def _create(cls, lims, creation_tag=None, udfs=None, **kwargs):
"""Create an instance from attributes and return it"""
if not udfs:
udfs = {}
instance = cls(lims, _create_new=True)
if creation_tag:
instance.root = ElementTree.Element(
nsmap(cls._PREFIX + ':' + creation_tag))
elif cls._TAG:
instance.root = ElementTree.Element(
nsmap(cls._PREFIX + ':' + cls._TAG))
else:
instance.root = ElementTree.Element(
nsmap(cls._PREFIX + ':' + cls.__name__.lower()))
for key in udfs:
instance.udf[key] = udfs[key]
for attribute in kwargs:
if hasattr(instance, attribute):
setattr(instance, attribute, kwargs.get(attribute))
else:
raise TypeError(
"%s create: got an unexpected keyword argument '%s'" %
(cls.__name__, attribute))
return instance
def _update_elems(self):
self._elems = []
if self._udt:
elem = self.rootnode.find(nsmap('udf:type'))
if elem is not None:
self._udt = elem.attrib['name']
self._elems = elem.findall(nsmap('udf:field'))
else:
tag = nsmap('udf:field')
for elem in self.rootnode.getchildren():
if elem.tag == tag:
self._elems.append(elem)
def _update_elems(self):
self._elems = []
if self._udt:
elem = self.rootnode.find(nsmap('udf:type'))
if elem is not None:
self._udt = elem.attrib['name']
self._elems = elem.findall(nsmap('udf:field'))
else:
tag = nsmap('udf:field')
for elem in self.rootnode.getchildren():
if elem.tag == tag:
self._elems.append(elem)
def _create(cls, lims, creation_tag=None, **kwargs):
"""Create an instance from attributes and return it"""
instance = cls(lims, _create_new=True)
if creation_tag:
instance.root = ElementTree.Element(nsmap(cls._PREFIX + ':' + creation_tag))
elif cls._TAG:
instance.root = ElementTree.Element(nsmap(cls._PREFIX + ':' + cls._TAG))
else:
instance.root = ElementTree.Element(nsmap(cls._PREFIX + ':' + cls.__name__.lower()))
for attribute in kwargs:
if hasattr(instance, attribute):
setattr(instance, attribute, kwargs.get(attribute))
else:
raise TypeError("%s create: got an unexpected keyword argument '%s'" % (cls.__name__, attribute))
return instance
def create(cls, lims, **kwargs):
"""Create an instance from attributes then post it to the LIMS"""
instance = cls(lims, _create_new=True)
if cls._TAG:
instance.root = ElementTree.Element(nsmap(cls._PREFIX + ":" + cls._TAG))
else:
instance.root = ElementTree.Element(nsmap(cls._PREFIX + ":" + cls.__name__.lower()))
for attribute in kwargs:
if hasattr(instance, attribute):
setattr(instance, attribute, kwargs.get(attribute))
else:
raise TypeError("%s create: got an unexpected keyword argument '%s'" % (cls.__name__, attribute))
data = lims.tostring(ElementTree.ElementTree(instance.root))
instance.root = lims.post(uri=lims.get_uri(cls._URI), data=data)
instance._uri = instance.root.attrib["uri"]
return instance
def set_udt(self, name):
assert isinstance(name, str)
if not self._udt:
raise AttributeError('cannot set name for a UDF dictionary')
self._udt = name
elem = self.rootnode.find(nsmap('udf:type'))
assert elem is not None
elem.set('name', name)
def set_udt(self, name):
assert isinstance(name, str)
if not self._udt:
raise AttributeError('cannot set name for a UDF dictionary')
self._udt = name
elem = self.rootnode.find(nsmap('udf:type'))
assert elem is not None
elem.set('name', name)
class Sample(Entity):
"Customer's sample to be analyzed; associated with a project."
