def loadPickledFiles():
global g_wnWords, g_wnWordIndex, g_wnDictPath, g_wnIndexPath, g_wnWordsPath
print "loading wn dictionary data files"
if not arsutils.fFileExists(g_wnDictPath):
print "WordNet dictionary file '%s' doesn't exist" % g_wnDictPath
return False
if not arsutils.fFileExists(g_wnIndexPath):
print "WordNet dictionary file '%s' doesn't exist" % g_wnIndexPath
return False
if not arsutils.fFileExists(g_wnWordsPath):
print "WordNet dictionary file '%s' doesn't exist" % g_wnWordsPath
return False
try:
fo = open(g_wnIndexPath, "rb")
g_wnWordIndex = cPickle.load(fo)
fo.close()
fo = open(g_wnWordsPath, "rb")
g_wnWords = cPickle.load(fo)
fo.close()
except Exception, ex:
print arsutils.exceptionAsStr(ex)
return False
def _merge_temps(self):
if not os.path.exists(ebooks.g_storage):
return
temps = [os.path.join(ebooks.g_storage, name) for name in os.listdir(ebooks.g_storage) if self.temp_file_pattern.match(name)]
if 0 == len(temps):
return
print "Merging temporary segments."
temps.sort()
try:
for temp in temps:
f = file(temp, "rb")
try:
data = cPickle.load(f)
finally:
f.close()
os.remove(temp)
for letter, index, book in data:
letter_data = self._data.get(letter, None)
if letter_data is None:
letter_data = [0, []]
self._data[letter] = letter_data
assert index <= letter_data[0]
letter_data[1].insert(index, book)
letter_data[0] += 1
except Exception, ex:
print exceptionAsStr(ex)
def loadPickledFiles():
global g_wnWords, g_wnWordIndex, g_wnDictPath, g_wnIndexPath, g_wnWordsPath
print "loading wn dictionary data files"
if not arsutils.fFileExists(g_wnDictPath):
print "WordNet dictionary file '%s' doesn't exist" % g_wnDictPath
return False
if not arsutils.fFileExists(g_wnIndexPath):
print "WordNet dictionary file '%s' doesn't exist" % g_wnIndexPath
return False
if not arsutils.fFileExists(g_wnWordsPath):
print "WordNet dictionary file '%s' doesn't exist" % g_wnWordsPath
return False
try:
fo = open(g_wnIndexPath, "rb")
g_wnWordIndex = cPickle.load(fo)
fo.close()
fo = open(g_wnWordsPath, "rb")
g_wnWords = cPickle.load(fo)
fo.close()
except Exception, ex:
print arsutils.exceptionAsStr(ex)
return False
def run(self):
try:
self.count = self._spider._spider_letter_range(self._letters)
except _StopSpidering:
return
except Exception, ex:
print exceptionAsStr(ex)
self._spider._finish.set()
def retrieveRequests():
global g_lastRequestLogId, g_dailyStats, g_modifiedDays
cursor = None
conn = getConnection()
try:
if None == g_lastRequestLogId:
sql = "SELECT request_id, user_id, DATE_FORMAT(log_date, '%Y-%m-%d'), free_p, request, result, error FROM request_log ORDER BY request_id;"
else:
sql = "SELECT request_id, user_id, DATE_FORMAT(log_date, '%Y-%m-%d'), free_p, request, result, error FROM request_log WHERE request_id > " + str(
g_lastRequestLogId) + " ORDER BY request_id;"
cursor = conn.cursor()
cursor.execute(sql)
processed = 0
prev_id = -1
while True:
row = cursor.fetchone()
if None == row:
break
reqData = RequestData()
reqData.request_id = row[0]
reqData.user_id = row[1]
reqData.log_date = row[2]
reqData.free_p = row[3]
reqData.request = row[4]
reqData.result = row[5]
reqData.error = row[
6] # it's either a number or None if there was no error
assert reqData.request_id > prev_id
prev_id = reqData.request_id
logDate = reqData.log_date
if g_dailyStats.has_key(logDate):
g_dailyStats[logDate].append(reqData)
else:
g_dailyStats[logDate] = [reqData]
if not g_modifiedDays.has_key(logDate):
g_modifiedDays[logDate] = 1
if reqData.request_id > g_lastRequestLogId:
g_lastRequestLogId = reqData.request_id
processed += 1
cursor.close()
# print "processed %d requests" % processed
except _mysql_exceptions.Error, ex:
if cursor:
cursor.close()
#log(SEV_HI, arsutils.exceptionAsStr(ex))
print "exception in retrieveRequests()"
print arsutils.exceptionAsStr(ex)
def retrieveRequests():
global g_lastRequestLogId, g_dailyStats, g_modifiedDays
cursor = None
conn = getConnection()
try:
if None == g_lastRequestLogId:
sql = "SELECT request_id, user_id, DATE_FORMAT(log_date, '%Y-%m-%d'), free_p, request, result, error FROM request_log ORDER BY request_id;"
else:
sql = "SELECT request_id, user_id, DATE_FORMAT(log_date, '%Y-%m-%d'), free_p, request, result, error FROM request_log WHERE request_id > " + str(g_lastRequestLogId) + " ORDER BY request_id;"
cursor = conn.cursor()
cursor.execute(sql)
processed = 0
prev_id = -1
while True:
row = cursor.fetchone()
if None == row:
break
reqData = RequestData()
reqData.request_id = row[0]
reqData.user_id = row[1]
reqData.log_date = row[2]
reqData.free_p = row[3]
reqData.request = row[4]
reqData.result = row[5]
reqData.error = row[6] # it's either a number or None if there was no error
assert reqData.request_id > prev_id
prev_id = reqData.request_id
logDate = reqData.log_date
if g_dailyStats.has_key(logDate):
g_dailyStats[logDate].append(reqData)
else:
g_dailyStats[logDate] = [reqData]
if not g_modifiedDays.has_key(logDate):
g_modifiedDays[logDate] = 1
if reqData.request_id > g_lastRequestLogId:
g_lastRequestLogId = reqData.request_id
processed += 1
cursor.close()
# print "processed %d requests" % processed
except _mysql_exceptions.Error, ex:
if cursor:
cursor.close()
#log(SEV_HI, arsutils.exceptionAsStr(ex))
print "exception in retrieveRequests()"
print arsutils.exceptionAsStr(ex)
def retrieveUsers():
global g_lastUserId, g_userStats
cursor = None
conn = getConnection()
try:
sql = "SELECT user_id, device_info, DATE_FORMAT(cookie_issue_date, '%Y-%m-%d'), reg_code, DATE_FORMAT(registration_date, '%Y-%m-%d'), disabled_p FROM users ORDER BY user_id;"
#sql = "SELECT user_id, device_info, DATE_FORMAT(cookie_issue_date, '%Y-%m-%d'), reg_code, DATE_FORMAT(registration_date, '%Y-%m-%d'), disabled_p FROM users WHERE user_id > " + str(g_lastUserId) + "ORDER BY user_id;";
cursor = conn.cursor()
cursor.execute(sql)
processed = 0
prev_id = -1
while True:
row = cursor.fetchone()
if None == row:
break
userData = UserData()
userData.user_id = row[0]
userData.device_info = row[1]
userData.cookie_issue_date = row[2]
userData.reg_code = row[3]
userData.registration_date = row[4]
userData.disabled_p = row[5]
if None == userData.registration_date:
userData.fRegistered = False
else:
userData.fRegistered = True
if userData.user_id > g_lastUserId:
userData.new_user_p = True
else:
userData.new_user_p = False
assert userData.user_id > prev_id
prev_id = userData.user_id
g_userStats.append(userData)
if userData.user_id > g_lastUserId:
g_lastUserId = userData.user_id
processed += 1
cursor.close()
# print "processed %d requests" % processed
except _mysql_exceptions.Error, ex:
if cursor:
cursor.close()
#log(SEV_HI, arsutils.exceptionAsStr(ex))
print "exception in retrieveUsers()"
print arsutils.exceptionAsStr(ex)
def _retrieve_xe():
global _g_xe_url
formData = {
"basecur": "USD",
"historical": "false",
"month": "1",
"day": "1",
"year": "2004",
"sort_by": "code",
"template": "ict-en"
}
encFormData = urllib.urlencode(formData)
headers = {
#"Host": getHostFromUrl(_g_xe_url),
"User-Agent":
"Mozilla/4.0 (compatible; MSIE 6.0; Windows NT 5.1; .NET CLR 1.1.4322)",
"Referer": _g_xe_url
}
request = urllib2.Request(_g_xe_url, encFormData, headers)
opener = urllib2.build_opener(urllib2.HTTPRedirectHandler())
htmlText = None
result = None
try:
result = opener.open(request)
htmlText = result.read()
except Exception, ex:
txt = exceptionAsStr(ex)
log(SEV_EXC,
"failed to retrieve data for '%s'\nreason:%s\n" % (_g_xe_url, txt))
def _spider_book_info(url, letter):
try:
html = getHttp(url, handleException = False)
soup = BeautifulSoup()
soup.feed(html)
h1 = soup.first("h1")
if h1 is None:
return None
assert h1 is not None
title = retrieveContents(h1).decode("iso-8859-1")
subtitle = None
author = None
code = None
labels = [retrieveContents(x) for x in soup.fetch("span", {"class": "title-label"})]
data = soup.fetch("span", {"class": "title-data"})
try:
index = labels.index("Subtitle")
subtitle = retrieveContents(data[index]).decode("iso-8859-1")
except ValueError:
pass
try:
index = labels.index("Author")
author = retrieveContents(data[index].first("a")).decode("iso-8859-1")
except ValueError:
pass
try:
index = labels.index("Language")
href = str(data[index].first("a", {"href": "/language.php?code=%"})["href"])
code = href[19:href.find("&", 19)].decode("iso-8859-1")
except ValueError:
pass
tid = soup.first("input", {"type": "hidden", "name": "tid"})
