如何使用scrapy的蜘蛛网名称指定日志文件名？

Question

I'm using scrapy,in my scrapy project,I created several spider classes,as the official document said,I used this way to specify log file name:

 def logging_to_file(file_name):
"""
@rtype: logging
@type file_name:str
@param file_name:
@return:
"""
import logging
from scrapy.utils.log import configure_logging
configure_logging(install_root_handler=False)
logging.basicConfig(
    filename=filename+'.txt',
    filemode='a',
    format='%(levelname)s: %(message)s',
    level=logging.DEBUG,

)
 Class Spider_One(scrapy.Spider):
      name='xxx1'
      logging_to_file(name)
  ......
 Class Spider_Two(scrapy.Spider):
      name='xxx2'
      logging_to_file(name)
  ......

Now,if I start Spider_One ,everything is correct!But,if I start Spider Two ,the log file of Spider Two will also be named with the name of Spider One !
I have searched many answers from google and stackoverflow,but unfortunately,none worked!
I am using python 2.7 & scrapy 1.1!
Hope anyone can help me!

Answer 1

It's because you initiate logging_to_file every time when you load up your package. You are using a class variable here where you should use instance variable.

When python loads in your package or module ir loads every class and so on.

class MyClass:
    # everything you do here is loaded everytime when package is loaded
    name = 'something'

    def __init__(self):
        # everything you do here is loaded ONLY when the object is created
        # using this class

To resolve your issue just move logging_to_file function call to your spiders __init__() method.

class MyClass(Spider):
    name = 'xx1'

    def __init__(self):
        super(MyClass, self).__init__()
        logging_to_file(self.name)