tut1 & tut2

mongodb_tut1.py

@@ -0,0 +1,12 @@
+from pymongo import MongoClient
+
+db = MongoClient('mongodb://143.215.138.132:27017')['big_data']
+
+sample = {'$sample': {'size': 3}}
+
+pipeline = [sample]
+
+query_result = db.tweet.aggregate(pipeline)
+
+for tweet in query_result:
+    print tweet['text']

mongodb_tut2.ipynb

@@ -0,0 +1,229 @@
+{
+ "cells": [
+  {
+   "cell_type": "markdown",
+   "metadata": {
+    "deletable": true,
+    "editable": true
+   },
+   "source": [
+    "## Timestamp Range"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {
+    "collapsed": true,
+    "deletable": true,
+    "editable": true
+   },
+   "outputs": [],
+   "source": [
+    "from pymongo import MongoClient\n",
+    "import matplotlib.pyplot as plt\n",
+    "import numpy as np\n",
+    "\n",
+    "db = MongoClient('mongodb://143.215.138.132:27017')['big_data']"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {
+    "collapsed": true,
+    "deletable": true,
+    "editable": true
+   },
+   "outputs": [],
+   "source": [
+    "# Sample Tweet:\n",
+    "# {   \"_id\" : ObjectId(\"5819377b8548676a7239588c\"),\n",
+    "#     \"guid\" : \"793613871756013568\",\n",
+    "#     \"text\" : \"#incrediblessuperdanceparty can you play all of the Spanish  songs?!\",\n",
+    "#     \"author_id\" : \"770418732665565185\",\n",
+    "#     \"author_name\" : \"Kayle Figueiredo\",\n",
+    "#     \"author_followers_count\" : 15,\n",
+    "#     \"timestamp\" : 1478047266,\n",
+    "#     \"lon\" : -83.804475,\n",
+    "#     \"lat\" : 27.698681999999998,\n",
+    "#     \"words\" : [ \"can\", \"you\", \"play\", \"all\", \"of\", \"the\", \"spanish\", \"songs\" ],\n",
+    "#     \"keywords\" : [ \"play\", \"songs\", \"spanish\" ],\n",
+    "#     \"hashtags\" : [ \"incrediblessuperdanceparty\" ],\n",
+    "#     \"mentions_id\" : [ ],\n",
+    "#     \"mentions_name\" : [ ],\n",
+    "#     \"urls\" : [ ],\n",
+    "#     \"favorite_count\" : 0,\n",
+    "#     \"retweet_count\" : 0 }"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {
+    "collapsed": true,
+    "deletable": true,
+    "editable": true
+   },
+   "outputs": [],
+   "source": [
+    "def find_first_timestamp():\n",
+    "    sort = {'$sort': {'timestamp': 1}}\n",
+    "    limit  = {'$limit': 1}\n",
+    "    pipeline = [sort, limit]\n",
+    "    return list(db.tweet_subset.aggregate(pipeline))[0]['timestamp']"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {
+    "collapsed": true,
+    "deletable": true,
+    "editable": true
+   },
+   "outputs": [],
+   "source": [
+    "def find_last_timestamp():\n",
+    "    sort = {'$sort': {'timestamp': -1}}\n",
+    "    limit  = {'$limit': 1}\n",
+    "    pipeline = [sort, limit]\n",
+    "    return list(db.tweet_subset.aggregate(pipeline))[0]['timestamp']"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {
+    "collapsed": true,
+    "deletable": true,
+    "editable": true
+   },
+   "outputs": [],
+   "source": [
+    "def get_timestamp_list():\n",
+    "    timestamp_list = []\n",
+    "    for tweet in list(db.tweet_subset.find()):\n",
+    "        timestamp_list.append(tweet['timestamp'])\n",
+    "    return timestamp_list"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {
+    "collapsed": false,
+    "deletable": true,
+    "editable": true
+   },
+   "outputs": [],
+   "source": [
+    "%time print find_first_timestamp()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {
+    "collapsed": false,
+    "deletable": true,
+    "editable": true
+   },
+   "outputs": [],
+   "source": [
+    "%time print find_last_timestamp()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {
+    "collapsed": false,
+    "deletable": true,
+    "editable": true
+   },
+   "outputs": [],
+   "source": [
+    "%time print min(get_timestamp_list())"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {
+    "collapsed": false,
+    "deletable": true,
+    "editable": true
+   },
+   "outputs": [],
+   "source": [
+    "%time print max(get_timestamp_list())"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {
+    "deletable": true,
+    "editable": true
+   },
+   "source": [
+    "## Histogram"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {
+    "collapsed": false,
+    "deletable": true,
+    "editable": true
+   },
+   "outputs": [],
+   "source": [
+    "timestamp_list = get_timestamp_list()\n",
+    "\n",
+    "bins = np.linspace(find_first_timestamp(), find_last_timestamp(), 100)\n",
+    "\n",
+    "plt.hist(timestamp_list, bins=bins)\n",
+    "\n",
+    "plt.title('Tweet Timeline')\n",
+    "plt.xlabel('Timestamp')\n",
+    "plt.ylabel('Number of Tweets')\n",
+    "\n",
+    "plt.show()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {
+    "collapsed": true,
+    "deletable": true,
+    "editable": true
+   },
+   "outputs": [],
+   "source": []
+  }
+ ],
+ "metadata": {
+  "kernelspec": {
+   "display_name": "Python 2",
+   "language": "python",
+   "name": "python2"
+  },
+  "language_info": {
+   "codemirror_mode": {
+    "name": "ipython",
+    "version": 2
+   },
+   "file_extension": ".py",
+   "mimetype": "text/x-python",
+   "name": "python",
+   "nbconvert_exporter": "python",
+   "pygments_lexer": "ipython2",
+   "version": "2.7.13"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 2
+}