Source code for galaxy.tools.actions
import json
import logging
import os
import re
from abc import abstractmethod
from json import dumps
from typing import (
Any,
cast,
Dict,
List,
Optional,
Set,
TYPE_CHECKING,
Union,
)
from packaging.version import Version
from galaxy import model
from galaxy.exceptions import (
AuthenticationRequired,
ItemAccessibilityException,
RequestParameterInvalidException,
)
from galaxy.job_execution.actions.post import ActionBox
from galaxy.managers.context import ProvidesHistoryContext
from galaxy.model import (
HistoryDatasetAssociation,
Job,
LibraryDatasetDatasetAssociation,
WorkflowRequestInputParameter,
)
from galaxy.model.base import transaction
from galaxy.model.dataset_collections.builder import CollectionBuilder
from galaxy.model.none_like import NoneDataset
from galaxy.objectstore import ObjectStorePopulator
from galaxy.tools.parameters import update_dataset_ids
from galaxy.tools.parameters.basic import (
DataCollectionToolParameter,
DataToolParameter,
SelectToolParameter,
)
from galaxy.tools.parameters.workflow_utils import RuntimeValue
from galaxy.tools.parameters.wrapped import (
LegacyUnprefixedDict,
WrappedParameters,
)
from galaxy.util import ExecutionTimer
from galaxy.util.template import fill_template
if TYPE_CHECKING:
from galaxy.model import DatasetInstance
from galaxy.tool_util.parser.output_objects import ToolOutput
log = logging.getLogger(__name__)
[docs]class ToolExecutionCache:
"""An object mean to cache calculation caused by repeatedly evaluting
the same tool by the same user with slightly different parameters.
"""
[docs] def __init__(self, trans):
self.trans = trans
self.current_user_roles = trans.get_current_user_roles()
self.chrom_info = {}
self.cached_collection_elements = {}
[docs] def get_chrom_info(self, tool_id, input_dbkey):
genome_builds = self.trans.app.genome_builds
custom_build_hack_get_len_from_fasta_conversion = tool_id != "CONVERTER_fasta_to_len"
if custom_build_hack_get_len_from_fasta_conversion and input_dbkey in self.chrom_info:
return self.chrom_info[input_dbkey]
chrom_info_pair = genome_builds.get_chrom_info(
input_dbkey,
trans=self.trans,
custom_build_hack_get_len_from_fasta_conversion=custom_build_hack_get_len_from_fasta_conversion,
)
if custom_build_hack_get_len_from_fasta_conversion:
self.chrom_info[input_dbkey] = chrom_info_pair
return chrom_info_pair
[docs]class ToolAction:
"""
The actions to be taken when a tool is run (after parameters have
been converted and validated).
"""
[docs] @abstractmethod
def execute(self, tool, trans, incoming=None, set_output_hid=True, **kwargs):
pass
[docs]class DefaultToolAction(ToolAction):
"""Default tool action is to run an external command"""
produces_real_jobs = True
def _collect_input_datasets(
self,
tool,
param_values,
trans: ProvidesHistoryContext,
history,
current_user_roles=None,
dataset_collection_elements=None,
collection_info=None,
):
"""
Collect any dataset inputs from incoming. Returns a mapping from
parameter name to Dataset instance for each tool parameter that is
of the DataToolParameter type.
"""
if current_user_roles is None:
current_user_roles = trans.get_current_user_roles()
input_datasets = LegacyUnprefixedDict()
all_permissions: Dict[str, Set[str]] = {}
def record_permission(action, role_id):
if action not in all_permissions:
all_permissions[action] = set()
all_permissions[action].add(role_id)
def visitor(input, value, prefix, prefixed_name: str, parent=None, **kwargs):
def process_dataset(data, formats=None):
if not data or isinstance(data, RuntimeValue):
return None
if formats is None:
formats = input.formats
data = getattr(data, "hda", data)
direct_match, target_ext, converted_dataset = data.find_conversion_destination(formats)
if not direct_match and target_ext:
if converted_dataset:
data = converted_dataset
else:
data = data.get_converted_dataset(trans, target_ext, target_context=parent, history=history)
input_name = prefixed_name
# Checked security of whole collection all at once if mapping over this input, else
# fetch dataset details for this input from the database.
if collection_info and collection_info.is_mapped_over(input_name):
action_tuples = collection_info.map_over_action_tuples(input_name)
if not trans.user_is_admin and not trans.app.security_agent.can_access_datasets(
current_user_roles, action_tuples
):
raise ItemAccessibilityException(
"User does not have permission to use a dataset provided for input."
)
for action, role_id in action_tuples:
record_permission(action, role_id)
else:
if not trans.user_is_admin and not trans.app.security_agent.can_access_dataset(
current_user_roles, data.dataset
):
raise ItemAccessibilityException(
f"User does not have permission to use dataset ({data.name}) provided for input."
