l = [i.groupdict() for i in p.finditer(content)]
l=l[1:]
if len(l):return l
else:return None
def pageinfo(src):
'''
获取一个页面的详细信息
返回对于的字典列表
name:图片的名字
cutaddr:缩小的浏览图
picaddr:实际图片的地址
'''
d=os.path.split(src)[0]
try:
url = urllib2.urlopen(src)
content = url.read()#.decode('utf-8')
except:
print 'error'
return None
#find all the pictures info in a page
p = re.compile(r'
def nextpageaddr(src):
'''
从页面的html源码中获取下一个页面地址的名称,最后一页返回None
'''
content=readsrc(src)
p = re.compile(r'.*?')
r = p.search(content)
if r:
return os.path.dirname(src)+"/"+r.group(1)
else:
return None
def picinfoaddr(src):
'''
参数相册图集的html代码
返回全部图片的相对地址
'''
content=readsrc(src)
p = re.compile(r'
#title
temp=''
r=re.search('
#keyword
temp=''
r=re.search('',content)
if r:
temp = r.group(1)
info['kw']=temp
#type
r=re.findall('
if r:
info['type']=':'.join(r)
else:
info['type']=''
r=re.search('
for index,i in enumerate(r):
r[index]=i[0:i.rfind('_')]
# print r[index]
info['pic']=r
return info
import threading
class mthread(threading.Thread):
def __init__(self,tp,addr,lock):
threading.Thread.__init__(self)
# self.doc = minidom.Document()
self.doc=minidom.Document()
self.tp=tp
self.lock=lock
self.addr=addr
self.thread_stop=False
self.picdoc=None
def run(self):
self.picdoc = self.doc.createElement('urlclass')
# print self.tp
self.picdoc.setAttribute('type',self.tp)
# self.doc.appendChild(self.picdoc)
m=pageinfo(self.addr)
while self.addr:
for i in m:
# print i['picaddr']
picaddr=picinfoaddr(i['picaddr'])
# print picaddr
info=parseinfo(readsrc(picaddr))
name=info['title']
picture=doc.createElement('picture')
title = doc.createElement('title')
title.appendChild(doc.createTextNode(info['title']))
picture.appendChild(title)
キーワード = doc.createElement('keywords')
keyword.appendChild(doc.createTextNode(info['kw']))
pic.appendChild(keyword)
tp = doc.createElement('pictype')
tp.appendChild(doc.createTextNode(info['type']))
pic.appendChild(tp)
cuturl = doc.createElement('piccut')
cuturl.appendChild(doc.createTextNode(i['cutaddr']))
pic.appendChild(cuturl)
urls = doc.createElement('urls')
self.lock.acquire()
print 'ダウンロード中',name
self.lock.rel easy()
info の picurl[ 'pic']:
singleurl=doc.createElement('url')
singleurl.appendChild(doc.createTextNode(picurl+'.jpg'))
urls.app endChild(singleurl)
picture.appendChild(urls)
self.picdoc.appendChild(picture)
m=pageinfo(self.addr)
self.addr=nextpageaddr(self.addr)
# f = open('c:\'+self.tp+'.xml','w')
# f.write(doc.toprettyxml(indent = ''))
# f.close()
def stop(self):
self.thread_stop=True
path='C:\pict\'#ダウンロードの経路
#import sys
sys.exit(12)
content=readsrc('http://photos.pcgames. com.cn/cate/3/1.html')
r=pictype(content)
lt=[]
doc = minidom.Document()
root=doc.createElement('url_resource ')
root.setAttribute('type','url')
root.setAttribute('urltype','image')
root.setAttribute('imgfmt','jpg')
doc.appendChild(root)
lock=threading.RLock()
for iaddr in r:
print 'ダウンロード タイプ: ',iaddr['name']
addr=iaddr['addr' ]
th=mthread(iaddr['name'],addr,lock)
lt.append(th)
th.start()
for t in lt:
t.join ()
root.appendChild(t.picdoc)
print 'write'
f = open('c:\'+'urls'+'.xml','w')
f.write(doc.toprettyxml(indent = ''))
f.close()
print doc.toprettyxml()
print 'end'