_URI = 'samples'
_PREFIX = 'smp'
name = StringDescriptor('name')
date_received = StringDescriptor('date-received')
date_completed = StringDescriptor('date-completed')
project = EntityDescriptor('project', Project)
submitter = EntityDescriptor('submitter', Researcher)
# artifact: defined below
udf = UdfDictionaryDescriptor()
udt = UdtDictionaryDescriptor()
notes = EntityListDescriptor('note', Note)
files = EntityListDescriptor(nsmap('file:file'), File)
externalids = ExternalidListDescriptor()
# biosource XXX
@classmethod
def create(cls, lims, container, position, **kwargs):
"""Create an instance of Sample from attributes then post it to the LIMS
Udfs can be sent in with the kwarg `udfs`. It should be a dictionary-like.
"""
if not isinstance(container, Container):
raise TypeError('%s is not of type Container' % container)
udfs = kwargs.pop("udfs", None)
instance = super(Sample, cls)._create(lims,
creation_tag='samplecreation',
**kwargs)
location = ElementTree.SubElement(instance.root, 'location')
ElementTree.SubElement(location, 'container', dict(uri=container.uri))
position_element = ElementTree.SubElement(location, 'value')
position_element.text = position
# NOTE: This is a quick fix. I assume that it must be possible to initialize samples
# with UDFs
for key, value in udfs.items():
attrib = {
"name": key,
"xmlns:udf": "http://genologics.com/ri/userdefined",
}
udf = ElementTree.SubElement(instance.root,
'udf:field',
attrib=attrib)
udf.text = value
data = lims.tostring(ElementTree.ElementTree(instance.root))
instance.root = lims.post(uri=lims.get_uri(cls._URI), data=data)
instance._uri = instance.root.attrib['uri']
return instance
class Project(Entity):
"Project concerning a number of samples; associated with a researcher."
_URI = 'projects'
_PREFIX = 'prj'
name = StringDescriptor('name')
open_date = StringDescriptor('open-date')
close_date = StringDescriptor('close-date')
invoice_date = StringDescriptor('invoice-date')
researcher = EntityDescriptor('researcher', Researcher)
udf = UdfDictionaryDescriptor()
udt = UdtDictionaryDescriptor()
files = EntityListDescriptor(nsmap('file:file'), File)
externalids = ExternalidListDescriptor()
class ProcessTypeParameter(object):
instance = None
name = None
root = None
tag = 'parameter'
string = StringDescriptor('string')
run_program_per_event = StringDescriptor('run-program-per-event')
channel = StringDescriptor('channel')
invocation_type = StringDescriptor('invocation-type')
file = EntityListDescriptor(nsmap('file:file'), File)
def __init__(self, pt_instance, node):
self.instance = pt_instance
self.root = node
self.name = self.root.attrib['name']
def get(self):
pass
class Sample(Entity):
"Customer's sample to be analyzed; associated with a project."
_URI = 'samples'
_TAG = 'sample'
_PREFIX = 'smp'
name = StringDescriptor('name')
date_received = StringDescriptor('date-received')
date_completed = StringDescriptor('date-completed')
project = EntityDescriptor('project', Project)
submitter = EntityDescriptor('submitter', Researcher)
# artifact: defined below
udf = UdfDictionaryDescriptor()
udt = UdtDictionaryDescriptor()
notes = EntityListDescriptor('note', Note)
files = EntityListDescriptor(nsmap('file:file'), File)
externalids = ExternalidListDescriptor()
# biosource XXX
control_type = EntityDescriptor('control-type', Controltype)
@classmethod
def create(cls, lims, container, position, udfs=None, **kwargs):
"""Create an instance of Sample from attributes then post it to the LIMS"""
if udfs is None:
udfs = {}
if not isinstance(container, Container):
raise TypeError('%s is not of type Container' % container)
instance = super(Sample, cls)._create(lims,
creation_tag='samplecreation',
udfs=udfs,
**kwargs)
location = ElementTree.SubElement(instance.root, 'location')
ElementTree.SubElement(location, 'container', dict(uri=container.uri))
position_element = ElementTree.SubElement(location, 'value')
position_element.text = position
data = lims.tostring(ElementTree.ElementTree(instance.root))
instance.root = lims.post(uri=lims.get_uri(cls._URI), data=data)
instance._uri = instance.root.attrib['uri']
return instance
class Artifact(Entity):
"Any process input or output; analyte or file."
