Merge pull request #2068 from jupyter-naas/2066-python-get-release-da…

…tes-from-pypi

feat: python release dates

PyPI/PyPI_Get_release_dates_from_package.ipynb

@@ -0,0 +1,314 @@
+{
+ "cells": [
+  {
+   "cell_type": "markdown",
+   "id": "72a216be-10dc-4c1d-a3fa-a9ce992d1c27",
+   "metadata": {
+    "papermill": {},
+    "tags": []
+   },
+   "source": [
+    "<img width=\"10%\" alt=\"Naas\" src=\"https://landen.imgix.net/jtci2pxwjczr/assets/5ice39g4.png?w=160\"/>"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "25f38b3f-5cb5-4700-ba73-fb24646dd3ca",
+   "metadata": {
+    "papermill": {},
+    "tags": []
+   },
+   "source": [
+    "# PyPI - Get release dates from package\n",
+    "<a href=\"https://app.naas.ai/user-redirect/naas/downloader?url=https://raw.githubusercontent.com/jupyter-naas/awesome-notebooks/master/PyPI/PyPI_Get_number_of_downloads_any_package.ipynb\" target=\"_parent\"><img src=\"https://naasai-public.s3.eu-west-3.amazonaws.com/Open_in_Naas_Lab.svg\"/></a><br><br><a href=\"https://bit.ly/3JyWIk6\">Give Feedbacks</a> | <a href=\"https://app.naas.ai/user-redirect/naas/downloader?url=https://raw.githubusercontent.com/jupyter-naas/awesome-notebooks/master/Naas/Naas_Start_data_product.ipynb\" target=\"_parent\">Generate Data Product</a>"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "4da75a0d-6fba-4115-8667-429956d5ffa0",
+   "metadata": {
+    "papermill": {},
+    "tags": []
+   },
+   "source": [
+    "**Tags:** #pypi #downloads #package #operations #analytics #plotly #html #csv #image #png"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "ab9c2b57-cd13-4086-8a8c-676006cdaebf",
+   "metadata": {
+    "papermill": {},
+    "tags": []
+   },
+   "source": [
+    "**Author:** [Mardiat-Iman](https://www.linkedin.com/in/mardiat-iman-ibrahim-imam-726027262)"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "963551f0-c489-4f73-b4d0-7699892623bf",
+   "metadata": {
+    "papermill": {},
+    "tags": []
+   },
+   "source": [
+    "**Last update:** 2023-07-27 (Created: 2023-07-27)"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "naas-description",
+   "metadata": {
+    "papermill": {},
+    "tags": [
+     "description"
+    ]
+   },
+   "source": [
+    "**Description:** This notebook get the release dates a package from the Python Package Index (PyPI) and plot a Barchart and Scatter Plot to display the release by month. \n",
+    "\n",
+    "NB: We have noticed that sometimes not all versions are accessible via this endpoint in comparison with the website. Please let us know if you manage to find a solution to this issue, we would appreciate."
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "922c974a-a4f5-41b1-af0e-83db3798dee4",
+   "metadata": {
+    "papermill": {},
+    "tags": []
+   },
+   "source": [
+    "## Input"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "ff478c09-fc56-440d-9aaf-4b6f018f3395",
+   "metadata": {
+    "papermill": {},
+    "tags": []
+   },
+   "source": [
+    "### Import libraries"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "93b77ec7-1f35-4f2e-9770-2bfef3ee3752",
+   "metadata": {
+    "papermill": {},
+    "tags": []
+   },
+   "outputs": [],
+   "source": [
+    "import requests\n",
+    "import numpy as np\n",
+    "import matplotlib.pyplot as plt\n",
+    "import matplotlib.dates as mdates\n",
+    "from dateutil.parser import parse"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "50b727c1-81d4-4976-8a14-2ff4cfb418b6",
+   "metadata": {
+    "papermill": {},
+    "tags": []
+   },
+   "source": [
+    "### Setup variables"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "2c16f635-fbb6-4921-9777-4a4013dcfa37",
+   "metadata": {
+    "papermill": {},
+    "tags": []
+   },
+   "outputs": [],
+   "source": [
+    "# Allow the user to input the package name\n",
+    "user_input = input(\"Enter the package name to fetch release dates from PyPI: \")\n",
+    "\n",
+    "# Extract the package name (remove version number if present)\n",
+    "package_name = user_input.split()[0]\n",
+    "print('Package:', package_name)"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "36f9d891-dd69-49e0-9288-ba6cf1566198",
+   "metadata": {
+    "papermill": {},
+    "tags": []
+   },
+   "source": [
+    "## Model"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "2beb4527-eb8c-40c3-8e99-a49c80018e3f",
+   "metadata": {},
+   "source": [
+    "### Get releases dates"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "22bcc4b1-bb81-4053-81fe-19675d4c0f19",
