def get_item(marc_no, status=0):
dict = {}
test = PinYin()
test.load_word('word.data')
hm = requests.get(ourl + 'item.php?marc_no=' + str(marc_no)).text.encode(
encoder).decode('utf8').replace(' ', '')
parser = HTMLParser.HTMLParser()
s1 = parser.unescape(hm)
static = re.findall('<div id="book_info">(.*?)<div class="clear"></div>',
s1, re.S)[0]
booklist = re.findall('<dl class="booklist">(.*?)</dl>', static, re.S)
for each in booklist:
pm = re.findall('<dt>(.*?)</dt>', each, re.S)[0]
if pm == '':
continue
st = re.findall('<dd>(.*?)</dd>', each, re.S)[0]
try:
st1 = re.findall('>(.*?)</a>', st, re.S)[0]
except:
st1 = st
pms = test.hanzi2pinyin_split(string=pm, split="",
firstcode=True).replace('/', '')
dict[pms] = st1
if status == 1:
print pm,
print st1
return dict
def main():
mysql = MySQLHander()
p = PinYin()
p.load_word()
with open('video.json') as json_file:
alldata = json.load(json_file)
for data in alldata:
sql = "INSERT INTO resource VALUES (null, '', '{title}', '{desc}', '{thumb}', '{url}',{duration}, {vister}, {likes},{creat_time}, '{up_time}')".format(title=data['name'].encode('utf-8', 'ignore'), desc=data['name'].encode('utf-8', 'ignore'), thumb=data['thumb'], url=data['url'],duration=random.randint(80, 120), vister=random.randint(4500, 9999), likes=random.randint(500, 2000), creat_time=time.time(), up_time=time.strftime('%Y-%m-%d %H:%M:%S', time.localtime(time.time())))
rid = mysql.insert(sql)
for tag in xrange(1, 2):
tagname = data["tag{0}".format(tag)].encode('utf-8', 'ignore')
sql = "SELECT id from category WHERE cname='{0}'".format(tagname)
mysql.query(sql)
result = mysql.fetchOneRow()
if not result:
ename = p.hanzi2pinyin_split(string=tagname, split="-").replace('-', '')
sql = "INSERT INTO category values(null, '{ename}', '{cname}', {time})".format(ename=ename, cname=tagname, time=int(time.time()))
tagid = mysql.insert(sql)
sql = "INSERT INTO category_mapping values(null, {cid}, {rid}, {time})".format(cid=tagid, rid=rid, time=int(time.time()))
mysql.insert(sql)
else:
tagid = result[0]
sql = "INSERT INTO category_mapping values(null, {cid}, {rid}, {time})".format(cid=tagid, rid=rid, time=int(time.time()))
mysql.insert(sql)
for pic in data['screen']:
sql = "INSERT INTO screenshots values(null, {rid}, '{pic}', {time})".format(rid=rid, pic=pic, time=int(time.time()))
mysql.insert(sql)
print "{0} done".format(data['thumb'])
mysql.close()
def getciyun(self):
# 得到词云回答者信息
test1 = PinYin()
test1.load_word()
str1 = str(test1.hanzi2pinyin_split(string=str(self.aa.topic), split="-"))
path1 = 'F:/zhihu/answer/people_qb.txt'
cloud.ciyun1(path1,str1+'people')
#得到词云，问题信息
path2='F:/zhihu/answer/question_top10.txt'
cloud.ciyun1(path2,str1+'question')
path2 = 'F:/zhihu/answer/p_location.txt'
cloud.ciyun1(path2,str1+'slocation')
def draw_frame(faces, img, gray, move):
global xdeg
global ydeg
global fps
global time_t
if move == 2:
steering_control(faces, img)
# Draw a rectangle around every face
for (x, y, w, h) in faces:
cv2.rectangle(img, (x, y), (x + w, y + h), (200, 255, 0), 2)