)
permissions = trans.app.security_agent.get_permissions(data.dataset)
for action, roles in permissions.items():
for role in roles:
record_permission(action.action, model.cached_id(role))
return data
if isinstance(input, DataToolParameter):
if isinstance(value, list):
# If there are multiple inputs with the same name, they
# are stored as name1, name2, ...
for i, v in enumerate(value):
processed_dataset = process_dataset(v)
if i == 0:
# Allow copying metadata to output, first item will be source.
input_datasets[prefixed_name] = processed_dataset
input_datasets.set_legacy_alias(new_key=prefixed_name, old_key=prefix + input.name)
input_datasets[prefixed_name + str(i + 1)] = processed_dataset
input_datasets.set_legacy_alias(
new_key=prefixed_name + str(i + 1), old_key=prefix + input.name + str(i + 1)
)
conversions = []
for conversion_name, conversion_extensions, conversion_datatypes in input.conversions:
new_data = process_dataset(input_datasets[prefixed_name + str(i + 1)], conversion_datatypes)
if not new_data or new_data.datatype.matches_any(conversion_datatypes):
input_datasets[prefixed_name[: -len(input.name)] + conversion_name + str(i + 1)] = (
new_data
)
input_datasets.set_legacy_alias(
new_key=prefixed_name[: -len(input.name)] + conversion_name + str(i + 1),
old_key=prefix + conversion_name + str(i + 1),
)
conversions.append((conversion_name, new_data))
else:
raise Exception(
f"A path for explicit datatype conversion has not been found: {input_datasets[prefixed_name + str(i + 1)].extension} --/--> {conversion_extensions}"
)
if parent:
parent[input.name][i] = input_datasets[prefixed_name + str(i + 1)]
for conversion_name, conversion_data in conversions:
# allow explicit conversion to be stored in job_parameter table
parent[conversion_name][
i
] = conversion_data.id # a more robust way to determine JSONable value is desired
else:
param_values[input.name][i] = input_datasets[prefixed_name + str(i + 1)]
for conversion_name, conversion_data in conversions:
# allow explicit conversion to be stored in job_parameter table
param_values[conversion_name][
i
] = conversion_data.id # a more robust way to determine JSONable value is desired
else:
input_datasets[prefixed_name] = process_dataset(value)
input_datasets.set_legacy_alias(new_key=prefixed_name, old_key=prefix + input.name)
conversions = []
for conversion_name, conversion_extensions, conversion_datatypes in input.conversions:
new_data = process_dataset(input_datasets[prefixed_name], conversion_datatypes)
if not new_data or new_data.datatype.matches_any(conversion_datatypes):
input_datasets[prefix + conversion_name] = new_data
conversions.append((conversion_name, new_data))
else:
raise Exception(
f"A path for explicit datatype conversion has not been found: {input_datasets[prefixed_name].extension} --/--> {conversion_extensions}"
)
target_dict = parent
if not target_dict:
target_dict = param_values
target_dict[input.name] = input_datasets[prefixed_name]
for conversion_name, conversion_data in conversions:
# allow explicit conversion to be stored in job_parameter table
target_dict[conversion_name] = (
conversion_data.id
) # a more robust way to determine JSONable value is desired
elif isinstance(input, DataCollectionToolParameter):
if not value:
return
collection = None
child_collection = False
if hasattr(value, "child_collection"):
# if we are mapping a collection over a tool, we only require the child_collection
child_collection = True
collection = value.child_collection
else:
# else the tool takes a collection as input so we need everything
collection = value.collection
action_tuples = collection.dataset_action_tuples
if not trans.user_is_admin and not trans.app.security_agent.can_access_datasets(
current_user_roles, action_tuples
):
raise ItemAccessibilityException(
"User does not have permission to use a dataset provided for input."
)
for action, role_id in action_tuples:
record_permission(action, role_id)
_, extensions = collection.dataset_states_and_extensions_summary
conversion_required = False
for ext in extensions:
if ext:
datatype = trans.app.datatypes_registry.get_datatype_by_extension(ext)
if not datatype:
raise RequestParameterInvalidException(
f"Extension '{ext}' unknown, cannot use dataset collection as input"
)
if not datatype.matches_any(input.formats):
conversion_required = True
break
processed_dataset_dict = {}
for i, v in enumerate(collection.dataset_instances):
processed_dataset = None
if conversion_required:
processed_dataset = process_dataset(v)
if processed_dataset is not v:
processed_dataset_dict[v] = processed_dataset
input_datasets[prefixed_name + str(i + 1)] = processed_dataset or v
input_datasets.set_legacy_alias(
new_key=prefixed_name + str(i + 1), old_key=prefix + input.name + str(i + 1)
)
if conversion_required:
collection_type_description = (
trans.app.dataset_collection_manager.collection_type_descriptions.for_collection_type(
collection.collection_type
)
)
collection_builder = CollectionBuilder(collection_type_description)
collection_builder.replace_elements_in_collection(
template_collection=collection,
replacement_dict=processed_dataset_dict,
)
new_collection = collection_builder.build()
if child_collection:
value.child_collection = new_collection
else:
value.collection = new_collection
elif isinstance(input, SelectToolParameter) and isinstance(value, HistoryDatasetAssociation):
input_datasets[prefixed_name] = value
tool.visit_inputs(param_values, visitor)
return input_datasets, all_permissions
[docs] def collect_input_dataset_collections(self, tool, param_values):
def append_to_key(the_dict: LegacyUnprefixedDict, key, legacy_key, value):
if key not in the_dict:
the_dict[key] = []
the_dict.set_legacy_alias(new_key=key, old_key=legacy_key)
the_dict[key].append(value)
input_dataset_collections = LegacyUnprefixedDict()
def visitor(input, value, prefix, parent=None, prefixed_name=None, **kwargs):
if isinstance(input, DataToolParameter):
values = value
if not isinstance(values, list):
values = [value]
for i, value in enumerate(values):
if isinstance(value, model.HistoryDatasetCollectionAssociation) or isinstance(
value, model.DatasetCollectionElement
):
append_to_key(input_dataset_collections, prefixed_name, prefix + input.name, (value, True))
target_dict = parent
if not target_dict:
target_dict = param_values
# This is just a DataToolParameter, so replace this
# collection with individual datasets. Database will still
# record collection which should be enought for workflow
# extraction and tool rerun.
if isinstance(value, model.DatasetCollectionElement):
if value.child_collection:
# if we are mapping a collection over a tool, we only require the child_collection
dataset_instances = value.child_collection.dataset_instances
else:
continue
else:
# else the tool takes a collection as input so we need everything
dataset_instances = value.collection.dataset_instances
if i == 0:
target_dict[input.name] = []
target_dict[input.name].extend(dataset_instances)
elif isinstance(input, DataCollectionToolParameter):
append_to_key(input_dataset_collections, prefixed_name, prefix + input.name, (value, False))
tool.visit_inputs(param_values, visitor)
return input_dataset_collections
def _check_access(self, tool, trans):
assert tool.allow_user_access(trans.user), f"User ({trans.user}) is not allowed to access this tool."
def _collect_inputs(self, tool, trans, incoming, history, current_user_roles, collection_info):
"""Collect history as well as input datasets and collections."""
# Set history.
if not history:
history = tool.get_default_history_by_trans(trans, create=True)
# Track input dataset collections - but replace with simply lists so collect
# input datasets can process these normally.
inp_dataset_collections = self.collect_input_dataset_collections(tool, incoming)
# Collect any input datasets from the incoming parameters
inp_data, all_permissions = self._collect_input_datasets(
tool,
incoming,
trans,
history=history,
current_user_roles=current_user_roles,
collection_info=collection_info,
)
preserved_tags = {}
preserved_hdca_tags = {}
# grab tags from incoming HDAs
for data in inp_data.values():
if not data:
continue
for tag in data.auto_propagated_tags:
preserved_tags[tag.value] = tag
# grab tags from incoming HDCAs
for collection_pairs in inp_dataset_collections.values():
for collection, _ in collection_pairs:
# if sub-collection mapping, this will be an DC not an HDCA
# (e.g. part of collection not a collection instance) and thus won't have tags.
if hasattr(collection, "tags"):
for tag in collection.auto_propagated_tags:
preserved_hdca_tags[tag.value] = tag
preserved_tags.update(preserved_hdca_tags)
return history, inp_data, inp_dataset_collections, preserved_tags, preserved_hdca_tags, all_permissions
[docs] def execute(
self,
tool,
trans,
incoming=None,
return_job=False,
set_output_hid=True,
history=None,
job_params=None,
rerun_remap_job_id=None,
execution_cache=None,
dataset_collection_elements=None,
completed_job=None,
collection_info=None,
job_callback=None,
preferred_object_store_id=None,
flush_job=True,
skip=False,
):
"""
Executes a tool, creating job and tool outputs, associating them, and
submitting the job to the job queue. If history is not specified, use
trans.history as destination for tool's output datasets.
"""
trans.check_user_activation()
incoming = incoming or {}
self._check_access(tool, trans)
app = trans.app
if execution_cache is None:
execution_cache = ToolExecutionCache(trans)
current_user_roles = execution_cache.current_user_roles
(
history,
inp_data,
inp_dataset_collections,
preserved_tags,
preserved_hdca_tags,
all_permissions,
) = self._collect_inputs(tool, trans, incoming, history, current_user_roles, collection_info)
# Build name for output datasets based on tool name and input names
on_text = self._get_on_text(inp_data)
# format='input" previously would give you a random extension from
# the input extensions, now it should just give "input" as the output
# format.
input_ext = "data" if Version(str(tool.profile)) < Version("16.04") else "input"
input_dbkey = incoming.get("dbkey", "?")
for name, data in reversed(list(inp_data.items())):
if not data:
data = NoneDataset(datatypes_registry=app.datatypes_registry)
continue
# Convert LDDA to an HDA.
if isinstance(data, LibraryDatasetDatasetAssociation) and not completed_job:
data = data.to_history_dataset_association(None)
inp_data[name] = data
if Version(str(tool.profile)) < Version("16.04"):
input_ext = data.ext
if data.dbkey not in [None, "?"]:
input_dbkey = data.dbkey
identifier = getattr(data, "element_identifier", None)
if identifier is not None:
incoming[f"{name}|__identifier__"] = identifier
# Collect chromInfo dataset and add as parameters to incoming
(chrom_info, db_dataset) = execution_cache.get_chrom_info(tool.id, input_dbkey)
if db_dataset:
inp_data.update({"chromInfo": db_dataset})
incoming["chromInfo"] = chrom_info
if not completed_job:
# Determine output dataset permission/roles list
if all_permissions:
output_permissions = app.security_agent.guess_derived_permissions(all_permissions)
else:
# No valid inputs, we will use history defaults
output_permissions = app.security_agent.history_get_default_permissions(history)
# Add the dbkey to the incoming parameters
incoming["dbkey"] = input_dbkey
incoming["__input_ext"] = input_ext
# wrapped params are used by change_format action and by output.label; only perform this wrapping once, as needed
wrapped_params = self._wrapped_params(trans, tool, incoming, inp_data)
out_data: Dict[str, DatasetInstance] = {}
