class HTMLClient:
  #獲取html網(wǎng)頁源碼
  def GetPage(self, url):
    #user_agent = 'Mozilla/4.0 (compatible; MSIE 5.5; Windows NT)'
    user_agent = 'Mozilla/5.0 (X11; Linux i686) AppleWebKit/537.36 (KHTML, like Gecko) Ubuntu Chromium/34.0.1847.116 Chrome/34.0.1847.116 Safari/537.36'
    headers = { 'User-Agent' : user_agent }
    req = urllib.request.Request(url, None, headers)
    try:
      res = urllib.request.urlopen(req)
      return res.read().decode("utf-8")
    except urllib.error.HTTPError as e:
      return None
  #獲取網(wǎng)絡圖片并保存在程序運行目錄下
  def GetPic(self, url):
    user_agent = 'Mozilla/4.0 (compatible; MSIE 5.5; Windows NT)'
    headers = { 'User-Agent' : user_agent }
    req = urllib.request.Request(url, None, headers)
    try:
      res = urllib.request.urlopen(req)
      return res.read()
    except urllib.error.HTTPError as e:
      return None

class MYHTMLParser(HTMLParser):
  def __init__(self, docfile):
    HTMLParser.__init__(self)
    self.docfile = docfile
    self.doc = Document(docfile)
    self.myclient = HTMLClient()
    self.text = ''
    self.title = False
    self.isdescription = False
    self.picList=[]
  #根據(jù)標簽頭類型決定標簽內(nèi)容的格式
  def handle_starttag(self, tag, attrs):
    #print "Encountered the beginning of a %s tag" % tag
    self.title = False
    self.isdescription = False
    #<h1>標簽說明其中的內(nèi)容是標題
    if re.match(r'h(\d)', tag):
      self.title = True
    #圖片的處理比較復雜，首先需要找到對應的圖片的url，然后下載并寫入doc中
    #下載的圖片格式如果有問題，docx模塊會報錯，因此重新定義異常處理
    #圖片名稱需要記錄下來，在文檔保存后要自動刪除
    if tag == "img":
      if len(attrs) == 0: pass
      else:
        for (variable, value) in attrs:
          if variable == "src":
            #此處圖片url類型為[http://url/pic.img!200*200]
            #不同網(wǎng)站圖片類型不同，因此當作不同處理
            picdata = self.myclient.GetPic(value.split('!')[0])
            if picdata == None:
              pass
            else:
              pictmp = value.split('/')[-1].split('!')[0]
              picfix = value.split('/')[-1].split('!')[-1]
              with open(pictmp, 'wb') as pic:
                pic.write(bytes(picdata))
                pic.close()
              try:
                if picfix[0:1] == 'c':
                  self.doc.add_picture(pictmp, width=Inches(4.5))
                else:
                  self.doc.add_picture(pictmp)#, width=Inches(2.25))
              except docx.image.exceptions.UnexpectedEndOfFileError as e:
                print(e)
              self.picList.append(pictmp)
    #javascript腳本
    if tag == 'script':
      self.isdescription = True
  def handle_data(self, data):
    if self.title == True:
      if self.text != '':
        self.doc.add_paragraph(self.text)
      self.text = ''
      self.doc.add_heading(data, level=2)
    if self.isdescription == False:
      self.text += data
  def handle_endtag(self, tag):
    #if tag == 'br' or tag == 'p' or tag == 'div':
    if self.text != '':
      self.doc.add_paragraph(self.text)
      self.text = ''
  def complete(self, html):
    self.feed(html)
    self.doc.save(self.docfile)
    for item in self.picList:
      if os.path.exists(item):
        os.remove(item)