def __init__(self, tool, param):
self.tool = tool
self.name = param.get("name")
self.label = util.xml_text(param, "label")
self.help = util.xml_text(param, "help")
self.html = "no html set"
self.validators = []
for elem in param.findall("validator"):
self.validators.append(validation.Validator.from_element(elem))
def __init__( self, tool, param ):
self.tool = tool
self.name = param.get("name")
self.label = util.xml_text(param, "label")
self.help = util.xml_text(param, "help")
self.html = "no html set"
self.validators = []
for elem in param.findall("validator"):
self.validators.append( validation.Validator.from_element( elem ) )
def parse_requirements_from_xml(xml_root):
"""
>>> from xml.etree import ElementTree
>>> def load_requirements( contents ):
... contents_document = '''<tool><requirements>%s</requirements></tool>'''
... root = ElementTree.fromstring( contents_document % contents )
... return parse_requirements_from_xml( root )
>>> reqs, containers = load_requirements('''<requirement>bwa</requirement>''')
>>> reqs[0].name
'bwa'
>>> reqs[0].version is None
True
>>> reqs[0].type
'package'
>>> reqs, containers = load_requirements('''<requirement type="binary" version="1.3.3">cufflinks</requirement>''')
>>> reqs[0].name
'cufflinks'
>>> reqs[0].version
'1.3.3'
>>> reqs[0].type
'binary'
"""
requirements_elem = xml_root.find("requirements")
requirement_elems = []
if requirements_elem is not None:
requirement_elems = requirements_elem.findall('requirement')
requirements = []
for requirement_elem in requirement_elems:
name = xml_text(requirement_elem)
type = requirement_elem.get("type", DEFAULT_REQUIREMENT_TYPE)
version = requirement_elem.get("version", DEFAULT_REQUIREMENT_VERSION)
requirement = ToolRequirement(name=name, type=type, version=version)
requirements.append(requirement)
container_elems = []
if requirements_elem is not None:
container_elems = requirements_elem.findall('container')
containers = []
for container_elem in container_elems:
identifier = xml_text(container_elem)
type = container_elem.get("type", DEFAULT_CONTAINER_TYPE)
container = ContainerDescription(identifier=identifier, type=type)
containers.append(container)
return requirements, containers
def parse_requirements_from_xml( xml_root ):
"""
>>> from xml.etree import ElementTree
>>> def load_requirements( contents ):
... contents_document = '''<tool><requirements>%s</requirements></tool>'''
... root = ElementTree.fromstring( contents_document % contents )
... return parse_requirements_from_xml( root )
>>> reqs, containers = load_requirements('''<requirement>bwa</requirement>''')
>>> reqs[0].name
'bwa'
>>> reqs[0].version is None
True
>>> reqs[0].type
'package'
>>> reqs, containers = load_requirements('''<requirement type="binary" version="1.3.3">cufflinks</requirement>''')
>>> reqs[0].name
'cufflinks'
>>> reqs[0].version
'1.3.3'
>>> reqs[0].type
'binary'
"""
requirements_elem = xml_root.find( "requirements" )
requirement_elems = []
if requirements_elem is not None:
requirement_elems = requirements_elem.findall( 'requirement' )
requirements = []
for requirement_elem in requirement_elems:
name = xml_text( requirement_elem )
type = requirement_elem.get( "type", DEFAULT_REQUIREMENT_TYPE )
version = requirement_elem.get( "version", DEFAULT_REQUIREMENT_VERSION )
requirement = ToolRequirement( name=name, type=type, version=version )
requirements.append( requirement )
container_elems = []
if requirements_elem is not None:
container_elems = requirements_elem.findall( 'container' )
containers = []
for container_elem in container_elems:
identifier = xml_text( container_elem )
type = container_elem.get( "type", DEFAULT_CONTAINER_TYPE )
container = ContainerDescription( identifier=identifier, type=type )
containers.append( container )
return requirements, containers
def _parse_output(
self,
data_elem,
tool,
default_format="data",
default_format_source=None,
default_metadata_source="",
):
output = ToolOutput(data_elem.get("name"))
output_format = data_elem.get("format", default_format)
auto_format = string_as_bool(data_elem.get("auto_format", "false"))
if auto_format and output_format != "data":
raise ValueError("Setting format and auto_format is not supported at this time.")