_URI = 'artifacts'
_TAG = 'artifact'
_PREFIX = 'art'
name = StringDescriptor('name')
type = StringDescriptor('type')
output_type = StringDescriptor('output-type')
parent_process = EntityDescriptor('parent-process', Process)
volume = StringDescriptor('volume')
concentration = StringDescriptor('concentration')
qc_flag = StringDescriptor('qc-flag')
location = LocationDescriptor('location')
working_flag = BooleanDescriptor('working-flag')
samples = EntityListDescriptor('sample', Sample)
udf = UdfDictionaryDescriptor()
files = EntityListDescriptor(nsmap('file:file'), File)
reagent_labels = ReagentLabelList()
# artifact_flags XXX
# artifact_groups XXX
def input_artifact_list(self):
"""Returns the input artifact ids of the parrent process."""
input_artifact_list = []
try:
for tuple in self.parent_process.input_output_maps:
if tuple[1]['limsid'] == self.id:
input_artifact_list.append(tuple[0]['uri']) # ['limsid'])
except:
pass
return input_artifact_list
def get_state(self):
"Parse out the state value from the URI."
parts = urlparse(self.uri)
params = parse_qs(parts.query)
try:
return params['state'][0]
except (KeyError, IndexError):
return None
@property
def container(self):
"The container where the artifact is located, or None"
try:
return self.location[0]
except:
return None
def stateless(self):
"returns the artefact independently of it's state"
parts = urlparse(self.uri)
if 'state' in parts[4]:
stateless_uri = urlunparse(
[parts[0], parts[1], parts[2], parts[3], '', ''])
return Artifact(self.lims, uri=stateless_uri)
else:
return self
# XXX set_state ?
state = property(get_state)
stateless = property(stateless)
def _get_workflow_stages_and_statuses(self):
self.get()
result = []
rootnode = self.root.find('workflow-stages')
for node in rootnode.findall('workflow-stage'):
result.append((Stage(self.lims, uri=node.attrib['uri']),
node.attrib['status'], node.attrib['name']))
return result
workflow_stages_and_statuses = property(_get_workflow_stages_and_statuses)
class Process(Entity):
"Process (instance of Processtype) executed producing ouputs from inputs."
_URI = 'processes'
_PREFIX = 'prc'
type = EntityDescriptor('type', Processtype)
date_run = StringDescriptor('date-run')
technician = EntityDescriptor('technician', Researcher)
protocol_name = StringDescriptor('protocol-name')
input_output_maps = InputOutputMapList()
udf = UdfDictionaryDescriptor()
udt = UdtDictionaryDescriptor()
files = EntityListDescriptor(nsmap('file:file'), File)
process_parameter = StringDescriptor('process-parameter')
instrument = EntityDescriptor('instrument', Instrument)
# process_parameters XXX
def outputs_per_input(self,
inart,
ResultFile=False,
SharedResultFile=False,
Analyte=False):
"""Getting all the output artifacts related to a particual input artifact"""
inouts = [
io for io in self.input_output_maps if io[0]['limsid'] == inart
]
if ResultFile:
inouts = [
io for io in inouts if io[1]['output-type'] == 'ResultFile'
]
elif SharedResultFile:
inouts = [
io for io in inouts
if io[1]['output-type'] == 'SharedResultFile'
]
elif Analyte:
inouts = [io for io in inouts if io[1]['output-type'] == 'Analyte']
outs = [io[1]['uri'] for io in inouts]
return outs
def input_per_sample(self, sample):
"""gettiung all the input artifacts dereved from the specifyed sample"""
ins_all = self.all_inputs()
ins = []
for inp in ins_all:
for samp in inp.samples:
if samp.name == sample and inp not in ins:
ins.append(inp)
return ins
def all_inputs(self, unique=True, resolve=False):
"""Retrieving all input artifacts from input_output_maps
if unique is true, no duplicates are returned.