+   "metadata": {
+    "papermill": {},
+    "tags": []
+   },
+   "outputs": [],
+   "source": [
+    "def get_release_dates(package_name):\n",
+    "    # Init \n",
+    "    release_dates = []\n",
+    "    \n",
+    "    # Request\n",
+    "    url = f\"https://pypi.org/pypi/{package_name}/json\"\n",
+    "    response = requests.get(url)\n",
+    "    \n",
+    "    # Response\n",
+    "    if response.status_code == 200:\n",
+    "        data = response.json()\n",
+    "        releases = data.get(\"releases\", {})\n",
+    "        release_dates = [parse(release_info[0][\"upload_time\"]) for release_info in releases.values() if release_info]\n",
+    "        if not releases:\n",
+    "            print(f\"No releases found for package '{package_name}'.\")\n",
+    "        else:\n",
+    "            for version, release_info in releases.items():\n",
+    "                if release_info and isinstance(release_info, list):\n",
+    "                    latest_release = max(release_info, key=lambda x: x[\"upload_time\"])\n",
+    "                    release_date = latest_release[\"upload_time\"]\n",
+    "                    print(f\"Version: {version}, Release Date: {release_date}\")\n",
+    "    else:\n",
+    "        print(f\"Error: Package '{package_name}' not found or error in fetching data.\")\n",
+    "        print(\"Raw JSON Response:\", response.text)\n",
+    "    return release_dates\n",
+    "\n",
+    "# Call the function with the package name to get the release dates\n",
+    "release_dates = get_release_dates(package_name)\n",
+    "print(\"Release dates found:\", len(release_dates))"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "b7077755-36d0-4e95-a581-86c5764cc383",
+   "metadata": {
+    "papermill": {},
+    "tags": []
+   },
+   "source": [
+    "## Output"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "d0b568d5-3904-4dbd-b1e8-476ad4ae9a61",
+   "metadata": {
+    "papermill": {},
+    "tags": []
+   },
+   "source": [
+    "### Plotting a Barchart and Scatter Plot for the release dates"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "70713839-101a-4d57-af31-3340bc13b857",
+   "metadata": {
+    "papermill": {},
+    "tags": []
+   },
+   "outputs": [],
+   "source": [
+    "# Check if release dates were obtained successfully\n",
+    "if release_dates:\n",
+    "    # 1. Sort release dates in ascending order\n",
+    "    release_dates.sort()\n",
+    "\n",
+    "    # 2. Bar Chart - Releases per Month\n",
+    "    release_months = [date.strftime('%Y-%m') for date in release_dates]\n",
+    "    unique_months, month_counts = np.unique(release_months, return_counts=True)\n",
+    "    unique_months = [parse(month + '-01') for month in unique_months]  # Convert back to datetime objects\n",
+    "    fig, ax = plt.subplots(figsize=(30, 15), constrained_layout=True)\n",
+    "    ax.set(title=f\"Bar Chart - Releases per Month for {package_name.capitalize()}\")\n",
+    "    ax.bar(unique_months, month_counts, color=\"tab:red\", width=20)\n",
+    "    ax.xaxis.set_major_locator(mdates.MonthLocator())\n",
+    "    ax.xaxis.set_major_formatter(mdates.DateFormatter(\"%b %Y\"))\n",
+    "    ax.set_xlabel(\"Month\")\n",
+    "    ax.set_ylabel(\"Number of Releases\")\n",
+    "    plt.xticks(rotation=45, ha=\"right\")\n",
+    "    ax.spines[[\"top\", \"right\"]].set_visible(False)\n",
+    "    plt.savefig(\"bar_chart.png\")  # Save the figure\n",
+    "    plt.show()\n",
+    "else:\n",
+    "    print(\"No release dates found for the specified package.\")"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "d0396b1b-00d3-45c0-af1d-af07daa6f07c",
+   "metadata": {
+    "papermill": {},
+    "tags": []
+   },
+   "source": [
+    "### Save the figure"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "52b4015d-f6ab-4bb8-a0a5-f7ac5d824bbc",
+   "metadata": {
+    "papermill": {},
+    "tags": []
+   },
+   "outputs": [],
+   "source": [
+    "fig.savefig(\"bar_chart.png\")"
+   ]
+  }
+ ],
+ "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": "3.9.6"
+  },
+  "naas": {
+   "notebook_id": "8da37e2aaf56e0b26e7ee696dc16f492d45a2e204454444b107880aac4b15824",
+   "notebook_path": "PyPI/PyPI_Get_number_of_downloads_any_package.ipynb"
+  },
+  "papermill": {
+   "default_parameters": {},
+   "environment_variables": {},
+   "parameters": {},
+   "version": "2.3.4"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 5
+}