From 493679d565ea88bac25c7f050702f6839eb840ad Mon Sep 17 00:00:00 2001
From: Mika <kontakt@donau2space.de>
Date: Mon, 30 Mar 2026 16:33:37 +0000
Subject: [PATCH] Add data_analysis/src/data_analysis/core.py

---
 data_analysis/src/data_analysis/core.py | 60 +++++++++++++++++++++++++
 1 file changed, 60 insertions(+)
 create mode 100644 data_analysis/src/data_analysis/core.py

diff --git a/data_analysis/src/data_analysis/core.py b/data_analysis/src/data_analysis/core.py
new file mode 100644
index 0000000..47da5f2
--- /dev/null
+++ b/data_analysis/src/data_analysis/core.py
@@ -0,0 +1,60 @@
+from typing import List, Dict, Any
+import pandas as pd
+import statistics
+import logging
+
+
+logging.basicConfig(level=logging.INFO)
+logger = logging.getLogger(__name__)
+
+
+def _validate_input(run_data_list: List[Dict[str, Any]]) -> None:
+    if not isinstance(run_data_list, list):
+        raise TypeError("run_data_list must be a list of dictionaries")
+    for idx, item in enumerate(run_data_list):
+        if not isinstance(item, dict):
+            raise TypeError(f"Item at index {idx} must be a dict")
+        required_fields = {"epoch_ms", "run_id", "retry_tail_p99", "band_width"}
+        if not required_fields.issubset(item.keys()):
+            missing = required_fields - item.keys()
+            raise ValueError(f"Missing required fields: {missing}")
+        if not isinstance(item["epoch_ms"], int):
+            raise TypeError("epoch_ms must be int")
+        if not isinstance(item["run_id"], str):
+            raise TypeError("run_id must be str")
+        if not isinstance(item["retry_tail_p99"], (int, float)):
+            raise TypeError("retry_tail_p99 must be numeric")
+        if not isinstance(item["band_width"], (int, float)):
+            raise TypeError("band_width must be numeric")
+
+
+def analyze_data(run_data_list: List[Dict[str, Any]]) -> Dict[str, Any]:
+    """Analysiert eine Liste von Laufdaten und berechnet Median, IQR und Stabilitätsindikator."""
+    _validate_input(run_data_list)
+
+    if not run_data_list:
+        raise ValueError("run_data_list darf nicht leer sein")
+
+    df = pd.DataFrame(run_data_list)
+
+    try:
+        value_series = df["retry_tail_p99"].astype(float)
+        median_val = float(value_series.median())
+        q75 = value_series.quantile(0.75)
+        q25 = value_series.quantile(0.25)
+        iqr_val = float(q75 - q25)
+    except Exception as e:
+        logger.error("Fehler bei der Berechnung: %s", e)
+        raise
+
+    # Einfache Stabilitätsregel: relative IQR < 10% vom Median
+    stability = bool((median_val != 0) and (iqr_val / abs(median_val) < 0.1))
+
+    result = {
+        "median": median_val,
+        "iqr": iqr_val,
+        "stability": stability,
+    }
+
+    assert all(k in result for k in ("median", "iqr", "stability")), "Output fields missing"
+    return result