executorlib 1.10.2__tar.gz → 1.10.3__tar.gz

{executorlib-1.10.2 → executorlib-1.10.3}/PKG-INFO

@@ -1,6 +1,6 @@
-Metadata-Version: 2.4
+Metadata-Version: 2.5
 Name: executorlib
-Version: 1.10.2
+Version: 1.10.3
 Summary: Up-scale python functions for high performance computing (HPC) with executorlib.
 Project-URL: Homepage, https://github.com/pyiron/executorlib
 Project-URL: Documentation, https://executorlib.readthedocs.io
@@ -49,31 +49,31 @@ Classifier: Programming Language :: Python :: 3.14
 Classifier: Topic :: Scientific/Engineering :: Physics
 Requires-Python: <3.15,>3.9
 Requires-Dist: cloudpickle<=3.1.2,>=2.0.0
-Requires-Dist: pyzmq<=27.1.0,>=25.0.0
+Requires-Dist: pyzmq<=27.2.0,>=25.0.0
 Provides-Extra: all
 Requires-Dist: h5py<=3.16.0,>=3.6.0; extra == 'all'
 Requires-Dist: ipython<=9.9.0,>=7.33.0; extra == 'all'
 Requires-Dist: mpi4py<=4.1.2,>=3.1.4; extra == 'all'
 Requires-Dist: networkx<=3.6.1,>=2.8.8; extra == 'all'
-Requires-Dist: pydantic<=2.13.4,>=2.5.3; extra == 'all'
-Requires-Dist: pygraphviz<=1.14,>=1.10; extra == 'all'
-Requires-Dist: pysqa==0.4.4; extra == 'all'
+Requires-Dist: pydantic<=2.13.5,>=2.5.3; extra == 'all'
+Requires-Dist: pygraphviz<=2.0.1,>=1.10; extra == 'all'
+Requires-Dist: pysqa==1.0.0; extra == 'all'
 Provides-Extra: cache
 Requires-Dist: h5py<=3.16.0,>=3.6.0; extra == 'cache'
 Provides-Extra: cluster
 Requires-Dist: h5py<=3.16.0,>=3.6.0; extra == 'cluster'
-Requires-Dist: pysqa==0.4.4; extra == 'cluster'
+Requires-Dist: pysqa==1.0.0; extra == 'cluster'
 Provides-Extra: graph
 Requires-Dist: networkx<=3.6.1,>=2.8.8; extra == 'graph'
-Requires-Dist: pygraphviz<=1.14,>=1.10; extra == 'graph'
+Requires-Dist: pygraphviz<=2.0.1,>=1.10; extra == 'graph'
 Provides-Extra: graphnotebook
 Requires-Dist: ipython<=9.9.0,>=7.33.0; extra == 'graphnotebook'
 Requires-Dist: networkx<=3.6.1,>=2.8.8; extra == 'graphnotebook'
-Requires-Dist: pygraphviz<=1.14,>=1.10; extra == 'graphnotebook'
+Requires-Dist: pygraphviz<=2.0.1,>=1.10; extra == 'graphnotebook'
 Provides-Extra: mpi
 Requires-Dist: mpi4py<=4.1.2,>=3.1.4; extra == 'mpi'
 Provides-Extra: validation
-Requires-Dist: pydantic<=2.13.4,>=2.5.3; extra == 'validation'
+Requires-Dist: pydantic<=2.13.5,>=2.5.3; extra == 'validation'
 Description-Content-Type: text/markdown
 
 # executorlib

{executorlib-1.10.2 → executorlib-1.10.3}/pyproject.toml

@@ -1,9 +1,9 @@
 [build-system]
 requires = [
-    "hatchling>=1.27.0,<=1.31.0",
+    "hatchling>=1.27.0,<=1.32.0",
     "hatch-vcs==0.5.0",
     "cloudpickle>=2.0.0,<=3.1.2",
-    "pyzmq>=25.0.0,<=27.1.0",
+    "pyzmq>=25.0.0,<=27.2.0",
 ]
 build-backend = "hatchling.build"
 
@@ -31,7 +31,7 @@ classifiers = [
 ]
 dependencies = [
     "cloudpickle>=2.0.0,<=3.1.2",
-    "pyzmq>=25.0.0,<=27.1.0",
+    "pyzmq>=25.0.0,<=27.2.0",
 ]
 dynamic = ["version"]
 