elif auto_format:
output_format = "_sniff_"
output.format = output_format
output.change_format = data_elem.findall("change_format")
output.format_source = data_elem.get("format_source", default_format_source)
output.default_identifier_source = data_elem.get("default_identifier_source", 'None')
output.metadata_source = data_elem.get("metadata_source", default_metadata_source)
output.parent = data_elem.get("parent", None)
output.label = xml_text(data_elem, "label")
output.count = int(data_elem.get("count", 1))
output.filters = data_elem.findall('filter')
output.tool = tool
output.from_work_dir = data_elem.get("from_work_dir", None)
output.hidden = string_as_bool(data_elem.get("hidden", ""))
output.actions = ToolOutputActionGroup(output, data_elem.find('actions'))
output.dataset_collector_descriptions = dataset_collector_descriptions_from_elem(data_elem, legacy=self.legacy_defaults)
return output
def parse_column_spec(self, config_element):
"""
Parse column definitions, which can either be a set of 'column' elements
with a name and index (as in dynamic options config), or a shorthand
comma separated list of names in order as the text of a 'column_names'
element.
A column named 'value' is required.
"""
self.columns = {}
if config_element.find('columns') is not None:
column_names = util.xml_text(config_element.find('columns'))
column_names = [n.strip() for n in column_names.split(',')]
for index, name in enumerate(column_names):
self.columns[name] = index
self.largest_index = index
else:
self.largest_index = 0
for column_elem in config_element.findall('column'):
name = column_elem.get('name', None)
assert name is not None, "Required 'name' attribute missing from column def"
index = column_elem.get('index', None)
assert index is not None, "Required 'index' attribute missing from column def"
index = int(index)
self.columns[name] = index
if index > self.largest_index:
self.largest_index = index
empty_field_value = column_elem.get('empty_field_value', None)
if empty_field_value is not None:
self.empty_field_values[name] = empty_field_value
assert 'value' in self.columns, "Required 'value' column missing from column def"
if 'name' not in self.columns:
self.columns['name'] = self.columns['value']
def _parse_collection(collection_elem):
name = collection_elem.get("name")
label = xml_text(collection_elem, "label")
default_format = collection_elem.get("format", "data")
collection_type = collection_elem.get("type", None)
collection_type_source = collection_elem.get("type_source", None)
collection_type_from_rules = collection_elem.get(
"type_from_rules", None)
structured_like = collection_elem.get("structured_like", None)
inherit_format = False
inherit_metadata = False
if structured_like:
inherit_format = string_as_bool(
collection_elem.get("inherit_format", None))
inherit_metadata = string_as_bool(
collection_elem.get("inherit_metadata", None))
default_format_source = collection_elem.get("format_source", None)
default_metadata_source = collection_elem.get(
"metadata_source", "")
filters = collection_elem.findall('filter')
dataset_collector_descriptions = None
if collection_elem.find("discover_datasets") is not None:
dataset_collector_descriptions = dataset_collector_descriptions_from_elem(
collection_elem, legacy=False)
structure = ToolOutputCollectionStructure(
collection_type=collection_type,
collection_type_source=collection_type_source,
collection_type_from_rules=collection_type_from_rules,
structured_like=structured_like,
dataset_collector_descriptions=dataset_collector_descriptions,
)
output_collection = ToolOutputCollection(
name,
structure,
label=label,
filters=filters,
default_format=default_format,
inherit_format=inherit_format,
inherit_metadata=inherit_metadata,
default_format_source=default_format_source,
default_metadata_source=default_metadata_source,
)
outputs[output_collection.name] = output_collection
for data_elem in collection_elem.findall("data"):
_parse(
data_elem,
default_format=default_format,
default_format_source=default_format_source,
default_metadata_source=default_metadata_source,
)
for data_elem in collection_elem.findall("data"):
output_name = data_elem.get("name")
data = data_dict[output_name]
assert data
del data_dict[output_name]
output_collection.outputs[output_name] = data
output_collections[name] = output_collection
def _parse_output(
self,
data_elem,
tool,
default_format="data",
default_format_source=None,
default_metadata_source="",
):
output = ToolOutput( data_elem.get("name") )
output_format = data_elem.get("format", default_format)
auto_format = string_as_bool( data_elem.get( "auto_format", "false" ) )
if auto_format and output_format != "data":
raise ValueError("Setting format and auto_format is not supported at this time.")
