feat(harvest): notebook to help compare conversions

Author: thekaveman

notebooks/verify-harvest.ipynb

@@ -0,0 +1,179 @@
+{
+  "cells": [
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {},
+      "outputs": [],
+      "source": [
+        "import os\n",
+        "\n",
+        "import pandas as pd"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {},
+      "outputs": [],
+      "source": [
+        "google_file = os.environ[\"HARVEST_GOOGLE_DATA\"]\n",
+        "local_file = os.environ[\"HARVEST_DATA\"]\n",
+        "\n",
+        "df_google = pd.read_csv(google_file)\n",
+        "df_local = pd.read_csv(local_file)"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {},
+      "outputs": [],
+      "source": [
+        "df_google.info()"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {},
+      "outputs": [],
+      "source": [
+        "df_local.info()"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {},
+      "outputs": [],
+      "source": [
+        "df_google.columns.to_list()"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {},
+      "outputs": [],
+      "source": [
+        "df_local.columns.to_list()"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {},
+      "outputs": [],
+      "source": [
+        "df_local.columns.equals(df_google.columns)"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {},
+      "outputs": [],
+      "source": [
+        "df_google[\"Hours\"].sum()"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {},
+      "outputs": [],
+      "source": [
+        "df_local[\"Hours\"].sum()"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {},
+      "outputs": [],
+      "source": [
+        "len(df_google)"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {},
+      "outputs": [],
+      "source": [
+        "len(df_local)"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {},
+      "outputs": [],
+      "source": [
+        "google_groups = df_google.groupby([\"Project\", \"Last name\"])\n",
+        "google_sums = {}\n",
+        "for group in google_groups.groups:\n",
+        "    project, last_name = group\n",
+        "    group_df = google_groups.get_group(group)\n",
+        "    google_sums[group] = group_df['Hours'].sum()"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {},
+      "outputs": [],
+      "source": [
+        "local_groups = df_local.groupby([\"Project\", \"Last name\"])\n",
+        "local_sums = {}\n",
+        "for group in local_groups.groups:\n",
+        "    project, last_name = group\n",
+        "    group_df = local_groups.get_group(group)\n",
+        "    local_sums[group] = group_df['Hours'].sum()"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {},
+      "outputs": [],
+      "source": [
+        "len(local_sums) == len(google_sums)"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {},
+      "outputs": [],
+      "source": [
+        "for k,v in local_sums.items():\n",
+        "    print(k)\n",
+        "    assert k in google_sums\n",
+        "    assert v == google_sums[k]"
+      ]
+    }
+  ],
+  "metadata": {
+    "kernelspec": {
+      "display_name": "Python 3",
+      "language": "python",
+      "name": "python3"
+    },
+    "language_info": {
+      "codemirror_mode": {
+        "name": "ipython",
+        "version": 3
+      },
+      "file_extension": ".py",
+      "mimetype": "text/x-python",
+      "name": "python",
+      "nbconvert_exporter": "python",
+      "pygments_lexer": "ipython3",
+      "version": "3.11.6"
+    }
+  },
+  "nbformat": 4,
+  "nbformat_minor": 2
+}