The correct way for Python to get the directory where the script is located

C:test
  [dir] getpath
    [file] path.py
    [dir]  sub
      [file] sub_path.py

For example,

C: There is also a directory named sub in the test directory;

C: There is getpath.py in the test directory, sub_path.py in the sub directory, getpath.py calls sub_path.py;

We Execute getpath.py under C:test. if we were If sys.path[0] is used in sub_path.py, what is actually obtained is the directory path "C:test" where getpath.py is located, because Python virtual The machine starts execution from getpath.py. If you want to get the path of sub_path.py, then you have to do this:
os.path.split(os.path.realpath(__file__))[0]

Verification:

Then we execute python getpath/path under C:test. py, at this time, the values ​​corresponding to various usages in sub_path.py are actually:
os.getcwd() "C:test", which takes the starting execution directory
sys.path[0] or sys.argv[0 ] "C:testgetpath", takes the directory where the initially executed script is located
os.path.split(os.path.realpath(__file__))[0] "C:testgetpathsub", takes the directory where __file__ is located The directory where the file sub_path.py is located