More docstrings and prints

scripts/cli.py

@@ -3,11 +3,11 @@
 import shutil
 import subprocess
 import sys
-import typer
-from typing_extensions import Annotated
 from typing import Optional
 
+import typer
 from perfcapture.workload import discover_workloads
+from typing_extensions import Annotated
 
 app = typer.Typer()
 
@@ -60,7 +60,7 @@ def bench(
             " vmtouch. Or run with the --keep-cache option, which does not call vmtouch.")
 
     workloads = discover_workloads(recipe_path)
-    print(f"Found {len(workloads)} Workload(s) in {recipe_path}")
+    print(f"Found {len(workloads)} Workload class(es) in {recipe_path}")
 
     # Filter workloads (if necessary).
     if selected_workloads:
@@ -69,9 +69,11 @@ def bench(
 
     # Prepare datasets (if necessary).
     all_datasets = set([workload.dataset for workload in workloads])
+    print(f"Found {len(all_datasets)} Dataset object(s).")
     for dataset in all_datasets:
         dataset.set_path(data_path)
         if not dataset.already_exists():
+            print(f"Creating dataset for {dataset}")
             dataset.prepare()
 
     # Run the workloads!

src/perfcapture/dataset.py

@@ -9,14 +9,25 @@ class Dataset(abc.ABC):
     
     Datasets are read by `Workload`s.
     """
-    def set_path(self, base_data_path: pathlib.Path):
-        self.path = base_data_path / self.name
+    def set_path(self, base_data_path: pathlib.Path) -> None:
+        self._path = base_data_path / self.name
+
+    @property
+    def path(self) -> pathlib.Path:
+        try:
+            return self._path
+        except Exception as e:
+            e.add_note("Run `Dataset.set_path()` before attempting to access `Dataset.path`!")
+            raise
 
     @property
     def name(self) -> str:
         """The name of this dataset.
         
         Must be unique amongst all the datasets used in the benchmark suite.
+        
+        The default implementation will use the name of the class. Override
+        this method if you wish to set a custom name.
         """
         return self.__class__.__name__
 

src/perfcapture/workload.py

@@ -1,8 +1,8 @@
 import abc
 import inspect
 import pathlib
-from perfcapture.dataset import Dataset
 
+from perfcapture.dataset import Dataset
 from perfcapture.utils import load_module_from_filename, path_not_empty
 
 
@@ -14,7 +14,10 @@ def __init__(self):
 
     @abc.abstractmethod
     def init_dataset(self) -> Dataset:
-        """Initialises and returns a concrete Dataset objects."""
+        """Initialises and returns a concrete Dataset objects.
+        
+        Use this method to assign a Dataset object to this Workload.
+        """
 
     @abc.abstractmethod
     def run(self) -> dict[str, object]: