Project: ThruDB

Revision: 358

Author: thai

Date: 22 Apr 2008 05:24:07

Diff at Trac: http://trac.thrudb.org/changeset/358

• /trunk/tutorial/tweetsearch/py/settings.py

  ......	@@ -4,7 +4,7 @@
  4	4	TEMPLATE_DEBUG = DEBUG
  5	5
  6	6	ADMINS = (
  7		('Thai Duong', 'thai@meetaa.com'),
  	7	('Thai Duong', 'thaidn@gmail.com'),
  8	8	)
  9	9
  10	10	MANAGERS = ADMINS
  ......	@@ -45,7 +45,7 @@
  45	45	# URL prefix for admin media -- CSS, JavaScript and images. Make sure to use a
  46	46	# trailing slash.
  47	47	# Examples: "http://foo.com/media/", "/media/".
  48		ADMIN_MEDIA_PREFIX = ''
  	48	ADMIN_MEDIA_PREFIX = '/media/'
  49	49
  50	50	# Make this unique, and don't share it with anybody.
  51	51	SECRET_KEY = '4bfqq(kaufg*un4rb_z8_%f%f094l8)6x)rn0l*4sf!(ur_%gx'

• /trunk/tutorial/tweetsearch/py/settings.pyc

  svn:mime-type = application/octet-stream

• /trunk/tutorial/tweetsearch/py/readme.txt

  ......	@@ -1,36 +1,38 @@
  1		HOW TO RUN
  	1	==========================
  	2	= HOW TO RUN
  	3	==========================
  2	4
  3		1. install django (http://www.djangoproject.com), apache, mod_python, and other necessary components
  	5	0. install django, see http://www.djangoproject.com for details
  4	6
  5		2. catch the tweet by running ./search/twitter.py, you can see the processs in /tmp/twitter.log
  	7	1. install thrudb, see http://www.thrudb.org for details
  6	8
  7		3. config apache with mod_python. Here's a sample conf:
  	9	2. install thrudex/thrudoc python libraries:
  8	10
  9		<VirtualHost *>
  10		ServerName tweetsearch.local
  11		DocumentRoot /path/to/thrudb/tutorial/tweetsearch/py
  12		<Location "/">
  13		SetHandler python-program
  14		PythonHandler django.core.handlers.modpython
  15		SetEnv DJANGO_SETTINGS_MODULE py.settings
  16		SetEnv PYTHON_EGG_CACHE /tmp
  17		PythonDebug On
  18		PythonPath "['/path/to/thrudb/tutorial/tweetsearch/py'] + sys.path"
  19		</Location>
  	11	$ cd thrudb/tutorial
  	12	$ make
  	13	$ sudo cp -pvr gen-py/Thrudex gen-py/Thrudoc /usr/lib/python2.5/site-packages/
  20	14
  21		</VirtualHost>
  	15	if you use python2.4, the last command should look like:
  	16	$ sudo cp -pvr gen-py/Thrudex gen-py/Thrudoc /usr/lib/python2.4/site-packages/
  22	17
  23		remember add this line to /etc/hosts:
  24		127.0.0.1 tweetsearch.local
  	18	3. review thrudex/thrudoc configuration in thrudex.conf and thrudoc.conf respectively. if everything's okie, let's start thrudb:
  25	19
  26		4. start apache, thrudex and thrudoc
  	20	$ cd thrudb/tutorial
  	21	$ ./thrudbctl start
  27	22
  28		5. that's it! Contact me if you have any problem.
  	23	4. start grabbing tweets from http://www.twitter.com:
  	24	$ cd thrudb/tutorial/tweetsearch/py
  	25	$ ./search/twitter.py
  29	26
  30		Thai Duong (thaidn@gmail.com).
  	27	The application will be running as a daemon. You can see the processs in /tmp/twitter.log, thrudex.log, and thrudoc.log
  31	28
  	29	5. start the django application
  32	30
  	31	$ cd thrudb/tutorial/tweetsearch/py
  	32	$ python manage.py runserver
  33	33
  	34	the application is available at http://localhost:8000/
  34	35
  	36	6. that's it! Contact me if you have any problem.
  35	37
  36
  	38	Thai Duong (thaidn@gmail.com).

• /trunk/tutorial/tweetsearch/py/urls.py

  ......	@@ -7,5 +7,5 @@
  7	7	# Uncomment this for admin:
  8	8	# (r'^admin/', include('django.contrib.admin.urls')),
  9	9	# catch all
  10		(r'^.*$', 'py.search.views.search')
  	10	(r'^.*$', 'py.search.views.search'),
  11	11	)

• /trunk/tutorial/tweetsearch/py/search/twitter.py

  ......	@@ -27,7 +27,7 @@
  27	27	THRUDOC_BUCKET = "tweets";
  28	28	THRUDEX_INDEX = "tweets";
  29	29
  30		class TweetCatcher(object):
  	30	class TweetManager(object):
  31	31	def __init__(self, since_id=None):
  32	32	self.connect_to_thrudoc()
  33	33	self.connect_to_thrudex()
  ......	@@ -71,7 +71,7 @@
  71	71	def save_tweet(self, tweet):
  72	72	self.thrudoc.put(THRUDOC_BUCKET, str(tweet["id"]), cjson.encode(tweet))
  73	73
  74		def run(self):
  	74	def grab_tweet(self):
  75	75	while True:
  76	76	# the random paramater used to avoid http caching by upstream provider
  77	77	url = "http://twitter.com/statuses/public_timeline.json?since_id=%s&r=%s" % (self.since_id, random())
  ......	@@ -96,27 +96,6 @@
  96	96	print e
  97	97	continue
  98	98	print "loaded %s tweets, last since_id %s" % (self.count, self.since_id)
  99
  100		class TweetManager(object):
  101		def __init__(self):
  102		self.connect_to_thrudoc()
  103		self.connect_to_thrudex()
  104
  105		def connect_to_thrudoc(self):
  106		socket = TSocket('localhost', THRUDOC_PORT)
  107		transport = TFramedTransport(socket)
  108		protocol = TBinaryProtocol(transport)
  109		self.thrudoc = Thrudoc.Client(protocol)
  110		transport.open()
  111		self.thrudoc.admin("create_bucket", THRUDOC_BUCKET)
  112
  113		def connect_to_thrudex(self):
  114		socket = TSocket('localhost', THRUDEX_PORT)
  115		transport = TFramedTransport(socket)
  116		protocol = TBinaryProtocol(transport)
  117		self.thrudex = Thrudex.Client(protocol)
  118		transport.open()
  119		self.thrudex.admin("create_index", THRUDEX_INDEX)
  120	99
  121	100	def search_tweet(self, terms, offset=0, limit=10):
  122	101	q = ThrudexTypes.SearchQuery()
  ......	@@ -147,12 +126,10 @@
  147	126	doc.bucket = THRUDOC_BUCKET
  148	127	doc.key = ele.key
  149	128	docs.append(doc)
  150
  151		return docs
  152
  	129	return docs
  153	130
  154	131	if __name__ == "__main__":
  155	132	import daemonize as dm
  156	133	dm.daemonize('/dev/null','/tmp/twitter.log','/tmp/twitter.log')
  157		tc = TweetCatcher()
  158		tc.run()
  	134	tc = TweetManager()
  	135	tc.grab_tweet()