Warning

This document is for an in-development version of Galaxy. You can alternatively view this page in the latest release if it exists or view the top of the latest release's documentation.

Source code for galaxy.tool_util.cwl.util

"""Client-centric CWL-related utilities.

Used to share code between the Galaxy test framework
and other Galaxy CWL clients (e.g. Planemo)."""

import abc
import hashlib
import io
import json
import os
import tarfile
import tempfile
import urllib.parse
from collections import namedtuple
from typing import (
    Any,
    BinaryIO,
    Callable,
    Dict,
    List,
    Optional,
    Tuple,
)

import yaml
from typing_extensions import (
    Literal,
    TypedDict,
)

from galaxy.util import (
    str_removeprefix,
    unicodify,
)

STORE_SECONDARY_FILES_WITH_BASENAME = True
SECONDARY_FILES_EXTRA_PREFIX = "__secondary_files__"
SECONDARY_FILES_INDEX_PATH = "__secondary_files_index.json"


[docs]def set_basename_and_derived_properties(properties, basename): properties["basename"] = basename properties["nameroot"], properties["nameext"] = os.path.splitext(basename) return properties
OutputPropertiesType = TypedDict( "OutputPropertiesType", { "class": str, "location": Optional[str], "path": Optional[str], "listing": Optional[List[Any]], "basename": Optional[str], "nameroot": Optional[str], "nameext": Optional[str], "secondaryFiles": List[Any], "checksum": str, "size": int, }, total=False, )
[docs]def output_properties( path: Optional[str] = None, content: Optional[bytes] = None, basename=None, pseudo_location=False, ) -> OutputPropertiesType: checksum = hashlib.sha1() properties: OutputPropertiesType = {"class": "File", "checksum": "", "size": 0} f: BinaryIO if path is not None: properties["path"] = path f = open(path, "rb") else: if content is None: raise Exception("If no 'path', must provide 'content'.") f = io.BytesIO(content) try: contents = f.read(1024 * 1024) filesize = 0 while contents: checksum.update(contents) filesize += len(contents) contents = f.read(1024 * 1024) finally: f.close() properties["checksum"] = f"sha1${checksum.hexdigest()}" properties["size"] = filesize set_basename_and_derived_properties(properties, basename) _handle_pseudo_location(properties, pseudo_location) return properties
def _handle_pseudo_location(properties, pseudo_location): if pseudo_location: properties["location"] = properties["basename"]
[docs]def abs_path_or_uri(path_or_uri: str, relative_to: str) -> str: """Return the absolute path if this isn't a URI, otherwise keep the URI the same.""" if "://" in path_or_uri: return path_or_uri abs_path_ = os.path.abspath(os.path.join(relative_to, path_or_uri)) _ensure_file_exists(abs_path_) return abs_path_
[docs]def abs_path(path_or_uri: str, relative_to: str) -> str: """Return the absolute path if this is a file:// URI or a local path.""" if path_or_uri.startswith("file://"): # The path after file:// must be absolute abs_path_ = path_or_uri[len("file://") :] else: index = path_or_uri.find("://") if index != -1: raise ValueError(f"Unsupported URI scheme: {path_or_uri[: index + 3]}") abs_path_ = os.path.abspath(os.path.join(relative_to, path_or_uri)) _ensure_file_exists(abs_path_) return abs_path_
[docs]def path_or_uri_to_uri(path_or_uri: str) -> str: if "://" not in path_or_uri: return f"file://{path_or_uri}" else: return path_or_uri
[docs]def galactic_job_json( job: Dict[str, Any], test_data_directory: str, upload_func: Callable[["UploadTarget"], Dict[str, Any]], collection_create_func: Callable[[List[Dict[str, Any]], str], Dict[str, Any]], tool_or_workflow: Literal["tool", "workflow"] = "workflow", ) -> Tuple[Dict[str, Any], List[Dict[str, Any]]]: """Adapt a CWL job object to the Galaxy API. CWL derived tools in Galaxy can consume a job description sort of like CWL job objects via the API but paths need to be replaced with datasets and records and arrays with collection references. This function will stage files and modify the job description to adapt to these changes for Galaxy. """ datasets: List[Dict[str, Any]] = [] dataset_collections: List[Dict[str, Any]] = [] def response_to_hda(target: UploadTarget, upload_response: Dict[str, Any]) -> Dict[str, str]: assert isinstance(upload_response, dict), upload_response assert "outputs" in upload_response, upload_response assert len(upload_response["outputs"]) > 0, upload_response dataset = upload_response["outputs"][0] datasets.append(dataset) dataset_id = dataset["id"] return {"src": "hda", "id": dataset_id} def upload_file(file_path: str, secondary_files: Optional[str], **kwargs) -> Dict[str, str]: file_path = abs_path_or_uri(file_path, test_data_directory) target = FileUploadTarget(file_path, secondary_files, **kwargs) upload_response = upload_func(target) return response_to_hda(target, upload_response) def upload_file_literal(contents: str, **kwd) -> Dict[str, str]: target = FileLiteralTarget(contents, **kwd) upload_response = upload_func(target) return response_to_hda(target, upload_response) def upload_tar(file_path: str) -> Dict[str, str]: file_path = abs_path_or_uri(file_path, test_data_directory) target = DirectoryUploadTarget(file_path) upload_response = upload_func(target) return response_to_hda(target, upload_response) def upload_file_with_composite_data(file_path: Optional[str], composite_data, **kwargs) -> Dict[str, str]: if file_path is not None: file_path = abs_path_or_uri(file_path, test_data_directory) composite_data_resolved = [] for cd in composite_data: composite_data_resolved.append(abs_path_or_uri(cd, test_data_directory)) target = FileUploadTarget(file_path, composite_data=composite_data_resolved, **kwargs) upload_response = upload_func(target) return response_to_hda(target, upload_response) def upload_object(the_object: Any) -> Dict[str, str]: target = ObjectUploadTarget(the_object) upload_response = upload_func(target) return response_to_hda(target, upload_response) def replacement_item(value, force_to_file: bool = False): is_dict = isinstance(value, dict) item_class = None if not is_dict else value.get("class", None) is_file = item_class == "File" is_directory = item_class == "Directory" is_collection = item_class == "Collection" # Galaxy extension. if force_to_file: if is_file: return replacement_file(value) else: return upload_object(value) if isinstance(value, list): return replacement_list(value) elif not isinstance(value, dict): if tool_or_workflow == "workflow": # All inputs represented as dataset or collection parameters return upload_object(value) else: return value if is_file: return replacement_file(value) elif is_directory: return replacement_directory(value) elif is_collection: return replacement_collection(value) else: return replacement_record(value) def replacement_file(value): if value.get("galaxy_id"): return {"src": "hda", "id": str(value["galaxy_id"])} file_path = value.get("location") or value.get("path") # format to match output definitions in tool, where did filetype come from? filetype = value.get("filetype", None) or value.get("format", None) composite_data_raw = value.get("composite_data", None) kwd = {} if "tags" in value: kwd["tags"] = value.get("tags") if "dbkey" in value: kwd["dbkey"] = value.get("dbkey") if "decompress" in value: kwd["decompress"] = value["decompress"] if composite_data_raw: composite_data = [] for entry in composite_data_raw: path = None if isinstance(entry, dict): path = entry.get("location", None) or entry.get("path", None) else: path = entry composite_data.append(path) rval_c = upload_file_with_composite_data(None, composite_data, filetype=filetype, **kwd) return rval_c if file_path is None: contents = value.get("contents") if contents is not None: return upload_file_literal(contents, **kwd) return value secondary_files = value.get("secondaryFiles", []) secondary_files_tar_path = None if secondary_files: tmp = tempfile.NamedTemporaryFile(delete=False) tf = tarfile.open(fileobj=tmp, mode="w:") order: List[str] = [] index_contents = {"order": order} for secondary_file in secondary_files: secondary_file_path = secondary_file.get("location", None) or secondary_file.get("path", None) assert secondary_file_path, f"Invalid secondaryFile entry found [{secondary_file}]" full_secondary_file_path = os.path.join(test_data_directory, secondary_file_path) basename = secondary_file.get("basename") or os.path.basename(secondary_file_path) order.append(unicodify(basename)) tf.add(full_secondary_file_path, os.path.join(SECONDARY_FILES_EXTRA_PREFIX, basename)) tmp_index = tempfile.NamedTemporaryFile(delete=False, mode="w") json.dump(index_contents, tmp_index) tmp_index.close() tf.add(tmp_index.name, SECONDARY_FILES_INDEX_PATH) tf.close() secondary_files_tar_path = tmp.name return upload_file(file_path, secondary_files_tar_path, filetype=filetype, **kwd) def replacement_directory(value: Dict[str, Any]) -> Dict[str, Any]: file_path = value.get("location", None) or value.get("path", None) if file_path is None: return value if not os.path.isabs(file_path): file_path = os.path.join(test_data_directory, file_path) tmp = tempfile.NamedTemporaryFile(delete=False) tf = tarfile.open(fileobj=tmp, mode="w:") tf.add(file_path, ".") tf.close() return upload_tar(tmp.name) def replacement_list(value) -> Dict[str, str]: collection_element_identifiers = [] for i, item in enumerate(value): dataset = replacement_item(item, force_to_file=True) collection_element = dataset.copy() collection_element["name"] = str(i) collection_element_identifiers.append(collection_element) # TODO: handle nested lists/arrays collection = collection_create_func(collection_element_identifiers, "list") dataset_collections.append(collection) hdca_id = collection["id"] return {"src": "hdca", "id": hdca_id} def to_elements(value, rank_collection_type: str) -> List[Dict[str, Any]]: collection_element_identifiers = [] assert "elements" in value elements = value["elements"] is_nested_collection = ":" in rank_collection_type for element in elements: if not is_nested_collection: # flat collection dataset = replacement_item(element, force_to_file=True) collection_element = dataset.copy() collection_element["name"] = element["identifier"] collection_element_identifiers.append(collection_element) else: # nested collection sub_collection_type = rank_collection_type[rank_collection_type.find(":") + 1 :] collection_element = { "name": element["identifier"], "src": "new_collection", "collection_type": sub_collection_type, "element_identifiers": to_elements(element, sub_collection_type), } collection_element_identifiers.append(collection_element) return collection_element_identifiers def replacement_collection(value: Dict[str, Any]) -> Dict[str, str]: if value.get("galaxy_id"): return {"src": "hdca", "id": str(value["galaxy_id"])} assert "collection_type" in value collection_type = value["collection_type"] elements = to_elements(value, collection_type) collection = collection_create_func(elements, collection_type) dataset_collections.append(collection) hdca_id = collection["id"] return {"src": "hdca", "id": hdca_id} def replacement_record(value): collection_element_identifiers = [] for record_key, record_value in value.items(): if not isinstance(record_value, dict) or record_value.get("class") != "File": dataset = replacement_item(record_value, force_to_file=True) collection_element = dataset.copy() else: dataset = upload_file(record_value["location"], None) collection_element = dataset.copy() collection_element["name"] = record_key collection_element_identifiers.append(collection_element) collection = collection_create_func(collection_element_identifiers, "record") dataset_collections.append(collection) hdca_id = collection["id"] return {"src": "hdca", "id": hdca_id} replace_keys = {} for key, value in job.items(): replace_keys[key] = replacement_item(value) job.update(replace_keys) return job, datasets
def _ensure_file_exists(file_path: str) -> None: if not os.path.exists(file_path): template = "File [%s] does not exist - parent directory [%s] does %sexist, cwd is [%s]" parent_directory = os.path.dirname(file_path) message = template % ( file_path, parent_directory, "" if os.path.exists(parent_directory) else "not ", os.getcwd(), ) raise Exception(message)
[docs]class UploadTarget: @abc.abstractmethod def __str__(self) -> str: pass
[docs]class FileLiteralTarget(UploadTarget):
[docs] def __init__(self, contents: str, **kwargs) -> None: self.contents = contents self.properties = kwargs
def __str__(self) -> str: return f"FileLiteralTarget[contents={self.contents}] with {self.properties}"
[docs]class FileUploadTarget(UploadTarget):
[docs] def __init__( self, path: Optional[str], secondary_files: Optional[str] = None, composite_data: Optional[List[str]] = None, **kwargs, ) -> None: self.path = path self.secondary_files = secondary_files self.composite_data = composite_data self.properties = kwargs
def __str__(self) -> str: return f"FileUploadTarget[path={self.path}] with {self.properties}"
[docs]class ObjectUploadTarget(UploadTarget):
[docs] def __init__(self, the_object: Any) -> None: self.object = the_object self.properties: Dict = {}
def __str__(self) -> str: return f"ObjectUploadTarget[object={self.object}] with {self.properties}"
[docs]class DirectoryUploadTarget(UploadTarget):
[docs] def __init__(self, tar_path: str) -> None: self.tar_path = tar_path
def __str__(self) -> str: return f"DirectoryUploadTarget[tar_path={self.tar_path}]"
GalaxyOutput = namedtuple("GalaxyOutput", ["history_id", "history_content_type", "history_content_id", "metadata"])
[docs]def tool_response_to_output(tool_response, history_id, output_name): for output in tool_response["outputs"]: if output["output_name"] == output_name: return GalaxyOutput(history_id, "dataset", output["id"], None) for output_collection in tool_response["output_collections"]: if output_collection["output_name"] == output_name: return GalaxyOutput(history_id, "dataset_collection", output_collection["id"], None) raise Exception(f"Failed to find output with label [{output_name}]")
[docs]def invocation_to_output(invocation, history_id, output_name): if output_name in invocation["outputs"]: dataset = invocation["outputs"][output_name] return GalaxyOutput(history_id, "dataset", dataset["id"], None) elif output_name in invocation["output_collections"]: collection = invocation["output_collections"][output_name] return GalaxyOutput(history_id, "dataset_collection", collection["id"], None) elif output_name in invocation["output_values"]: output_value = invocation["output_values"][output_name] return GalaxyOutput(None, "raw_value", output_value, None) raise Exception(f"Failed to find output with label [{output_name}] in [{invocation}]")
[docs]def output_to_cwl_json( galaxy_output, get_metadata, get_dataset, get_extra_files, pseudo_location=False, ): """Convert objects in a Galaxy history into a CWL object. Useful in running conformance tests and implementing the cwl-runner interface via Galaxy. """ def element_to_cwl_json(element): object = element["object"] content_type = object.get("history_content_type") metadata = None if content_type is None: content_type = "dataset_collection" metadata = element["object"] metadata["history_content_type"] = content_type element_output = GalaxyOutput( galaxy_output.history_id, content_type, object["id"], metadata, ) return output_to_cwl_json( element_output, get_metadata, get_dataset, get_extra_files, pseudo_location=pseudo_location ) output_metadata = galaxy_output.metadata if output_metadata is None: output_metadata = get_metadata(galaxy_output.history_content_type, galaxy_output.history_content_id) def dataset_dict_to_json_content(dataset_dict): if "content" in dataset_dict: return json.loads(dataset_dict["content"]) else: with open(dataset_dict["path"]) as f: return json.load(f) if galaxy_output.history_content_type == "raw_value": return galaxy_output.history_content_id elif output_metadata["history_content_type"] == "dataset": ext = output_metadata["file_ext"] if ext == "expression.json": dataset_dict = get_dataset(output_metadata) return dataset_dict_to_json_content(dataset_dict) else: file_or_directory = "Directory" if ext == "directory" else "File" secondary_files = [] if file_or_directory == "File": dataset_dict = get_dataset(output_metadata) properties = output_properties(pseudo_location=pseudo_location, **dataset_dict) basename = properties["basename"] extra_files = get_extra_files(output_metadata) found_index = False for extra_file in extra_files: if extra_file["class"] == "File": path = extra_file["path"] if