sslog2csv.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "import ujson as json\n",
    "from csv import DictWriter\n",
    "\n",
    "sslogfile = \"/Users/sm/Code/linedatasimulator/testdata.sslog\"\n",
    "staticfields = ['timestamp', 'counter', 'running', 'sslog_type']\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "with open(sslogfile) as sslog:\n",
    "    lines = [json.loads(line) for line in sslog.readlines()]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "sources = {}\n",
    "for line in lines:\n",
    "    for item in line:\n",
    "        for v in item.values():\n",
    "            sources[v['source']] = sources.get(v['source'], {'dynfields': set()})\n",
    "            sources[v['source']]['dynfields'] = dynamicvals[v['source']]['dynfields'].union(set(v['fieldvalues'].keys()))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "for sourcename, source in sources.items():\n",
    "    source['headerfields'] = sorted(list(source['dynfields']))\n",
    "    source['csvfile'] = open('testsslog_{}.csv'.format(sourcename), 'w+')\n",
    "    source['csvwriter'] = DictWriter(source['csvfile'], fieldnames=staticfields + source['headerfields'])\n",
    "    source['csvwriter'].writeheader()\n",
    "    "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "for line in lines:\n",
    "    for item in line:\n",
    "        for v in item.values():\n",
    "            sourcename = v['source']\n",
    "            source = sources[sourcename]\n",
    "            \n",
    "            row = {k: v[k] for k in staticfields}\n",
    "            row.update({k: v['fieldvalues'].get(k, {}).get('value', None) for k in source['dynfields']})\n",
    "            \n",
    "            source['csvwriter'].writerow(row)\n",
    "            \n",
    "for source in sources.values():\n",
    "    source['csvfile'].close()\n",
    "        \n",
    "        \n",
    "        \n",
    "\n",
    "\n",
    "        \n",
    "            "
   ]
  }
 ],
 "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": "2.7.10"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 1
}