elif auto_format:
output_format = "_sniff_"
output.format = output_format
output.change_format = data_elem.findall("change_format")
output.format_source = data_elem.get("format_source", default_format_source)
output.metadata_source = data_elem.get("metadata_source", default_metadata_source)
output.parent = data_elem.get("parent", None)
output.label = xml_text( data_elem, "label" )
output.count = int( data_elem.get("count", 1) )
output.filters = data_elem.findall( 'filter' )
output.tool = tool
output.from_work_dir = data_elem.get("from_work_dir", None)
output.hidden = string_as_bool( data_elem.get("hidden", "") )
output.actions = ToolOutputActionGroup( output, data_elem.find( 'actions' ) )
output.dataset_collector_descriptions = dataset_collector_descriptions_from_elem( data_elem, legacy=self.legacy_defaults )
return output
def parse_column_spec(self, config_element):
"""
Parse column definitions, which can either be a set of 'column' elements
with a name and index (as in dynamic options config), or a shorthand
comma separated list of names in order as the text of a 'column_names'
element.
A column named 'value' is required.
"""
self.columns = {}
if config_element.find('columns') is not None:
column_names = util.xml_text(config_element.find('columns'))
column_names = [n.strip() for n in column_names.split(',')]
for index, name in enumerate(column_names):
self.columns[name] = index
self.largest_index = index
else:
self.largest_index = 0
for column_elem in config_element.findall('column'):
name = column_elem.get('name', None)
assert name is not None, "Required 'name' attribute missing from column def"
index = column_elem.get('index', None)
assert index is not None, "Required 'index' attribute missing from column def"
index = int(index)
self.columns[name] = index
if index > self.largest_index:
self.largest_index = index
empty_field_value = column_elem.get('empty_field_value', None)
if empty_field_value is not None:
self.empty_field_values[name] = empty_field_value
assert 'value' in self.columns, "Required 'value' column missing from column def"
if 'name' not in self.columns:
self.columns['name'] = self.columns['value']
def parse( self, root ):
# Get the name
self.name = root.get( "name" )
if not self.name:
raise Exception, "Missing external_service_type 'name'"
# Get the UNIQUE id for the tool
self.id = root.get( "id" )
if not self.id:
raise Exception, "Missing external_service_type 'id'"
self.config_version = root.get( "version" )
if not self.config_version:
self.config_version = '1.0.0'
self.description = util.xml_text(root, "description")
self.version = util.xml_text( root.find( "version" ) )
# parse the form
self.form_definition = form_factory.from_elem( root.find( 'form' ) )
self.parse_data_transfer_settings( root )
self.parse_run_details( root )
#external services actions
self.actions = ExternalServiceActionsGroup.from_elem( root.find( 'actions' ), parent=self )
def parse( self, root ):
# Get the name
self.name = root.get( "name" )
if not self.name:
raise Exception( "Missing external_service_type 'name'" )
# Get the UNIQUE id for the tool
self.id = root.get( "id" )
if not self.id:
raise Exception( "Missing external_service_type 'id'" )
self.config_version = root.get( "version" )
if not self.config_version:
self.config_version = '1.0.0'
self.description = util.xml_text(root, "description")
self.version = util.xml_text( root.find( "version" ) )
# parse the form
self.form_definition = form_factory.from_elem( root.find( 'form' ) )
self.parse_data_transfer_settings( root )
self.parse_run_details( root )
# external services actions
self.actions = ExternalServiceActionsGroup.from_elem( root.find( 'actions' ), parent=self )
def container_from_element(container_elem):
identifier = xml_text(container_elem)
type = container_elem.get("type", DEFAULT_CONTAINER_TYPE)
resolve_dependencies = asbool(container_elem.get("resolve_dependencies", DEFAULT_CONTAINER_RESOLVE_DEPENDENCIES))
