{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Usage\n",
    "    1 - run Data loader for the dataframes,\n",
    "    1 - run Plotting, \n",
    "    2 - run Main part,\n",
    "\n",
    "Plotting: The main plotting function: plotMetrika(). It gets a dataframe and group it (ref_df). The plotting function (label_group_bar_table) use the groupped dataframe. \n",
    "The plotting is based on the stackoverflow code (https://stackoverflow.com/questions/50997997/pandas-bar-plot-hierarchical-labeling-alternative-version )\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Main part"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "06cb0e8b8fd849769de80d22d1b22599",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(Button(description='cubic-vs-bbr2', style=ButtonStyle()), Button(description='bbr21-vs-bbr22', …"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "BWParam: bw1000mbps\n",
      "RTT param: None\n",
      "['bbr-100' 'cubic-100' 'bbr-10' 'cubic-10']\n",
      "CC: cubic-vs-bbr2\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/usr/local/lib/python3.6/dist-packages/ipykernel_launcher.py:334: UserWarning: This figure was using constrained_layout==True, but that is incompatible with subplots_adjust and or tight_layout: setting constrained_layout==False. \n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1500x400 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "['bbr-100' 'cubic-100' 'bbr-10' 'cubic-10']\n",
      "CC: cubic-vs-bbr2\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/usr/local/lib/python3.6/dist-packages/ipykernel_launcher.py:334: UserWarning: This figure was using constrained_layout==True, but that is incompatible with subplots_adjust and or tight_layout: setting constrained_layout==False. \n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1500x400 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "['bbr-100' 'cubic-100' 'bbr-10' 'cubic-10']\n",
      "CC: cubic-vs-bbr2\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/usr/local/lib/python3.6/dist-packages/ipykernel_launcher.py:334: UserWarning: This figure was using constrained_layout==True, but that is incompatible with subplots_adjust and or tight_layout: setting constrained_layout==False. \n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1500x400 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# imports\n",
    "%matplotlib inline\n",
    "import matplotlib.pyplot as plt\n",
    "import sys\n",
    "import os, shutil\n",
    "\n",
    "#import logging\n",
    "#logging.basicConfig(level=logging.DEBUG)\n",
    "\n",
    "\n",
    "from itertools import groupby  \n",
    "\n",
    "import glob\n",
    "from zipfile import ZipFile\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "import json\n",
    "\n",
    "from ipywidgets import Button, HBox, VBox\n",
    "from IPython.display import clear_output\n",
    "\n",
    "# the zip file with the measurment data\n",
    "zipFilename = \"fifo_ppv_gsp_pie.zip\"\n",
    "#zipFilename = \"buff_percent.zip\"\n",
    "\n",
    "# filter params,  rtt_param can be None\n",
    "rtt_param = \"rtt100\"\n",
    "bw_param = \"bw1000mbps\"\n",
    "\n",
    "# list of the skipped schedulers\n",
    "#skipSched = []\n",
    "skipSched = ['gspmp50', 'pie50']\n",
    "\n",
    "#mixed=False\n",
    "mixed=True\n",
    "\n",
    "# use mixed sceanrios\n",
    "if mixed:\n",
    "    zipFilename = \"mix_full.zip\"\n",
    "    rtt_param = None # \"rtt10-8x10-2x100\"\n",
    "\n",
    "buttons = []\n",
    "\n",
    "def multiplot(btn):\n",
    "    global buttons\n",
    "    c= btn.description\n",
    "    clear_output()\n",
    "    display(HBox(buttons))\n",
    "    \n",
    "    print(\"BWParam: \"+bw_param)\n",
    "    print(\"RTT param: \"+str(rtt_param))\n",
    "    \n",
    "    plotMetrika(\"jains\", c)\n",