"""
# if the process has no input, that is not standard and we want to know about it
try:
ids = [io[0]['limsid'] for io in self.input_output_maps]
except TypeError:
logger.error("Process ", self, " has no input artifacts")
raise TypeError
if unique:
ids = list(frozenset(ids))
if resolve:
return self.lims.get_batch(
[Artifact(self.lims, id=id) for id in ids if id is not None])
else:
return [Artifact(self.lims, id=id) for id in ids if id is not None]
def all_outputs(self, unique=True, resolve=False):
"""Retrieving all output artifacts from input_output_maps
if unique is true, no duplicates are returned.
"""
# Given how ids is structured, io[1] might be None : some process don't have an output.
ids = [
io[1]['limsid'] for io in self.input_output_maps
if io[1] is not None
]
if unique:
ids = list(frozenset(ids))
if resolve:
return self.lims.get_batch(
[Artifact(self.lims, id=id) for id in ids if id is not None])
else:
return [Artifact(self.lims, id=id) for id in ids if id is not None]
def shared_result_files(self):
"""Retreve all resultfiles of output-generation-type PerAllInputs."""
artifacts = self.all_outputs(unique=True)
return [a for a in artifacts if a.output_type == 'SharedResultFile']
def result_files(self):
"""Retreve all resultfiles of output-generation-type perInput."""
artifacts = self.all_outputs(unique=True)
return [a for a in artifacts if a.output_type == 'ResultFile']
def analytes(self):
"""Retreving the output Analytes of the process, if existing.
If the process is not producing any output analytes, the input
analytes are returned. Input/Output is returned as a information string.
Makes aggregate processes and normal processes look the same."""
info = 'Output'
artifacts = self.all_outputs(unique=True)
analytes = [a for a in artifacts if a.type == 'Analyte']
if len(analytes) == 0:
artifacts = self.all_inputs(unique=True)
analytes = [a for a in artifacts if a.type == 'Analyte']
info = 'Input'
return analytes, info
def parent_processes(self):
"""Retrieving all parent processes through the input artifacts"""
return [i_a.parent_process for i_a in self.all_inputs(unique=True)]
def output_containers(self):
"""Retrieve all unique output containers"""
cs = []
for o_a in self.all_outputs(unique=True):
if o_a.container:
cs.append(o_a.container)
return list(frozenset(cs))
@property
def step(self):
"""Retrive the Step coresponding to this process. They share the same id"""
return Step(self.lims, id=self.id)
def __get__(self, instance, cls):
instance.get()
result = []
for node in instance.root.findall(nsmap('ri:externalid')):
result.append((node.attrib.get('id'), node.attrib.get('uri')))
return result
def __setitem__(self, key, value):
self._lookup[key] = value
for node in self._elems:
if node.attrib['name'] != key: continue
vtype = node.attrib['type'].lower()
if value is None:
pass
elif vtype == 'string':
if not self._is_string(value):
raise TypeError('String UDF requires str or unicode value')
elif vtype == 'str':
if not self._is_string(value):
raise TypeError('String UDF requires str or unicode value')
elif vtype == 'text':
if not self._is_string(value):
raise TypeError('Text UDF requires str or unicode value')
elif vtype == 'numeric':
if not isinstance(value, (int, float, Decimal)):
raise TypeError('Numeric UDF requires int or float value')
value = str(value)
elif vtype == 'boolean':
if not isinstance(value, bool):
raise TypeError('Boolean UDF requires bool value')
value = value and 'true' or 'false'
elif vtype == 'date':
if not isinstance(value, datetime.date): # Too restrictive?