assert tid is not None
book_id = tid["value"].decode("iso-8859-1")
print (u"%s: \"%s\"" % (author, title)).encode("iso-8859-1", "ignore")
sel = soup.first("select", {"name": "book"})
assert sel is not None
opts = sel.fetch("option")
formats = []
for opt in opts:
try:
format = retrieveContents(opt).split()[0]
if format not in ebooks.FORMATS:
continue
val = opt["value"]
formats.append((format, val))
except Exception, ex:
log(SEV_EXC, exceptionAsStr(ex))
formats.sort()
return (url, title, subtitle, author, book_id, code, formats)
def _update_cache():
global _g_currency_cache, _g_currency_cache_lock, _g_retrieve_functions, _g_tracked_currencies, _g_cache_update_interval
t = threading.Timer(_g_cache_update_interval, _update_cache)
t.start()
out = {}
tracked = {}
tracked.update(_g_tracked_currencies)
for func in _g_retrieve_functions:
if 0 == len(tracked):
break
try:
res, data = func()
if RESULTS_DATA != res:
log(SEV_MED, "currency parser: %s returned result: %d" % (str(func), res))
continue
for item in data.iteritems():
key, value = item
#if key in ["KRW"]:
# print "Func: %s" % (str(func))
# print "Key: %s Value: %s " % (key, str(value))
if tracked.has_key(key) and 0 != value:
out[key] = value
del tracked[key]
except Exception, ex:
log(SEV_EXC, exceptionAsStr(ex))
def _retrieve_xe():
global _g_xe_url
formData = {
"basecur": "USD",
"historical": "false",
"month": "1",
"day": "1",
"year": "2004",
"sort_by": "code",
"template": "ict-en"
}
encFormData = urllib.urlencode(formData)
headers = {
#"Host": getHostFromUrl(_g_xe_url),
"User-Agent": "Mozilla/4.0 (compatible; MSIE 6.0; Windows NT 5.1; .NET CLR 1.1.4322)",
"Referer": _g_xe_url
}
request = urllib2.Request(_g_xe_url, encFormData, headers)
opener = urllib2.build_opener(urllib2.HTTPRedirectHandler())
htmlText = None
result = None
try:
result = opener.open(request)
htmlText = result.read()
except Exception, ex:
txt = exceptionAsStr(ex)
log(SEV_EXC, "failed to retrieve data for '%s'\nreason:%s\n" % (_g_xe_url, txt))
def retrieveHttpResponseWithRedirectHandleExceptionRetry(url, retryCount=3):
while True:
try:
#log(SEV_LOW, "retrieveHttpResponseWithRedirectHandleExceptionRetry: %s\n" % url)
status, reason, responseText = retrieveHttpResponseWithRedirection(
url)
except socket.error, (err, txt):
retryCount -= 1
#txt = exceptionAsStr(ex)
log(
SEV_EXC,
"failed to retrieve data for '%s'\nsocket error:%d, %s\n" %
(url, err, txt))
if retryCount < 0:
log(
SEV_EXC,
"failed to retrieve data for '%s'\ntoo many socket errors\n"
% (url))
return None
continue
# TODO: add handling of urllib2.URLError?
# File "C:\Python22\lib\urllib2.py", line 809, in do_open
# raise URLError(err)
# URLError: <urlopen error (10060, 'Operation timed out')>
except Exception, ex:
txt = exceptionAsStr(ex)
log(SEV_EXC,
"failed to retrieve data for '%s'\nreason:%s\n" % (url, txt))
return None
def _update_cache():
global _g_currency_cache, _g_currency_cache_lock, _g_retrieve_functions, _g_tracked_currencies, _g_cache_update_interval
t = threading.Timer(_g_cache_update_interval, _update_cache)
t.start()
out = {}
tracked = {}
tracked.update(_g_tracked_currencies)
for func in _g_retrieve_functions:
if 0 == len(tracked):
break
try:
res, data = func()
if RESULTS_DATA != res:
log(
SEV_MED, "currency parser: %s returned result: %d" %
(str(func), res))
continue
for item in data.iteritems():
key, value = item
#if key in ["KRW"]:
# print "Func: %s" % (str(func))
# print "Key: %s Value: %s " % (key, str(value))
if tracked.has_key(key) and 0 != value:
out[key] = value
del tracked[key]
except Exception, ex:
log(SEV_EXC, exceptionAsStr(ex))
def retrieveUsers():
global g_lastUserId, g_userStats
cursor = None
conn = getConnection()
try:
sql = "SELECT user_id, device_info, DATE_FORMAT(cookie_issue_date, '%Y-%m-%d'), reg_code, DATE_FORMAT(registration_date, '%Y-%m-%d'), disabled_p FROM users ORDER BY user_id;";
#sql = "SELECT user_id, device_info, DATE_FORMAT(cookie_issue_date, '%Y-%m-%d'), reg_code, DATE_FORMAT(registration_date, '%Y-%m-%d'), disabled_p FROM users WHERE user_id > " + str(g_lastUserId) + "ORDER BY user_id;";
cursor = conn.cursor()
cursor.execute(sql)
processed = 0