    "    plotMetrika(\"sum_thr\", c)\n",
    "    plotMetrika(\"class_thr\", c)\n",
    "\n",
    "def saveImages(btn):\n",
    "    global buttons\n",
    "    clear_output()\n",
    "    display(HBox(buttons))\n",
    "    for c in df_total.cc.unique():\n",
    "        plotMetrika(\"jains\", c, True)\n",
    "        plotMetrika(\"sum_thr\", c, True)\n",
    "        plotMetrika(\"class_thr\", c, True)\n",
    "        plotMetrika(\"loss\", c, True)\n",
    "        plotMetrika(\"avg_delay\", c, True)\n",
    "        plotMetrika(\"max_delay\", c, True)\n",
    "    \n",
    "\n",
    "def loadButtons():\n",
    "    global df_total, df_not_total, buttons\n",
    "    buttons = []\n",
    "    \n",
    "    for i in df_total.cc.unique():\n",
    "        button = Button(description=i)\n",
    "        button.on_click(multiplot)\n",
    "        buttons.append(button)\n",
    "    button = Button(description=\"Load Data\")\n",
    "    button.on_click(loadData)\n",
    "    buttons.append(button)\n",
    "    button = Button(description=\"Save Images\")\n",
    "    button.on_click(saveImages)\n",
    "    buttons.append(button)\n",
    "    display(HBox(buttons))\n",
    "        \n",
    "    \n",
    "def loadData(btn):\n",
    "    global df_total, df_not_total, buttons\n",
    "    clear_output()\n",
    "    if not readJsons(zipFilename): # if every data is good clear the screen\n",
    "        clear_output()\n",
    "        \n",
    "    print(df_total.count())\n",
    "    print(df_total.cc.unique())\n",
    "    loadButtons()\n",
    "\n",
    "loadButtons()\n",
    "\n",
    "\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Dataframe saver"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "BufferStudyCsv/buff_percent.total.csv\n",
      "BufferStudyCsv/buff_percent.not_total.csv\n"
     ]
    }
   ],
   "source": [
    "csvFilename = 'BufferStudyCsv/'+zipFilename.replace(\"zip\",\"\")\n",
    "print(csvFilename+\"total.csv\")\n",
    "df_total.to_csv(index=False,path_or_buf=csvFilename+\"total.csv\")\n",
    "print(csvFilename+\"not_total.csv\")\n",
    "df_not_total.to_csv(index=False,path_or_buf=csvFilename+\"not_total.csv\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Data loader functions"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "scrolled": false
   },
   "outputs": [],
   "source": [
    "#df header cc;fn;rtt;bdp;bw;sched;fname;avg_thr;avg_delay;max_delay;loss;loss_percent;jains;sum_thr;util\n",
    "df_total = pd.DataFrame(columns=['cc','fn','rtt','bdp',\n",
    "                           'bw','sched','jains','sum_thr','util'])\n",
    "df_not_total = pd.DataFrame(columns=['cc','fn','rtt','bdp',\n",
    "                           'bw','sched',\n",
    "                           'avg_thr','avg_delay',\n",
    "                           'max_delay','loss','loss_percent',\n",
    "                            'sum_thr', 'flownum','cname',\n",
    "                           ])\n",
    "\n",
    "\n",
    "def jains(bws):\n",
    "    return sum(bws) ** 2 / (sum([i ** 2 for i in bws]) * len(bws))\n",
    "\n",
    "# hacks for design, and orders\n",
    "def folderNameSplitter(_folder):\n",
    "    cc, fn, rtt, bdp, bw, sched = _folder.split('_')\n",
    "    fn = ('002' if fn == 'fn1' else '010' if fn=='fn5' else '020' if fn=='fn10' else '100' if fn=='fn50' else '040' if fn=='fn20' else fn)\n",
    "    bdp = bdp.replace('bdp','')\n",
    "    sched = ('--1TailDrop' if 'fifo' in sched else sched)\n",
    "    sched = ('--4PPV' if 'ppv' in sched else sched)\n",
    "    sched = ('--2PIE' if 'pie75' in sched else sched)\n",
    "    sched = ('--3GSP' if 'gspmp75' in sched else sched)\n",
    "    if cc == 'cubic-vs-bbr2':\n",
    "        rtt = ('2-2-8-8'if \"rtt10-8x10-2x100\" in rtt else '4-4-16-16' if \"rtt10-16x10-4x100\" in rtt else rtt)\n",