input_collections = LegacyUnprefixedDict({k: v[0][0] for k, v in inp_dataset_collections.items()})
input_collections._legacy_mapping = inp_dataset_collections._legacy_mapping
output_collections = OutputCollections(
trans,
history,
tool=tool,
tool_action=self,
input_collections=input_collections,
dataset_collection_elements=dataset_collection_elements,
on_text=on_text,
incoming=incoming,
params=wrapped_params.params,
job_params=job_params,
tags=preserved_tags,
hdca_tags=preserved_hdca_tags,
)
async_tool = tool.tool_type == "data_source_async"
def handle_output(name, output, hidden=None):
if async_tool and name in incoming:
# HACK: output data has already been created as a result of the async controller
dataid = incoming[name]
data = trans.sa_session.get(HistoryDatasetAssociation, dataid)
assert data is not None
out_data[name] = data
else:
ext = determine_output_format(
output,
wrapped_params.params,
inp_data,
inp_dataset_collections,
input_ext,
python_template_version=tool.python_template_version,
execution_cache=execution_cache,
)
create_datasets = True
dataset = None
if completed_job:
for output_dataset in completed_job.output_datasets:
if output_dataset.name == name:
create_datasets = False
completed_data = output_dataset.dataset
dataset = output_dataset.dataset.dataset
break
data = app.model.HistoryDatasetAssociation(
extension=ext, dataset=dataset, create_dataset=create_datasets, flush=False
)
if create_datasets:
from_work_dir = output.from_work_dir
if from_work_dir is not None:
data.dataset.created_from_basename = os.path.basename(from_work_dir)
if hidden is None:
hidden = output.hidden
if not hidden and dataset_collection_elements is not None: # Mapping over a collection - hide datasets
hidden = True
if hidden:
data.visible = False
if dataset_collection_elements is not None and name in dataset_collection_elements:
dataset_collection_elements[name].hda = data
trans.sa_session.add(data)
if not completed_job:
trans.app.security_agent.set_all_dataset_permissions(
data.dataset, output_permissions, new=True, flush=False
)
data.copy_tags_to(preserved_tags.values())
# This may not be necessary with the new parent/child associations
data.designation = name
# Copy metadata from one of the inputs if requested.
# metadata source can be either a string referencing an input
# or an actual object to copy.
metadata_source = output.metadata_source
if metadata_source:
if isinstance(metadata_source, str):
metadata_source = inp_data.get(metadata_source)
if metadata_source is not None:
data.init_meta(copy_from=metadata_source)
else:
data.init_meta()
# Take dbkey from LAST input
data.dbkey = str(input_dbkey)
# Set state
if completed_job:
data.blurb = completed_data.blurb
data.peek = completed_data.peek
data._metadata = completed_data._metadata
else:
data.blurb = "queued"
# Set output label
data.name = self.get_output_name(
output, data, tool, on_text, trans, incoming, history, wrapped_params.params, job_params
)
# Store output
out_data[name] = data
if output.actions:
# Apply pre-job tool-output-dataset actions; e.g. setting metadata, changing format
output_action_params = dict(out_data)
output_action_params.update(wrapped_params.params)
output_action_params["__python_template_version__"] = tool.python_template_version
output.actions.apply_action(data, output_action_params)
# Flush all datasets at once.
return data
child_dataset_names = set()
for name, output in tool.outputs.items():
if not filter_output(tool, output, incoming):
handle_output_timer = ExecutionTimer()
if output.collection:
if completed_job and dataset_collection_elements and name in dataset_collection_elements:
# Output collection is mapped over and has already been copied from original job
continue
collections_manager = app.dataset_collection_manager
element_identifiers: List[Dict[str, Union[str, List[Dict[str, Union[str, List[Any]]]]]]] = []
# mypy doesn't yet support recursive type definitions
known_outputs = output.known_outputs(input_collections, collections_manager.type_registry)
# Just to echo TODO elsewhere - this should be restructured to allow
# nested collections.
for output_part_def in known_outputs:
# Add elements to top-level collection, unless nested...
current_element_identifiers = element_identifiers
current_collection_type = output.structure.collection_type
for parent_id in output_part_def.parent_ids or []:
# TODO: replace following line with formal abstractions for doing this.
current_collection_type = ":".join(current_collection_type.split(":")[1:])
name_to_index = {
value["name"]: index for (index, value) in enumerate(current_element_identifiers)
}
if parent_id not in name_to_index:
if parent_id not in current_element_identifiers:
index = len(current_element_identifiers)
current_element_identifiers.append(
dict(
name=parent_id,
collection_type=current_collection_type,
src="new_collection",
element_identifiers=[],
)
)
else:
index = name_to_index[parent_id]
current_element_identifiers = cast(
List[
Dict[
str,
Union[str, List[Dict[str, Union[str, List[Any]]]]],
]
],
current_element_identifiers[index]["element_identifiers"],
)
effective_output_name = output_part_def.effective_output_name
child_dataset_names.add(effective_output_name)
element = handle_output(effective_output_name, output_part_def.output_def, hidden=True)
history.stage_addition(element)
