iwla/plugins/display/robot_bandwidth.py

# -*- coding: utf-8 -*-
#
# Copyright Grégory Soutadé 2016

# This file is part of iwla

# iwla is free software: you can redistribute it and/or modify
# it under the terms of the GNU General Public License as published by
# the Free Software Foundation, either version 3 of the License, or
# (at your option) any later version.
#
# iwla is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
# GNU General Public License for more details.
#
# You should have received a copy of the GNU General Public License
# along with iwla.  If not, see <http://www.gnu.org/licenses/>.
#

import time

from iwla import IWLA
from iplugin import IPlugin
from display import *

"""
Display hook

Display top 10 robot bandwidth use

Plugin requirements :
    None

Conf values needed :
    display_visitor_ip*
    create_all_robot_bandwidth_page*

Output files :
    OUTPUT_ROOT/year/month/top_robots_bandwidth.html
    OUTPUT_ROOT/year/month/index.html

Statistics creation :
    None

Statistics update :
    None

Statistics deletion :
    None
"""

class IWLADisplayRobotBandwidth(IPlugin):
    def __init__(self, iwla):
        super(IWLADisplayRobotBandwidth, self).__init__(iwla)
        self.API_VERSION = 1
        self.display_visitor_ip = self.iwla.getConfValue('display_visitor_ip', False)
        self.create_all_pages = self.iwla.getConfValue('create_all_robot_bandwidth_page', True)

    def load(self):
        return True

    def hook(self):
        display = self.iwla.getDisplay()
        hits = self.iwla.getCurrentVisits()

        bandwidths = []
        for (k, super_hit) in hits.items():
            if not self.iwla.isRobot(super_hit):
                continue
            bandwidths.append((super_hit, super_hit['bandwidth'][0]))
        bandwidths.sort(key=lambda tup: tup[1], reverse=True)
        
        # All in a page
        if self.create_all_pages:
            title = createCurTitle(self.iwla, u'Robots bandwidth')
            filename = 'top_robots_bandwidth.html'
            path = self.iwla.getCurDisplayPath(filename)

            page = display.createPage(title, path, self.iwla.getConfValue('css_path', []))
            table = display.createBlock(DisplayHTMLBlockTable, title, [self.iwla._(u'Host'), self.iwla._(u'Bandwidth'), self.iwla._(u'Last seen')])
            table.setColsCSSClass(['', 'iwla_bandwidth', ''])
            for (super_hit, bandwidth) in bandwidths:
                address = super_hit['remote_addr']
                if self.display_visitor_ip and\
                   super_hit.get('dns_name_replaced', False):
                    address = '%s [%s]' % (address, super_hit['remote_ip'])

                row = [
                    address,
                    bytesToStr(bandwidth),
                    time.asctime(super_hit['last_access'])
                ]
                table.appendRow(row)
            page.appendBlock(table)

            display.addPage(page)

        title = self.iwla._(u'Robots bandwidth')
        if self.create_all_pages:
            link = '<a href=\'%s\'>%s</a>' % (filename, self.iwla._(u'All robots bandwidth'))
            title = '%s - %s' % (title, link)

        # Top in index
        index = self.iwla.getDisplayIndex()
        table = display.createBlock(DisplayHTMLBlockTable, title, [self.iwla._(u'Host'), self.iwla._(u'Bandwidth'), self.iwla._(u'Last seen')])
        table.setColsCSSClass(['', 'iwla_bandwidth', ''])

        for (super_hit, bandwidth) in bandwidths[:10]:
            address = super_hit['remote_addr']
            if self.display_visitor_ip and\
                    super_hit.get('dns_name_replaced', False):
                address = '%s [%s]' % (address, super_hit['remote_ip'])

            row = [
                address,
                bytesToStr(bandwidth),
                time.asctime(super_hit['last_access'])
                ]
            table.appendRow(row)
        index.appendBlock(table)