#-----rec-face
roi = gray[x:x + w, y:y + h]
try:
roi = cv2.resize(roi, (200, 200), interpolation=cv2.INTER_LINEAR)
params = model.predict(roi)
if params[1] < 500.0:
#print (names[params[0]])
#pec = (' %.2f' % (params[1]))
#sign = names[params[0]] + pec
pyin = PinYin()
pyin.load_word()
pname = names[params[0]]
change_identity(pname)
#pyin.hanzi2pinyin(string = pname)
pname = pyin.hanzi2pinyin_split(string=pname, split='')
s = ''
for p in pname:
s = s + p
sign = ("%s %.2f" % (s, params[1]))
# print(sign)
cv2.putText(img, sign, (x, y - 2), cv2.FONT_HERSHEY_SIMPLEX,
0.5, (0, 0, 255), 2)
#img = cv2ImgAddText(img, sign , x , y - 2, (0, 0, 255), 20)
#img = change_cv2_draw(img,sign,(x, y + 2), 20 , 'firebrick' )
except:
continue
# Calculate and show the FPS
fps = fps + 1
sfps = fps / (time.time() - t_start)
cv2.putText(img, "FPS : " + str(int(sfps)), (10, 15),
cv2.FONT_HERSHEY_SIMPLEX, 0.5, (0, 0, 255), 2)
cv2.imshow("recognize-face", img)
def mallSearch():
kw = MyRequest.getParams('kw')
if kw:
kw = addslashes(kw)
malls = Mall.query.filter(Mall.keyword.like('%'+kw+'%')).all()
keyword = Keyword.query.filter_by(keyword=kw).first()
if keyword:
if len(malls) > keyword.items:
keyword.items = len(malls)
if timetodate(keyword.updatetime,0) == timetodate(g.siteTime,0):
keyword.month_search += 1
else:
keyword.month_search = 1
if timetodate(keyword.updatetime,8) == timetodate(g.siteTime,8):
keyword.week_search += 1
else:
keyword.week_search = 1
if timetodate(keyword.updatetime,3) == timetodate(g.siteTime,3):
keyword.today_search += 1
else:
keyword.today_search = 1
keyword.total_search += 1
keyword.updatetime = g.siteTime
else:
keyword = Keyword()
keyword.keyword = kw
keyword.items = len(malls)
keyword.updatetime = g.siteTime
keyword.month_search = 1
keyword.week_search = 1
keyword.today_search = 1
keyword.total_search = 1
py = PinYin(g.rootpath+os.path.sep+'assets'+os.path.sep+'word.data')
py.load_word()
keyword.letter = py.hanzi2pinyin_split(string=kw, split=" ")
db.session.add(keyword)
db.session.commit()
rewords = Keyword.query.filter(Keyword.keyword.like('%'+kw+'%')).limit(10).all()
return render_template('mall/mallSearch.html',malls=malls,g=g,kw=kw,rewords=rewords)
else:
return redirect(url_for('index'))
class changetopinyin:
wf_dict={}
def __init__(self):
self.test=PinYin()
self.test.load_word()
def change(self,filename):
with open(filename,'r') as ff:
for item in ff.readlines():
word,fre=item.split(' ')[0],int(item.split(' ')[1])
wf=word_fre(word,fre)
self.addtodict(wf)
for item in self.wf_dict.itervalues():
i=0
for ii in item:
try:
item[i]=(ii.word,ii.fre)
except Exception as e:
print(e)
i+=1
self.save('pinyin_dict2')
def addtodict(self,wf):
pp=self.test.hanzi2pinyin_split(wf.word,'_')
if(pp in self.wf_dict):
if(len(self.wf_dict[pp])<=5):
heapq.heappush(self.wf_dict[pp],wf)
else:
heapq.heappushpop(self.wf_dict[pp],wf)
else:
self.wf_dict[pp]=[]
self.wf_dict[pp].append(wf)
def save(self,filename):
with open(filename,'w') as wff:
try:
wff.write(json.dumps(self.wf_dict))
except Exception as e:
print(e)
class changetopinyin:
wf_dict = {}
def __init__(self):
self.test = PinYin()
self.test.load_word()
def change(self, filename):
with open(filename, 'r') as ff:
for item in ff.readlines():
word, fre = item.split(' ')[0], int(item.split(' ')[1])
wf = word_fre(word, fre)
self.addtodict(wf)
for item in self.wf_dict.itervalues():
i = 0
for ii in item:
try:
item[i] = (ii.word, ii.fre)
except Exception as e:
print(e)
i += 1
self.save('pinyin_dict2')
def addtodict(self, wf):
pp = self.test.hanzi2pinyin_split(wf.word, '_')
if (pp in self.wf_dict):
if (len(self.wf_dict[pp]) <= 5):
heapq.heappush(self.wf_dict[pp], wf)
else:
heapq.heappushpop(self.wf_dict[pp], wf)
else:
self.wf_dict[pp] = []
self.wf_dict[pp].append(wf)
def save(self, filename):
with open(filename, 'w') as wff:
try:
wff.write(json.dumps(self.wf_dict))
except Exception as e:
print(e)
class correction:
def __init__(self):
self.pp=PinYin()
self.pp.load_word()
with open('pinyin_dict','r') as ff:
line=ff.readline()
self.jj_dict=json.loads(line)
ff.close()
def correct(self,phrase_list):
termlist=[]
flag=False
newplist=self.recompose(phrase_list)
'''for item in newplist:
for item2 in item:
print(item2.encode('utf-8'))'''
for nnlist in newplist:
i=0
tmp_correct=[]
correct_num=[]
for item in nnlist:
py=self.pp.hanzi2pinyin_split(item,'_')
tmp=[]
tmp_correct.append(tmp)
if(py in self.jj_dict):
for item2 in self.jj_dict[py]:
tmp_correct[i].append(item2)
else:
tmp_correct[i].append((item,1))
correct_num.append(0)
i+=1
length=len(tmp_correct)
notend=True
while notend:
i=0
tmpstr=''
score=0
for j in xrange(0,length):
tmps=tmp_correct[j][correct_num[j]][0]
tmpstr+=tmps
score+=int(tmp_correct[j][correct_num[j]][1])*len(tmps)**7
termlist.append(tup(tmpstr,score))
correct_num[0]+=1
while correct_num[i]>=len(tmp_correct[i]):
correct_num[i]=0
if i<length-1:
correct_num[i+1]+=1
i+=1
else:
notend=False
result_list=self.sscore(termlist)
comstr=''
for item in phrase_list:
comstr+=item
if result_list[0]==comstr:
result_list=[]
else:
if comstr in result_list:
result_list.pop(result_list.index(comstr))
return result_list
def sscore(self,termlist):
heap=[]
result_list=[]
for item in termlist:
heap.append(item)
heapq.heapify(heap)
while len(heap) > 5:
a=heapq.heappop(heap)
while len(heap) > 0:
result_list.append(heapq.heappop(heap).term)
result_list.reverse()
return result_list
def recompose(self,phrase_list):
position=[]
attach={}
cpl=[] #the consequence:list of list
i=0 #position of single word
for item in phrase_list:
if(len(item)==1):
position.append(i)
attach[i]=0
i+=1
notend=True
length=len(position)
if length>0:
while notend:
gap=0
tmp_list=copy.deepcopy(phrase_list)
tmp_position=copy.deepcopy(position)
pi=0
while pi < len(tmp_position):
item2=tmp_position[pi]
if(attach[item2]==0):
if item2-1-gap>=0:
tmp_list[item2-1-gap]+=tmp_list[item2-gap]
k=tmp_position.index(item2)
tmp_position.pop(k)
tmp_list.pop(item2-gap)
gap+=1
else:
pi+=1
'''while k < len(tmp_position):
tmp_position[k]-=gap