prev_id = -1
while True:
row = cursor.fetchone()
if None == row:
break
userData = UserData()
userData.user_id = row[0]
userData.device_info = row[1]
userData.cookie_issue_date = row[2]
userData.reg_code = row[3]
userData.registration_date = row[4]
userData.disabled_p = row[5]
if userData.user_id > g_lastUserId:
userData.new_user_p = True
else:
userData.new_user_p = False
assert userData.user_id > prev_id
prev_id = userData.user_id
g_userStats.append(userData)
if userData.user_id > g_lastUserId:
g_lastUserId = userData.user_id
processed += 1
cursor.close()
# print "processed %d requests" % processed
except _mysql_exceptions.Error, ex:
if cursor:
cursor.close()
#log(SEV_HI, arsutils.exceptionAsStr(ex))
print "exception in retrieveUsers()"
print arsutils.exceptionAsStr(ex)
def convertArticle(term, text):
try:
text = text.replace('__NOTOC__', '')
text = fixSup2(text)
text = removeImageRx(text)
# remove categories. TODO: provide a better support for categories
# i.e. we remember categories on the server and client can display
# all articles in a given category
# text=replaceRegExp(text, categoryRe, '')
text = replaceWikiMacros(text)
# remove remaining templates. TODO: better support for templates
# in wikipedia template text is replaced by a page from Template:
# namespace
text = replaceRegExp(text, wikiTemplateRe, '')
text = text.replace('\r', '')
text = replaceRegExp(
text, commentRe, ''
) # This should be safe, as it's illegal in html to nest comments
text = stripTagBlocks(text, 'div')
text = stripTagBlocks(text, 'table')
text = stripBlocks(text, r'\{\|', r'\|\}')
text = replaceRegExp(text, scriptRe, '')
text = replaceTagList(text, ['b', 'strong'], "'''")
text = replaceTagList(text, ['em', 'i', 'cite'], "''")
text = replaceTagList(text, ['hr'], '----')
text = replaceTagList(text, ['p'], '<br>')
text = replaceTagList(text, [
'dfn', 'code', 'samp', 'kbd', 'var', 'abbr', 'acronym',
'blockquote', 'q', 'pre', 'ins', 'del', 'dir', 'menu', 'img',
'object', 'big', 'span', 'applet', 'font', 'basefont', 'tr', 'td',
'table', 'center', 'div'
], '')
text = replaceRegExp(text, badLinkRe, '', supportedLanguagesRe())
text = entities.convertNamedEntities(term, text)
text = entities.convertNumberedEntities(term, text)
text = stripMultipleNewLines(text)
text = text.strip()
text += '\n'
return text
except Exception, ex:
print "Exception while converting term: ", term
print arsutils.exceptionAsStr(ex)
return ''
def run(self):
global g_lupyIndex
print "Thread start (dict lupy index)"
try:
g_lupyIndex.initialize()
except Exception, ex:
txt = arsutils.exceptionAsStr(ex)
log(SEV_EXC, "exception in lupy index dictionary\n%s\n" % (txt))
def run(self):
global g_lupyIndex
print "Thread start (dict lupy index)"
try:
g_lupyIndex.initialize()
except Exception, ex:
txt = arsutils.exceptionAsStr(ex)
log(SEV_EXC, "exception in lupy index dictionary\n%s\n" % (txt))
def retrieveHttpResponseHandleException(url):
#log(SEV_LOW, "retrieveHttpResponseHandleException: %s\n" % url)
try:
status, reason, responseText = retrieveHttpResponse(url)
except Exception, ex:
txt = exceptionAsStr(ex)
log(SEV_EXC, "failed to retrieve data for '%s'\nreason:%s\n" % (url, txt))
return None
def runTestCase(self, prepAndTest):
testName, prepFun, testFun=prepAndTest
try:
request=prepFun()
response=retrieveResponse(self.address, request)
self.extractFields(response)
try:
testFun()
stdout.write('.')
except TestAssertionFailed, ex:
print "\n--------------------------------------------------------------------------------"
print "Test case %s FAILED: %s" % (testName, ex.cause)
print "--------------------------------------------------------------------------------"
except Exception, ex:
print "\n--------------------------------------------------------------------------------"
print "Test case %s caused Exception.\n" % (testName)
print exceptionAsStr(ex)
print "--------------------------------------------------------------------------------"
def runTestCase(self, prepAndTest):
testName, prepFun, testFun = prepAndTest
try:
request = prepFun()
response = retrieveResponse(self.address, request)
self.extractFields(response)
try:
testFun()
stdout.write('.')