# TODO: this shouldn't exist in the top-level of the history at all
# but for now we are still working around that by hiding the contents
# there.
# Following hack causes dataset to no be added to history...
trans.sa_session.add(element)
current_element_identifiers.append(
{
"__object__": element,
"name": output_part_def.element_identifier,
}
)
if output.dynamic_structure:
assert not element_identifiers # known_outputs must have been empty
element_kwds = dict(elements=collections_manager.ELEMENTS_UNINITIALIZED)
else:
element_kwds = dict(element_identifiers=element_identifiers)
output_collections.create_collection(
output=output, name=name, completed_job=completed_job, **element_kwds
)
log.info(f"Handled collection output named {name} for tool {tool.id} {handle_output_timer}")
else:
handle_output(name, output)
log.info(f"Handled output named {name} for tool {tool.id} {handle_output_timer}")
add_datasets_timer = tool.app.execution_timer_factory.get_timer(
"internals.galaxy.tools.actions.add_datasets",
"Added output datasets to history",
)
# Add all the top-level (non-child) datasets to the history unless otherwise specified
for name, data in out_data.items():
if name not in incoming and name not in child_dataset_names:
# don't add already existing datasets, i.e. async created
history.stage_addition(data)
history.add_pending_items(set_output_hid=set_output_hid)
log.info(add_datasets_timer)
job_setup_timer = ExecutionTimer()
# Create the job object
job, galaxy_session = self._new_job_for_session(trans, tool, history)
if skip:
job.state = job.states.SKIPPED
for output_collection in output_collections.out_collections.values():
output_collection.mark_as_populated()
for hdca in output_collections.out_collection_instances.values():
hdca.visible = False
hdca.collection.mark_as_populated()
object_store_populator = ObjectStorePopulator(trans.app, trans.user)
for data in out_data.values():
data.set_skipped(object_store_populator)
job.preferred_object_store_id = preferred_object_store_id
self._record_inputs(trans, tool, job, incoming, inp_data, inp_dataset_collections)
self._record_outputs(job, out_data, output_collections)
# execute immediate post job actions and associate post job actions that are to be executed after the job is complete
if job_callback:
job_callback(job)
if job_params:
job.params = dumps(job_params)
if completed_job:
job.set_copied_from_job_id(completed_job.id)
trans.sa_session.add(job)
# Remap any outputs if this is a rerun and the user chose to continue dependent jobs
# This functionality requires tracking jobs in the database.
if app.config.track_jobs_in_database and rerun_remap_job_id is not None:
self._remap_job_on_rerun(
trans=trans,
galaxy_session=galaxy_session,
rerun_remap_job_id=rerun_remap_job_id,
current_job=job,
out_data=out_data,
)
log.info(f"Setup for job {job.log_str()} complete, ready to be enqueued {job_setup_timer}")
# Some tools are not really executable, but jobs are still created for them ( for record keeping ).
# Examples include tools that redirect to other applications ( epigraph ). These special tools must
# include something that can be retrieved from the params ( e.g., REDIRECT_URL ) to keep the job
# from being queued.
if "REDIRECT_URL" in incoming:
# Get the dataset - there should only be 1
for name in inp_data.keys():
dataset = inp_data[name]
redirect_url = tool.parse_redirect_url(dataset, incoming)
# GALAXY_URL should be include in the tool params to enable the external application
# to send back to the current Galaxy instance
GALAXY_URL = incoming.get("GALAXY_URL", None)
assert GALAXY_URL is not None, "GALAXY_URL parameter missing in tool config."
redirect_url += f"&GALAXY_URL={GALAXY_URL}"
# Job should not be queued, so set state to ok
job.set_state(app.model.Job.states.OK)
job.info = f"Redirected to: {redirect_url}"
trans.sa_session.add(job)
with transaction(trans.sa_session):
trans.sa_session.commit()
trans.response.send_redirect(redirect_url)
else:
if flush_job:
# Set HID and add to history.
job_flush_timer = ExecutionTimer()
with transaction(trans.sa_session):
trans.sa_session.commit()
log.info(f"Flushed transaction for job {job.log_str()} {job_flush_timer}")
return job, out_data, history
def _remap_job_on_rerun(
self,
trans: ProvidesHistoryContext,
galaxy_session: Optional[model.GalaxySession],
rerun_remap_job_id: int,
current_job: Job,
out_data,
):
"""
Re-connect dependent datasets for a job that is being rerun (because it failed initially).
If a job fails, the user has the option to try the job again with changed parameters.
To be able to resume jobs that depend on this jobs output datasets we change the dependent's job
input datasets to be those of the job that is being rerun.
"""
old_job = trans.sa_session.get(Job, rerun_remap_job_id)
if not old_job:
# I don't think that can really happen
raise RequestParameterInvalidException("rerun_remap_job_id parameter is invalid")
old_tool = trans.app.toolbox.get_tool(old_job.tool_id, exact=False)
new_tool = trans.app.toolbox.get_tool(current_job.tool_id, exact=False)
if old_tool and new_tool and old_tool.old_id != new_tool.old_id:
# If we currently only have the old or new tool installed we'll find the other tool anyway with `exact=False`.
# If we don't have the tool at all we'll fail anyway, no need to worry here.
raise RequestParameterInvalidException(
f"Old tool id ({old_job.tool_id}) does not match rerun tool id ({current_job.tool_id})"
)
if trans.user is not None:
if old_job.user_id != trans.user.id:
raise RequestParameterInvalidException(
"Cannot remap job dependencies for job not created by current user."
)
elif trans.user is None and galaxy_session:
if old_job.session_id != galaxy_session.id:
raise RequestParameterInvalidException(
"Cannot remap job dependencies for job not created by current user."
