【杂】pyqt+webkit 访问dom对象

【杂】pyqt+webkit 访问dom对象

import sys 

from PyQt4.QtCore import *

from PyQt4.QtGui import *

from PyQt4.QtWebKit import *

start_url = 'http://searchex.yixun.com/?YTAG=1.100090000'

def save_file(data,file='temp.txt'):

f = open(file,'w')

f.write(data)

f.close()

web = None

def onDone():

f = open('scrapy.html','w')

page = web.page().mainFrame().toHtml().toUtf8().data()

f.write(page) 

f.close()

print "Done ..."

e = web.page().mainFrame().documentElement().findFirst("h3[class=m_classbox_hd]")

print e.tagName()

#print dir(e)

print e.toPlainText().toUtf8().data() #.decode('gbk') #.decode('utf-8') #.encode('utf-8')

save_file(e.toPlainText().toUtf8().data())

def onStart():

print "Started..." 

app = QApplication(sys.argv)

web = QWebView()

web.page().mainFrame().loadStarted.connect(onStart)

web.page().mainFrame().loadFinished.connect(onDone)

web.load(QUrl(start_url))

#print web.page().mainFrame().toHtml()

print 'end'

web.show()

sys.exit(app.exec_())