txtv

commit 6cdda6bfd09e6328582ef1b87cbae0f928c9e71e
parent 88cf01cbc68460f12d17c4752f669f57d29e8882
Author: Isak Lindhé <isak.e.lindhe@gmail.com>
Date:   Sun, 20 Jan 2019 16:24:32 +0100

experimenting with content listing

Diffstat:
A
notebooks/listing_test.ipynb
 | 
129
+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

1 file changed, 129 insertions(+), 0 deletions(-)
diff --git a/notebooks/listing_test.ipynb b/notebooks/listing_test.ipynb
@@ -0,0 +1,129 @@
+{
+ "cells": [
+  {
+   "cell_type": "code",
+   "execution_count": 1,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "import txtv\n",
+    "import listing as ls\n",
+    "import bs4\n",
+    "import re"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 2,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "inr = txtv.get_page(101)[0]"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 3,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      " 101 SVT Text        Torsdag 17 jan 2019\n",
+      " \u001b[34m \u001b[0m\u001b[34m                                      \u001b[0m\n",
+      " \u001b[34m \u001b[0m\u001b[34m                           \u001b[0m\u001b[34m           \u001b[0m\n",
+      " \u001b[34m \u001b[0m\u001b[34m                          \u001b[0m\u001b[34m \u001b[0m\u001b[34mSVT Text   \u001b[0m\n",
+      "                                        \n",
+      " \u001b[34m \u001b[0m\u001b[34m \u001b[0m\u001b[34m* = efter kl 12     Nyhetsrullen 188 \u001b[0m\n",
+      "                                        \u001b[0m\n",
+      "* S och SD har ökat sedan valet......106\u001b[0m\n",
+      "* Opinionen enligt Novus mätning....160-\u001b[0m\n",
+      "                                        \u001b[0m\n",
+      "* Novus: Förtroenderas för politiker 107\u001b[0m\n",
+      "* SD- och M-väljare mest kritiska....108\u001b[0m\n",
+      "                                        \u001b[0m\n",
+      "* Fler MP-profiler hoppar av.........109\u001b[0m\n",
+      "                                        \u001b[0m\n",
+      "* Migrationsministern S EU-kandidat..110\u001b[0m\n",
+      "                                        \u001b[0m\n",
+      "* Höjt skadestånd efter terrordådet..111\u001b[0m\n",
+      "                                        \u001b[0m\n",
+      "* Städade utan tillstånd hos polisen 112\u001b[0m\n",
+      "                                        \u001b[0m\n",
+      "                                        \u001b[0m\n",
+      "          \u001b[2m \u001b[0m\u001b[2m \u001b[0mFler rubriker 102  \u001b[0m \u001b[0m        \u001b[0m\n",
+      " \u001b[34m \u001b[0m\u001b[34m \u001b[0m\u001b[34mUtrikes 104  Sport 300  Innehåll 700 \u001b[0m\n"
+     ]
+    }
+   ],
+   "source": [
+    "txtv.show_page(inr)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 5,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "['S och SD har ökat sedan valet......106',\n",
+       " 'Opinionen enligt Novus mätning....160-',\n",
+       " 'Novus: Förtroenderas för politiker 107',\n",
+       " 'SD- och M-väljare mest kritiska....108',\n",
+       " 'Fler MP-profiler hoppar av.........109',\n",
+       " 'Migrationsministern S EU-kandidat..110',\n",
+       " 'Höjt skadestånd efter terrordådet..111',\n",
+       " 'Städade utan tillstånd hos polisen 112']"
+      ]
+     },
+     "execution_count": 5,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "entries = []\n",
+    "for n in inr.children:\n",
+    "    if isinstance(n, str):\n",
+    "        #print(n, end='')\n",
+    "        pass\n",
+    "    elif isinstance(n, bs4.element.Tag):\n",
+    "        if 'bgB' not in n.attrs['class'] and 'bgY' not in n.attrs['class']:\n",
+    "            entries.append(n.get_text())\n",
+    "entries = [e for e in entries if not re.fullmatch(' *', e)]\n",
+    "entries"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": []
+  }
+ ],
+ "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": "3.7.2"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 2
+}