@@ -43,28 +43,28 @@ Repository = "https://github.com/pyiron/executorlib"
 [project.optional-dependencies]
 cache = ["h5py>=3.6.0,<=3.16.0"]
 graph = [
-    "pygraphviz>=1.10,<=1.14",
+    "pygraphviz>=1.10,<=2.0.1",
     "networkx>=2.8.8,<=3.6.1",
 ]
 graphnotebook = [
-    "pygraphviz>=1.10,<=1.14",
+    "pygraphviz>=1.10,<=2.0.1",
     "networkx>=2.8.8,<=3.6.1",
     "ipython>=7.33.0,<=9.9.0",
 ]
 mpi = ["mpi4py>=3.1.4,<=4.1.2"]
 cluster = [
-    "pysqa==0.4.4",
+    "pysqa==1.0.0",
     "h5py>=3.6.0,<=3.16.0",
 ]
 validation = [
-    "pydantic>=2.5.3,<=2.13.4",
+    "pydantic>=2.5.3,<=2.13.5",
 ]
 all = [
     "mpi4py>=3.1.4,<=4.1.2",
-    "pysqa==0.4.4",
+    "pysqa==1.0.0",
     "h5py>=3.6.0,<=3.16.0",
-    "pydantic>=2.5.3,<=2.13.4",
-    "pygraphviz>=1.10,<=1.14",
+    "pydantic>=2.5.3,<=2.13.5",
+    "pygraphviz>=1.10,<=2.0.1",
     "networkx>=2.8.8,<=3.6.1",
     "ipython>=7.33.0,<=9.9.0",
 ]
@@ -102,6 +102,8 @@ ignore = [
     "PLR0912",
     # Too many arguments in function definition
     "PLR0913",
+    # Too many statements
+    "PLR0915",
     # Too many not-keyword-only arguments,
     "PLR0917",
     # Magic value used in comparison

{executorlib-1.10.2 → executorlib-1.10.3}/src/executorlib/__init__.py

@@ -44,6 +44,35 @@ def get_cache_data(cache_directory: str) -> list[dict]:
     return get_cache_data(cache_directory=cache_directory)
 
 
+def get_cache_data_queue(
+    cache_directory: str,
+    queue_type: Optional[str] = None,
+    config_directory: Optional[str] = None,
+) -> list[dict]:
+    """
+    Collect all HDF5 files in the cache directory and check their status in the queue system
+
+    Args:
+        cache_directory (str): The directory to store cache files.
+        queue_type (str, optional): The type of the queue system ["slurm", "flux"].
+        config_directory (str, optional): The directory containing the configuration for the queue system.
+
+    Returns:
+        list[dict]: List of dictionaries each representing on of the HDF5 files in the
+                    cache directory with their status in the queue system.
+    """
+    from executorlib.standalone.hdf import get_cache_data
+    from executorlib.task_scheduler.file.spawner_pysqa import (
+        get_queue_system_cache_data,
+    )
+
+    return get_queue_system_cache_data(
+        cache_dict=get_cache_data(cache_directory=cache_directory),
+        queue_type=queue_type,
+        config_directory=config_directory,
+    )
+
+
 def get_future_from_cache(
     cache_directory: str,
     cache_key: str,

{executorlib-1.10.2 → executorlib-1.10.3}/src/executorlib/_version.py

@@ -18,7 +18,7 @@ version_tuple: tuple[int | str, ...]
 commit_id: str | None
 __commit_id__: str | None
 
-__version__ = version = '1.10.2'
-__version_tuple__ = version_tuple = (1, 10, 2)
+__version__ = version = '1.10.3'
+__version_tuple__ = version_tuple = (1, 10, 3)
 
 __commit_id__ = commit_id = None

{executorlib-1.10.2 → executorlib-1.10.3}/src/executorlib/standalone/interactive/communication.py

@@ -150,9 +150,11 @@ class SocketInterface:
         """
         result = None
         if self._spawner.poll():
-            result = self.send_and_receive_dict(
+            output = self.send_and_receive_dict(
                 input_dict={"shutdown": True, "wait": wait}
-            )["result"]
+            )
+            if "result" in output:
+                result = output["result"]
             self._spawner.shutdown(wait=wait)
         self._reset_socket()
         return result

{executorlib-1.10.2 → executorlib-1.10.3}/src/executorlib/standalone/serialize.py

@@ -34,7 +34,7 @@ def serialize_funct(
     fn_kwargs: Optional[dict] = None,
     resource_dict: Optional[dict] = None,
     cache_key: Optional[str] = None,
-) -> tuple[str, dict]:
+) -> tuple[str, dict, Optional[Exception]]:
     """
     Serialize a function and its arguments and keyword arguments into an HDF5 file.
 