)
else:
raise AuthenticationRequired("Authentication required to remap job dependencies")
# Need to flush here so that referencing outputs by id works
session = trans.sa_session()
try:
session.expire_on_commit = False
with transaction(session):
session.commit()
finally:
session.expire_on_commit = True
try:
# Start by hiding current job outputs before taking over the old job's (implicit) outputs.
current_job.hide_outputs(flush=False)
# Duplicate PJAs before remap.
for pjaa in old_job.post_job_actions:
current_job.add_post_job_action(pjaa.post_job_action)
if old_job.workflow_invocation_step:
replacement_dict = {}
for parameter in old_job.workflow_invocation_step.workflow_invocation.input_parameters:
if parameter.type == WorkflowRequestInputParameter.types.REPLACEMENT_PARAMETERS:
replacement_dict[parameter.name] = parameter.value
for pja in old_job.workflow_invocation_step.workflow_step.post_job_actions:
# execute immediate actions here, with workflow context.
if pja.action_type in ActionBox.immediate_actions:
ActionBox.execute(trans.app, trans.sa_session, pja, current_job, replacement_dict)
for p in old_job.parameters:
if p.name.endswith("|__identifier__"):
current_job.parameters.append(p.copy())
remapped_hdas = self.__remap_data_inputs(old_job=old_job, current_job=current_job)
for jtod in old_job.output_datasets:
for job_to_remap, jtid in [(jtid.job, jtid) for jtid in jtod.dataset.dependent_jobs]:
if (trans.user is not None and job_to_remap.user_id == trans.user.id) or (
trans.user is None and galaxy_session and job_to_remap.session_id == galaxy_session.id
):
self.__remap_parameters(job_to_remap, jtid, jtod, out_data)
trans.sa_session.add(job_to_remap)
trans.sa_session.add(jtid)
job_to_remap.resume()
jtod.dataset.visible = False
trans.sa_session.add(jtod)
for jtodc in old_job.output_dataset_collection_instances:
# Update JobToOutputDatasetCollectionAssociation to the current job
jtodc.job = current_job
hdca = jtodc.dataset_collection_instance
hdca.collection.replace_failed_elements(remapped_hdas)
if hdca.implicit_collection_jobs:
for job in hdca.implicit_collection_jobs.jobs:
if job.job_id == old_job.id:
job.job_id = current_job.id
hdca.update()
for jtoidca in old_job.output_dataset_collections:
jtoidca.dataset_collection.replace_failed_elements(remapped_hdas)
except Exception:
log.exception("Cannot remap rerun dependencies.")
def __remap_data_inputs(self, old_job, current_job):
"""Record output datasets from old_job and build a dictionary that maps the old output HDAs to the new output HDAs."""
remapped_hdas = {}
old_output_datasets = {jtod.name: jtod.dataset for jtod in old_job.output_datasets}
for jtod in current_job.output_datasets:
remapped_hdas[old_output_datasets[jtod.name]] = jtod.dataset
return remapped_hdas
def __remap_parameters(self, job_to_remap, jtid, jtod, out_data):
input_values = {p.name: json.loads(p.value) for p in job_to_remap.parameters if p.value is not None}
old_dataset_id = jtod.dataset_id
new_dataset_id = out_data[jtod.name].id
input_values = update_dataset_ids(input_values, {old_dataset_id: new_dataset_id}, src="hda")
for p in job_to_remap.parameters:
if p.name in input_values:
p.value = json.dumps(input_values[p.name])
jtid.dataset = out_data[jtod.name]
jtid.dataset.hid = jtod.dataset.hid
log.info(f"Job {job_to_remap.id} input HDA {jtod.dataset.id} remapped to new HDA {jtid.dataset.id}")
def _wrapped_params(self, trans, tool, incoming, input_datasets=None):
wrapped_params = WrappedParameters(trans, tool, incoming, input_datasets=input_datasets)
return wrapped_params
def _get_on_text(self, inp_data):
input_names = []
for data in reversed(list(inp_data.values())):
if getattr(data, "hid", None):
input_names.append(f"data {data.hid}")
return on_text_for_names(input_names)
def _new_job_for_session(self, trans, tool, history):
job = trans.app.model.Job()
job.galaxy_version = trans.app.config.version_major
galaxy_session = None
if hasattr(trans, "get_galaxy_session"):
galaxy_session = trans.get_galaxy_session()
# If we're submitting from the API, there won't be a session.
if isinstance(galaxy_session, trans.model.GalaxySession):
job.session_id = model.cached_id(galaxy_session)
if trans.user is not None:
job.user_id = model.cached_id(trans.user)
job.user = trans.user
if history:
job.history_id = model.cached_id(history)
job.tool_id = tool.id
try:
# For backward compatibility, some tools may not have versions yet.
job.tool_version = tool.version
except AttributeError:
job.tool_version = "1.0.0"
job.dynamic_tool = tool.dynamic_tool
return job, galaxy_session
def _record_inputs(self, trans, tool, job, incoming, inp_data, inp_dataset_collections):
# FIXME: Don't need all of incoming here, just the defined parameters
# from the tool. We need to deal with tools that pass all post
# parameters to the command as a special case.
reductions: Dict[str, List[str]] = {}
for name, dataset_collection_info_pairs in inp_dataset_collections.items():
for dataset_collection, reduced in dataset_collection_info_pairs:
if reduced:
if name not in reductions:
reductions[name] = []
reductions[name].append(dataset_collection)
# TODO: verify can have multiple with same name, don't want to lose traceability
if isinstance(dataset_collection, model.HistoryDatasetCollectionAssociation):
job.add_input_dataset_collection(name, dataset_collection)
elif isinstance(dataset_collection, model.DatasetCollectionElement):
job.add_input_dataset_collection_element(name, dataset_collection)
# If this an input collection is a reduction, we expanded it for dataset security, type
# checking, and such, but the persisted input must be the original collection
# so we can recover things like element identifier during tool command evaluation.
def restore_reduction_visitor(input, value, prefix, parent=None, prefixed_name=None, **kwargs):
if prefixed_name in reductions and isinstance(input, DataToolParameter):
target_dict = parent
if not target_dict:
target_dict = incoming
target_dict[input.name] = []
for reduced_collection in reductions[prefixed_name]:
if hasattr(reduced_collection, "child_collection"):
target_dict[input.name].append({"id": model.cached_id(reduced_collection), "src": "dce"})
else:
target_dict[input.name].append({"id": model.cached_id(reduced_collection), "src": "hdca"})
if reductions:
tool.visit_inputs(incoming, restore_reduction_visitor)
for name, value in tool.params_to_strings(incoming, trans.app).items():
job.add_parameter(name, value)
self._record_input_datasets(trans, job, inp_data)
def _record_outputs(self, job, out_data, output_collections):
out_collections = output_collections.out_collections
out_collection_instances = output_collections.out_collection_instances
for name, dataset in out_data.items():
job.add_output_dataset(name, dataset)
for name, dataset_collection in out_collections.items():
job.add_implicit_output_dataset_collection(name, dataset_collection)
for name, dataset_collection_instance in out_collection_instances.items():
job.add_output_dataset_collection(name, dataset_collection_instance)
dataset_collection_instance.job = job
def _record_input_datasets(self, trans, job, inp_data):
for name, dataset in inp_data.items():
# TODO: figure out why can't pass dataset_id here.
job.add_input_dataset(name, dataset=dataset)
[docs] def get_output_name(
self,
output,
dataset=None,
tool=None,
on_text=None,
trans=None,
incoming=None,
history=None,
params=None,
job_params=None,
):
if output.label:
params["tool"] = tool
params["on_string"] = on_text
return fill_template(output.label, context=params, python_template_version=tool.python_template_version)
else:
return self._get_default_data_name(
dataset,
tool,
on_text=on_text,
trans=trans,
incoming=incoming,
history=history,
params=params,
job_params=job_params,
)
def _get_default_data_name(
self, dataset, tool, on_text=None, trans=None, incoming=None, history=None, params=None, job_params=None, **kwd
):
name = tool.name
if on_text:
name += f" on {on_text}"
return name
[docs]class OutputCollections:
"""Keeps track of collections (DC or HDCA) created by actions.
Actions do fairly different things depending on whether we are creating
just part of an collection or a whole output collection (mapping_over_collection
parameter).
"""
[docs] def __init__(
self,
trans,
history,
tool,
tool_action,
input_collections,
dataset_collection_elements,
on_text,
incoming,
params,
job_params,
tags,
hdca_tags,
):
self.trans = trans
self.tag_handler = trans.tag_handler
self.history = history
self.tool = tool
self.tool_action = tool_action
self.input_collections = input_collections
self.dataset_collection_elements = dataset_collection_elements
self.on_text = on_text
self.incoming = incoming
self.params = params
self.job_params = job_params
self.out_collections = {}
self.out_collection_instances = {}
self.tags = tags # all inherited tags
self.hdca_tags = hdca_tags # only tags inherited from input HDCAs
[docs] def create_collection(
self, output, name, collection_type=None, completed_job=None, propagate_hda_tags=True, **element_kwds
):
input_collections = self.input_collections
collections_manager = self.trans.app.dataset_collection_manager
collection_type = collection_type or output.structure.collection_type
if collection_type is None:
collection_type_source = output.structure.collection_type_source
if collection_type_source is None:
# TODO: Not a new problem, but this should be determined
# sooner.
raise Exception("Could not determine collection type to create.")
if collection_type_source not in input_collections:
raise Exception(f"Could not find collection type source with name [{collection_type_source}].")
# Using the collection_type_source string we get the DataCollectionToolParameter
data_param = self.tool.inputs
groups = collection_type_source.split("|")
for group in groups:
values = group.split("_")
if values[-1].isdigit():
key = "_".join(values[0:-1])
# We don't care about the repeat index, we just need to find the correct DataCollectionToolParameter
else:
key = group
if isinstance(data_param, dict):
data_param = data_param.get(key)
else:
data_param = data_param.inputs.get(key)
collection_type_description = data_param._history_query(self.trans).can_map_over(
input_collections[collection_type_source]
)
if collection_type_description:
collection_type = collection_type_description.collection_type
else:
collection_type = input_collections[collection_type_source].collection.collection_type
if "elements" in element_kwds:
def check_elements(elements):
if hasattr(elements, "items"): # else it is ELEMENTS_UNINITIALIZED object.
for value in elements.values():
# Either a HDA (if) or a DatasetCollection or a recursive dict.
if getattr(value, "history_content_type", None) == "dataset":
assert value.history is not None or value.history_id is not None
elif hasattr(value, "dataset_instances"):
for dataset in value.dataset_instances:
assert dataset.history is not None or dataset.history_id is not None
else:
assert value["src"] == "new_collection"
check_elements(value["elements"])
elements = element_kwds["elements"]
check_elements(elements)
if self.dataset_collection_elements is not None:
dc = collections_manager.create_dataset_collection(
self.trans, collection_type=collection_type, **element_kwds
)
if name in self.dataset_collection_elements:
self.dataset_collection_elements[name].child_collection = dc
# self.trans.sa_session.add(self.dataset_collection_elements[name])
self.out_collections[name] = dc
else:
hdca_name = self.tool_action.get_output_name(
output,
None,
self.tool,
self.on_text,
self.trans,
self.incoming,
self.history,
self.params,
self.job_params,
)
hdca = collections_manager.create(
self.trans,
self.history,
name=hdca_name,
collection_type=collection_type,
trusted_identifiers=True,
tags=self.tags if propagate_hda_tags else self.hdca_tags,
set_hid=False,
flush=False,
completed_job=completed_job,
output_name=name,
**element_kwds,
)
