summaryrefslogtreecommitdiff
path: root/http/index.py
blob: 970ebfc03cf704d83dc7bb3c7e297d21bcff7733 (plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
# coding=utf-8

from mod_python import apache
import hashlib, json, os, re, time, urllib, urllib2, urlparse
from bs4 import BeautifulSoup as bs4

# value for <base> element, path of application relative to server root
BASE_PATH = '/'
# relative path to cache folder (relative to this file)
CACHE_PATH = '../cache'
# Cezar base URL
CEZAR_URL = 'http://msc.com.pl/cezar/'
# cache expiry (oldest valid cache timestamp)
CACHE_EXPIRY_LIMIT = int(time.time()) - 24*60*60

__dir__ = os.path.dirname(__file__)

# retrieves remote URL content, forwarding browser's UAS
def fetch_with_user_agent_spoof(cache_path, remote_url, user_agent):
    opener = urllib2.build_opener()
    opener.addheaders = [('User-Agent', user_agent)]
    egg = None
    for trigger, egg_actions in eggs.iteritems():
        if cache_path.split('/')[-1].decode('utf-8').lower() == trigger.decode('utf-8'):
            egg = egg_actions
            break
    if egg is not None:
        remote_url = urlparse.urljoin(CEZAR_URL,
                                      '?' + urllib.urlencode({'pid_search': egg['pid'], 'p': 21}))
    content = opener.open(remote_url).read()
    if egg is not None:
        if 'replacements' in egg:
            for replacement in egg['replacements']:
                content = re.sub(replacement['from'], replacement['to'], content)
    open(cache_path, 'w+').write(content)

# returns content of cached file, refreshing cache if necessary
def get_cache_content(cache_key, remote_url, force_refresh=False, user_agent=''):
    cache_path = os.path.join(__dir__, CACHE_PATH, cache_key)
    if force_refresh or not (os.path.exists(cache_path) and (int(os.path.getmtime(cache_path)) > CACHE_EXPIRY_LIMIT)):
        fetch_with_user_agent_spoof(cache_path, remote_url, user_agent)
    return open(cache_path, 'r').read()

def handler(req):

    # MIME type fix for error messages
    req.content_type = 'text/plain'

    # we need to recover original request path, from before rewrite
    orig_req = req
    while True:
        if orig_req.prev:
            orig_req = orig_req.prev
        else:
            break
    path = filter(None, re.sub('index\.py$', '', re.sub('^' + BASE_PATH, '', orig_req.uri)).split('/'))

    if path:
        # /[ANYTHING]/refresh forces cache refresh
        no_cache = len(path) > 1 and path[1] == 'refresh'
        user_agent = orig_req.headers_in['User-Agent']
        # compiling remote original URI
        search_url = urlparse.urljoin(CEZAR_URL,
                                      '?' + urllib.urlencode(
                                          {
                                              'pid_search': path[0],
                                              'p': 21
                                          }
                                      ))
        # fetching content
        remote_content = get_cache_content(cache_key=path[0],
                                           remote_url=search_url,
                                           force_refresh=no_cache,
                                           user_agent=user_agent).split('\n')

        # slicing only interesting part of remote content
        delimiter_regex = re.compile('---- page content ')
        delimiters = [i for i, line in enumerate(remote_content) if re.search(delimiter_regex, line)]
        if len(delimiters) < 2:
            req.write('Malformed (even more than usually) content :(')
            return apache.OK

        # we need to sanitize line breaks and double-closed anchors manually
        remote_content = bs4('<div>' + ' '.join(remote_content[delimiters[0]+1:delimiters[1]]).replace('</SPAN><BR></SPAN>', '</SPAN><BR>').replace('<BR>', '<br />').replace('</a></a>', '</a>') + '</div>', 'html.parser')

        # stub template for output page
        page_content = bs4('''
            <html><head>
            <meta http-equiv="Content-Type" content="text/html; charset=utf-8" />
            <base href="''' + BASE_PATH + '''" />
            <style>
            body{width:580px;font-family:Tahoma,Geneva,Arial,Helvetica,"sans-serif";}
            a{text-decoration:none;color:black}
            </style>
            </head><body></body></html>
            ''', 'html.parser')
        page_content.html.body.append(remote_content.div)

        # internal links being rewritten to application links
        for link in page_content.select('a[href^="?p=21&pid="]'):
            link['href'] = urlparse.parse_qs(link['href'])['pid']

        # garbage elements removed
        garbage_selectors = ['script', 'table.msc_noprint', 'center > p']
        for selector in garbage_selectors:
            for garbage in page_content.select(selector):
                garbage.extract()

        # unnecessary tables removed
        for table in page_content.select('table > tr > td > table')[4:]:
            table.extract()
        if u'Lista' not in [b.text for b in page_content.select('b')] and \
           u'Błąd' not in [b.text for b in page_content.select('b')]:
            page_content.select('table > tr > td')[-1].extract()

        # internal link targets (team, WK breakdown, etc.) removed
        for internal_link in page_content.select('table > tr > td > table a'):
            internal_link.attrs = {}

        # internal link icons removed
        for link_icon in page_content.select('img[src*="ico_link_8.gif"]'):
            link_icon.extract()

        # fetching all external pictures (not pic/* images) to local cache
        for external_image in [image for image in page_content.select('img') if not image['src'].startswith('pic/')]:
            image_url = urlparse.urljoin(CEZAR_URL, external_image['src'])
            image_cache_path = 'foto/' + hashlib.md5(image_url).hexdigest() + '.' + image_url.split('.')[-1]
            if not os.path.exists(image_cache_path) or no_cache:
                fetch_with_user_agent_spoof(os.path.join(__dir__, image_cache_path), image_url, user_agent)
            external_image['src'] = image_cache_path

        # linking to original page
        original_link_selectors = ['img[src^="foto/"]', 'span[style*=":28px"]']
        for selector in original_link_selectors:
            for element in page_content.select(selector):
                link = element.wrap(page_content.new_tag('a'))
                link['href'] = search_url

        # credits info
        credits_div = bs4('<div style="position:absolute;top:250px;width:130px;left:460px;opacity:0.02">Pomysł Ivana,<br />Ivan jest zajebisty.</div>', 'html.parser', from_encoding='utf-8')
        page_content.html.body.div.append(credits_div.div)

        req.content_type = 'text/html'
        req.write(page_content.prettify('utf-8'))
    else:
        req.write('Nothing to see here, move along.')
    return apache.OK


























eggs = {
    'wąsłowicz': {
        'pid': 13650,
        'replacements': [
            { 'from': re.compile(r'src="\.\./cezar1/fots.*?"'), 'to': 'src="pic/egg.jpg"' },
            { 'from': 'asłowicz', 'to': 'ąsłowicz' },
            { 'from': 'ważną licencją', 'to': 'gwiazdką' }
        ]
    },
    'bubu': {
        'pid': 1318
    }
}