simple parsing of hpt rules into preauthoring autocomplete templates

hpt_rule_parse.ipynb

{
 "metadata": {
  "name": ""
 },
 "nbformat": 3,
 "nbformat_minor": 0,
 "worksheets": [
  {
   "cells": [
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "import re, codecs, json"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 1
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "def parse_rule_template(tokens):\n",
      "#     {'text': 'Chris X', template: [{'text': 'Chris' }, {'type': 'nonterminal'}]},\n",
      "    template_obj = []\n",
      "    for tok in tokens:\n",
      "        if tok == u'[X][X]':\n",
      "            template_obj.append({'type': 'nonterminal'})\n",
      "        else:\n",
      "            template_obj.append({'text': tok})\n",
      "            \n",
      "    return template_obj"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 2
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "def map_template_to_string(template):\n",
      "    string_rep = []\n",
      "    for node in template:\n",
      "        if 'type' in node:\n",
      "            string_rep.append('X')\n",
      "        else:\n",
      "            string_rep.append(node['text'])\n",
      "    return ' '.join(string_rep)"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 3
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "def parse_hpt_rules(lines):\n",
      "    \n",
      "    def parse_rule(hpt_rule):\n",
      "        # remove the final [X], then split into units\n",
      "        source_rule_units = re.sub(r' \\[X\\]$', '', hpt_rule).split()\n",
      "        rule_template = parse_rule_template(source_rule_units)\n",
      "        template_string = map_template_to_string(rule_template)\n",
      "        return {'text': template_string, 'template': rule_template}\n",
      "    \n",
      "    return [parse_rule(line) for line in lines]"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 4
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "rule_file = '/home/chris/projects/preauthoring_ui/data/Alchmey/filter/uniquerule'\n",
      "with codecs.open(rule_file, encoding='utf8') as hpt_rules:\n",
      "    sample_rules = [l.strip() for l in hpt_rules]\n",
      "\n",
      "ui_rules = parse_hpt_rules(sample_rules)"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 5
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "ui_rules[:5]"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 6,
       "text": [
        "[{'template': [{'text': u\"'s\"}], 'text': u\"'s\"},\n",
        " {'template': [{'text': u\"'s\"}, {'type': 'nonterminal'}, {'text': u'.'}],\n",
        "  'text': u\"'s X .\"},\n",
        " {'template': [{'text': u\"'s\"}, {'type': 'nonterminal'}], 'text': u\"'s X\"},\n",
        " {'template': [{'text': u\"'s\"}, {'text': u'consent'}], 'text': u\"'s consent\"},\n",
        " {'template': [{'text': u\"'s\"},\n",
        "   {'text': u'consent'},\n",
        "   {'type': 'nonterminal'},\n",
        "   {'text': u'.'}],\n",
        "  'text': u\"'s consent X .\"}]"
       ]
      }
     ],
     "prompt_number": 6
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "# dump the parsed rules\n",
      "json_representation = json.dumps(ui_rules, indent=4)\n",
      "output = '/home/chris/projects/preauthoring_ui/data/Alchmey/filter/uniquerule.json'\n",
      "with codecs.open(output, 'w', encoding='utf8') as out:\n",
      "    out.write(json_representation)"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 8
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [],
     "language": "python",
     "metadata": {},
     "outputs": []
    }
   ],
   "metadata": {}
  }
 ]
}