    "    else:\n",
    "        rtt = ('4-16'if \"rtt10-8x10-2x100\" in rtt else '8-32' if \"rtt10-16x10-4x100\" in rtt else rtt)\n",
    "\n",
    "    return cc, fn, rtt, bdp, bw, sched\n",
    "        \n",
    "def readJsons(zipfile):\n",
    "    global df_total, df_not_total\n",
    "    df_total.drop(df_total.index, inplace=True)\n",
    "    df_not_total.drop(df_not_total.index, inplace=True)\n",
    "    error = False\n",
    "    bandwidths = {}\n",
    "    loss = {}\n",
    "    fileNum = 0\n",
    "    with ZipFile(zipfile) as myzip:\n",
    "        for f in myzip.namelist():\n",
    "            fileNum +=1\n",
    "            print(\".\",end=\"\")\n",
    "            if fileNum % 125 == 0:\n",
    "                print(\"\")\n",
    "            info = myzip.getinfo(f)\n",
    "            if info.is_dir():\n",
    "                continue\n",
    "            if not f.endswith('.txt'):\n",
    "                continue\n",
    "            if not 'iperf' in f:\n",
    "                continue\n",
    "\n",
    "            with myzip.open(f,'r') as fd:\n",
    "                folder, fname = f.split(\"/\")[-2:]\n",
    "                cc, fn, rtt, bdp, bw, sched = folderNameSplitter(folder)\n",
    "                if sched in skipSched:\n",
    "                    continue\n",
    "                j = json.load(fd)\n",
    "                if \"error\" in j:\n",
    "                    print(\"ERROR: \",f)\n",
    "                    error = True\n",
    "                    continue\n",
    "\n",
    "                # not total data\n",
    "\n",
    "                sum_throughput = 0.0\n",
    "                avg_delay = 0.0\n",
    "                max_delay = 0.0\n",
    "                #loss = 0.0\n",
    "                flownum = len(j[\"end\"][\"streams\"])\n",
    "                for flow in j[\"end\"][\"streams\"]:\n",
    "                    sum_throughput += flow[\"receiver\"][\"bits_per_second\"]\n",
    "                    avg_delay += flow[\"sender\"][\"mean_rtt\"]\n",
    "                    max_delay = max(max_delay,flow[\"sender\"][\"max_rtt\"])\n",
    "                    #loss += flow[\"sender\"][\"retransmits\"]\n",
    "                    \n",
    "                    # init jains data\n",
    "                    if not folder in bandwidths:\n",
    "                        bandwidths[folder] = []\n",
    "                    bandwidths[folder].append(flow[\"receiver\"][\"bits_per_second\"])\n",
    "\n",
    "                \n",
    "                avg_throughput = sum_throughput / flownum\n",
    "                avg_delay /= flownum\n",
    "\n",
    "                #loss_percent = (loss * 1514) / (10**6 * 180 * (avg_throughput) / 8) * 100 #TODO: pssible bug(s)\n",
    "               \n",
    "                if (len(fname.split(\"_rtt\")) >=2):\n",
    "                    base_rtt = int(fname.split(\"_rtt\")[1][:-4])\n",
    "                else:\n",
    "                    base_rtt = int(rtt[3:])\n",
    "                df_not_total = df_not_total.append({\n",
    "                    'cc':cc,        #folder alapjan\n",
    "                    'fn':fn,        #filename alapjan\n",
    "                    'rtt':rtt,\n",
    "                    'bdp':bdp,\n",
    "                    'bw':bw,\n",
    "                    'sched':sched,\n",
    "\n",
    "                    'fname':fname,\n",
    "                    'avg_thr':avg_throughput,\n",
    "                    'avg_delay':(avg_delay/1e3)-base_rtt,\n",
    "                    'max_delay':(max_delay/1e3)-base_rtt,\n",
    "                    \n",
    "                    'sum_thr':sum_throughput,\n",
    "                    'flownum':flownum,      #based on iperf \n",