except TestAssertionFailed, ex:
print "\n--------------------------------------------------------------------------------"
print "Test case %s FAILED: %s" % (testName, ex.cause)
print "--------------------------------------------------------------------------------"
except Exception, ex:
print "\n--------------------------------------------------------------------------------"
print "Test case %s caused Exception.\n" % (testName)
print exceptionAsStr(ex)
print "--------------------------------------------------------------------------------"
def retrieveHttpResponseHandleException(url):
#log(SEV_LOW, "retrieveHttpResponseHandleException: %s\n" % url)
try:
status, reason, responseText = retrieveHttpResponse(url)
except Exception, ex:
txt = exceptionAsStr(ex)
log(SEV_EXC,
"failed to retrieve data for '%s'\nreason:%s\n" % (url, txt))
return None
def _getHttpHandleExceptionRetry(url, postData, handleRedirect, dbgLevel, referer, retryCount, cookieJar):
assert retryCount > 0
while retryCount > 0:
try:
htmlTxt = _getHttpHelper(url, postData, handleRedirect, dbgLevel, referer, cookieJar)
return htmlTxt
except socket.error, (err,txt):
txt = exceptionAsStr(ex)
log(SEV_EXC, "failed to retrieve data for '%s'\nsocket error:%d, %s\n" % (url, err, txt))
retryCount -= 1
def retrieveInternational(code):
global _g_retrieve_international
for func in _g_retrieve_international:
try:
res, data = func(code)
if RETRIEVE_FAILED != res and UNKNOWN_FORMAT != res:
return res, data
except Exception, ex:
txt = exceptionAsStr(ex)
log(SEV_EXC, "failed to parse data\nreason:%s\n" % (txt))
def retrieveAreaCodeByCity(city, state):
global _g_retrieve_areaCodeByCity
for func in _g_retrieve_areaCodeByCity:
try:
res, data = func(city, state)
if RETRIEVE_FAILED != res and UNKNOWN_FORMAT != res:
return res, data
except Exception, ex:
txt = exceptionAsStr(ex)
log(SEV_EXC, "failed to parse data\nreason:%s\n" % (txt))
def retrievePerson(firstName,lastName,cityOrZip,state):
global _g_retrieve_person
for func in _g_retrieve_person:
try:
res, data = func(firstName,lastName,cityOrZip,state)
if RETRIEVE_FAILED != res and UNKNOWN_FORMAT != res:
return res, data
except Exception, ex:
txt = exceptionAsStr(ex)
log(SEV_EXC, "failed to parse data\nreason:%s\n" % (txt))
def retrieveReversePhone(xxx,yyy,zzzz):
global _g_retrieve_reversePhone
for func in _g_retrieve_reversePhone:
try:
res, data = func(xxx,yyy,zzzz)
if RETRIEVE_FAILED != res and UNKNOWN_FORMAT != res:
return res, data
except Exception, ex:
txt = exceptionAsStr(ex)
log(SEV_EXC, "failed to parse data\nreason:%s\n" % (txt))
def retrieveBusiness(name,cityOrZip,state,surrounding,categoryOrName):
global _g_retrieve_business
for func in _g_retrieve_business:
try:
res, data = func(name,cityOrZip,state,surrounding,categoryOrName)
if res not in [RETRIEVE_FAILED, UNKNOWN_FORMAT]:
return res, data
except Exception, ex:
txt = exceptionAsStr(ex)
log(SEV_EXC, "failed to parse data\nreason:%s\n" % (txt))
def find(self, book_id):
self._lock.acquire()
file_name = None
try:
file_name = self._cache.get(book_id, None)
if file_name is None:
self._lock.release()
return None;
f = file(file_name, "rb")
self._lock.release()
return f
except Exception, ex:
log(SEV_EXC, exceptionAsStr(ex))
if file_name is not None:
try:
os.remove(file_name)
except Exception, ex1:
log(SEV_EXC, exceptionAsStr(ex1))
pass
def utf8ToLatin1(text):
decoded = text
try:
decoded = text.decode("utf_8")
except ValueError, ex:
sys.stdout.write("exception while decoding utf-8\n")
sys.stdout.write("%s\n" % arsutils.exceptionAsStr(ex))
sys.stdout.write('\n')
sys.stdout.write(text[:240])
sys.stdout.write('\n')
return text
def utf8ToLatin1(text):
decoded = text
try:
decoded = text.decode("utf_8")
except ValueError, ex:
sys.stdout.write("exception while decoding utf-8\n")
sys.stdout.write("%s\n" % arsutils.exceptionAsStr(ex))
sys.stdout.write("\n")
sys.stdout.write(text[:240])
sys.stdout.write("\n")
return text
def add(self, book_id, file_name):
self._lock.acquire()
try:
cached_name = self._cache.get(book_id, None)
if cached_name is not None:
if cached_name == file_name:
return
else:
self._cache[book_id] = file_name
self._pickle_out()
try:
os.remove(cached_name)
except Exception, ex:
log(SEV_EXC, exceptionAsStr(ex))
else:
def initDictionary():
global g_wnFo, g_wnDictPath, g_wnDefReadLock, g_random, g_fInitialized, g_fDisabled
