WIP - chart

[dotfiles.git] / data / issues.ipynb

diff --git a/data/issues.ipynb b/data/issues.ipynb
index b731ad103201a471bf606afa37caa395c483d885..735326e771234919ff27dd251456e45fbfdf1a3c 100644 (file)
--- a/data/issues.ipynb
+++ b/data/issues.ipynb
@@ -13,6 +13,18 @@
     "data_files = glob.glob(os.path.expanduser('~\\\\Data\\\\*.json'))\n"
    ]
   },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "import os\n",
+    "\n",
+    "for file in data_files:\n",
+    "\tprint(f\"{file}\", os.path.getsize(file))\n"
+   ]
+  },
   {
    "cell_type": "code",
    "execution_count": 5,
@@ -41,13 +53,138 @@
     "\n",
     "data_sets = []\n",
     "\n",
-    "file = data_files[0]\n",
+    "file = data_files[6]\n",
     "print(file)\n",
     "\n",
     "f = open(file, encoding='utf-16')\n",
-    "data = pd.read_json(f)\n",
+    "df = pd.read_json(f)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 26,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "<class 'pandas._libs.tslibs.timestamps.Timestamp'>\n"
+     ]
+    }
+   ],
+   "source": [
+    "print(type(df.closed[0]))"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "import numpy as np\n",
+    "df['count'] = np.zeros(len(df))\n",
+    "\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "print(df.milestone[22].get('title'))"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 68,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "closed_months = pd.to_datetime(df['closedAt'], format='ISO8601')\n",
+    "df['closed_month'] = closed_months.dt.month"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 69,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "df['mile'] = [d.get('title') if d else None for d in df.milestone]"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 39,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "for_plot = df.groupby(df.closed.dt.month)['url'].count()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 70,
+   "metadata": {},
+   "outputs": [],
+   "source": [
     "\n",
-    "print(data)\n"
+    "grp_df = df.groupby(['mile', 'closed_month']).count().reset_index()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "print(grp_df)"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "#### Plot of closed issues by month for one project\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "for_plot.plot(kind='bar')"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 75,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "import seaborn as sns\n",
+    "import matplotlib.pyplot as plt\n",
+    "sns.set_theme()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "plt.figure(figsize=(10, 6))\n",
+    "sns.lineplot(x='closed_month', y='url', hue='mile', data=grp_df, marker='o', ci=None)\n",
+    "plt.xlabel('Closed Month')\n",
+    "plt.ylabel('ClosedAt')\n",
+    "plt.title('Stacked Line Chart: Milestone Effort by Issue Close Date')\n",
+    "plt.xticks(rotation=45)\n",
+    "plt.legend(title='Mile', loc='upper left')\n",
+    "plt.show()"
    ]
   }
  ],