    "                    'cname':('cubic' if 'cubic' in fname else 'bbr')  +\"-\"+  str(base_rtt),\n",
    "                    \n",
    "                    #'loss':loss,\n",
    "                    #'loss_percent':loss_percent\n",
    "                }, ignore_index=True)\n",
    "                \n",
    "                \n",
    "                # init total data\n",
    "                if not folder in loss:\n",
    "                    loss[folder] = (1.0 - j[\"end\"][\"sum_received\"][\"bits_per_second\"]/j[\"end\"][\"sum_sent\"][\"bits_per_second\"]) * 100.0\n",
    "                \n",
    "    for folder in bandwidths:\n",
    "        cc, fn, rtt, bdp, bw, sched = folderNameSplitter(folder)\n",
    "\n",
    "        _jains = jains(bandwidths[folder])\n",
    "        _sum_thr = sum(bandwidths[folder])/1e6\n",
    "        _util = _sum_thr / float(bw[2:-4])\n",
    "        \n",
    "        df_total = df_total.append({\n",
    "            'cc':cc,\n",
    "            'fn':fn,\n",
    "            'rtt':rtt,\n",
    "            'bdp':bdp,\n",
    "            'bw':bw,\n",
    "            'sched':sched,\n",
    "\n",
    "            'jains':_jains,\n",
    "            'sum_thr':_sum_thr,\n",
    "            'util':_util,\n",
    "            'loss': loss[folder]\n",
    "        }, ignore_index=True)\n",
    "    return error"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Plotting"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "scrolled": false
   },
   "outputs": [],
   "source": [
    "# source: https://stackoverflow.com/questions/50997997/pandas-bar-plot-hierarchical-labeling-alternative-version\n",
    "\n",
    "#15,5\n",
    "\n",
    "figsize=(15,4)\n",
    "\n",
    "\n",
    "params = {'legend.fontsize': 'x-large',\n",
    "            'figure.figsize': figsize,   #6,2\n",
    "            'axes.labelsize': 'x-large',\n",
    "            'axes.titlesize':'x-large',\n",
    "            'xtick.labelsize':'x-large',\n",
    "            'ytick.labelsize':'x-large',\n",
    "            'figure.autolayout':False,\n",
    "            'figure.constrained_layout.use':True,\n",
    "#            'text.usetex':True,\n",
    "            'legend.framealpha':None,\n",
    "         }\n",
    "\n",
    "class_colors = {\n",
    "        'cubic-10':'bX',\n",
    "        'cubic-100':'y^',\n",
    "        'bbr-10':'go',\n",
    "        'bbr-100':'m*',\n",
    "        }\n",
    "\n",
    "schedulers = [\n",
    "        ('pie50','c*'),\n",
    "        ('gspmp75','y^'),\n",
    "        ('gsp','bX'),\n",
    "        ('ppv','go'),\n",
    "        ('gspmp50','r^'),\n",
    "        ('pie75','m*'),\n",
    "        ('fifo','bX'),\n",
    "        ('TailDrop','bX'),\n",
    "        ('GSP','y^'),\n",
    "        ('PPV','go'),\n",
    "        ('PIE','m*'),\n",
    "        ('--1TailDrop','bX'),\n",
    "        ('--2PIE','m*'),\n",
    "        ('--3GSP','y^'),\n",
    "        ('--4PPV','go'),\n",
    "        ('gspRmp20','r^'),\n",
    "        ('pieRmp20','c*'),\n",
    "        ]\n",
    "\n",
    "yaxisLabel={\n",
    "    \"jains\":\"Jain's Fairness\",\n",
    "    \"sum_thr\":\"Goodput [Mbps]\",\n",
    "    \"class_thr\":\"Relative Goodput\",\n",
    "    \"loss\":\"Packet Loss\",\n",
    "    \"avg_delay\":\"Delay [ms]\",\n",
    "    \"max_delay\":\"Delay [ms]\",\n",
    "}\n",
    "\n",
    "\n",
    "\n",
    "levelLables = []\n",
    "\n",
    "\n",
    "def add_line(ax, xpos, ypos,last_ypos,maxLenLabel, _linestyle='-'):\n",
    "    line = plt.Line2D([xpos, xpos], [ypos , last_ypos],\n",
    "                  transform=ax.transAxes, color='darkslategrey',zorder=0,linestyle=_linestyle)\n",
    "        #'default', 'steps', 'steps-pre', 'steps-mid', 'steps-post'\n",
    "    line.set_clip_on(False)\n",
    "    ax.add_line(line)\n",