#print(tmp_position[k])
attach[tmp_position[k]]=attach[tmp_position[k]+gap]
k+=1'''
else:
if attach[item2]==1:
if item2+1-gap<len(tmp_list):
tmp_list[item2+1-gap]=tmp_list[item2-gap]+tmp_list[item2+1-gap]
k=tmp_position.index(item2)
tmp_position.pop(k)
tmp_list.pop(item2-gap)
gap+=1
if item2+1 in tmp_position:
tmp_position.pop(tmp_position.index(item2+1))
else:
pi+=1
else:
pi+=1
'''while k < len(tmp_position):
tmp_position[k]-=gap
attach[tmp_position[k]]=attach[tmp_position[k]+gap]
k+=1'''
'''flag=True
for item3 in tmp_list:
if len(item3)==1:
flag=False
if flag:'''
if tmp_list not in cpl:
cpl.append(tmp_list)
attach[position[0]]+=1 #每次变换一个
i=0
while attach[position[i]]>=3:
attach[position[i]]=0
if i<length-1:
attach[position[i+1]]+=1
i+=1
else:
notend=False
else:
cpl.append(phrase_list)
return cpl
#!/usr/bin/env python # -*- coding:utf-8 -*- from pinyin import PinYin test = PinYin() test.load_word() string = "钓鱼岛是中国的" print "out: %s" % test.hanzi2pinyin_split(string=string, split="-")
u"爱奇艺",
u"腾讯视频",
u"qq",
u"熊猫tv",
u"快递",
u'4399',
}
word2pinyin = PinYin()
word2pinyin.load_word()
alphabet = {'a':1, 'b':1, 'c':1, 'd':1, 'e':1, 'f':1, 'g':1,
'h':1, 'i':1, 'j':1, 'k':1, 'l':1, 'm':1, 'n':1,
'o':1, 'p':1, 'q':1, 'r':1, 's':1, 't':1,
'u':1, 'v':1, 'w':1, 'x':1, 'y':1, 'z':1}
def hanzi2pinyi(word):
result = []
for hanzi in word:
if hanzi.lower() in alphabet:
result.append(hanzi.lower())
else:
result.append(word2pinyin.hanzi2pinyin(hanzi))
return ''.join(result)
if __name__ == '__main__':
for word in white_list:
print word2pinyin.hanzi2pinyin_split(word)
#!/usr/bin/env python
# -*- coding:utf-8 -*-
from pinyin import PinYin
import sys
test = PinYin('application/libraries/pinyin.py/word.data')
test.load_word()
print test.hanzi2pinyin_split(string=sys.argv[1], split='_')
class correction:
def __init__(self):
self.pp = PinYin()
self.pp.load_word()
with open('pinyin_dict', 'r') as ff:
line = ff.readline()
self.jj_dict = json.loads(line)
ff.close()
def correct(self, phrase_list):
termlist = []
flag = False
newplist = self.recompose(phrase_list)
'''for item in newplist:
for item2 in item:
print(item2.encode('utf-8'))'''
for nnlist in newplist:
i = 0
tmp_correct = []
correct_num = []
for item in nnlist:
py = self.pp.hanzi2pinyin_split(item, '_')
tmp = []
tmp_correct.append(tmp)
if (py in self.jj_dict):
for item2 in self.jj_dict[py]:
tmp_correct[i].append(item2)
else:
tmp_correct[i].append((item, 1))
correct_num.append(0)
i += 1
length = len(tmp_correct)
notend = True
while notend:
i = 0
tmpstr = ''
score = 0
for j in xrange(0, length):
tmps = tmp_correct[j][correct_num[j]][0]
tmpstr += tmps
score += int(
tmp_correct[j][correct_num[j]][1]) * len(tmps)**7
termlist.append(tup(tmpstr, score))
correct_num[0] += 1
while correct_num[i] >= len(tmp_correct[i]):
correct_num[i] = 0
if i < length - 1:
correct_num[i + 1] += 1
i += 1
else:
notend = False
result_list = self.sscore(termlist)
comstr = ''
for item in phrase_list:
comstr += item
if result_list[0] == comstr:
result_list = []
else:
if comstr in result_list:
result_list.pop(result_list.index(comstr))
return result_list
def sscore(self, termlist):
heap = []
result_list = []
for item in termlist:
heap.append(item)
heapq.heapify(heap)
while len(heap) > 5:
a = heapq.heappop(heap)
while len(heap) > 0:
result_list.append(heapq.heappop(heap).term)
result_list.reverse()
return result_list
def recompose(self, phrase_list):
position = []
attach = {}
cpl = [] #the consequence:list of list
i = 0 #position of single word
for item in phrase_list:
if (len(item) == 1):
position.append(i)
attach[i] = 0
i += 1
notend = True
length = len(position)
if length > 0:
while notend:
gap = 0
tmp_list = copy.deepcopy(phrase_list)
tmp_position = copy.deepcopy(position)
pi = 0
while pi < len(tmp_position):
item2 = tmp_position[pi]
if (attach[item2] == 0):
if item2 - 1 - gap >= 0:
tmp_list[item2 - 1 - gap] += tmp_list[item2 - gap]
k = tmp_position.index(item2)
tmp_position.pop(k)
tmp_list.pop(item2 - gap)
gap += 1
else:
pi += 1
'''while k < len(tmp_position):
tmp_position[k]-=gap
#print(tmp_position[k])
attach[tmp_position[k]]=attach[tmp_position[k]+gap]
k+=1'''
else:
if attach[item2] == 1:
if item2 + 1 - gap < len(tmp_list):
tmp_list[item2 + 1 - gap] = tmp_list[
item2 - gap] + tmp_list[item2 + 1 - gap]
k = tmp_position.index(item2)
tmp_position.pop(k)
tmp_list.pop(item2 - gap)
gap += 1
if item2 + 1 in tmp_position:
tmp_position.pop(
tmp_position.index(item2 + 1))
else:
pi += 1
else:
pi += 1
'''while k < len(tmp_position):
tmp_position[k]-=gap
attach[tmp_position[k]]=attach[tmp_position[k]+gap]
k+=1'''
'''flag=True
for item3 in tmp_list:
if len(item3)==1:
flag=False
if flag:'''
if tmp_list not in cpl:
cpl.append(tmp_list)
attach[position[0]] += 1 #每次变换一个
i = 0
while attach[position[i]] >= 3:
attach[position[i]] = 0
if i < length - 1:
attach[position[i + 1]] += 1
i += 1
else:
notend = False
else:
cpl.append(phrase_list)
return cpl
positionTypes[name] = nid
# parse file and import data to db
fd = open(employee_file, 'r')
for line in fd.readlines():
value = {}
line = line.strip()
items = line.split('\t')
name = items[1]
no = items[2]
if no in nos.keys():
continue
username = h2p.hanzi2pinyin_split(string=name, split='')
i = 0
while username in userNames.keys():
i = i + 1
username = username + str(i)
departmentName = items[3]
if departmentName not in departmentTypes.keys():
cursor.execute("insert into DeptType (name) values(%s)", (departmentName))
cursor.execute("select id from DeptType where name=%s", (departmentName))
id = cursor.fetchone()[0]
departmentTypes[departmentName] = id
department = departmentTypes[departmentName]
positionName = items[4]
if positionName not in positionTypes.keys():
# -*- coding: utf-8 -*- # from models import Cnword from pinyin import PinYin test = PinYin() test.load_word() print test.hanzi2pinyin(string='钓鱼岛是中国的') print test.hanzi2pinyin_split(string='钓鱼岛是中国的')
def hanziToPinyin(hanzi):
test = PinYin()
test.load_word()
return test.hanzi2pinyin_split(string=hanzi, split="_")
