commit b1f6ef7e5528f5863f9ad20a0f0add965fe561d8
Author: Mika <kontakt@donau2space.de>
Date:   Tue Dec 9 14:56:49 2025 +0000

    Add python_sampling_tool/src/python_sampling_tool/main.py

diff --git a/python_sampling_tool/src/python_sampling_tool/main.py b/python_sampling_tool/src/python_sampling_tool/main.py
new file mode 100644
index 0000000..df5daca
--- /dev/null
+++ b/python_sampling_tool/src/python_sampling_tool/main.py
@@ -0,0 +1,75 @@
+import json
+import random
+import argparse
+import os
+from pathlib import Path
+from typing import List, Dict, Any
+
+
+def perform_stratified_sampling(sample_size: int, num_jobs: int) -> List[Dict[str, Any]]:
+    """Generate stratified samples for CI jobs.
+
+    Each job receives an approximately equal share of the sample size.
+    Returns list of Sample-like dicts with job_id and data_points.
+    """
+    if num_jobs <= 0:
+        raise ValueError("num_jobs must be positive")
+    if sample_size <= 0:
+        raise ValueError("sample_size must be positive")
+
+    base = sample_size // num_jobs
+    remainder = sample_size % num_jobs
+    samples = []
+    for i in range(num_jobs):
+        data_points = base + (1 if i < remainder else 0)
+        sample = {
+            "job_id": f"job_{i+1}",
+            "data_points": data_points
+        }
+        samples.append(sample)
+    random.shuffle(samples)
+    return samples
+
+
+def _load_config(path: Path) -> Dict[str, Any]:
+    with path.open("r", encoding="utf-8") as f:
+        cfg = json.load(f)
+    if not isinstance(cfg, dict):
+        raise ValueError("Config JSON must be an object")
+    return cfg
+
+
+def _save_output(samples: List[Dict[str, Any]], path: Path) -> None:
+    path.parent.mkdir(parents=True, exist_ok=True)
+    with path.open("w", encoding="utf-8") as f:
+        json.dump(samples, f, indent=2)
+
+
+def main():
+    parser = argparse.ArgumentParser(description="Stratified sampling tool for Mini CI experiments.")
+    parser.add_argument("--config", required=True, help="Path to sampling config JSON")
+    parser.add_argument("--output", required=False, help="Output JSON file path")
+    args = parser.parse_args()
+
+    config_path = Path(args.config)
+    cfg = _load_config(config_path)
+
+    try:
+        sample_size = int(cfg.get("sample_size", 0))
+        num_jobs = int(cfg.get("num_jobs", 0))
+    except (TypeError, ValueError):
+        raise ValueError("Invalid sample_size or num_jobs in config")
+
+    samples = perform_stratified_sampling(sample_size, num_jobs)
+
+    if args.output:
+        output_path = Path(args.output)
+    else:
+        output_path = Path("output/stratified_samples.json")
+    _save_output(samples, output_path)
+
+    print(f"Generated {len(samples)} stratified samples -> {output_path}")
+
+
+if __name__ == "__main__":
+    main()