    "\n",
    "def label_len(my_index,level):\n",
    "    labels = my_index.get_level_values(level)\n",
    "    return [(k, sum(1 for i in g)) for k,g in groupby(labels)]\n",
    "\n",
    "def max_label_len(my_index,level):\n",
    "    if level == 0:\n",
    "        return 3\n",
    "    elif level < my_index.nlevels:\n",
    "        labels = my_index.get_level_values(level)\n",
    "        maxLength = max([len(str(replaceLabel(k))) for k,g in groupby(labels)])\n",
    "        return maxLength\n",
    "    else:\n",
    "        return 1\n",
    "\n",
    "def replaceLabel(label):\n",
    "    dic = {\n",
    "        \"iperf\":\"\",\n",
    "        \".txt\":\"\",\n",
    "        \"002\":\"2\",\n",
    "        \"010\":\"10\",\n",
    "        \"020\":\"20\",\n",
    "        \"040\":\"40\",\n",
    "#        \"gspmp75\":\"GSP\",\n",
    "#        \"ppv\":\"PPV\",\n",
    "#        \"pie75\":\"PIE\",\n",
    "#        \"fifo\":\"TailDrop\",\n",
    "        \"--1\":\"\",\n",
    "        \"--2\":\"\",\n",
    "        \"--3\":\"\",\n",
    "        \"--4\":\"\",\n",
    "        \n",
    "        \"fn\":\"N\",\n",
    "        \"bdp\":\"Buffer\",\n",
    "        \"sched\":\"AQM\",\n",
    "        \"rtt\":\"N\",\n",
    "    }\n",
    "    \n",
    "    for i, j in dic.items():\n",
    "        label = label.replace(i, j)\n",
    "    return label\n",
    "\n",
    "def label_group_bar_table(ax, df):\n",
    "    ypos = 0  #-.1\n",
    "    last_ypos = ypos\n",
    "    scale = 1./df.index.size\n",
    "    for level in range(df.index.nlevels)[::-1]:\n",
    "        pos = 0\n",
    "        maxLenLabel = max_label_len(df.index,level)\n",
    "        if level==0 or (level==1 and df.index.nlevels>2):\n",
    "            rot = 0\n",
    "            maxLenLabel=3\n",
    "        else:\n",
    "            rot=90\n",
    "        ypos -= 0.014*(10.0/float(figsize[1]))*maxLenLabel # 5re jo: 0.028*maxLenLabel  # ez 10-es magassagra jo: 0.014*maxLenLabel\n",
    "        for label, rpos in label_len(df.index,level):\n",
    "            lxpos = (pos + .5 * rpos)*scale\n",
    "            fontsize = 'x-large'\n",
    "            if level==2 and df.index.nlevels==3:\n",
    "                fontsize='large'\n",
    "            ax.text(lxpos, ypos, replaceLabel(label), ha='center', transform=ax.transAxes, rotation = rot,fontsize=fontsize) #14\n",
    "            # grid plotting\n",
    "            # if the level is 3 the we do not need the max level and level 2 is smaller\n",
    "            if df.index.nlevels==3:\n",
    "                if level==df.index.nlevels:\n",
    "                    pass\n",
    "                else:\n",
    "                    if level == 0:\n",
    "                        add_line(ax, pos*scale , ypos, 1.0, maxLenLabel)\n",
    "                    elif level == 1:\n",
    "                        add_line(ax, pos*scale , ypos, 1.0, maxLenLabel,_linestyle='--')\n",
    "                    elif level == 2:\n",
    "                        pass\n",
    "                    else:\n",
    "                        add_line(ax, pos*scale, ypos, last_ypos, maxLenLabel)\n",
    "            elif df.index.nlevels==2 and level == 0:\n",
    "                add_line(ax, pos*scale , ypos, 1.0, maxLenLabel)\n",
    "            else:\n",
    "                add_line(ax, pos*scale, ypos, last_ypos, maxLenLabel)\n",
    "            pos += rpos\n",
    "        ax.text(-0.018-(0.004*len(replaceLabel(levelLabels[level]))),last_ypos+((ypos-last_ypos)), replaceLabel(levelLabels[level]), ha='center', transform=ax.transAxes, rotation = 0,fontsize='x-large') #14\n",
    "        add_line(ax, pos*scale , ypos, 1.0, maxLenLabel)\n",
    "        last_ypos = ypos\n",
    "        #ypos -= 0.02*maxLenLabel\n",
    "\n",