# name here is name of the output element - not name
# of the hdca.
self.history.stage_addition(hdca)
self.out_collection_instances[name] = hdca
[docs]def on_text_for_names(input_names):
# input_names may contain duplicates... this is because the first value in
# multiple input dataset parameters will appear twice once as param_name
# and once as param_name1.
unique_names = []
for name in input_names:
if name not in unique_names:
unique_names.append(name)
input_names = unique_names
# Build name for output datasets based on tool name and input names
if len(input_names) == 0:
on_text = ""
elif len(input_names) == 1:
on_text = input_names[0]
elif len(input_names) == 2:
on_text = "{} and {}".format(*input_names)
elif len(input_names) == 3:
on_text = "{}, {}, and {}".format(*input_names)
else:
on_text = "{}, {}, and others".format(*input_names[:2])
return on_text
[docs]def filter_output(tool, output, incoming):
for filter in output.filters:
try:
if not eval(filter.text.strip(), globals(), incoming):
return True # do not create this dataset
except Exception as e:
log.debug(f"Tool {tool.id} output {output.name}: dataset output filter ({filter.text}) failed: {e}")
return False
[docs]def get_ext_or_implicit_ext(hda):
if hda.implicitly_converted_parent_datasets:
# implicitly_converted_parent_datasets is a list of ImplicitlyConvertedDatasetAssociation
# objects, and their type is the target_ext, so this should be correct even if there
# are multiple ImplicitlyConvertedDatasetAssociation objects (meaning 2 datasets had been converted
# to produce a dataset with the required datatype)
return hda.implicitly_converted_parent_datasets[0].type
return hda.ext
[docs]def determine_output_format(
output: "ToolOutput",
parameter_context,
input_datasets,
input_dataset_collections,
random_input_ext,
python_template_version="3",
execution_cache=None,
):
"""Determines the output format for a dataset based on an abstract
description of the output (galaxy.tool_util.parser.ToolOutput), the parameter
wrappers, a map of the input datasets (name => HDA), and the last input
extensions in the tool form.
TODO: Make the input extension used deterministic instead of random.
"""
# the type should match the input
ext = output.format
if ext == "input":
if input_datasets and random_input_ext in {"data", "auto"}:
# Probably dealing with an implicitly converted dataset
try:
first_input_dataset = next(iter(input_datasets.values()))
random_input_ext = get_ext_or_implicit_ext(first_input_dataset)
except Exception:
pass
ext = random_input_ext
format_source = output.format_source
if format_source is not None and format_source in input_datasets:
try:
input_dataset = input_datasets[output.format_source]
ext = get_ext_or_implicit_ext(input_dataset)
except Exception:
pass
elif format_source is not None:
element_index = None
collection_name = format_source
if re.match(r"^[^\[\]]*\[[^\[\]]*\]$", format_source):
collection_name, element_index = format_source[0:-1].split("[")
# Treat as json to interpret "forward" vs 0 with type
# Make it feel more like Python, single quote better in XML also.
element_index = element_index.replace("'", '"')
element_index = json.loads(element_index)
if collection_name in input_dataset_collections:
try:
input_collection = input_dataset_collections[collection_name][0][0]
input_collection_collection = input_collection.collection
if element_index is None:
# just pick the first HDA
input_dataset = input_collection_collection.dataset_instances[0]
else:
try:
input_element = input_collection_collection[element_index]
except KeyError:
if execution_cache:
dataset_elements = execution_cache.cached_collection_elements.get(
input_collection_collection.id
)
if dataset_elements is None:
dataset_elements = execution_cache.cached_collection_elements[
input_collection_collection.id
] = input_collection_collection.dataset_elements
else:
dataset_elements = input_collection_collection.dataset_elements
for element in dataset_elements:
if element.element_identifier == element_index:
input_element = element
break
input_dataset = input_element.element_object
ext = get_ext_or_implicit_ext(input_dataset)
except Exception as e:
log.debug("Exception while trying to determine format_source: %s", e)
# process change_format tags
if output.change_format:
for change_format_model in output.change_format:
input_check = change_format_model.get("input")
if input_check is not None:
try:
if (
fill_template(
input_check, context=parameter_context, python_template_version=python_template_version
)
== change_format_model["value"]
):
if change_format_model["format"]:
return change_format_model["format"]
except Exception:
# bad tag input value; possibly referencing a param within a different conditional when block or other nonexistent grouping construct
continue
else:
input_dataset_check = change_format_model.get("input_dataset")
if input_dataset_check is not None:
dataset = input_datasets.get(input_dataset_check)
# At this point check is a HistoryDatasetAssociation object.
check_format = change_format_model["format"] or ext
check_value = change_format_model["value"]
check_attribute = change_format_model["check_attribute"]
if dataset is not None and check_value is not None and check_attribute is not None:
# See if the attribute to be checked belongs to the HistoryDatasetAssociation object.
if hasattr(dataset, check_attribute):
if str(getattr(dataset, check_attribute)) == str(check_value):
return check_format
# See if the attribute to be checked belongs to the metadata associated with the
# HistoryDatasetAssociation object.
if dataset.metadata is not None:
metadata_value = dataset.metadata.get(check_attribute)
if metadata_value is not None:
if str(metadata_value) == str(check_value):
return check_format
return ext