From 95e38b2ba318f25e3511999139f708460a61c937 Mon Sep 17 00:00:00 2001
From: frankknoll <Knoll_Frank@web.de>
Date: Sat, 5 Feb 2022 15:01:27 +0100
Subject: [PATCH] refactoring

---
 HowBadIsMyBatch.ipynb | 66 +++++++++++++++++++++++--------------------
 1 file changed, 36 insertions(+), 30 deletions(-)

diff --git a/HowBadIsMyBatch.ipynb b/HowBadIsMyBatch.ipynb
index b587a747521..6a2efd5a818 100644
--- a/HowBadIsMyBatch.ipynb
+++ b/HowBadIsMyBatch.ipynb
@@ -131,6 +131,26 @@
     "        return self.dataFrame[\"VAX_DOSE_SERIES\"].str.contains(dose) if dose is not None else True\n"
    ]
   },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "c62cfaff",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "class AggregationHelper:\n",
+    "\n",
+    "    @staticmethod\n",
+    "    def aggregateFlattenColumnsRenameColumns(dataFrame, aggFunctionsByColumn, columnNameMappingsDict):\n",
+    "        aggregatedDataFrame = dataFrame.agg(aggFunctionsByColumn)\n",
+    "        AggregationHelper._flattenColumns(aggregatedDataFrame)\n",
+    "        return aggregatedDataFrame.rename(columns = columnNameMappingsDict)\n",
+    "\n",
+    "    @staticmethod\n",
+    "    def _flattenColumns(dataFrame):\n",
+    "        dataFrame.columns = [\"_\".join(a) for a in dataFrame.columns.to_flat_index()]\n"
+   ]
+  },
   {
    "cell_type": "code",
    "execution_count": null,
@@ -146,16 +166,14 @@
     "        self.dataFrame = dataFrame        \n",
     "\n",
     "    def createBatchCodeTable(self):\n",
-    "        batchCodeTable = self.dataFrame.groupby('VAX_LOT').agg(\n",
-    "            {\n",
+    "        batchCodeTable = AggregationHelper.aggregateFlattenColumnsRenameColumns(\n",
+    "            dataFrame = self.dataFrame.groupby('VAX_LOT'),\n",
+    "            aggFunctionsByColumn = {\n",
     "                'DIED': ['sum', 'size'],\n",
     "                'L_THREAT': 'sum',\n",
     "                'DISABLE': 'sum'\n",
-    "            })\n",
-    "        self._flattenColumns(batchCodeTable)\n",
-    "        batchCodeTable = batchCodeTable.rename(\n",
-    "            columns =\n",
-    "            {\n",
+    "            },\n",
+    "            columnNameMappingsDict = {\n",
     "                \"DIED_size\": \"ADRs\",\n",
     "                \"DIED_sum\": \"DEATHS\",\n",
     "                \"L_THREAT_sum\": \"LIFE THREATENING ILLNESSES\",\n",
@@ -165,18 +183,16 @@
     "\n",
     "    # create table from https://www.howbadismybatch.com/combined.html\n",
     "    def createSevereEffectsBatchCodeTable(self):\n",
-    "        batchCodeTable = self.dataFrame.groupby('VAX_LOT').agg(\n",
-    "            {\n",
+    "        batchCodeTable = AggregationHelper.aggregateFlattenColumnsRenameColumns(\n",
+    "            dataFrame = self.dataFrame.groupby('VAX_LOT'),\n",
+    "            aggFunctionsByColumn = {\n",
     "                'DIED': ['sum', 'size'],\n",
     "                'L_THREAT': 'sum',\n",
     "                'DISABLE': 'sum',\n",
     "                'HOSPITAL': 'sum',\n",
     "                'ER_VISIT': 'sum'\n",
-    "            })\n",
-    "        self._flattenColumns(batchCodeTable)\n",
-    "        batchCodeTable = batchCodeTable.rename(\n",
-    "            columns =\n",
-    "            {\n",
+    "            },\n",
+    "            columnNameMappingsDict = {\n",
     "                \"DIED_size\": \"ADRs\",\n",
     "                \"DIED_sum\": \"DEATHS\",\n",
     "                \"L_THREAT_sum\": \"LIFE THREATENING ILLNESSES\",\n",
@@ -204,9 +220,6 @@
     "        manufacturerByBatchCodeTable = manufacturerByBatchCodeTable.drop_duplicates(subset = ['VAX_LOT'])\n",
     "        return manufacturerByBatchCodeTable.set_index('VAX_LOT')\n",
     "\n",
-    "    def _flattenColumns(self, batchCodeTable):\n",
-    "        batchCodeTable.columns = [\"_\".join(a) for a in batchCodeTable.columns.to_flat_index()]\n",
-    "\n",
     "\n",
     "class BatchCodeTableFactory:\n",
     "\n",
@@ -233,28 +246,21 @@
     "    \n",
     "    @staticmethod\n",
     "    def getDoseTable(dataFrame):\n",
-    "        doseTable = dataFrame.groupby('VAX_DOSE_SERIES').agg(\n",
-    "            {\n",
+    "        doseTable = AggregationHelper.aggregateFlattenColumnsRenameColumns(\n",
+    "            dataFrame = dataFrame.groupby('VAX_DOSE_SERIES'),\n",
+    "            aggFunctionsByColumn = {\n",
     "                'DIED': ['sum', 'size'],\n",
     "                'L_THREAT': 'sum',\n",
     "                'DISABLE': 'sum'\n",
-    "            })\n",
-    "        DoseAnalysis._flattenColumns(doseTable)\n",
-    "        doseTable = doseTable.rename(\n",
-    "            columns =\n",
-    "            {\n",
+    "            },\n",
+    "            columnNameMappingsDict = {\n",
     "                \"DIED_size\": \"Total reports\",\n",
     "                \"DIED_sum\": \"Deaths\",\n",
     "                \"L_THREAT_sum\": \"Life Threatening Illnesses\",\n",
     "                \"DISABLE_sum\": \"Disabilities\"\n",
     "            })[['Total reports', 'Deaths', 'Disabilities', 'Life Threatening Illnesses']]\n",
     "        doseTable['Severe reports (%)'] = (doseTable['Deaths'] + doseTable['Disabilities'] + doseTable['Life Threatening Illnesses']) / doseTable['Total reports'] * 100\n",
-    "        return doseTable\n",
-    "\n",
-    "    # FK-TODO: DRY with BatchCodeTableHelper\n",
-    "    @staticmethod\n",
-    "    def _flattenColumns(batchCodeTable):\n",
-    "        batchCodeTable.columns = [\"_\".join(a) for a in batchCodeTable.columns.to_flat_index()]\n"
+    "        return doseTable\n"
    ]
   },
   {