global g_thFo, g_thDictPath, g_thDefReadLock
if g_fInitialized:
return
g_fInitialized = True
g_fDisabled = True
if not loadPickledFiles():
return
try:
assert None == g_wnFo
g_wnFo = open(g_wnDictPath, "rb")
assert None == g_wnDefReadLock
g_wnDefReadLock = Lock()
assert None == g_thFo
g_thFo = open(g_thDictPath, "rb")
assert None == g_thDefReadLock
g_thDefReadLock = Lock()
assert None == g_random
g_random = random.Random()
g_random.seed()
except Exception, ex:
print arsutils.exceptionAsStr(ex)
return
def _getHttpHandleExceptionRetry(url, postData, handleRedirect, dbgLevel,
referer, retryCount, cookieJar):
assert retryCount > 0
while retryCount > 0:
try:
htmlTxt = _getHttpHelper(url, postData, handleRedirect, dbgLevel,
referer, cookieJar)
return htmlTxt
except socket.error, (err, txt):
txt = exceptionAsStr(ex)
log(
SEV_EXC,
"failed to retrieve data for '%s'\nsocket error:%d, %s\n" %
(url, err, txt))
retryCount -= 1
def initDictionary():
global g_wnFo, g_wnDictPath, g_wnDefReadLock, g_random, g_fInitialized, g_fDisabled
global g_thFo, g_thDictPath, g_thDefReadLock
if g_fInitialized:
return
g_fInitialized = True
g_fDisabled = True
if not loadPickledFiles():
return
try:
assert None == g_wnFo
g_wnFo = open(g_wnDictPath, "rb")
assert None == g_wnDefReadLock
g_wnDefReadLock = Lock()
assert None == g_thFo
g_thFo = open(g_thDictPath, "rb")
assert None == g_thDefReadLock
g_thDefReadLock = Lock()
assert None == g_random
g_random = random.Random()
g_random.seed()
except Exception, ex:
print arsutils.exceptionAsStr(ex)
return
def retrieveHttpResponseWithRedirectHandleExceptionRetry(url,retryCount=3):
while True:
try:
#log(SEV_LOW, "retrieveHttpResponseWithRedirectHandleExceptionRetry: %s\n" % url)
status, reason, responseText = retrieveHttpResponseWithRedirection(url)
except socket.error, (err,txt):
retryCount -= 1
#txt = exceptionAsStr(ex)
log(SEV_EXC, "failed to retrieve data for '%s'\nsocket error:%d, %s\n" % (url, err, txt))
if retryCount < 0:
log(SEV_EXC, "failed to retrieve data for '%s'\ntoo many socket errors\n" % (url))
return None
continue
# TODO: add handling of urllib2.URLError?
# File "C:\Python22\lib\urllib2.py", line 809, in do_open
# raise URLError(err)
# URLError: <urlopen error (10060, 'Operation timed out')>
except Exception, ex:
txt = exceptionAsStr(ex)
log(SEV_EXC, "failed to retrieve data for '%s'\nreason:%s\n" % (url, txt))
return None
return False
if not arsutils.fFileExists(g_thWordsPath):
print "Thesaurus dictionary file '%s' doesn't exist" % g_thWordsPath
return False
try:
fo = open(g_thIndexPath, "rb")
g_thWordIndex = cPickle.load(fo)
fo.close()
fo = open(g_thWordsPath, "rb")
g_thWords = cPickle.load(fo)
fo.close()
except Exception, ex:
print arsutils.exceptionAsStr(ex)
return False
print "Finished loading Thesaurus files"
return True
def initDictionary():
global g_wnFo, g_wnDictPath, g_wnDefReadLock, g_random, g_fInitialized, g_fDisabled
global g_thFo, g_thDictPath, g_thDefReadLock
if g_fInitialized:
return
g_fInitialized = True
g_fDisabled = True
if not loadPickledFiles():
return
try:
def _retrieveHours(self, hours, fast):
cursor = self._conn.cursor()
try:
start = time.mktime((self._date.tm_year, self._date.tm_mon, self._date.tm_mday, 0 , 0, 0, -1, -1, -1))
end = start + 3600 * 24
cursor.execute("SELECT date FROM zap2it_cached_data WHERE provider_id = %d AND date >= '%s' AND date < '%s'" %
(self._provider, time.strftime("%Y-%m-%d %H:%M:%S", time.localtime(start)), time.strftime("%Y-%m-%d %H:%M:%S", time.localtime(end))))
for row in cursor:
h = row[0].timetuple()[3]
hours.remove(h)
finally:
cursor.close()
# if all data is cached, return None and use _retrieveFromDatabase() instead
if 0 == len(hours):
return None;
opener = None
if not self._has_grid:
response, opener = _zap2it_retrieve_grid(self._jar, self._zipCode, self._provider)
self._has_grid = True
response.close()
else:
opener = opener = _zap2it_opener(self._jar)
if not fast:
for h in hours:
self._events[h] = threading.Event()
date = time.localtime(time.mktime((self._date.tm_year, self._date.tm_mon, self._date.tm_mday, h , 0, 0, -1, -1, -1)))
_g_zap2it_cache_manager._addActiveRetriever(_zap2it_retriever_key(self._provider, date), self)
rows = 0
duration = 1
if fast:
rows = 20
duration = 3
hours = [hours[0]]
out = []
formData = {
"displayType": "Text",
