def has_similar(self, url):
'''
'''
if url == "":
return (False, [], [])
if self.has_node(url):
return (True, [], self.search_url_index(url))
else:
max_rate = 0
url_split_list, host_po = up.url_split(url)
max_list = []
max_url = ""
rate = 0
for my_url in self.treeContent:
if fuzz.ratio(
os.path.splitext(my_url)[1],
os.path.splitext(url)[1]) < SIMILAR_THRESHOLD:
# if not has the same expend name
continue
if (fuzz.ratio(url, my_url) / 100.0) < self.SIMILAR_THRESHOLD:
continue
my_url_list, my_host_po = up.url_split(my_url)
rate, dismatch_list = up.url_list_compare(
url_split_list, my_url_list)
rate = max(fuzz.ratio(url, my_url) / 100.0, rate)
if max_rate < rate:
max_rate = rate
max_list = dismatch_list
max_url = my_url
if max_rate > self.SIMILAR_THRESHOLD and max_url != "":
print "SIMILAR URL(Tree.has_similar):\n", max_url, "\n", url
return (True, dismatch_list, self.search_url_index(max_url))
return (False, max_list, [])
def has_similar(self, url):
'''
'''
if url == "":
return (False,[], [])
if self.has_node( url):
return (True,[], self.search_url_index(url))
else:
max_rate = 0
url_split_list, host_po = up.url_split( url)
max_list =[]
max_url = ""
rate = 0
for my_url in self.treeContent:
if fuzz.ratio( os.path.splitext( my_url)[1], os.path.splitext( url)[1])<SIMILAR_THRESHOLD:
# if not has the same expend name
continue
if (fuzz.ratio(url, my_url)/100.0)< self.SIMILAR_THRESHOLD:
continue
my_url_list, my_host_po = up.url_split(my_url)
rate, dismatch_list = up.url_list_compare(url_split_list, my_url_list)
rate = max( fuzz.ratio(url, my_url)/100.0, rate)
if max_rate < rate:
max_rate = rate
max_list = dismatch_list
max_url = my_url
if max_rate > self.SIMILAR_THRESHOLD and max_url != "":
print "SIMILAR URL(Tree.has_similar):\n",max_url,"\n",url
return (True, dismatch_list, self.search_url_index(max_url))
return (False, max_list, [])
def find_similar_and_replace(self,
url,
indSet,
context={
'front': ' ',
'back': ' '
}):
if url == "":
return ""
else:
max_rate = 0
max_ind = -2
url_split_list, url_host_po = up.url_split(url)
for ind in indSet:
my_url = self.treeContent[ind]
if fuzz.ratio(
os.path.splitext(my_url)[1],
os.path.splitext(url)[1]) < SIMILAR_THRESHOLD:
# if not has the same expend name
continue
DIYrate, dis_list = up.similar_ratio(url, my_url)
rate = max(
fuzz.ratio(url, my_url) / 100.0 + 0.001, DIYrate + 0.001)
if rate < self.SIMILAR_THRESHOLD or rate <= max_rate:
continue
front, back, start = self.get_surround_text(ind)
context_similar_ratio = (fuzz.ratio(front, context['front'])/100.0+0.001) \
* (fuzz.ratio(back, context['back'])/100.0+0.001)
rate = rate * context_similar_ratio
if rate < self.SIMILAR_THRESHOLD or rate <= max_rate:
continue
else:
max_rate = rate
max_ind = ind
# max_ind is the one that most similar to the given url
# replace
if max_rate == 0 or max_rate < self.SIMILAR_THRESHOLD or max_ind < 0:
print "No similar"
return ""
url_split_list, host_po = up.url_split(url)
simi_url_list, simi_url_host_po = up.url_split(
self.treeContent[max_ind])
DIYrate, dismatch_list = up.url_list_compare(
url_split_list, simi_url_list)
tmp_url = url
for item in dismatch_list:
tmp_url = up.replace_url(tmp_url, item)
print "SIMILAR URL(Tree.find_similar_and_replace):\n", self.treeContent[
max_ind], "\n", url
print "replace:", tmp_url
# if len(dismatch_list)>=2:
# print "Multi-Replacement:\n",url
# print self.treeContent[ max_ind]
# input('Multi-Replacement:\n')
return tmp_url
def find_similar_and_replace(self, url, indSet, context={'front':' ', 'back':' '}):
if url == "":
return ""
else:
max_rate = 0
max_ind = -2
url_split_list, url_host_po = up.url_split( url)
for ind in indSet:
my_url = self.treeContent[ ind]
if fuzz.ratio( os.path.splitext( my_url)[1], os.path.splitext( url)[1])<SIMILAR_THRESHOLD:
# if not has the same expend name
continue
DIYrate, dis_list = up.similar_ratio( url, my_url)
rate = max( fuzz.ratio(url, my_url)/100.0 +0.001, DIYrate+0.001)
if rate < self.SIMILAR_THRESHOLD or rate <=max_rate:
continue
front, back, start = self.get_surround_text( ind)
context_similar_ratio = (fuzz.ratio(front, context['front'])/100.0+0.001) \
* (fuzz.ratio(back, context['back'])/100.0+0.001)
rate = rate* context_similar_ratio
if rate < self.SIMILAR_THRESHOLD or rate <=max_rate:
continue
else:
max_rate = rate
max_ind = ind
# max_ind is the one that most similar to the given url
# replace
if max_rate == 0 or max_rate< self.SIMILAR_THRESHOLD or max_ind<0:
print "No similar"
return ""
url_split_list, host_po = up.url_split( url)
simi_url_list, simi_url_host_po = up.url_split( self.treeContent[ max_ind])
DIYrate, dismatch_list = up.url_list_compare(url_split_list, simi_url_list)
tmp_url = url
for item in dismatch_list:
tmp_url = up.replace_url(tmp_url, item)
print "SIMILAR URL(Tree.find_similar_and_replace):\n",self.treeContent[ max_ind],"\n",url
print "replace:",tmp_url
# if len(dismatch_list)>=2:
# print "Multi-Replacement:\n",url
# print self.treeContent[ max_ind]
# input('Multi-Replacement:\n')
return tmp_url
