API tools faq
paste
Advertisement
Guest User

browscap.py

a guest
Dec 10th, 2011
285
0
Never
Add comment
Not a member of Pastebin yet? Sign Up, it unlocks many cool features!
Python 9.49 KB | None | 0 0
raw download clone embed print report
  1. from ConfigParser import SafeConfigParser as ConfigParser
  2. import re
  3. import os
  4. import sys
  5.  
  6.  
  7. __all__ = [
  8.    "BrowserCapabilities"
  9. ]
  10.  
  11. MOBILE_UA=re.compile(".*(320x240|240x320|windows\sce|up.browser|up.link|mmp|symbian|smartphone|midp|wap|phone|vodafone|o2|pocket|mobile|pda|psp).*",re.IGNORECASE)
  12. MOBILE_UA_START=('acs-','alav','alca','amoi','audi','aste','avan','benq','bird','blac','blaz','brew','cell','cldc','cmd-','dang','doco','eric','hipt','htc','inno','ipaq','java','jigs','kddi','keji','leno','lg-c','lg-d','lg-g','lge-','maui','maxo','midp','mits','mmef','mobi','mot-','moto','mwbp','nec-','newt','noki','oper','opwv','palm','pana','pant','pdxg','phil','play','pluc','port','prox','qtek','qwap','sage','sams','sany','sch-','sec-','send','seri','sgh-','shar','sie-','siem','smal','smar','sony','sph-','symb','t-mo','teli','tim-','tosh','tsm-','upg1','upsi','vk-v','voda','wap-','wapa','wapi','wapp','wapr','webc','winw','winw','xda','xda-')
  13. BOT_UA=re.compile('.*(spyder|spider|googlebot|robot|bing|crawler).*',re.IGNORECASE)
  14.  
  15. def find_on_path(filename):
  16.    """Given a search path, find file
  17.   """
  18.    file_found = 0
  19.    paths = sys.path
  20.    for path in paths:
  21.        if os.path.exists(os.path.join(path, filename)):
  22.           return path
  23.    return None
  24.  
  25. BC_PATH=find_on_path('browscap.ini')
  26.  
  27. def mobile_useragent_check(useragent):
  28.     if MOBILE_UA.match(useragent) or useragent[:4] in MOBILE_UA_START:
  29.         return True
  30.     return False
  31.  
  32. def bot_useragent_check(useragent):
  33.     if BOT_UA.match(useragent):
  34.         return True
  35.     return False
  36.  
  37.  
  38. class Browser(object):
  39.     def __init__(self, capabilities):
  40.         self.lazy_flag = True
  41.         self.cap = capabilities
  42.  
  43.  
  44.     def parse(self):
  45.         for name, value in self.cap.items():
  46.             if name in ["tables", "aol", "javaapplets",
  47.                        "activexcontrols", "backgroundsounds",
  48.                        "vbscript", "win16", "javascript", "cdf",
  49.                        "wap", "crawler", "netclr", "beta",
  50.                         "iframes", "frames", "stripper", "wap"]:
  51.                 self.cap[name] = (value.strip().lower() == "true")
  52.             elif name in ["ecmascriptversion", "w3cdomversion"]:
  53.                 self.cap[name] = float(value)
  54.             elif name in ["css"]:
  55.                 self.cap[name] = int(value)
  56.             else:
  57.                 self.cap[name] = value
  58.         self.lazy_flag = False
  59.  
  60.  
  61.     def __repr__(self):
  62.         if self.lazy_flag: self.parse()
  63.         return repr(self.cap)
  64.  
  65.  
  66.     def get(self, name, default=None):
  67.         if self.lazy_flag: self.parse()
  68.         try:
  69.             return self[name]
  70.         except KeyError:
  71.             return default
  72.  
  73.  
  74.     def __getitem__(self, name):
  75.         if self.lazy_flag: self.parse()
  76.         return self.cap[name.lower()]
  77.  
  78.  
  79.     def keys(self):
  80.         return self.cap.keys()
  81.  
  82.  
  83.     def items(self):
  84.         if self.lazy_flag: self.parse()
  85.         return self.cap.items()
  86.  
  87.  
  88.     def values(self):
  89.         if self.lazy_flag: self.parse()
  90.         return self.cap.values()
  91.    
  92.  
  93.     def __len__(self):
  94.         return len(self.cap)
  95.  
  96.  
  97.     def supports(self, feature):
  98.         value = self.cap.get(feature)
  99.         if value == None:
  100.             return False
  101.         return value
  102.  
  103.  
  104.     def features(self):
  105.         l = []
  106.         for f in ["tables", "frames", "iframes", "javascript",
  107.                   "cookies", "w3cdomversion", "wap"]:
  108.             if self.supports(f):
  109.                 l.append(f)
  110.         if self.supports_java():
  111.             l.append("java")
  112.         if self.supports_activex():
  113.             l.append("activex")
  114.         css = self.css_version()
  115.         if css > 0:
  116.             l.append("css1")
  117.         if css > 1:
  118.             l.append("css2")
  119.         return l
  120.  
  121.  
  122.     def supports_tables(self):
  123.         return self.supports("frames")
  124.  
  125.     def supports_iframes(self):
  126.         return self.supports("iframes")
  127.  
  128.  
  129.     def supports_frames(self):
  130.         return self.supports("frames")
  131.  
  132.  
  133.     def supports_java(self):
  134.         return self.supports("javaapplets")
  135.  
  136.  
  137.     def supports_javascript(self):
  138.         return self.supports("javascript")
  139.  
  140.  
  141.     def supports_vbscript(self):
  142.         return self.supports("vbscript")
  143.  
  144.  
  145.     def supports_activex(self):
  146.         return self.supports("activexcontrols")
  147.  
  148.  
  149.     def supports_cookies(self):
  150.         return self.supports("cookies")
  151.  
  152.  
  153.     def supports_wap(self):
  154.         return self.supports("wap")
  155.  
  156.  
  157.     def css_version(self):
  158.         return self.get("css", 0)
  159.  
  160.  
  161.     def version(self):
  162.         major = self.get("majorver")
  163.         minor = self.get("minorver")
  164.         if major and minor:
  165.             return (major, minor)
  166.         elif major:
  167.             return (major, None)
  168.         elif minor:
  169.             return (None, minor)
  170.         else:
  171.             ver = self.get("version")
  172.             if ver and "." in ver:
  173.                 return tuple(ver.split(".", 1))
  174.             elif ver:
  175.                 return (ver, None)
  176.             else:
  177.                 return (None, None)
  178.  
  179.  
  180.     def dom_version(self):
  181.         return self.get("w3cdomversion", 0)
  182.  
  183.  
  184.     def is_bot(self):
  185.         ca=getattr(self,'_bot_cache',None)
  186.         if ca is None:
  187.             ca = bot_useragent_check(self.name()) or (self.get("crawler") == True)
  188.             self._bot_cache=ca
  189.         return ca
  190.  
  191.  
  192.     def is_mobile(self):
  193.         ca=getattr(self,'_mobile_cache',None)
  194.         if ca is None:
  195.             ca = mobile_useragent_check(self.name()) or (self.get("ismobiledevice") == True)
  196.             self._mobile_cache=ca
  197.         return ca
  198.  
  199.    
  200.     def name(self):
  201.         return self.get("browser")
  202.  
  203.  
  204.  
  205.  
  206. class BrowserCapabilities(object):
  207.  
  208.  
  209.     def __new__(cls, *args, **kwargs):
  210.         # Only create one instance of this clas
  211.         if "instance" not in cls.__dict__:
  212.             cls.instance = object.__new__(cls, *args, **kwargs)
  213.         return cls.instance
  214.  
  215.  
  216.     def __init__(self):
  217.         self.cache = {}
  218.         self.parse()
  219.  
  220.  
  221.     def parse(self):
  222.         cfg = ConfigParser()
  223.         files = ("browscap.ini", "bupdate.ini")
  224.         read_ok = cfg.read([os.path.join(BC_PATH, name) for name in files])
  225.         if len(read_ok) == 0:
  226.             raise IOError, "Could not read browscap.ini, " + \
  227.                   "please get it from http://www.GaryKeith.com"
  228.         self.sections = []
  229.         self.items = {}
  230.         self.browsers = {}
  231.         parents = set()
  232.         for name in cfg.sections():
  233.             qname = name
  234.             for unsafe in list("^$()[].-"):
  235.                 qname = qname.replace(unsafe, "\%s" % unsafe)
  236.             qname = qname.replace("?", ".").replace("*", ".*?")
  237.             qname = "^%s$" % qname
  238.             sec_re = re.compile(qname)
  239.             sec = dict(regex=qname)
  240.             sec.update(cfg.items(name))
  241.             p = sec.get("parent")
  242.             if p: parents.add(p)
  243.             self.browsers[name] = sec
  244.             if name not in parents:
  245.                 self.sections.append(sec_re)
  246.             self.items[sec_re] = sec
  247.  
  248.  
  249.     def query(self, useragent):
  250.         b = self.cache.get(useragent)
  251.         if b: return b
  252.  
  253.         for sec_pat in self.sections:
  254.             if sec_pat.match(useragent):
  255.                 browser = dict(agent=useragent)
  256.                 browser.update(self.items[sec_pat])
  257.                 parent = browser.get("parent")
  258.                 while parent:
  259.                     items = self.browsers[parent]
  260.                     for key, value in items.items():
  261.                         if key not in browser.keys():
  262.                             browser[key] = value
  263.                         elif key == "browser" and value != "DefaultProperties":
  264.                             browser["category"] = value # Wget, Godzilla -> Download Managers
  265.                     parent = items.get("parent")
  266.                 if browser.get("browser") != "Default Browser":
  267.                     b = Browser(browser)
  268.                     self.cache[useragent] = b
  269.                     return b
  270.         self.cache[useragent] = None
  271.  
  272.  
  273.     __call__ = query
  274.  
  275.  
  276.  
  277. def test():
  278.     bc = BrowserCapabilities()
  279.     for agent in [
  280.         "Mozilla/5.0 (compatible; Konqueror/3.5; Linux; X11; de) KHTML/3.5.2 (like Gecko) Kubuntu 6.06 Dapper",
  281.         "Mozilla/5.0 (X11; U; Linux i686; de; rv:1.8.0.5) Gecko/20060731 Ubuntu/dapper-security Firefox/1.5.0.5",
  282.         "Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.7.12) Gecko/20060216 Debian/1.7.12-1.1ubuntu2",
  283.         "Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.8.0.5) Gecko/20060731 Ubuntu/dapper-security Epiphany/2.14 Firefox/1.5.0.5",
  284.         "Opera/9.00 (X11; Linux i686; U; en)",
  285.         "Wget/1.10.2",
  286.         "Mozilla/5.0 (X11; Linux i686; U;) Gecko/20051128 Kazehakase/0.3.3 Debian/0.3.3-1",
  287.         "Mozilla/5.0 (X11; U; Linux i386) Gecko/20063102 Galeon/1.3test",
  288.         "Mozilla/4.0 (compatible; MSIE 6.0; Windows 98)" # Tested under Wine
  289.       ]:
  290.         b = bc(agent)
  291.         if not b:
  292.             print "!", agent
  293.         else:
  294.             print b.name(), b.version(), b.get("category", ""), b.features()
  295.  
  296.  
  297. def update():
  298.     import urllib
  299.     urllib.urlretrieve("http://browsers.garykeith.com/stream.asp?BrowsCapINI",
  300.                        "browscap.ini")
  301.  
  302.  
  303. if __name__ == "__main__":
  304.     import sys, os
  305.     bc_filename = os.path.join(BC_PATH, "browscap.ini")
  306.     if not os.path.exists(bc_filename) or "-update" in sys.argv[1:]:
  307.         print "Downloading browser database to %r..." % bc_filename,
  308.         update()
  309.         print "done"
  310.     test()
  311.  
  312.  
Advertisement
Add Comment
Please, Sign In to add comment
Advertisement
Public Pastes
Advertisement
We use cookies for various purposes including analytics. By continuing to use Pastebin, you agree to our use of cookies as described in the Cookies Policy.  OK, I Understand
Not a member of Pastebin yet?
Sign Up, it unlocks many cool features!