shell = container_elem.get("shell", DEFAULT_CONTAINER_SHELL)
container = ContainerDescription(
identifier=identifier,
type=type,
resolve_dependencies=resolve_dependencies,
shell=shell,
)
return container
def container_from_element(container_elem):
identifier = xml_text(container_elem)
type = container_elem.get("type", DEFAULT_CONTAINER_TYPE)
resolve_dependencies = asbool(container_elem.get("resolve_dependencies", DEFAULT_CONTAINER_RESOLVE_DEPENDENCIES))
shell = container_elem.get("shell", DEFAULT_CONTAINER_SHELL)
container = ContainerDescription(
identifier=identifier,
type=type,
resolve_dependencies=resolve_dependencies,
shell=shell,
)
return container
def _parse_expression_output(self, output_elem, tool, **kwds):
output_type = output_elem.get("type")
from_expression = output_elem.get("from")
output = ToolExpressionOutput(
output_elem.get("name"),
output_type,
from_expression,
)
output.path = output_elem.get("value")
output.label = xml_text(output_elem, "label")
output.hidden = string_as_bool(output_elem.get("hidden", ""))
output.actions = ToolOutputActionGroup(output, output_elem.find('actions'))
output.dataset_collector_descriptions = []
return output
def parse_requirements_from_xml(xml_root):
"""
>>> from galaxy.util import parse_xml_string
>>> def load_requirements(contents):
... contents_document = '''<tool><requirements>%s</requirements></tool>'''
... root = parse_xml_string(contents_document % contents)
... return parse_requirements_from_xml(root)
>>> reqs, containers = load_requirements('''<requirement>bwa</requirement>''')
>>> reqs[0].name
'bwa'
>>> reqs[0].version is None
True
>>> reqs[0].type
'package'
>>> reqs, containers = load_requirements('''<requirement type="binary" version="1.3.3">cufflinks</requirement>''')
>>> reqs[0].name
'cufflinks'
>>> reqs[0].version
'1.3.3'
>>> reqs[0].type
'binary'
"""
requirements_elem = xml_root.find("requirements")
requirement_elems = []
if requirements_elem is not None:
requirement_elems = requirements_elem.findall('requirement')
requirements = ToolRequirements()
for requirement_elem in requirement_elems:
name = xml_text(requirement_elem)
type = requirement_elem.get("type", DEFAULT_REQUIREMENT_TYPE)
version = requirement_elem.get("version", DEFAULT_REQUIREMENT_VERSION)
requirement = ToolRequirement(name=name, type=type, version=version)
requirements.append(requirement)
container_elems = []
if requirements_elem is not None:
container_elems = requirements_elem.findall('container')
containers = [container_from_element(c) for c in container_elems]
return requirements, containers
def _parse_output(
self,
data_elem,
tool,
default_format="data",
default_format_source=None,
default_metadata_source="",
):
output = galaxy.tools.ToolOutput( data_elem.get("name") )
output.format = data_elem.get("format", default_format)
output.change_format = data_elem.findall("change_format")
output.format_source = data_elem.get("format_source", default_format_source)
output.metadata_source = data_elem.get("metadata_source", default_metadata_source)
output.parent = data_elem.get("parent", None)
output.label = xml_text( data_elem, "label" )
output.count = int( data_elem.get("count", 1) )
output.filters = data_elem.findall( 'filter' )
output.tool = tool
output.from_work_dir = data_elem.get("from_work_dir", None)
output.hidden = string_as_bool( data_elem.get("hidden", "") )
output.actions = galaxy.tools.ToolOutputActionGroup( output, data_elem.find( 'actions' ) )
output.dataset_collectors = output_collect.dataset_collectors_from_elem( data_elem )
return output
def _parse_output(
self,
data_elem,
tool,
default_format="data",
default_format_source=None,
default_metadata_source="",
):
output = galaxy.tools.ToolOutput( data_elem.get("name") )
output.format = data_elem.get("format", default_format)
output.change_format = data_elem.findall("change_format")
output.format_source = data_elem.get("format_source", default_format_source)
output.metadata_source = data_elem.get("metadata_source", default_metadata_source)