raise TypeError('Date UDF requires datetime.date value')
value = str(value)
elif vtype == 'uri':
if not self._is_string(value):
raise TypeError(
'URI UDF requires str or punycode (unicode) value')
value = str(value)
else:
raise NotImplemented("UDF type '%s'" % vtype)
if not isinstance(value, str):
if not self._is_string(value):
value = str(value).encode('UTF-8')
node.text = value
break
else: # Create new entry; heuristics for type
if self._is_string(value):
vtype = '\n' in value and 'Text' or 'String'
elif isinstance(value, bool):
vtype = 'Boolean'
value = value and 'true' or 'false'
elif isinstance(value, (int, float, Decimal)):
vtype = 'Numeric'
value = str(value)
elif isinstance(value, datetime.date):
vtype = 'Date'
value = str(value)
else:
raise NotImplementedError("Cannot handle value of type '%s'"
" for UDF" % type(value))
if self._udt:
root = self.rootnode.find(nsmap('udf:type'))
else:
root = self.rootnode
elem = ElementTree.SubElement(root,
nsmap('udf:field'),
type=vtype,
name=key)
if not isinstance(value, str):
if not self._is_string(value):
value = str(value).encode('UTF-8')
elem.text = value
#update the internal elements and lookup with new values
self._update_elems()
self._prepare_lookup()
def __get__(self, instance, cls):
instance.get()
result = []
for node in instance.root.findall(nsmap('ri:externalid')):
result.append((node.attrib.get('id'), node.attrib.get('uri')))
return result
def __setitem__(self, key, value):
self._lookup[key] = value
for node in self._elems:
if node.attrib['name'] != key: continue
vtype = node.attrib['type'].lower()
if value is None:
pass
elif vtype == 'string':
if not self._is_string(value):
raise TypeError('String UDF requires str or unicode value')
elif vtype == 'str':
if not self._is_string(value):
raise TypeError('String UDF requires str or unicode value')
elif vtype == 'text':
if not self._is_string(value):
raise TypeError('Text UDF requires str or unicode value')
elif vtype == 'numeric':
if not isinstance(value, (int, float, Decimal)):
raise TypeError('Numeric UDF requires int or float value')
value = str(value)
elif vtype == 'boolean':
if not isinstance(value, bool):
raise TypeError('Boolean UDF requires bool value')
value = value and 'true' or 'false'
elif vtype == 'date':
if not isinstance(value, datetime.date): # Too restrictive?
raise TypeError('Date UDF requires datetime.date value')
value = str(value)
elif vtype == 'uri':
if not self._is_string(value):
raise TypeError('URI UDF requires str or punycode (unicode) value')
value = str(value)
else:
raise NotImplemented("UDF type '%s'" % vtype)
if not isinstance(value, str):
if not self._is_string(value):
value = str(value).encode('UTF-8')
node.text = value
break
else: # Create new entry; heuristics for type
if self._is_string(value):
vtype = '\n' in value and 'Text' or 'String'
elif isinstance(value, bool):
vtype = 'Boolean'
value = value and 'true' or 'false'
elif isinstance(value, (int, float, Decimal)):
vtype = 'Numeric'
value = str(value)
elif isinstance(value, datetime.date):
vtype = 'Date'
value = str(value)
else:
raise NotImplementedError("Cannot handle value of type '%s'"
" for UDF" % type(value))
if self._udt:
root = self.rootnode.find(nsmap('udf:type'))
else:
root = self.rootnode
elem = ElementTree.SubElement(root,
nsmap('udf:field'),
type=vtype,
name=key)
if not isinstance(value, str):
if not self._is_string(value):
value = str(value).encode('UTF-8')
elem.text = value
#update the internal elements and lookup with new values
self._update_elems()
self._prepare_lookup()
