finished scrapping all the data

.ipynb_checkpoints/project_notebook-checkpoint.ipynb

@@ -0,0 +1,130 @@
+{
+ "cells": [
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "## Imported modules"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 1,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "import json\n",
+    "import scrapping\n",
+    "import psycopg2\n",
+    "from psycopg2.extras import execute_batch\n",
+    "import re\n",
+    "import logging\n",
+    "import datetime\n",
+    "import multiprocessing\n",
+    "import os\n",
+    "logging.basicConfig(\n",
+    "    level=logging.INFO, filename=f'{datetime.datetime.now().strftime(\"%Y-%m-%d_%H-%M-%S\")}.log', filemode='w',\n",
+    "    format='%(name)s - %(levelname)s - %(message)s', datefmt='%d-%b-%y %H:%M:%S')\n",
+    "# from psycopg2.extensions import register_adapter\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 29,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "conn = psycopg2.connect(\"dbname='dh' user='dh' host='dh.saret.tk' password='qwerty'\")\n",
+    "connection = conn.cursor()\n",
+    "with open('project_list') as f:\n",
+    "    for project in f.read().split('\\n'):\n",
+    "        # connection = connection.execute(\"insert into raw_texts values (%(id_text)s, %(project_name)s, %(raw_text)s)\", scrap)\n",
+    "        try:\n",
+    "            scrap = scrapping.get_raw_english_texts_of_project(project)\n",
+    "        except Exception:\n",
+    "            logging.error(f\"Error in {project}:{Exception}\")\n",
+    "        try:\n",
+    "            execute_batch(\n",
+    "                connection,\n",
+    "                \"insert into raw_texts values (%(id_text)s, %(project_name)s, %(raw_text)s) ON CONFLICT DO NOTHING\",\n",
+    "                scrap)\n",
+    "            conn.commit()\n",
+    "        except Exception:\n",
+    "            with open(project, 'w') as f:\n",
+    "                f.write(json.dumps(scrap))\n",
+    "            logging.error(f\"Error in {project}:{Exception}\")"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 3,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "conn = psycopg2.connect(\"dbname='dh' user='dh' host='dh.saret.tk' password='qwerty'\")\n",
+    "connection = conn.cursor()\n",
+    "logging.basicConfig(\n",
+    "    level=logging.INFO, filename=f'./now.log',\n",
+    "    format='%(name)s - %(levelname)s - %(message)s')\n",
+    "logging.info('Start')\n",
+    "with open('project_list', encoding='unicode-escape') as f:\n",
+    "    # project = f.read().split('\\n')[2]\n",
+    "    for project in f.read().split('\\n'):\n",
+    "        # connection = connection.execute(\"insert into raw_texts values (%(id_text)s, %(project_name)s, %(raw_text)s)\", scrap)\n",
+    "        if os.path.exists(project):\n",
+    "            with open(project) as f:\n",
+    "                logging.info(f'reading {project}')\n",
+    "                scrap = json.loads(f.read())\n",
+    "            # scrap = json.loads(project)\n",
+    "                execute_batch(\n",
+    "                    connection, \"insert into raw_texts values (%(id_text)s, %(project_name)s, %(raw_text)s) ON CONFLICT DO NOTHING\",\n",
+    "                    scrap)\n",
+    "                conn.commit()\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# with open('cams') as file:\n",
+    "conn = psycopg2.connect(\"dbname='dh' user='dh' host='dh.saret.tk' password='qwerty'\")\n",
+    "connection = conn.cursor()\n",
+    "scrap = scrapping.get_raw_english_texts_of_project('saao')\n",
+    "execute_batch(\n",
+    "    connection, \"insert into raw_texts values (%(id_text)s, %(project_name)s, %(raw_text)s) ON CONFLICT DO NOTHING\",\n",
+    "    scrap)\n",
+    "conn.commit()\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "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": "3.9.13"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 2
+}