@@ -56,7 +56,8 @@ def serialize_funct(
                                    overwritten by setting the cache_key.
 
     Returns:
-        Tuple[str, dict]: A tuple containing the task key and the serialized data.
+        Tuple[str, dict, Optional[Exception]]: A tuple containing the task key, the serialized data and an
+                                               exception if the serialization failed.
 
     """
     if fn_args is None:
@@ -68,21 +69,25 @@ def serialize_funct(
     if cache_key is not None:
         task_key = cache_key
     else:
-        binary_all = cloudpickle.dumps(
-            {
-                "fn": fn,
-                "args": fn_args,
-                "kwargs": fn_kwargs,
-            }
-        )
-        task_key = _get_function_name(fn=fn) + _get_hash(binary=binary_all)
+        try:
+            binary_all = cloudpickle.dumps(
+                {
+                    "fn": fn,
+                    "args": fn_args,
+                    "kwargs": fn_kwargs,
+                }
+            )
+        except Exception as e:
+            return "", {}, e
+        else:
+            task_key = _get_function_name(fn=fn) + _get_hash(binary=binary_all)
     data = {
         "fn": fn,
         "args": fn_args,
         "kwargs": fn_kwargs,
         "resource_dict": resource_dict,
     }
-    return task_key, data
+    return task_key, data, None
 
 
 def _get_hash(binary: bytes) -> str:

{executorlib-1.10.2 → executorlib-1.10.3}/src/executorlib/task_scheduler/file/shared.py

@@ -129,7 +129,7 @@ def execute_tasks_h5(
                     executor_kwargs=executor_kwargs,
                 )
             )
-            task_key, data_dict = serialize_funct(
+            task_key, data_dict, serialize_exception = serialize_funct(
                 fn=task_dict["fn"],
                 fn_args=task_args,
                 fn_kwargs=task_kwargs,
@@ -137,7 +137,9 @@ def execute_tasks_h5(
                 cache_key=cache_key,
             )
             data_dict["error_log_file"] = error_log_file
-            if task_key not in memory_dict:
+            if serialize_exception is not None:
+                task_dict["future"].set_exception(serialize_exception)
+            elif task_key not in memory_dict:
                 if os.path.join(
                     cache_directory, task_key + "_o.h5"
                 ) not in get_cache_files(cache_directory=cache_directory):
@@ -156,25 +158,31 @@ def execute_tasks_h5(
                                 )
                             )
                         task_dependent_lst = []
-                    process_dict[task_key] = execute_function(
-                        command=get_cache_execute_command(
+                    try:
+                        process_dict[task_key] = execute_function(
+                            command=get_cache_execute_command(
+                                file_name=file_name,
+                                cores=task_resource_dict["cores"],
+                                backend=backend,
+                                exclusive=task_resource_dict.get("exclusive", False),
+                                openmpi_oversubscribe=task_resource_dict.get(
+                                    "openmpi_oversubscribe", False
+                                ),
+                                pmi_mode=pmi_mode,
+                            ),
                             file_name=file_name,
-                            cores=task_resource_dict["cores"],
+                            data_dict=data_dict,
+                            task_dependent_lst=task_dependent_lst,
+                            resource_dict=task_resource_dict,
+                            config_directory=pysqa_config_directory,
                             backend=backend,
-                            exclusive=task_resource_dict.get("exclusive", False),
-                            openmpi_oversubscribe=task_resource_dict.get(
-                                "openmpi_oversubscribe", False
-                            ),
-                            pmi_mode=pmi_mode,
-                        ),
-                        file_name=file_name,
-                        data_dict=data_dict,
-                        task_dependent_lst=task_dependent_lst,
-                        resource_dict=task_resource_dict,
-                        config_directory=pysqa_config_directory,
-                        backend=backend,
-                        cache_directory=cache_directory,
-                    )
+                            cache_directory=cache_directory,
+                        )
+                    except Exception as e:
+                        task_dict["future"].set_exception(e)
+                        memory_dict[task_key] = task_dict["future"]
+                        future_queue.task_done()
+                        continue
                 file_name = os.path.join(cache_directory, task_key + "_o.h5")
                 file_name_dict[task_key] = file_name
                 queue_id = get_queue_id(file_name=file_name)