"duration": str(duration),
"startDay": time.strftime("%m/%d/%Y", self._date),
"category": "0",
"station": "0",
"rowdisplay": str(rows),
"goButton": "GO"
}
stations = {}
programs = {}
for h in hours:
formData["startTime"] = str(h)
date = time.localtime(time.mktime((self._date.tm_year, self._date.tm_mon, self._date.tm_mday, h , 0, 0, -1, -1, -1)))
encData = urllib.urlencode(formData)
request = urllib2.Request(_g_zap2it_listings_url, encData)
request.add_header("Referer", _g_zap2it_grid_url)
response = opener.open(request)
htmlText = None
try:
_zap2it_tracker(self._jar)
contentLength = long(response.info()["Content-Length"])
htmlText = response.read(contentLength)
if fast:
out.extend(_zap2it_parse_listings(None, htmlText, date, self._provider, stations, programs))
else:
_zap2it_parse_listings(self._conn, htmlText, date, self._provider, stations, programs)
except Exception, ex:
# todo: log exception
print exceptionAsStr(ex)
f = file(time.strftime('tvlistings-%Y%m%dT%H%M%S.html'), 'wb')
f.write(htmlText)
f.close()
pass
response.close()
if not fast:
self._events[h].set()
_g_zap2it_cache_manager._removeActiveRetriever(_zap2it_retriever_key(self._provider, date))
return False
if not arsutils.fFileExists(g_thWordsPath):
print "Thesaurus dictionary file '%s' doesn't exist" % g_thWordsPath
return False
try:
fo = open(g_thIndexPath, "rb")
g_thWordIndex = cPickle.load(fo)
fo.close()
fo = open(g_thWordsPath, "rb")
g_thWords = cPickle.load(fo)
fo.close()
except Exception, ex:
print arsutils.exceptionAsStr(ex)
return False
print "Finished loading Thesaurus files"
return True
def initDictionary():
global g_wnFo, g_wnDictPath, g_wnDefReadLock, g_random, g_fInitialized, g_fDisabled
global g_thFo, g_thDictPath, g_thDefReadLock
if g_fInitialized:
return
g_fInitialized = True
g_fDisabled = True
if not loadPickledFiles():
return
try:
def iterWikipediaArticles(sqlFileName, limit=None, fUseCache=False, fRecreateCache=False):
# if limit:
# assert fUseCache==False
print "fUseCache %d, fRecreateCache=%d" % (fUseCache, fRecreateCache)
cacheWriter = None
fReallyUseCache = False
if fRecreateCache:
cacheWriter = ArticleCacheWriter(sqlFileName)
cacheWriter.open()
else:
if fUseCache and fCacheExists(sqlFileName):
fReallyUseCache = True
else:
cacheWriter = ArticleCacheWriter(sqlFileName)
cacheWriter.open()
lang = os.path.basename(sqlFileName)[:2]
print "database dump language: ", lang
isUtf8 = False
if lang in wikiToDbConvert.g_utf8Languages:
isUtf8 = True
if isUtf8:
print "performing UTF-8 to Latin-1 conversion"
if fReallyUseCache:
fileName = getIdxFileName(sqlFileName)
print "getting articles from cache %s" % fileName
fo = open(fileName, "rb")
count = 0
while True:
title = fo.readline()
if len(title) == 0:
break
if fIsRedirectLine(title):
redirect = fo.readline()
title = title.strip()
if title == REDIRECT_MARK:
# need this to remove stupid redirecto of 0xa0=>Space_(punctuation)
print "title after stripping is equal to '%s' (REDIRECT_MARK), so skipping" % REDIRECT_MARK
continue
title = title[len(REDIRECT_MARK) :]
if len(title) == 0:
print "title after stripping is empty string, so skipping '%s'" % redirect
continue
article = WikipediaArticleRedirect(title, redirect.strip())
else:
title = title.strip()
line = fo.readline()
if len(title) == 0:
print "title after stripping is empty string, so skipping '%s'" % line.strip()
continue
lineParts = line.split(",")
try:
ns = int(lineParts[0])
assert ns == NS_MAIN
txtOffset = int(lineParts[1])
txtLen = int(lineParts[2])
md5Hash = lineParts[3]
viewCount = int(lineParts[4])
article = WikipediaArticleFromCache(sqlFileName, title, ns, txtOffset, txtLen, md5Hash, viewCount)
except ValueError, ex:
# in en 2004-09-17 db we have an error in ns = int(lineParts[0]), so just ignore it
print "exception in iterWikipediaArticles"
print arsutils.exceptionAsStr(ex)
print "title:_%s_" % title
print "line:_%s_" % line
print "lineParts[0]=%s" % lineParts[0]
continue
yield article
count += 1
if limit and count > limit:
break
fo.close()
return
def iterWikipediaArticles(sqlFileName,
limit=None,
fUseCache=False,
fRecreateCache=False):
#if limit:
# assert fUseCache==False
print "fUseCache %d, fRecreateCache=%d" % (fUseCache, fRecreateCache)
cacheWriter = None
fReallyUseCache = False
if fRecreateCache:
cacheWriter = ArticleCacheWriter(sqlFileName)
cacheWriter.open()
else:
if fUseCache and fCacheExists(sqlFileName):
fReallyUseCache = True
else:
cacheWriter = ArticleCacheWriter(sqlFileName)
cacheWriter.open()
lang = os.path.basename(sqlFileName)[:2]
print "database dump language: ", lang
isUtf8 = False
if lang in wikiToDbConvert.g_utf8Languages:
isUtf8 = True
if isUtf8:
print "performing UTF-8 to Latin-1 conversion"
if fReallyUseCache:
fileName = getIdxFileName(sqlFileName)
print "getting articles from cache %s" % fileName
fo = open(fileName, "rb")
count = 0
while True:
title = fo.readline()
if len(title) == 0:
break
if fIsRedirectLine(title):
redirect = fo.readline()
title = title.strip()
if title == REDIRECT_MARK:
#need this to remove stupid redirecto of 0xa0=>Space_(punctuation)
print "title after stripping is equal to '%s' (REDIRECT_MARK), so skipping" % REDIRECT_MARK
continue
title = title[len(REDIRECT_MARK):]
if len(title) == 0:
print "title after stripping is empty string, so skipping '%s'" % redirect
continue
article = WikipediaArticleRedirect(title, redirect.strip())
else:
title = title.strip()
line = fo.readline()
if len(title) == 0:
print "title after stripping is empty string, so skipping '%s'" % line.strip(
)
continue
lineParts = line.split(",")
try:
ns = int(lineParts[0])
assert ns == NS_MAIN
txtOffset = int(lineParts[1])
txtLen = int(lineParts[2])
md5Hash = lineParts[3]
viewCount = int(lineParts[4])
article = WikipediaArticleFromCache(
sqlFileName, title, ns, txtOffset, txtLen, md5Hash,
viewCount)
except ValueError, ex:
# in en 2004-09-17 db we have an error in ns = int(lineParts[0]), so just ignore it
print "exception in iterWikipediaArticles"
print arsutils.exceptionAsStr(ex)
print "title:_%s_" % title
print "line:_%s_" % line
print "lineParts[0]=%s" % lineParts[0]
continue
yield article
count += 1
if limit and count > limit:
break
fo.close()
return
def convertArticle(term, text):
try:
text = text.replace("__NOTOC__", "")
text = fixSup2(text)
text = removeImageRx(text)
# remove categories. TODO: provide a better support for categories
# i.e. we remember categories on the server and client can display
# all articles in a given category
# text=replaceRegExp(text, categoryRe, '')
text = replaceWikiMacros(text)
# remove remaining templates. TODO: better support for templates
# in wikipedia template text is replaced by a page from Template:
# namespace
text = replaceRegExp(text, wikiTemplateRe, "")
text = text.replace("\r", "")
text = replaceRegExp(text, commentRe, "") # This should be safe, as it's illegal in html to nest comments
text = stripTagBlocks(text, "div")
text = stripTagBlocks(text, "table")
text = stripBlocks(text, r"\{\|", r"\|\}")
text = replaceRegExp(text, scriptRe, "")
text = replaceTagList(text, ["b", "strong"], "'''")
text = replaceTagList(text, ["em", "i", "cite"], "''")
text = replaceTagList(text, ["hr"], "----")
text = replaceTagList(text, ["p"], "<br>")
text = replaceTagList(
text,
[
"dfn",
"code",
"samp",
"kbd",
"var",
"abbr",
"acronym",
"blockquote",
"q",
"pre",
"ins",
"del",
"dir",
"menu",
"img",
"object",
"big",
"span",
"applet",
"font",
"basefont",
"tr",
"td",
"table",
"center",
"div",
],
"",
)
text = replaceRegExp(text, badLinkRe, "", supportedLanguagesRe())
text = entities.convertNamedEntities(term, text)
text = entities.convertNumberedEntities(term, text)
text = stripMultipleNewLines(text)
text = text.strip()
text += "\n"
return text
except Exception, ex:
print "Exception while converting term: ", term
print arsutils.exceptionAsStr(ex)
return ""
def spider_last_modified():
return os.path.getmtime(_g_spider_data_path)
def reindex(index, data = None):
data = _load_spider_data()
for letter, letter_data in data.iteritems():
print "Indexing %d books for letter '%s'." % (letter_data[0], letter)
for book in letter_data[1]:
url, title, subtitle, author, book_id, code, formats = book
title = _decode(title)
subtitle = _decode(subtitle)
author = _transform_author(_decode(author))
book_id = _decode(book_id)
code = _decode(code)
formats.sort()
index.index_ebook(title, subtitle, author, book_id, formats, PROVIDER_ID, code)
try:
import psyco
psyco.bind(_Spider._merge_temps)
psyco.bind(_Spider._spider_letter_range)
except Exception, ex:
print exceptionAsStr(ex)
except ImportError:
print "psyco not available. You should consider using it (http://psyco.sourceforge.net/)"
if __name__ == "__main__":
spider()
