{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "from string import ascii_letters\n",
    "from sklearn.feature_extraction.text import TfidfVectorizer\n",
    "import requests\n",
    "import re\n",
    "from sklearn.cluster import DBSCAN\n",
    "from tkinter import *"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "def parsString(s):\n",
    "    resS=\"\"\n",
    "    tmpS=\"\"\n",
    "    s=s.lower()\n",
    "    for ch in s:\n",
    "        if (ch in ascii_letters):\n",
    "            tmpS+=ch\n",
    "        else:\n",
    "            #resS+=tmpS\n",
    "            #tmpS=ch\n",
    "            if (tmpS!=\"\"):\n",
    "                resS+=' '\n",
    "                resS+=tmpS\n",
    "            else:\n",
    "                resS+=' '\n",
    "                resS+=ch\n",
    "            tmpS=\"\"\n",
    "    if tmpS==\"\":\n",
    "        resS+=tmpS\n",
    "    #print(resS)\n",
    "    return resS\n",
    "    "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "def readDataset():\n",
    "    f=open('dataset', 'r')\n",
    "    it=0;\n",
    "    vectorsS=[]\n",
    "    for i in f:\n",
    "        vectorsS.append(parsString(i))\n",
    "        it+=1\n",
    "        #if (it>1000):\n",
    "        #    break\n",
    "    f.close()\n",
    "    return vectorsS"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "def makeDataset():\n",
    "    f=open('dataset', 'w')\n",
    "    for i in range(200):\n",
    "        responce=requests.get('https://slot-ml.ptsecurity.com/api/v1/users/ad828642388ee5a23f0c90483520309204ea3d53/vectors/?random')\n",
    "        vect=str(responce.text)[str(responce.text).find(\"vector\")+10:str(responce.text).find(\"\\\"id\\\":\")-4]\n",
    "        #id=str(responce.text)[str(responce.text).find(\"\\\"id\\\":\")+7:-4]\n",
    "        #print(rpost, mpost)\n",
    "        f.write(vect+'\\n')\n",
    "        #if (i%10==0):\n",
    "        #    print(i)\n",
    "    f.close()\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "def train(dataset):\n",
    "    # Obtain some string samples.\n",
    "    # Get a char-based vectorizer with (1,2) n-gram range.\n",
    "    vectorizer = TfidfVectorizer(analyzer='word', ngram_range=(1, 1))\n",
    "    # Vectorize the samples.\n",
    "    vectors = vectorizer.fit_transform(dataset)\n",
    "    #print(vectorizer.get_feature_names())\n",
    "    #print(vectors)\n",
    "    return vectors"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "def clusterization(vectors):\n",
    "    model = DBSCAN(eps=0.5).fit(vectors)\n",
    "    return model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "def sendSol(vectorsString):\n",
    "    #f=open('dataset', 'w')\n",
    "    ok=0\n",
    "    for i in range(100):\n",
    "        responce=requests.get('https://slot-ml.ptsecurity.com/api/v1/users/c34f6b9b747ebe438da878a07a318892a988c06f/vectors/?random')\n",
    "        vect=str(responce.text)[str(responce.text).find(\"vector\")+10:str(responce.text).find(\"\\\"id\\\":\")-4]\n",
    "        id=str(responce.text)[str(responce.text).find(\"\\\"id\\\":\")+7:-4]\n",
    "        #print(responce.text)\n",
    "        #print(vect)\n",
    "        #print(id)    \n",
    "        rpost = ('https://slot-ml.ptsecurity.com/api/v1/users/c34f6b9b747ebe438da878a07a318892a988c06f/results/')\n",
    "        #rpost=('5.8.180.70:80/api/v1/users/c34f6b9b747ebe438da878a07a318892a988c06f/results/')\n",
    "        #mpost=('{ \\\"vector\\\": \\\"'+id+'\\\", \\\"class\\\": 1 }')\n",
    "        #vectorsString[i]=parsString(vect)\n",
    "        strv=parsString(vect)\n",
    "        #vectorsString.append(parsString(strv))\n",
    "        vectorsString[999]=strv\n",
    "        vectors=train(vectorsString)\n",
    "        #print(vectors)\n",
    "        model = clusterization(vectors)\n",
    "        cnt=0;\n",
    "        clas=model.labels_[999]\n",
    "        if(clas==-1):\n",
    "            if (ok*2<=i):\n",
    "                if (len(strv)>100):\n",
    "                    clas=-1\n",
    "                else:\n",
    "                    clas=-2\n",
    "            else:\n",
    "                clas=-4\n",
    "        mpost={}\n",
    "        mpost[\"vector\"]=id\n",
    "        mpost[\"class\"]=clas+3\n",
    "        #print(clas)\n",
    "        #print(rpost, mpost)\n",
    "        msg=requests.post(rpost, data=mpost)\n",
    "        if (str(msg)==\"<Response [200]>\"):\n",
    "            ok+=1\n",
    "        #if (i%10==0):\n",
    "        #    print(i, ok)\n",
    "    #f.close()\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "def check(strfr):\n",
    "    makeDataset()\n",
    "    vectorsString=readDataset()\n",
    "    #f=open('dataset', 'w')\n",
    "    ok=0\n",
    "    vect=str(strfr)\n",
    "    #print(responce.text)\n",
    "    #print(vect)\n",
    "    #print(id)    \n",
    "    #rpost = ('https://slot-ml.ptsecurity.com/api/v1/users/c34f6b9b747ebe438da878a07a318892a988c06f/results/')\n",
    "    #rpost=('5.8.180.70:80/api/v1/users/c34f6b9b747ebe438da878a07a318892a988c06f/results/')\n",
    "    #mpost=('{ \\\"vector\\\": \\\"'+id+'\\\", \\\"class\\\": 1 }')\n",
    "    #vectorsString[i]=parsString(vect)\n",
    "    strv=parsString(vect)\n",
    "    #vectorsString.append(parsString(strv))\n",
    "    vectorsString[999]=strv\n",
    "    vectors=train(vectorsString)\n",
    "    #print(vectors)\n",
    "    model = clusterization(vectors)\n",
    "    cnt=0;\n",
    "    clas=model.labels_[999]\n",
    "    return clas\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "def action(event):\n",
    "    ent.delete(\"1.0\", END)\n",
    "    ent.insert(\"1.0\", check(tf.get(\"1.0\",END)))\n",
    "root = Tk()\n",
    "tf=Text(root, height=8, width=70)\n",
    "tf.pack()\n",
    "ent = Text(root,height=1,width=3)\n",
    "ent.pack()\n",
    "but = Button(root)\n",
    "but[\"text\"]=\"Check\"\n",
    "\n",
    "but.bind(\"<Button-1>\", action)\n",
    "but.pack()\n",
    "root.mainloop()\n",
    "#makeDataset()\n",
    "#print(vectorsString)\n",
    "#vectors=train(vectorsString)\n",
    "#print(vectors)\n",
    "#model = clusterization(vectors)\n",
    "#sendSol(vectorsString)\n",
    "\n"
   ]
  }
 ],
 "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"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 1
}