{executorlib-1.10.2 → executorlib-1.10.3}/src/executorlib/task_scheduler/file/spawner_pysqa.py

@@ -2,6 +2,7 @@ import os
 from subprocess import CalledProcessError
 from typing import Optional
 
+from pandas import DataFrame
 from pysqa import QueueAdapter
 
 from executorlib.standalone.command_pysqa import pysqa_execute_command, pysqa_terminate
@@ -154,3 +155,58 @@ def terminate_task_in_cache(
             backend=backend,
         )
     os.remove(file_name)
+
+
+def get_queue_system_cache_data(
+    cache_dict: list[dict],
+    queue_type: Optional[str] = None,
+    config_directory: Optional[str] = None,
+) -> list[dict]:
+    """
+    Get the status of the queue system.
+
+    Args:
+        cache_dict (list[dict]): List of dictionaries each representing on of the HDF5 files in the cache directory.
+        queue_type (str, optional): The type of the queue system ["slurm", "flux"].
+        config_directory (str, optional): The directory containing the configuration for the queue system.
+
+    Returns:
+        cache_dict (list[dict]): List of dictionaries each representing on of the HDF5 files in the cache directory.
+    """
+    return _merge_cache_with_queue_status(
+        cache_dict=cache_dict,
+        dataframe=QueueAdapter(
+            directory=config_directory,
+            queue_type=queue_type,
+        ).get_queue_status(),
+    )
+
+
+def _merge_cache_with_queue_status(
+    cache_dict: list[dict], dataframe: DataFrame
+) -> list[dict]:
+    """
+    Merge the cache data with the queue status.
+
+    Args:
+        cache_dict (list[dict]): List of dictionaries each representing on of the HDF5 files in the cache directory.
+        dataframe (DataFrame): DataFrame containing the status of the queue system.
+
+    Returns:
+        cache_dict (list[dict]): List of dictionaries each representing on of the HDF5 files in the cache directory.
+    """
+    cache_updated_dict = []
+    for row in cache_dict:
+        if "output" in row and row["output"]:
+            row["status"] = "finished"
+        elif "queue_id" in row and row["queue_id"] in dataframe["jobid"].values:
+            row["status"] = dataframe[dataframe["jobid"] == row["queue_id"]][
+                "status"
+            ].values[-1]
+        elif "queue_id" not in row:
+            row["status"] = "running"
+        else:
+            row["status"] = "aborted"
+        cache_updated_dict.append(row)
+
+    return cache_updated_dict

{executorlib-1.10.2 → executorlib-1.10.3}/src/executorlib/task_scheduler/interactive/blockallocation.py

@@ -80,35 +80,37 @@ class BlockAllocationTaskScheduler(TaskSchedulerBase):
         executor_kwargs["restart_limit"] = restart_limit
         self._process_kwargs = executor_kwargs
         self._max_workers = max_workers
-        self_id = random.getrandbits(128)
-        self._self_id = self_id
+        self._self_id = random.getrandbits(128)
         _interrupt_bootup_dict[self._self_id] = False
-        alive_workers = [max_workers]
-        alive_workers_lock = Lock()
-        bootup_events = [Event() for _ in range(self._max_workers)]
-        bootup_events[0].set()
+        self._alive_workers = [max_workers]
+        self._alive_workers_lock = Lock()
+        self._bootup_events = [Event() for _ in range(self._max_workers)]
+        self._bootup_events[0].set()
         self._set_process(
             process=[
                 Thread(
                     target=_execute_multiple_tasks,
-                    kwargs=executor_kwargs
-                    | {
-                        "worker_id": worker_id,
-                        "stop_function": lambda: _interrupt_bootup_dict[self_id],
-                        "bootup_event": bootup_events[worker_id],
-                        "next_bootup_event": (
-                            bootup_events[worker_id + 1]
-                            if worker_id + 1 < self._max_workers
-                            else None
-                        ),
-                        "alive_workers": alive_workers,
-                        "alive_workers_lock": alive_workers_lock,
-                    },
+                    kwargs=self._worker_kwargs(worker_id),
                 )
                 for worker_id in range(self._max_workers)
             ],
         )
 