    "def getSchedulerData(metrika,c, sched):\n",
    "    global df_total, df_not_total\n",
    "    if metrika == 'loss_percent' or metrika == 'avg_delay' or metrika == 'max_delay':\n",
    "        df_tmp = df_not_total\n",
    "    else:\n",
    "        df_tmp = df_total\n",
    "    \n",
    "    df_tmp=df_tmp[df_tmp.bw == bw_param]\n",
    "    if rtt_param:\n",
    "        df_tmp=df_tmp[df_tmp.rtt == rtt_param]\n",
    "    \n",
    "    df_tmp = df_tmp[df_tmp.cc==c]\n",
    "    df_tmp = df_tmp[df_tmp.sched==sched]\n",
    "    _df_tmp = df_tmp.rename(columns={metrika: sched})\n",
    "    if metrika == 'loss_percent' or metrika == 'avg_delay' or metrika == 'max_delay':\n",
    "        if mixed:\n",
    "            if rtt_param:\n",
    "                df_tmp_rownum = _df_tmp[[sched,'fname','bdp']].groupby(['bdp','fname']).ngroup().unique()\n",
    "                df_tmp = _df_tmp[[sched,'fname','bdp','bw']].groupby(['bdp','fname']).sum()\n",
    "                _levelLabels = ['bdp','fname']\n",
    "            else:\n",
    "                df_tmp_rownum = _df_tmp[[sched,'fname','rtt','bdp']].groupby(['rtt','bdp','fname']).ngroup().unique()\n",
    "                df_tmp = _df_tmp[[sched,'fname','rtt','bdp','bw']].groupby(['rtt','bdp','fname']).sum()\n",
    "                _levelLabels = ['rtt','bdp','fname']\n",
    "        else:\n",
    "            if rtt_param:\n",
    "                df_tmp_rownum = _df_tmp[[sched,'fname','fn','bdp']].groupby(['fn','bdp','fname']).ngroup().unique()\n",
    "                df_tmp = _df_tmp[[sched,'fname','fn','bdp','bw']].groupby(['fn','bdp','fname']).sum()\n",
    "                _levelLabels = ['fn','bdp','fname']\n",
    "            else:\n",
    "                df_tmp_rownum = _df_tmp[[sched,'fname','rtt','fn','bdp']].groupby(['rtt','fn','bdp','fname']).ngroup().unique()\n",
    "                df_tmp = _df_tmp[[sched,'fname','rtt','fn','bdp','bw']].groupby(['rtt','fn','bdp','fname']).sum()\n",
    "                _levelLabels = ['rtt','fn','bdp','fname']\n",
    "    else:\n",
    "        if mixed:\n",
    "            if rtt_param:\n",
    "                df_tmp_rownum = _df_tmp[[sched,'bdp']].groupby(['bdp']).ngroup().unique()\n",
    "                df_tmp = _df_tmp[[sched,'bdp']].groupby(['bdp']).sum()\n",
    "                levelLabels = ['bdp']\n",
    "            else:\n",
    "                if metrika == 'jains':\n",
    "                    df_tmp_rownum = _df_tmp[[sched,'rtt','bdp']].groupby(['rtt','bdp']).ngroup().unique()\n",
    "                    df_tmp = _df_tmp[[sched,'rtt','bdp']].groupby(['rtt','bdp']).sum()\n",
    "                    _levelLabels = ['rtt','bdp']\n",
    "                else:\n",
    "                    df_tmp_rownum = _df_tmp[[sched,'fn','bdp']].groupby(['fn','bdp']).ngroup().unique()\n",
    "                    df_tmp = _df_tmp[[sched,'fn','bdp']].groupby(['fn','bdp']).sum()\n",
    "                    _levelLabels = ['fn','bdp']\n",
    "        else:\n",
    "            if rtt_param:\n",
    "                df_tmp_rownum = _df_tmp[[sched,'fn','bdp']].groupby(['fn','bdp']).ngroup().unique()\n",
    "                df_tmp = _df_tmp[[sched,'fn','bdp']].groupby(['fn','bdp']).sum()\n",
    "                _levelLabels = ['fn','bdp']\n",
    "            else:\n",
    "                df_tmp_rownum = _df_tmp[[sched,'rtt','fn','bdp']].groupby(['rtt','fn','bdp']).ngroup().unique()\n",
    "                df_tmp = _df_tmp[[sched,'rtt','fn','bdp']].groupby(['rtt','fn','bdp']).sum()\n",
    "                _levelLabels = ['rtt','fn','bdp']\n",
    "    return (df_tmp,len(df_tmp_rownum),_levelLabels)\n",
    "\n",
    "\n",
    "def plotMetrika(metrika, c, saveImage = False):\n",
    "    global schedulers, figsize, levelLabels\n",
    "        \n",