output.parent = data_elem.get("parent", None)
output.label = xml_text( data_elem, "label" )
output.count = int( data_elem.get("count", 1) )
output.filters = data_elem.findall( 'filter' )
output.tool = tool
output.from_work_dir = data_elem.get("from_work_dir", None)
output.hidden = string_as_bool( data_elem.get("hidden", "") )
output.actions = galaxy.tools.ToolOutputActionGroup( output, data_elem.find( 'actions' ) )
output.dataset_collectors = output_collect.dataset_collectors_from_elem( data_elem )
return output
def _parse_output(
self,
data_elem,
tool,
default_format="data",
default_format_source=None,
default_metadata_source="",
expression_type=None,
):
from_expression = data_elem.get("from")
output = ToolOutput(data_elem.get("name"), from_expression=from_expression)
output_format = data_elem.get("format", default_format)
auto_format = string_as_bool(data_elem.get("auto_format", "false"))
if auto_format and output_format != "data":
raise ValueError("Setting format and auto_format is not supported at this time.")
elif auto_format:
output_format = "_sniff_"
output.format = output_format
output.change_format = data_elem.findall("change_format")
output.format_source = data_elem.get("format_source", default_format_source)
output.default_identifier_source = data_elem.get("default_identifier_source", 'None')
output.metadata_source = data_elem.get("metadata_source", default_metadata_source)
output.parent = data_elem.get("parent", None)
output.label = xml_text(data_elem, "label")
output.count = int(data_elem.get("count", 1))
output.filters = data_elem.findall('filter')
output.tool = tool
output.from_work_dir = data_elem.get("from_work_dir", None)
if output.from_work_dir and getattr(tool, 'profile', 0) < 21.09:
# We started quoting from_work_dir outputs in 21.09.
# Prior to quoting, trailing spaces had no effect.
# This ensures that old tools continue to work.
output.from_work_dir = output.from_work_dir.strip()
output.hidden = string_as_bool(data_elem.get("hidden", ""))
output.actions = ToolOutputActionGroup(output, data_elem.find('actions'))
output.dataset_collector_descriptions = dataset_collector_descriptions_from_elem(data_elem, legacy=self.legacy_defaults)
return output
def parse_help(self):
return xml_text(self.input_elem, "help")
def parse_help(self):
return xml_text(self.input_elem, "help")
def parse_label(self):
return xml_text(self.input_elem, "label")
def parse_description(self):
return xml_text(self.root, "description")
def parse_outputs(self, tool):
out_elem = self.root.find("outputs")
outputs = odict()
output_collections = odict()
if out_elem is None:
return outputs, output_collections
data_dict = odict()
def _parse(data_elem, **kwds):
output_def = self._parse_output(data_elem, tool, **kwds)
data_dict[output_def.name] = output_def
return output_def
map(_parse, out_elem.findall("data"))
for collection_elem in out_elem.findall("collection"):
name = collection_elem.get( "name" )
label = xml_text( collection_elem, "label" )
default_format = collection_elem.get( "format", "data" )
collection_type = collection_elem.get( "type", None )
collection_type_source = collection_elem.get( "type_source", None )
structured_like = collection_elem.get( "structured_like", None )
inherit_format = False
inherit_metadata = False
if structured_like:
inherit_format = string_as_bool( collection_elem.get( "inherit_format", None ) )
inherit_metadata = string_as_bool( collection_elem.get( "inherit_metadata", None ) )
default_format_source = collection_elem.get( "format_source", None )
default_metadata_source = collection_elem.get( "metadata_source", "" )
filters = collection_elem.findall( 'filter' )
dataset_collector_descriptions = None
if collection_elem.find( "discover_datasets" ) is not None:
dataset_collector_descriptions = dataset_collector_descriptions_from_elem( collection_elem, legacy=False )
structure = ToolOutputCollectionStructure(
collection_type=collection_type,
collection_type_source=collection_type_source,
