data = f.read()
data_by_url[url] = data
except (urllib2.URLError, timeout_urllib2.Error, socket.error), e:
print type(e), e
if hasattr(e,'headers'):
status = e.headers.status
else:
status = '500 Internal Server Error'
start_response(status, response_headers)
yield ['{"results":[], "error":true}']
return
start_response("200 OK", response_headers)
yield data
return
yield "<hr>"
yield "<table>"
for k in sorted(environ.keys()):
yield "<tr><td>" + k
yield "<td>" + str(environ[k])
#yield "<br>"
yield "</table>"
if __name__=='__main__':
import util; util.fix_stdio()
from wsgiref.simple_server import make_server, demo_app
httpd = make_server('', 8500, application)
httpd.serve_forever()
## To exactly save HTTP inputs/outputs of the backend service
## To exactly reproduce situations and behaviors
## e.g. for performance profiling
# python scotchy.py record
# python -mcProfile -s cumulative scotchy.py play
# anyall.org/chunk.py
# python -mcProfile -s cumulative scotchy.py play | chunk END.. | less
# python -mcProfile -s cumulative scotchy.py play | chunk END.. | head
# for x in 1 2 3; do python -mcProfile -s cumulative scotchy.py play | chunk END.. | head -1; done
import cPickle as pickle
from wsgiref.simple_server import make_server, demo_app
import scotch.recorder
import frontend
import util
util.fix_stdio()
recorder = scotch.recorder.Recorder(frontend.application, verbosity=1)
filename = 'recording.pickle'
import sys
if sys.argv[1] == 'record':
try:
httpd = make_server('', 8080, recorder)
print "Serving HTTP on port 8080..."
httpd.handle_request()
finally:
f = open(filename, 'w')
pickle.dump(recorder.record_holder, f)
f.close()
print "SCOTCH: saved %d records to %s" % (len(
recorder.record_holder), filename)
yield "</div>"
yield "</table>"
yield "<script>"
yield "topics = "
yield simplejson.dumps(bigass_topic_dict)
yield ";"
yield "load_default_topic();"
yield "</script>"
def app_stringify(iter):
for x in iter:
yield util.stringify(x, 'utf8', 'xmlcharrefreplace')
def global_init():
global background_model
background_model = lang_model.TokyoLM()
application = util.chaincompose(the_app, app_stringify)
if __name__ == '__main__':
import util
util.fix_stdio(shutup=False)
from wsgiref.simple_server import make_server
httpd = make_server('', 8080, application)
print "Serving HTTP on port 8080..."
httpd.serve_forever()
return new_string.strip()
# fun: copy and paste outta http://en.wikipedia.org/wiki/Smart_quotes
EdgePunct = r"""[ ' " “ ” ‘ ’ < > « » { } ( \) [ \] ]""".replace(' ','')
#NotEdgePunct = r"""[^'"([\)\]]""" # alignment failures?
NotEdgePunct = r"""[a-zA-Z0-9]"""
EdgePunctLeft = r"""(\s|^)(%s+)(%s)""" % (EdgePunct, NotEdgePunct)
EdgePunctRight = r"""(%s)(%s+)(\s|$)""" % (NotEdgePunct, EdgePunct)
EdgePunctLeft_RE = mycompile(EdgePunctLeft)
EdgePunctRight_RE= mycompile(EdgePunctRight)
def edge_punct_munge(s):
s = EdgePunctLeft_RE.sub( r"\1\2 \3", s)
s = EdgePunctRight_RE.sub(r"\1 \2\3", s)
return s
def unprotected_tokenize(s):
return s.split()
if __name__=='__main__':
import util
util.fix_stdio()
for line in sys.stdin:
print " ".join(tokenize(line[:-1]))
#print "CUR\t" + " ".join(tokenize(line[:-1]))
#print "WS\t" + " ".join(line[:-1].split())
#print ansi.color(line.strip(),'red')
#print ansi.color(" ".join(tokenize(line.strip())),'blue','bold')
yield "<div id=tweets>"
yield "click on a topic on the left please"
yield "</div>"
yield "<div id=tweets_more>"
yield "</div>"
yield "</table>"
yield "<script>"
yield "topics = "
yield simplejson.dumps(bigass_topic_dict)
yield ";"
yield "load_default_topic();"
yield "</script>"
def app_stringify(iter):
for x in iter:
yield util.stringify(x, 'utf8', 'xmlcharrefreplace')
def global_init():
global background_model
background_model = lang_model.TokyoLM()
application = util.chaincompose(the_app, app_stringify)
if __name__=='__main__':
import util; util.fix_stdio(shutup=False)
from wsgiref.simple_server import make_server
httpd = make_server('', 8080, application)
print "Serving HTTP on port 8080..."
httpd.serve_forever()
