#
# main
#
if len(sys.argv) != 3:
print """Usage:
%s username password
""" % sys.argv[0]
exit(0)
# set up trends. These are the default API access codes for the demo instance.
key = "f9aaf865-b89a-444d-9070-38ec6666e539"
secret = "9e13e4ac-ad93-4c8f-a896-d5a937b84c8a"
c = StreamDrillClient("http://localhost:9669", key, secret)
# Create a trend.
#
# The reason we're storing host and path seperately is because this allows
# us to filter for the host.
c.create("twitter-links", "host:path", 10000, ("day", "hour", "minute"))
c.clear("twitter-links")
# This will create a link-out button linking to the real-time search page
# of Twitter for that link. Works like a charm (well, mostly)
c.setMeta("twitter-links", "linkTemplate", "http://twitter.com/search/realtime?q=http://$1$2")
def parsedate(ds):
return datetime.datetime(*email.utils.parsedate(ds)[:6])
for example:
%s http-blog.mikiobraun.de.log""".replace("%s", sys.argv[0]))
sys.exit(0)
site = "http://blog.mikiobraun.de/"
pageViews = "page-views"
referers = "referers"
visitors = "visitors"
# 65.55.215.69 - - [01/Aug/2013:00:02:07 +0200] "GET /robots.txt HTTP/1.1" 410 1129 "-" "msnbot-media/1.1 (+http://search.msn.com/msnbot.htm)"
# group 1 2 3 4 5 6 7 8 9
logline = re.compile(
r"([0-9]+\.[0-9]+\.[0-9]+\.[0-9]+) (\S+) (\S+) \[([^\]]+)\] \"GET ([^\"]+) HTTP/1\.1\" ([0-9]+) ([0-9]+) \"([^\"]+)\" \"([^\"]+)\""
)
client = StreamDrillClient("http://localhost:9669")
client.delete(pageViews)
client.delete(referers)
client.delete(visitors)
client.create(pageViews, "path", 1000, ("hour", "minute", "second"))
client.create(referers, "path:referer", 1000, ("hour", "minute", "second"))
client.create(visitors, "path:addr", 1000, ("hour", "minute", "second"))
stream = client.stream()
for line in open(sys.argv[1]):
result = logline.match(line)
if result:
#print(result.groups())
addr = result.group(1)
path = result.group(5)