    "    print(df_not_total[df_not_total.cc == c].cname.unique())\n",
    "    if metrika == \"class_thr\" and len(df_not_total[df_not_total.cc == c].cname.unique()) == 1:\n",
    "        return\n",
    "    \n",
    "    print(\"CC: \"+c)\n",
    "    sys.stdout.flush()\n",
    "\n",
    "    if not os.path.exists(\"BufferImages\"):\n",
    "        os.makedirs(\"BufferImages\")\n",
    "    if not os.path.exists(\"BufferImages/newPlot\"):\n",
    "        os.makedirs(\"BufferImages/newPlot\")\n",
    "    if not os.path.exists(\"BufferImages/paperImages\"):\n",
    "        os.makedirs(\"BufferImages/paperImages\")\n",
    "    scenarioFolder = zipFilename[:-4]\n",
    "    if not os.path.exists(\"BufferImages/newPlot/\"+scenarioFolder):\n",
    "        os.makedirs(\"BufferImages/newPlot/\"+scenarioFolder)\n",
    "    \n",
    "    plt.rcParams.update(plt.rcParamsDefault)\n",
    "    plt.rcParams.update(params)\n",
    "    fig = plt.figure(figsize=figsize)\n",
    "    ax = fig.add_subplot(111)\n",
    "\n",
    "    ref_df = None\n",
    "    \n",
    "    if metrika == \"class_thr\":\n",
    "        df_tmp=df_not_total[df_not_total.bw == bw_param]\n",
    "        if rtt_param:\n",
    "            df_tmp=df_tmp[df_tmp.rtt == rtt_param]\n",
    "        df_tmp = df_tmp[df_tmp.cc==c]\n",
    "    \n",
    "        df_tmp_max_scenario = df_tmp[['sum_thr','flownum','sched','rtt','fn','bdp']].groupby(['sched','bdp','fn','rtt'])\\\n",
    "                                    .agg({'sum_thr':'sum','flownum':'sum'}).reset_index()\n",
    "        df_tmp_max_scenario['x'] = df_tmp_max_scenario.apply(lambda row : row['sum_thr'] / row['flownum'], axis=1) \n",
    "        \n",
    "        def getCnValue2(row, _max_df,cn):\n",
    "                                xx_df = _max_df[\n",
    "                                    (_max_df.rtt == row.rtt) &\n",
    "                                    (_max_df.fn == row.fn) &\n",
    "                                    (_max_df.bdp == row.bdp) &\n",
    "                                    (_max_df.sched == row.sched)\n",
    "                                               ]\n",
    "                                if not xx_df.empty and row.cname == cn:\n",
    "                                    return row[cn]/xx_df['x'].iloc[0]\n",
    "                                else:\n",
    "                                    return np.NaN\n",
    "        \n",
    "        df_tmp_max_class = df_tmp[['sum_thr','flownum','sched','rtt','fn','bdp','cname']].groupby(['sched','bdp','fn','rtt','cname'])\\\n",
    "                                    .agg({'sum_thr':'sum','flownum':'sum'}).reset_index()\n",
    "        \n",
    "        i = 0\n",
    "        for cn in df_not_total.cname.unique():\n",
    "            df_tmp_max_class[cn] = df_tmp_max_class.apply (lambda row: (row.sum_thr / row.flownum), axis=1)\n",
    "            df_tmp_max_class[cn] = df_tmp_max_class.apply(getCnValue2, args=[df_tmp_max_scenario,cn], axis=1 )\n",
    "            _rownum = 0\n",
    "            \n",
    "            if mixed:\n",
    "                if rtt_param:\n",
    "                    _rownum = df_tmp_max_class[[cn,'sched','bdp']].groupby(['sched','bdp']).ngroup().unique()\n",
    "                    ref_df = df_tmp_max_class[[cn,'sched','bdp']].groupby(['sched','bdp']).mean()\n",
    "                    _levelLabels = ['sched','bdp']\n",
    "                else:\n",
    "                    _rownum = df_tmp_max_class[[cn,'sched','rtt','bdp']].groupby(['sched','rtt','bdp']).ngroup().unique()\n",
    "                    ref_df = df_tmp_max_class[[cn,'sched','rtt','bdp']].groupby(['sched','rtt','bdp']).mean()\n",
    "                    _levelLabels = ['sched','rtt','bdp']\n",
    "            else:\n",
    "                if rtt_param:\n",
