Shlukuje metadata z registru smluv

bez-cen.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "import xmltodict\n",
    "import pandas as pd\n",
    "import numpy as np"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "with open('dump_2016_07.xml') as f:\n",
    "    d = xmltodict.parse(f.read())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "z = d['dump']['zaznam']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "450"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(z)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "OrderedDict([('identifikator',\n",
       "              OrderedDict([('idSmlouvy', '9'), ('idVerze', '9')])),\n",
       "             ('odkaz', 'https://smlouvy.gov.cz/smlouva/9'),\n",
       "             ('casZverejneni', '2016-07-01T01:17:36+02:00'),\n",
       "             ('smlouva',\n",
       "              OrderedDict([('subjekt',\n",
       "                            OrderedDict([('datovaSchranka', 'c5kbvkw'),\n",
       "                                         ('nazev', 'Liberecký kraj'),\n",
       "                                         ('ico', '70891508'),\n",
       "                                         ('adresa',\n",
       "                                          'U Jezu 642/2a, 46001 Liberec, CZ')])),\n",
       "                           ('smluvniStrana',\n",
       "                            OrderedDict([('datovaSchranka', 'sxk8tap'),\n",
       "                                         ('nazev', 'GORDIC spol. s r.o.'),\n",
       "                                         ('ico', '47903783'),\n",
       "                                         ('adresa',\n",
       "                                          'Erbenova 2108/4, 58601, Jihlava, Jihlava'),\n",
       "                                         ('prijemce', '1')])),\n",
       "                           ('predmet',\n",
       "                            'rozšíření jádra IS Ginis o funkcionalitu zveřejňování smluv'),\n",
       "                           ('datumUzavreni', '2016-06-30'),\n",
       "                           ('cisloSmlouvy', '12-OBJ/893/2016'),\n",
       "                           ('schvalil', 'Ing. Pavel Tvrzník'),\n",
       "                           ('hodnotaBezDph', '80000')])),\n",
       "             ('prilohy',\n",
       "              OrderedDict([('priloha',\n",
       "                            OrderedDict([('nazevSouboru', '893_Gordic.pdf'),\n",
       "                                         ('hash',\n",
       "                                          OrderedDict([('@algoritmus',\n",
       "                                                        'sha256'),\n",
       "                                                       ('#text',\n",
       "                                                        'a407edf283dd009c4a0f580b44c1baf191563d44629c927331802f6263c0d299')])),\n",
       "                                         ('odkaz',\n",
       "                                          'https://smlouvy.gov.cz/smlouva/soubor/5/893_Gordic.pdf')]))])),\n",
       "             ('platnyZaznam', '1')])"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "z[0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "OrderedDict([('datovaSchranka', 'c5kbvkw'),\n",
       "             ('nazev', 'Liberecký kraj'),\n",
       "             ('ico', '70891508'),\n",
       "             ('adresa', 'U Jezu 642/2a, 46001 Liberec, CZ')])"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "z[0]['smlouva']['subjekt']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "res = pd.DataFrame(columns=['nazev', 'ico', 'bezdph', 'sdph', 'predmet'])\n",
    "for j, sml in enumerate(z):\n",
    "    res.loc[j] = [sml['smlouva']['subjekt']['nazev'], sml['smlouva']['subjekt']['ico'],\n",
    "                 sml['smlouva'].get('hodnotaBezDph', np.nan),\n",
    "                 sml['smlouva'].get('hodnotaVcetneDph', np.nan),\n",
    "                 sml['smlouva']['predmet']]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>nazev</th>\n",
       "      <th>ico</th>\n",
       "      <th>bezdph</th>\n",
       "      <th>sdph</th>\n",
       "      <th>predmet</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Liberecký kraj</td>\n",
       "      <td>70891508</td>\n",
       "      <td>80000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>rozšíření jádra IS Ginis o funkcionalitu zveře...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Zdravotnická záchranná služba Olomouckého kraj...</td>\n",
       "      <td>00849103</td>\n",
       "      <td>91308</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Podnájmní smlouva</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Zdravotnická záchranná služba Olomouckého kraj...</td>\n",
       "      <td>00849103</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Podnájemní smlouva</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                               nazev       ico bezdph sdph  \\\n",
       "0                                     Liberecký kraj  70891508  80000  NaN   \n",
       "1  Zdravotnická záchranná služba Olomouckého kraj...  00849103  91308  NaN   \n",
       "2  Zdravotnická záchranná služba Olomouckého kraj...  00849103    NaN  NaN   \n",
       "\n",
       "                                             predmet  \n",
       "0  rozšíření jádra IS Ginis o funkcionalitu zveře...  \n",
       "1                                  Podnájmní smlouva  \n",
       "2                                 Podnájemní smlouva  "
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "res.head(3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "uni = pd.DataFrame(index=res.ico.unique(), columns = ['nazev', 'smluv', 'bezcen'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "for ic in res.ico.unique():\n",
    "    nz = res.loc[res.ico == ic].nazev.iloc[0]\n",
    "    n = len(res.loc[res.ico == ic, :])\n",
    "    bc = res.loc[res.ico == ic, ['bezdph', 'sdph']].isnull().all(axis=1).sum()\n",
    "\n",
    "    uni.loc[ic] = [nz, n, bc]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 67,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "uni.sort_values(by='smluv', ascending=False).to_csv('bezcen.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 69,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "nazev     Liberecký krajZdravotnická záchranná služba Ol...\n",
       "smluv                                                   450\n",
       "bezcen                                                  119\n",
       "dtype: object"
      ]
     },
     "execution_count": 69,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "uni.sum()"
   ]
  },
  {
   "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": "3.5.2"
  },
  "widgets": {
   "state": {},
   "version": "1.1.2"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 0
}