spider/EasySpider
1
0
Fork 0
mirror of https://github.com/NaiboWang/EasySpider.git synced 2025-04-20 04:39:57 +08:00
Code Issues Packages Projects Releases Wiki Activity

replace variables

Browse Source
This commit is contained in:
NaiboWang-Alienware 2023-07-22 17:10:13 +08:00
parent 8c3a613014
commit c9eb5b0932
1 changed files with 7 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

8
ExecuteStage/easyspider_executestage.py
View File

@ -1071,6 +1071,9 @@ class BrowserThread(Thread):
break break
elif int(node["parameters"]["loopType"]) == 3: # 固定文本列表 elif int(node["parameters"]["loopType"]) == 3: # 固定文本列表
textList = node["parameters"]["textList"].split("\n") textList = node["parameters"]["textList"].split("\n")
if len(textList) == 1: # 如果固定文本列表只有一行，现在就可以替换变量
textList = replace_field_values(
node["parameters"]["textList"], self.outputParameters).split("\n")
for text in textList: for text in textList:
text = replace_field_values(text, self.outputParameters) text = replace_field_values(text, self.outputParameters)
# self.recordLog("当前循环文本|Current loop text:", text) # self.recordLog("当前循环文本|Current loop text:", text)
@ -1092,6 +1095,9 @@ class BrowserThread(Thread):
# tempList = node["parameters"]["textList"].split("\r\n") # tempList = node["parameters"]["textList"].split("\r\n")
urlList = list( urlList = list(
filter(isnotnull, node["parameters"]["textList"].split("\n"))) # 去空行 filter(isnotnull, node["parameters"]["textList"].split("\n"))) # 去空行
if len(urlList) == 1: # 如果固定网址列表只有一行，现在就可以替换变量
urlList = replace_field_values(
node["parameters"]["textList"], self.outputParameters).split("\n")
# urlList = [] # urlList = []
# for url in tempList: # for url in tempList:
# if url != "": # if url != "":
@ -1516,7 +1522,7 @@ class BrowserThread(Thread):
content = select_element.first_selected_option.text content = select_element.first_selected_option.text
except: except:
content = "" content = ""
elif p["contentType"] == 14: # 元素属性值 elif p["contentType"] == 14: # 元素属性值
attribute_name = p["JS"] attribute_name = p["JS"]
try: try:
content = element.get_attribute(attribute_name) content = element.get_attribute(attribute_name)