structured_like=structured_like,
dataset_collector_descriptions=dataset_collector_descriptions,
)
output_collection = ToolOutputCollection(
name,
structure,
label=label,
filters=filters,
default_format=default_format,
inherit_format=inherit_format,
inherit_metadata=inherit_metadata,
default_format_source=default_format_source,
default_metadata_source=default_metadata_source,
)
outputs[output_collection.name] = output_collection
for data_elem in collection_elem.findall("data"):
_parse(
data_elem,
default_format=default_format,
default_format_source=default_format_source,
default_metadata_source=default_metadata_source,
)
for data_elem in collection_elem.findall("data"):
output_name = data_elem.get("name")
data = data_dict[output_name]
assert data
del data_dict[output_name]
output_collection.outputs[output_name] = data
output_collections[ name ] = output_collection
for output_def in data_dict.values():
outputs[output_def.name] = output_def
return outputs, output_collections
def parse_hidden(self):
hidden = xml_text(self.root, "hidden")
if hidden:
hidden = string_as_bool(hidden)
return hidden
def parse_hidden(self):
hidden = xml_text(self.root, "hidden")
if hidden:
hidden = string_as_bool(hidden)
return hidden
def parse_description(self):
return xml_text(self.root, "description")
def parse_label(self):
return xml_text(self.input_elem, "label")
def parse_outputs(self, tool):
out_elem = self.root.find("outputs")
outputs = odict()
output_collections = odict()
if out_elem is None:
return outputs, output_collections
data_dict = odict()
def _parse(data_elem, **kwds):
output_def = self._parse_output(data_elem, tool, **kwds)
data_dict[output_def.name] = output_def
return output_def
for _ in out_elem.findall("data"):
_parse(_)
for collection_elem in out_elem.findall("collection"):
name = collection_elem.get("name")
label = xml_text(collection_elem, "label")
default_format = collection_elem.get("format", "data")
collection_type = collection_elem.get("type", None)
collection_type_source = collection_elem.get("type_source", None)
collection_type_from_rules = collection_elem.get("type_from_rules", None)
structured_like = collection_elem.get("structured_like", None)
inherit_format = False
inherit_metadata = False
if structured_like:
inherit_format = string_as_bool(collection_elem.get("inherit_format", None))
inherit_metadata = string_as_bool(collection_elem.get("inherit_metadata", None))
default_format_source = collection_elem.get("format_source", None)
default_metadata_source = collection_elem.get("metadata_source", "")
filters = collection_elem.findall('filter')
dataset_collector_descriptions = None
if collection_elem.find("discover_datasets") is not None:
dataset_collector_descriptions = dataset_collector_descriptions_from_elem(collection_elem, legacy=False)
structure = ToolOutputCollectionStructure(
collection_type=collection_type,
collection_type_source=collection_type_source,
collection_type_from_rules=collection_type_from_rules,
structured_like=structured_like,
dataset_collector_descriptions=dataset_collector_descriptions,
)
output_collection = ToolOutputCollection(
name,
structure,
label=label,
filters=filters,
default_format=default_format,
inherit_format=inherit_format,
inherit_metadata=inherit_metadata,
default_format_source=default_format_source,
default_metadata_source=default_metadata_source,
)
outputs[output_collection.name] = output_collection
for data_elem in collection_elem.findall("data"):
_parse(
data_elem,
default_format=default_format,
default_format_source=default_format_source,
default_metadata_source=default_metadata_source,
)
for data_elem in collection_elem.findall("data"):
output_name = data_elem.get("name")
data = data_dict[output_name]
assert data
del data_dict[output_name]
output_collection.outputs[output_name] = data
output_collections[name] = output_collection
for output_def in data_dict.values():
outputs[output_def.name] = output_def
return outputs, output_collections
