我有一個(gè)腳本，可以解析文檔并使用lxml更改表單值?，F(xiàn)在我想用硒導(dǎo)航到頁(yè)面并做到這一點(diǎn)。不幸的是，硒似乎不夠先進(jìn)，支持我在lxml中所做的一些事情，例如，“如果'attrName'in tag.attrib”然后我說(shuō)，“啊哈！，但是硒提供了webdriver.page_source，我可以得到html！”是的，有點(diǎn)...我可以得到它，然后用lxml解析它...但令人討厭的是，page_source是不可寫的，所以我無(wú)法堅(jiān)持我的更改！我仍然覺(jué)得我走在正確的道路上，但是有沒(méi)有辦法將修改后的html寫回瀏覽器？更新：我嘗試了幾種嘗試重寫所有頁(yè)面內(nèi)容的方法，但是firefox網(wǎng)絡(luò)驅(qū)動(dòng)程序開始拋出安全異常，就像它不喜歡我試圖做的事情的外觀一樣。下面我將發(fā)布我使用的替代方案。
1 回答
叮當(dāng)貓咪
TA貢獻(xiàn)1776條經(jīng)驗(yàn) 獲得超12個(gè)贊
我最終做的是創(chuàng)建一個(gè)類似lxml的中介類。這只實(shí)現(xiàn)了我需要的那些功能，但是，我仍然認(rèn)為我會(huì)分享作為其他嘗試做同樣事情的人的起點(diǎn)。
import lxml.etree
import collections.abc
class AttribWrapper(collections.abc.MutableMapping):
"""
Make something that acts like a dict to use as a mock lxml.etree.element attrib value
see also:
https://docs.python.org/3/reference/datamodel.html?emulating-container-types#emulating-container-types
"""
def __init__(self,seleniumTag):
self.seleniumTag=seleniumTag
while not hasattr(seleniumTag,'execute_script'):
seleniumTag=seleniumTag.parent
self.webdriver=seleniumTag
def hasAttribute(self,attrName):
"""
determine if a tag has an attribute
"""
try:
val=self.seleniumTag.get_attribute(attrName)
except Exception:
return False
return val is not None
def getAttribute(self,attrName,default=None):
"""
get a tag's attribute or default if not present
"""
try:
return self.seleniumTag.get_attribute(attrName)
except Exception:
pass
return default
def setAttribute(self,attrName,attrValue):
"""
get a tag's attribute
"""
self.webdriver.execute_script("arguments[0].setAttribute(arguments[1],arguments[2]);",self.seleniumTag,attrName,attrValue);
def __getitem__(self,key):
if key not in self.keys():
raise KeyError()
return self.getAttribute(key)
def __setitem__(self,key,value):
self.setAttribute(key,value)
def __delitem__(self,key):
self.webdriver.execute_script('arguments[0].attributes.removeNamedItem("%s");'%key,self.seleniumTag)
def __iter__(self):
items=[]
for k in self.keys():
v=self[k]
items.append((k,v))
return self.keys().__iter__()
def __len__(self):
return self.webdriver.execute_script('return arguments[0].attributes.length',self.seleniumTag)
def keys(self):
n=self.webdriver.execute_script('return arguments[0].attributes.length',self.seleniumTag)
names=[]
for idx in range(n):
result=self.webdriver.execute_script('return arguments[0].attributes.item(%d).nodeName'%idx,self.seleniumTag)
names.append(result)
return names
def __repr__(self):
ret=[]
for k,v in self.items():
ret.append('("%s":"%s")'%(k,v))
return "{%s}"%(', '.join(ret))
class LikeEtreeElement:
def __init__(self,seleniumTag):
self.seleniumTag=seleniumTag
self.attrib=AttribWrapper(seleniumTag)
while not hasattr(seleniumTag,'execute_script'):
seleniumTag=seleniumTag.parent
self.webdriver=seleniumTag
@property
def tag(self):
return self.seleniumTag.tag_name
def getchildren(self):
"""
get all child elements
(NOTE: this will dip into the html every single time just
in case things have changed.)
"""
return self.seleniumTag.find_elements_by_xpath('/*')
@property
def innerHTML(self):
return self.webdriver.execute_script('return arguments[0].innerHTML',self.seleniumTag)
@innerHTML.setter
def innerHTML(self,value):
value=str(value).replace('\\\\','\\').replace('\n','\\n').replace('"','\\"')
return self.webdriver.execute_script('arguments[0].innerHTML="%s"'%(value),self.seleniumTag)
def __getitem__(self,idx):
return self.getchildren()[idx]
def __iter__(self):
return self.getchildren().__iter__()
def __len__(self):
return len(self.getchildren())
添加回答
舉報(bào)
0/150
提交
取消