se_chat_scrape.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# SE Chat Scraper\n",
    "Because http://meta.stackexchange.com/questions/129374/obtaining-full-chat-transcripts"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "import time\n",
    "import requests\n",
    "\n",
    "MAX_MSG_COUNT = 500"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "def get_messages(room_id, before=None, n=MAX_MSG_COUNT, session=None):\n",
    "    \"\"\"SE Chat API endpoint\"\"\"\n",
    "    url = 'http://chat.stackexchange.com/chats/{}/events'.format(room_id)\n",
    "    params = {\n",
    "        'mode': 'Messages',\n",
    "        'msgCount': n,\n",
    "    }\n",
    "    if before is not None:\n",
    "        params['before'] = before\n",
    "    if session is None:\n",
    "        session = requests.session()\n",
    "        \n",
    "    messages = session.post(url, params).json()['events']\n",
    "    earliest = min(msg['message_id'] for msg in messages)\n",
    "    done = len(messages) != n\n",
    "    \n",
    "    return messages, earliest, done\n",
    "\n",
    "\n",
    "def get_all_messages(room_id, throttle=1):\n",
    "    \"\"\"\n",
    "    Get all messages from the provided room ID. Throttle controls the inter-\n",
    "    request delay to avoid being too obnoxious.\n",
    "    \"\"\"\n",
    "    messages = []\n",
    "\n",
    "    s = requests.session()\n",
    "\n",
    "    new_messages, earliest, done = get_messages(room_id, session=s)\n",
    "    messages.extend(new_messages)\n",
    "    while not done:\n",
    "        time.sleep(throttle)\n",
    "        new_messages, earliest, done = get_messages(room_id, before=earliest, session=s)\n",
    "        messages.extend(new_messages)\n",
    "        \n",
    "    return messages"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "# room_id = 27364 # esolang/PCG room\n",
    "room_id = 25038 # smaller JS\n",
    "\n",
    "messages = get_all_messages(room_id)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'content': 'Ok, legal!',\n",
       " 'event_type': 1,\n",
       " 'message_id': 22529004,\n",
       " 'room_id': 25038,\n",
       " 'time_stamp': 1435857785,\n",
       " 'user_id': 144907,\n",
       " 'user_name': 'gustavox'}"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "messages[-1]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "import collections"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "most_active_users = collections.Counter(msg['user_name'] for msg in messages)\n",
    "most_active_users = sorted(\n",
    "    (u[::-1] for u in most_active_users.items() if u[1] > 10), \n",
    "    reverse=True,\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[(296, 'KaduAmaral'),\n",
       " (210, 'Sergio'),\n",
       " (143, 'Luis'),\n",
       " (89, 'gustavox'),\n",
       " (88, 'Renan Rodrigues'),\n",
       " (74, 'Maicon Carraro'),\n",
       " (48, 'Xeoon'),\n",
       " (37, 'Joao Vitor Farias Scheuermann'),\n",
       " (24, 'Ricardo Henrique'),\n",
       " (19, 'Ricardo'),\n",
       " (16, 'stringnome'),\n",
       " (16, 'Alexandre C. Caus'),\n",
       " (15, 'Mayla Campos'),\n",
       " (14, 'ricardo'),\n",
       " (14, 'ctgPi'),\n",
       " (13, 'Julio Santos'),\n",
       " (12, 'Marciano.Andrade')]"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "most_active_users # some people changed name, but have same user_id...but not as pretty to display :P"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "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": "3.5.1"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 0
}

How can I get all of the messages, not just 500?

How can I get all of the messages, not just 500?

I think you'd have to look at the time of the last message, then set before to that.

How can I get all of the messages, not just 500?

I think you'd have to look at the time of the last message, then set before to that.

@Anonymous941 Do you mean the time of the most recent message?

@Anonymous941 Do you mean the time of the most recent message?

@Tech-Expert-Wizard I mean set before to the time of the last message (the 500th message) to get the next 500 messages.

How can I get all of the messages, not just 500?

Use get_all_messages, it repeatedly calls get_messages and sets before to before whatever the last 500 were.

eu estou nessa lista 🤔