Add draft for adding links to wikipedia #673

Author: TobiasNx

conf/addNWBibLinksToWikipedia.ipynb

@@ -0,0 +1,141 @@
+{
+ "cells": [
+  {
+   "cell_type": "markdown",
+   "id": "57a00977-79ce-4d17-b6cc-49dd54bb2af6",
+   "metadata": {},
+   "source": [
+    "## Import pywikibot and rdflib"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 2,
+   "id": "80e390c6-2cb3-4c6c-a260-7c21df6a0831",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "import pywikibot\n",
+    "import rdflib"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "c56365cc-f1ba-4aab-a432-341450258c48",
+   "metadata": {},
+   "source": [
+    "## Query wikidata for wikipedia pages of nwbib spatial concepts/locations"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 3,
+   "id": "0ccae4b0-e678-48f4-bd8e-a20b232c0220",
+   "metadata": {
+    "scrolled": true
+   },
+   "outputs": [],
+   "source": [
+    "g = rdflib.Graph()\n",
+    "qres = g.query(\n",
+    "        \"\"\"\n",
+    "PREFIX wd: <http://www.wikidata.org/entity/>\n",
+    "PREFIX wdt: <http://www.wikidata.org/prop/direct/>\n",
+    "PREFIX schema: <http://schema.org/>\n",
+    "SELECT ?articleName\n",
+    "WHERE {\n",
+    "    SERVICE <https://query.wikidata.org/sparql> {\n",
+    "      ?item wdt:P6814 ?nwbibId.\n",
+    "      ?article schema:about ?item.\n",
+    "      ?article schema:name ?articleName .\n",
+    "      ?article schema:isPartOf <https://de.wikipedia.org/>.\n",
+    "    }\n",
+    "  }\n",
+    "    \"\"\"\n",
+    ")\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 4,
+   "id": "fd2e6410-0ccc-4724-ad1c-5117aecd346d",
+   "metadata": {},
+   "outputs": [],
+   "source": []
+  },
+  {
+   "cell_type": "markdown",
+   "id": "e8f7a562-33de-4291-b345-8aab06b0227f",
+   "metadata": {},
+   "source": [
+    "## Iterate over the query results, check if page exists and get wikipedia pages"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "7db6d415-22c1-4209-8681-b8e694378c1a",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "file1 = open(\"missing.txt\", \"a\")\n",
+    "file2 = open(\"alreadyLinked.txt\", \"a\")\n",
+    "\n",
+    "for row in qres:\n",
+    "    page = pywikibot.Page(pywikibot.Site(), 'row.articleName')\n",
+    "    pageExists = page.exists()\n",
+    "    if  pageExists == True:\n",
+    "        pageText = page.get()\n",
+    "        if \"NWBib\" not in pageText:\n",
+    "            print(page.text)\n",
+    "            # TODO: HOW TO REPLACE THIS IN A SMART WAY\n",
+    "            # test = page.text + \"\\n* {{NWBib}}\"\n",
+    "            # page.put(test, summary='Bot: Test edit')\n",
+    "        else:\n",
+    "            file2.write(row.articleName + \" nwbiblink SCHON DA \\n\")\n",
+    "            print(row.articleName + \" nwbiblink SCHON DA \\n\")\n",
+    "    else:\n",
+    "            file1.write(\"Seite für \" + row.articleName + \" fehlt \\n\")\n",
+    "file1.close()\n",
+    "file2.close()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 3,
+   "id": "f69c6f63-b2c0-47aa-8fd4-a72310b789f9",
+   "metadata": {},
+   "outputs": [],
+   "source": []
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "3ed54705-47f7-4c84-a2f0-5282bd531deb",
+   "metadata": {},
+   "outputs": [],
+   "source": []
+  }
+ ],
+ "metadata": {
+  "kernelspec": {
+   "display_name": "Python 3 (ipykernel)",
+   "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": "3.12.3"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 5
+}