys=data_keys, time=100.0, uid=str(uuid.uuid4()))
insert_descriptor(run_start=rs2_uid, data_keys=data_keys, time=200.0, uid=str(uuid.uuid4()))
result1 = db(data_key="fork")
result2 = db(data_key="fork", start_time=150)
assert len(result1) == 2
assert len(result2) == 1
actual = result2[0]["start"]["uid"]
assert actual == str(rs2.uid)
def test_document_funcs_for_smoke():
global run_start_uid, descriptor_uid
# todo this next line will break once NSLS-II/metadatastore#142 is merged
run_start, = find_run_starts(uid=run_start_uid)
descriptors = [desc for desc in find_event_descriptors(uid=descriptor_uid)]
run_stop, = find_run_stops(uid=run_stop_uid)
documents = [run_start, run_stop]
documents.extend(descriptors)
attrs = ['__repr__', '__str__', '_repr_html_', ]
for doc, attr in product(documents, attrs):
getattr(doc, attr)()
def test_data_key():
rs1_uid = insert_run_start(time=100., scan_id=1,
owner='nedbrainard', beamline_id='example',
beamline_config=insert_beamline_config(
{}, time=0.))
rs2_uid = insert_run_start(time=200., scan_id=2,
owner='nedbrainard', beamline_id='example',
beamline_config=insert_beamline_config(
{}, time=0.))
rs1, = find_run_starts(uid=rs1_uid)
rs2, = find_run_starts(uid=rs2_uid)
data_keys = {'fork': {'source': '_', 'dtype': 'number'},
'spoon': {'source': '_', 'dtype': 'number'}}
evd1_uid = insert_event_descriptor(run_start=rs1_uid, data_keys=data_keys,
time=100.)
insert_event_descriptor(run_start=rs2_uid, data_keys=data_keys, time=200.)
result1 = db.find_headers(data_key='fork')
result2 = db.find_headers(data_key='fork', start_time=150)
assert_equal(len(result1), 2)
assert_equal(len(result2), 1)
actual = result2[0].run_start_uid
assert_equal(actual, str(rs2.uid))
def test_data_key():
rs1_uid = insert_run_start(time=100., scan_id=1,
owner='nedbrainard', beamline_id='example',
uid=str(uuid.uuid4()))
rs2_uid = insert_run_start(time=200., scan_id=2,
owner='nedbrainard', beamline_id='example',
uid=str(uuid.uuid4()))
rs1, = find_run_starts(uid=rs1_uid)
rs2, = find_run_starts(uid=rs2_uid)
data_keys = {'fork': {'source': '_', 'dtype': 'number'},
'spoon': {'source': '_', 'dtype': 'number'}}
insert_descriptor(run_start=rs1_uid, data_keys=data_keys,
time=100.,
uid=str(uuid.uuid4()))
insert_descriptor(run_start=rs2_uid, data_keys=data_keys, time=200.,
uid=str(uuid.uuid4()))
result1 = db(data_key='fork')
result2 = db(data_key='fork', start_time=150)
assert len(result1) == 2
assert len(result2) == 1
actual = result2[0]['start']['uid']
assert actual == str(rs2.uid)
def __getitem__(cls, key):
if isinstance(key, slice):
# Interpret key as a slice into previous scans.
if key.start is not None and key.start > -1:
raise ValueError("Slices must be negative. The most recent "
"run is referred to as -1.")
if key.stop is not None and key.stop > -1:
raise ValueError("Slices must be negative. The most recent "
"run is referred to as -1.")
if key.stop is not None:
stop = -key.stop
else:
stop = None
if key.start is None:
raise ValueError("Cannot slice infinitely into the past; "
"the result could become too large.")
start = -key.start
result = list(find_last(start))[stop::key.step]
header = Headers([Header.from_run_start(h) for h in result])
elif isinstance(key, int):
if key > -1:
# Interpret key as a scan_id.
gen = find_run_starts(scan_id=key)
try:
result = next(gen) # most recent match
except StopIteration:
raise ValueError("No such run found.")
header = Header.from_run_start(result)
else:
# Interpret key as the Nth last scan.
gen = find_last(-key)
for i in range(-key):
try:
result = next(gen)
except StopIteration:
raise IndexError(
"There are only {0} runs.".format(i))
header = Header.from_run_start(result)
elif isinstance(key, six.string_types):
# Interpret key as a uid (or the few several characters of one).
# First try searching as if we have the full uid.
results = list(find_run_starts(uid=key))
if len(results) == 0:
# No dice? Try searching as if we have a partial uid.
gen = find_run_starts(uid={'$regex': '{0}.*'.format(key)})
results = list(gen)
if len(results) < 1:
raise ValueError("No such run found.")
if len(results) > 1:
raise ValueError("That partial uid matches multiple runs. "
"Provide more characters.")
result, = results
header = Header.from_run_start(result)
elif isinstance(key, Iterable):
# Interpret key as a list of several keys. If it is a string
# we will never get this far.
return Headers([cls.__getitem__(k) for k in key])
else:
raise ValueError("Must give an integer scan ID like [6], a slice "
"into past scans like [-5], [-5:], or [-5:-9:2], "
"a list like [1, 7, 13], or a (partial) uid "
"like ['a23jslk'].")
return header
def find_headers(cls, **kwargs):
"""Given search criteria, find Headers describing runs.
This function returns a list of dictionary-like objects encapsulating
the metadata for a run -- start time, instruments uses, and so on.
In addition to the Parameters below, advanced users can specifiy
arbitrary queries that are passed through to mongodb.
Parameters
----------
start_time : time-like, optional
Include Headers for runs started after this time. Valid
"time-like" representations are:
- float timestamps (seconds since 1970), such as time.time()
- '2015'
- '2015-01'
- '2015-01-30'
- '2015-03-30 03:00:00'
- Python datetime objects, such as datetime.datetime.now()
stop_time: time-like, optional
Include Headers for runs started before this time. See
`start_time` above for examples.
beamline_id : str, optional
String identifier for a specific beamline
project : str, optional
Project name
owner : str, optional
The username of the logged-in user when the scan was performed
scan_id : int, optional
Integer scan identifier
uid : str, optional
Globally unique id string provided to metadatastore
_id : str or ObjectId, optional
The unique id generated by mongo
data_key : str, optional
The alias (e.g., 'motor1') or PV identifier of data source
Returns
-------
data : list
Header objects
Examples
--------
>>> find_headers(start_time='2015-03-05', stop_time='2015-03-10')
>>> find_headers(data_key='motor1')
>>> find_headers(data_key='motor1', start_time='2015-03-05')
"""
data_key = kwargs.pop('data_key', None)
run_start = find_run_starts(**kwargs)
if data_key is not None:
node_name = 'data_keys.{0}'.format(data_key)
query = {node_name: {'$exists': True}}
descriptors = []
for rs in run_start:
descriptor = find_event_descriptors(run_start=rs, **query)
for d in descriptor:
descriptors.append(d)
# query = {node_name: {'$exists': True},
# 'run_start_id': {'$in': [ObjectId(rs.id) for rs in run_start]}}
# descriptors = find_event_descriptors(**query)
result = []
known_uids = deque()
for descriptor in descriptors:
if descriptor.run_start.uid not in known_uids:
rs = descriptor.run_start
known_uids.append(rs.uid)
result.append(rs)
run_start = result
result = []
for rs in run_start:
result.append(Header.from_run_start(rs))
return Headers(result)