path == SECONDARY_FILES_INDEX_PATH: found_index = True if found_index: ec = get_dataset(output_metadata, filename=SECONDARY_FILES_INDEX_PATH) index = dataset_dict_to_json_content(ec) def dir_listing(dir_path): listing = [] for extra_file in extra_files: path = extra_file["path"] extra_file_class = extra_file["class"] extra_file_basename = os.path.basename(path) if os.path.join(dir_path, extra_file_basename) != path: continue if extra_file_class == "File": ec = get_dataset(output_metadata, filename=path) ec["basename"] = extra_file_basename ec_properties = output_properties(pseudo_location=pseudo_location, **ec) elif extra_file_class == "Directory": ec_properties = {} ec_properties["class"] = "Directory" ec_properties["location"] = ec_basename ec_properties["listing"] = dir_listing(path) else: raise Exception("Unknown output type encountered....") listing.append(ec_properties) return listing for basename in index["order"]: for extra_file in extra_files: path = extra_file["path"] if path != os.path.join(SECONDARY_FILES_EXTRA_PREFIX, basename or ""): continue extra_file_class = extra_file["class"] # This is wrong... if not STORE_SECONDARY_FILES_WITH_BASENAME: ec_basename = basename + os.path.basename(path) else: ec_basename = os.path.basename(path) if extra_file_class == "File": ec = get_dataset(output_metadata, filename=path) ec["basename"] = ec_basename ec_properties = output_properties(pseudo_location=pseudo_location, **ec) elif extra_file_class == "Directory": ec_properties = {} ec_properties["class"] = "Directory" ec_properties["location"] = ec_basename ec_properties["listing"] = dir_listing(path) else: raise Exception("Unknown output type encountered....") secondary_files.append(ec_properties) else: basename = output_metadata.get("created_from_basename") if not basename: basename = output_metadata.get("name") listing: List[OutputPropertiesType] = [] properties = { "class": "Directory", "basename": basename, "listing": listing, } extra_files = get_extra_files(output_metadata) for extra_file in extra_files: if extra_file["class"] == "File": path = extra_file["path"] ec = get_dataset(output_metadata, filename=path) ec["basename"] = os.path.basename(path) ec_properties = output_properties(pseudo_location=pseudo_location, **ec) listing.append(ec_properties) if secondary_files: properties["secondaryFiles"] = secondary_files return properties elif output_metadata["history_content_type"] == "dataset_collection": collection_type = output_metadata["collection_type"].split(":", 1)[0] if collection_type in ["list", "paired"]: rval_l = [] for element in output_metadata["elements"]: rval_l.append(element_to_cwl_json(element)) return rval_l elif collection_type == "record": rval_d = {} for element in output_metadata["elements"]: rval_d[element["element_identifier"]] = element_to_cwl_json(element) return rval_d return None else: raise NotImplementedError("Unknown history content type encountered")
[docs]def download_output(galaxy_output, get_metadata, get_dataset, get_extra_files, output_path): output_metadata = get_metadata(galaxy_output.history_content_type, galaxy_output.history_content_id) dataset_dict = get_dataset(output_metadata) with open(output_path, "wb") as fh: fh.write(dataset_dict["content"])
[docs]def guess_artifact_type(path): tool_or_workflow = "workflow" path, object_id = urllib.parse.urldefrag(path) with open(path) as f: document = yaml.safe_load(f) if "$graph" in document: # Packed document without a process object at the root objects = document["$graph"] if not object_id: object_id = "main" # default object id # Have to use str_removeprefix() instead of rstrip() because only the # first '#' should be removed from the object id matching_objects = [o for o in objects if str_removeprefix(o["id"], "#") == object_id] if len(matching_objects) == 0: raise Exception(f"No process object with id [{object_id}]") if len(matching_objects) > 1: raise Exception(f"Multiple process objects with id [{object_id}]") object_ = matching_objects[0] else: object_ = document tool_or_workflow = "tool" if object_["class"] != "Workflow" else "workflow" return tool_or_workflow