    "                    _rownum = df_tmp_max_class[[cn,'sched','fn','bdp']].groupby(['sched','fn','bdp']).ngroup().unique()\n",
    "                    ref_df = df_tmp_max_class[[cn,'sched','fn','bdp']].groupby(['sched','fn','bdp']).mean()\n",
    "                    _levelLabels = ['sched','fn','bdp']\n",
    "                else:\n",
    "                    _rownum = df_tmp_max_class[[cn,'sched','rtt','fn','bdp']].groupby(['rtt','sched','fn','bdp']).ngroup().unique()\n",
    "                    ref_df = df_tmp_max_class[[cn,'sched','rtt','fn','bdp']].groupby(['rtt','sched','fn','bdp']).mean()\n",
    "                    _levelLabels = ['rtt','sched','fn','bdp']\n",
    "\n",
    "            if ref_df.count()[cn] > 0:\n",
    "                ref_df.plot(style=class_colors[cn],stacked=False,ax=ax, subplots=True, ms=12)\n",
    "                rownum = len(_rownum)\n",
    "                levelLabels = _levelLabels\n",
    "\n",
    "            i += 1\n",
    "    else:\n",
    "        for sched in schedulers:\n",
    "            (df_tmp,rownum,_levLab) = getSchedulerData(metrika, c, sched[0])\n",
    "            if not df_tmp.empty:\n",
    "                df_tmp.plot(style=sched[1],stacked=False,ax=ax, subplots=True, ms=12)\n",
    "                ref_df = df_tmp\n",
    "                levelLabels = _levLab\n",
    "    \n",
    "            \n",
    "    if ref_df is None:\n",
    "        print(\"EMPTY (Check the rtt / bw filter)\")\n",
    "    else:\n",
    "        # COMMENT: PLOTTING START HERE!!!!!!!\n",
    "        #Below 3 lines remove default labels\n",
    "        labels = ['' for item in ax.get_xticklabels()]\n",
    "        ax.set_xticklabels(labels)\n",
    "        ax.set_xlabel('')\n",
    "        \n",
    "        handles, _labels = ax.get_legend_handles_labels()\n",
    "        ax.legend(labels=[replaceLabel(x) for x in _labels])\n",
    "\n",
    "        if metrika in yaxisLabel:\n",
    "            ax.set_ylabel(yaxisLabel[metrika])\n",
    "\n",
    "        # COMMENT: main label plotting function call \n",
    "        label_group_bar_table(ax, ref_df)\n",
    "\n",
    "        plt.xlim(left=-.5,right=len(ref_df)-0.5)\n",
    "        if metrika == 'jains':\n",
    "            plt.ylim(top=1.1,bottom=-.05)\n",
    "        plt.legend\n",
    "        \n",
    "        _rownum, _ = ref_df.shape\n",
    "        major_ticks = np.arange(0, rownum, 1)\n",
    "        ax.set_xticks(major_ticks)\n",
    "        ax.grid(which='major')\n",
    "        \n",
    "        # for presentation!!! we needed a zoomed version of the plot\n",
    "        #plt.ylim(top=2.2)\n",
    "        plt.gcf().subplots_adjust(bottom=.01*ref_df.index.nlevels)\n",
    "        \n",
    "        _outFilename = str(c)+\"_\"+str(metrika)+\"_\"+bw_param+('' if rtt_param is None else \"_\"+rtt_param)\n",
    "        \n",
    "        if saveImage:\n",
    "            #plt.gcf().subplots_adjust(bottom=.01*ref_df.index.nlevels)\n",
    "            print(\"SAVE: BufferImages/newPlot/\"+str(scenarioFolder)+\"/\"+_outFilename+\".png\")\n",
    "            fig.savefig(\"BufferImages/newPlot/\"+str(scenarioFolder)+\"/\"+_outFilename+\".png\",format=\"png\",bbox_inches=\"tight\")\n",
    "            print(\"SAVE: BufferImages/newPlot/\"+str(scenarioFolder)+\"/\"+_outFilename+\".eps\")\n",
    "            fig.savefig(\"BufferImages/newPlot/\"+str(scenarioFolder)+\"/\"+_outFilename+\".eps\",format=\"eps\",bbox_inches=\"tight\")\n",
    "            plt.close(fig)\n",
    "        else:\n",
    "            #plt.gcf().subplots_adjust(bottom=.01*ref_df.index.nlevels)\n",
    "            plt.show()\n",
    "            \n",
    "         \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": "3.6.8"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}