getRemoteSize

getremotesize.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "getRemoteSize",
      "private_outputs": true,
      "provenance": [],
      "authorship_tag": "ABX9TyOKcYmpN/LdJAoHdlpoflEL",
      "include_colab_link": true
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "language_info": {
      "name": "python"
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/gist/thanakijwanavit/17b7384237a80e598fe5eba952df0c11/getremotesize.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "qRO_caTQM2Bw"
      },
      "source": [
        "!pip install beartype"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "DM-IE4ItKKmQ"
      },
      "source": [
        "import pandas as pd\n",
        "from beartype import beartype\n",
        "import requests"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "5b_tddlfKNMs"
      },
      "source": [
        "url = 'https://inventory-bucket-dev-manual.s3-accelerate.amazonaws.com/pp8pcAY7E%2BSxHbchQ0zDh7gmgpg%3D?AWSAccessKeyId=ASIAVX4Z5TKDSNHIJOSR&Signature=WIniD29GK7nW97IW51%2BEg725pPA%3D&x-amz-security-token=IQoJb3JpZ2luX2VjEGQaDmFwLXNvdXRoZWFzdC0xIkcwRQIhAMAcEWMVtu37Q86MJsfq6xst3KmwdgE098UmmRhN%2FWprAiBIhVe%2Bzz9ZOeYwDEiGz33lnq14SN6kNlbz4SaTyH%2Fo8yrpAQid%2F%2F%2F%2F%2F%2F%2F%2F%2F%2F8BEAEaDDM5NDkyMjkyNDY3OSIMFLjmkMbNep3%2BfCcJKr0BMAaiZ9DOkVD26AoUHxgE6zQi1gz49jQ3I98iVdYRQ08SLZrPfJlM6xRzmaCO4KOpltYRGrqhjtiNHHiSRYb2TiL1cVF5BDJBBA5BoelLWeEdLZgcijV%2Bm4Mg3d05PVHP%2BumCxxpLwfDyY%2FADvs14GXbtaXhgrSNpNwI7%2BWht%2FgO%2B4%2FIwpgvVWsPpm%2Bi0wOOwCBy30t3Nc71I1%2BYxx0KR8hejmedOB3BsS%2BoniUqPHD24JcQzNrKqMWy2PhUqMOyy4IIGOuAByn30YJx3ReKi2Ni%2BNAhR4wB7kUUKW9ZEa0%2BIWw6hg1OPxsGkKViQDJQhJtwyK%2FhdgsGKzSjrVRwoqYgsOGmF%2Bt81tUsZlPuz94ZYcsz7zqitAv25ksRh3NjkYXZQNXvjX%2FzRY20bu3DK%2Bfxf2t6zZo0lRG7bMNPWx5LwXDZt%2BCcdaxuBOL1jvhZbRcn7yTbxjUMTrLzZKP37A6605%2BQpzUGd7ZgymVAH8ulPRNMtQXenvoA1jIDCXNOfTeSr%2BOPiWfqAZ69ghDoXyaxMDngqYMeO66bbwKuSKb6gC4YHSDo%3D&Expires=1616387910'\n",
        "r = requests.get(url)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "saaSMlo8Mg92"
      },
      "source": [
        "# Get decoded size"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "myqLC48iKW8H"
      },
      "source": [
        "import sys\n",
        "sys.getsizeof(r.content)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "fC7F9972KbGd"
      },
      "source": [
        "# Get raw response size"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "4m-PvPmYKqA9"
      },
      "source": [
        "def requestsSize(r:requests.Response):\n",
        "  def header_size(headers):\n",
        "      return sum(len(key) + len(value) + 4 for key, value in headers.items()) + 2\n",
        "  request_line_size = len(r.request.method) + len(r.request.path_url) + 12\n",
        "  request_size = request_line_size + header_size(r.request.headers) + int(r.request.headers.get('content-length', 0))\n",
        "  response_line_size = len(r.reason) + 15\n",
        "  response_size = response_line_size + header_size(r.headers) + int(r.headers.get('content-length', 0))\n",
        "  total_size = request_size + response_size\n",
        "  return total_size"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "NFJQYSVSLaHr"
      },
      "source": [
        "total_size"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "P9kMg5PML_EE"
      },
      "source": [
        "# get size from url"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "nThey5iEMv0d"
      },
      "source": [
        "@beartype\n",
        "def sizeFromUrl(url:str):\n",
        "  r = requests.get(url)\n",
        "  return requestsSize(r)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "0M9D0zJBM_oG"
      },
      "source": [
        "sizeFromUrl(url)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "4duKz7wiNDJD"
      },
      "source": [
        ""
      ],
      "execution_count": null,
      "outputs": []
    }
  ]
}