listing_test.ipynb (4019B)
1 { 2 "cells": [ 3 { 4 "cell_type": "code", 5 "execution_count": 1, 6 "metadata": {}, 7 "outputs": [], 8 "source": [ 9 "import txtv\n", 10 "import listing as ls\n", 11 "import bs4\n", 12 "import re" 13 ] 14 }, 15 { 16 "cell_type": "code", 17 "execution_count": 2, 18 "metadata": {}, 19 "outputs": [], 20 "source": [ 21 "inr = txtv.get_page(101)[0]" 22 ] 23 }, 24 { 25 "cell_type": "code", 26 "execution_count": 3, 27 "metadata": {}, 28 "outputs": [ 29 { 30 "name": "stdout", 31 "output_type": "stream", 32 "text": [ 33 " 101 SVT Text Torsdag 17 jan 2019\n", 34 " \u001b[34m \u001b[0m\u001b[34m \u001b[0m\n", 35 " \u001b[34m \u001b[0m\u001b[34m \u001b[0m\u001b[34m \u001b[0m\n", 36 " \u001b[34m \u001b[0m\u001b[34m \u001b[0m\u001b[34m \u001b[0m\u001b[34mSVT Text \u001b[0m\n", 37 " \n", 38 " \u001b[34m \u001b[0m\u001b[34m \u001b[0m\u001b[34m* = efter kl 12 Nyhetsrullen 188 \u001b[0m\n", 39 " \u001b[0m\n", 40 "* S och SD har ökat sedan valet......106\u001b[0m\n", 41 "* Opinionen enligt Novus mätning....160-\u001b[0m\n", 42 " \u001b[0m\n", 43 "* Novus: Förtroenderas för politiker 107\u001b[0m\n", 44 "* SD- och M-väljare mest kritiska....108\u001b[0m\n", 45 " \u001b[0m\n", 46 "* Fler MP-profiler hoppar av.........109\u001b[0m\n", 47 " \u001b[0m\n", 48 "* Migrationsministern S EU-kandidat..110\u001b[0m\n", 49 " \u001b[0m\n", 50 "* Höjt skadestånd efter terrordådet..111\u001b[0m\n", 51 " \u001b[0m\n", 52 "* Städade utan tillstånd hos polisen 112\u001b[0m\n", 53 " \u001b[0m\n", 54 " \u001b[0m\n", 55 " \u001b[2m \u001b[0m\u001b[2m \u001b[0mFler rubriker 102 \u001b[0m \u001b[0m \u001b[0m\n", 56 " \u001b[34m \u001b[0m\u001b[34m \u001b[0m\u001b[34mUtrikes 104 Sport 300 Innehåll 700 \u001b[0m\n" 57 ] 58 } 59 ], 60 "source": [ 61 "txtv.show_page(inr)" 62 ] 63 }, 64 { 65 "cell_type": "code", 66 "execution_count": 5, 67 "metadata": {}, 68 "outputs": [ 69 { 70 "data": { 71 "text/plain": [ 72 "['S och SD har ökat sedan valet......106',\n", 73 " 'Opinionen enligt Novus mätning....160-',\n", 74 " 'Novus: Förtroenderas för politiker 107',\n", 75 " 'SD- och M-väljare mest kritiska....108',\n", 76 " 'Fler MP-profiler hoppar av.........109',\n", 77 " 'Migrationsministern S EU-kandidat..110',\n", 78 " 'Höjt skadestånd efter terrordådet..111',\n", 79 " 'Städade utan tillstånd hos polisen 112']" 80 ] 81 }, 82 "execution_count": 5, 83 "metadata": {}, 84 "output_type": "execute_result" 85 } 86 ], 87 "source": [ 88 "entries = []\n", 89 "for n in inr.children:\n", 90 " if isinstance(n, str):\n", 91 " #print(n, end='')\n", 92 " pass\n", 93 " elif isinstance(n, bs4.element.Tag):\n", 94 " if 'bgB' not in n.attrs['class'] and 'bgY' not in n.attrs['class']:\n", 95 " entries.append(n.get_text())\n", 96 "entries = [e for e in entries if not re.fullmatch(' *', e)]\n", 97 "entries" 98 ] 99 }, 100 { 101 "cell_type": "code", 102 "execution_count": null, 103 "metadata": {}, 104 "outputs": [], 105 "source": [] 106 } 107 ], 108 "metadata": { 109 "kernelspec": { 110 "display_name": "Python 3", 111 "language": "python", 112 "name": "python3" 113 }, 114 "language_info": { 115 "codemirror_mode": { 116 "name": "ipython", 117 "version": 3 118 }, 119 "file_extension": ".py", 120 "mimetype": "text/x-python", 121 "name": "python", 122 "nbconvert_exporter": "python", 123 "pygments_lexer": "ipython3", 124 "version": "3.7.2" 125 } 126 }, 127 "nbformat": 4, 128 "nbformat_minor": 2 129 }