<tr class='row1'>
			<td id='shoutbox-shouts-td' valign='top' class='altrow'>
				<div id='shoutbox-shouts' style='height:406px;overflow:auto;'>
					<table id='shoutbox-shouts-table'>
						<tbody>
							<tr class='row2' id='shout-row-19835'>
	<td style='width: 1%; white-space: nowrap;'>
			<a href="#" class="at_member" data-store="kobzar" title="Insert member's name into Shout">@</a>&nbsp;
		<a hovercard-ref="member" hovercard-id="21788" data-ipb="noparse" data-ipb="noparse" class="_hovertrigger url fn name  " href='http://lanos.com.ua/forum/user/217-kobzar/' title='Просмотр профиля'><span itemprop="name"><span style='color:#000000;font-weight:bold;'>kobzar</span></span></a>
	</td>
	<td style='width: 1%; white-space: nowrap;'>:</td>
	<td style='width: 98%;'>
		<span class='right desc' title=''>(03/03/15 - 08:18:51)</span>
			<span class='shoutbox_text'>Фтарой!</span>
	</td>
</tr><tr class='row2' id='shout-row-1734'>
	<td style='width: 1%; white-space: nowrap;'>
			<a href="#" class="at_member" data-store="lyolik" title="Insert member's name into Shout">@</a>&nbsp;
		<a hovercard-ref="member" hovercard-id="574" data-ipb="noparse" data-ipb="noparse" class="_hovertrigger url fn name  " href='http://lanos.com.ua/forum/user/574-lyolik/' title='Просмотр профиля'><span itemprop="name"><span style='color:#0000ff;font-weight:bold;'>lyolik</span></span></a>
	</td>
	<td style='width: 1%; white-space: nowrap;'>:</td>
	<td style='width: 98%;'>
		<span class='right desc' title=''>(03/03/15 - 08:13:56)</span>
			<span class='shoutbox_text'>Утро доброе!</span>
	</td>
</tr>

# -*- coding: utf-8 -*-
from lxml import html
import requests
page = requests.get('http://lanos.com.ua/forum/shoutbox/')
tree = html.fromstring(page.text)
data = tree.xpath('//span[@class="right desc"]/text()')
mess = tree.xpath('//span[@class="shoutbox_text"]/text()')

<a hovercard-ref="member"...

import requests
from lxml.etree import XPath
from lxml.html import fromstring
def get_chat_rows():
    page = requests.get('http://lanos.com.ua/forum/shoutbox/')
    tree = fromstring(page.text)
    parts = XPath("//tr/*/a/@href|//tr/*/span/@style|//tr/*/span/text()|//tr/*/span/a/text()")
    parts = parts(tree)[1:]
    
    row = []
    for part in parts:
        if part == "#":
            yield row
            row[:] = []
            continue
        row.append(part)
        
for row in get_chat_rows():
    print(row[:3] + ["\n".join(row[3:])])