+    def _worker_kwargs(self, worker_id: int) -> dict:
+        self_id = self._self_id
+        return self._process_kwargs | {
+            "worker_id": worker_id,
+            "stop_function": lambda: _interrupt_bootup_dict[self_id],
+            "bootup_event": self._bootup_events[worker_id],
+            "next_bootup_event": (
+                self._bootup_events[worker_id + 1]
+                if worker_id + 1 < len(self._bootup_events)
+                else None
+            ),
+            "alive_workers": self._alive_workers,
+            "alive_workers_lock": self._alive_workers_lock,
+        }
+
     @property
     def max_workers(self) -> int:
         return self._max_workers
@@ -126,12 +128,20 @@ class BlockAllocationTaskScheduler(TaskSchedulerBase):
                         process for process in self._process if process.is_alive()
                     ]
             elif self._max_workers < max_workers:
+                old_max_workers = self._max_workers
+                self._bootup_events.extend(
+                    Event() for _ in range(max_workers - old_max_workers)
+                )
+                for idx in range(old_max_workers, max_workers):
+                    self._bootup_events[idx].set()
+                with self._alive_workers_lock:
+                    self._alive_workers[0] += max_workers - old_max_workers
                 new_process_lst = [
                     Thread(
                         target=_execute_multiple_tasks,
-                        kwargs=self._process_kwargs,
+                        kwargs=self._worker_kwargs(worker_id),
                     )
-                    for _ in range(max_workers - self._max_workers)
+                    for worker_id in range(old_max_workers, max_workers)
                 ]
                 for process_instance in new_process_lst:
                     process_instance.start()

{executorlib-1.10.2 → executorlib-1.10.3}/src/executorlib/task_scheduler/interactive/shared.py

@@ -101,13 +101,17 @@ def _execute_task_without_cache(
     Returns:
         bool: True if the task was submitted successfully, False otherwise.
     """
-    output = interface.send_and_receive_dict(input_dict=task_dict)
-    if "result" in output:
-        future_obj.set_result(output["result"])
-    elif isinstance(output["error"], ExecutorlibSocketError):
-        return False
+    try:
+        output = interface.send_and_receive_dict(input_dict=task_dict)
+    except Exception as e:
+        future_obj.set_exception(exception=e)
     else:
-        future_obj.set_exception(exception=output["error"])
+        if "result" in output:
+            future_obj.set_result(output["result"])
+        elif isinstance(output["error"], ExecutorlibSocketError):
+            return False
+        else:
+            future_obj.set_exception(exception=output["error"])
     return True
 
 
@@ -132,27 +136,34 @@ def _execute_task_with_cache(
     """
     from executorlib.standalone.hdf import dump, get_cache_files, get_output
 
-    task_key, data_dict = serialize_funct(
+    task_key, data_dict, serialize_exception = serialize_funct(
         fn=task_dict["fn"],
         fn_args=task_dict["args"],
         fn_kwargs=task_dict["kwargs"],
         resource_dict=task_dict.get("resource_dict", {}),
         cache_key=cache_key,
     )
-    file_name = os.path.abspath(os.path.join(cache_directory, task_key + "_o.h5"))
-    if file_name not in get_cache_files(cache_directory=cache_directory):
-        time_start = time.time()
-        output = interface.send_and_receive_dict(input_dict=task_dict)
-        if "result" in output:
-            data_dict["output"] = output["result"]
-            data_dict["runtime"] = time.time() - time_start
-            dump(file_name=file_name, data_dict=data_dict)
-            future_obj.set_result(output["result"])
-        elif isinstance(output["error"], ExecutorlibSocketError):
-            return False
-        else:
-            future_obj.set_exception(exception=output["error"])
+    if serialize_exception is not None:
+        future_obj.set_exception(exception=serialize_exception)
     else:
-        _, _, result = get_output(file_name=file_name)
-        future_obj.set_result(result)
+        file_name = os.path.abspath(os.path.join(cache_directory, task_key + "_o.h5"))
+        if file_name not in get_cache_files(cache_directory=cache_directory):
+            time_start = time.time()
+            try:
+                output = interface.send_and_receive_dict(input_dict=task_dict)
+            except Exception as e:
+                future_obj.set_exception(exception=e)
+                return True
+            if "result" in output:
+                data_dict["output"] = output["result"]
+                data_dict["runtime"] = time.time() - time_start
+                dump(file_name=file_name, data_dict=data_dict)
+                future_obj.set_result(output["result"])
+            elif isinstance(output["error"], ExecutorlibSocketError):
+                return False
+            else:
+                future_obj.set_exception(exception=output["error"])
+        else:
+            _, _, result = get_output(file_name=file_name)
+            future_obj.set_result(result)
     return True