全局上下文python中使用的全局关键字

Question

在python中，全局上下文中使用的全局键盘意味着什么？我知道如果在函数内部使用它，它可以修改全局变量，但我不知道当你用global声明一个全局变量时它是如何工作的，因为我认为当你在函数外部声明时，它们已经是全局变量了。如下代码所示：

from ctypes import *

import ctypes

import os
import cv2
import time
import Image
import argparse
import numpy as np
import thread as thread

from select import select
from evdev import InputDevice

import ArducamSDK

COLOR_BYTE2RGB = 49
CAMERA_MT9F002 = 0x4D091031
SensorShipAddr = 32
I2C_MODE_16_16  = 3
usbVid = 0x52cb
Width = 4384
Height = 3288
cfg ={"u32CameraType":CAMERA_MT9F002,
      "u32Width":Width,"u32Height":Height,
      "u32UsbVersion":1,
      "u8PixelBytes":1,
      "u16Vid":0x52cb,
      "u8PixelBits":8,
      "u32SensorShipAddr":SensorShipAddr,
      "emI2cMode":I2C_MODE_16_16 }
# *********** Here is where I'm talking about****************
global saveNum,saveFlag,downFlag,flag,handle,openFlag
openFlag = False
saveNum = 0
handle = {}
downFlag = False
flag = True
saveFlag = False


regArr= [[0x301A, 0x0018],

    [0x0103, 0x1   ],  

    [0x31AE, 0x301 ],  
    [0x0112, 0x0C0C],   
    [0x301A, 0x0010],   
    [0x3064, 0x0805],   
    [0x301E, 0x00A8],   
    [0x301A, 0x10DC],   
    [0x0104, 0x1   ],  
    [0x0100, 0x0   ],  

    [0x0300, 0x4   ],  
    [0x0302, 0x01  ],  
    [0x0304, 0x03  ],  
    [0x0306, 0x6E  ],  
    [0x0308, 0x0C  ],  
    [0x030A, 0x01  ],  

    [0x3016, 0x111 ],  
    [0x0344, 0x090 ],  
    [0x0348, 0x11AF],   
    [0x0346, 0x020 ],  
    [0x034A, 0xCF7 ],  
    [0x3040, 0x0041],   
    [0x0400, 0x0   ],  
    [0x0404, 0x10  ],  
    [0x034C, 0x1120],   
    [0x034E, 0xCD8 ],  
    [0x0342, 0x33C1],   
    [0x0340, 0x0D67],   
    [0x3010, 0x00CF],   
    [0x3012, 0x0010],   
    [0x3014, 0x1F50],   
    [0x3018, 0x0000],   

    [0x0104, 0x0],  
    [0x0100, 0x1],  
    [0x0304, 8],

    [0x301a, 0x5ccc],        
    [0x3012, 500],
    [0x0206, 33],
    [0x0208, 50],
    [0x020a, 50],
    [0x020c, 33],

    [0xffff, 0xffff],
    [0xffff, 0xffff]
]


ap = argparse.ArgumentParser()
ap.add_argument("--type",default = "jpg",required = False, help = "type to the image file")
ap.add_argument("--name",required = False, help = "name to the image file")
args = vars(ap.parse_args())

def readThread(threadName,read_Flag):
    global flag,handle
    data = {}
    cv2.namedWindow("MT9F002",1)
    while flag:
        if ArducamSDK.Py_ArduCam_availiable(handle) > 0:

            res,data = ArducamSDK.Py_ArduCam_read(handle,Width * Height)
            if res == 0:
                ArducamSDK.Py_ArduCam_del(handle)
            else:
                print "read data fail!"

        else:
            print "is not availiable"
        if len(data) >= Width * Height:
            show(data)
            flag = False
        else:
            print "data length is not enough!"
        if flag == False:       
            break

thread.start_new_thread( readThread,("Thread-2", flag,))

pass

def show(bufferData):
    global downFlag,saveFlag,saveNum
    image = Image.frombuffer("L",(Width,Height),bufferData)
    img = np.array(image)
    height,width = img.shape[:2]
    img2 = cv2.cvtColor(img,COLOR_BYTE2RGB)
        saveNum += 1

    if args["name"] != "":
                name = args["name"]   
        else:
                name = str(saveNum)
    if "bmp" == args["type"]:        
                name += ".bmp"
        if "png" == args["type"]:        
                name += ".png"
        if "jpg" == args["type"]:        
                name += ".jpg"
    cv2.imwrite(name,img2)  


def video():
    global flag,regArr,handle
    regNum = 0
    res,handle = ArducamSDK.Py_ArduCam_autoopen(cfg)
    capFlag = True
    if res == 0:
        openFlag = True
        print "device open success!"
        while (regArr[regNum][0] != 0xFFFF):
            ArducamSDK.Py_ArduCam_writeSensorReg(handle,regArr[regNum][0],regArr[regNum][1])
            regNum = regNum + 1
        res = ArducamSDK.Py_ArduCam_beginCapture(handle)

        if res == 0:
            print "transfer task create success!"
            while flag :
                                if capFlag:
                                        res = ArducamSDK.Py_ArduCam_capture(handle)
                                        if res != 0:
                                                print "capture fail!"
                                                break
                                        else:
                                                capFlag = False
                                time.sleep(0.4)
                if flag == False:       
                    break
        else:
            print "transfer task create fail!"
        res = ArducamSDK.Py_ArduCam_close(handle)
        if res == 0:
            openFlag = False
            print "device close success!"
        else:
            print "device close fail!"
    else:
        print "device open fail!"

if __name__ == "__main__":      
    video()

Answer 1

绝对没有什么。global提示编译器需要检查外部作用域中当前范围内代码的名称。因为全局范围是最外面的范围，所以没有任何影响。在

>>> import dis
>>> def foo():
...   global bar
...   bar = 1
...   baz = 2
... 
>>> dis.dis(foo)
  3           0 LOAD_CONST               1 (1)
              3 STORE_GLOBAL             0 (bar)

  4           6 LOAD_CONST               2 (2)
              9 STORE_FAST               0 (baz)
             12 LOAD_CONST               0 (None)
             15 RETURN_VALUE        
>>> def quux():
...   global bar
...   print bar
...   baz = 1
...   print baz
... 
>>> dis.dis(quux)
  3           0 LOAD_GLOBAL              0 (bar)
              3 PRINT_ITEM          
              4 PRINT_NEWLINE       

  4           5 LOAD_CONST               1 (1)
              8 STORE_FAST               0 (baz)

  5          11 LOAD_FAST                0 (baz)
             14 PRINT_ITEM          
             15 PRINT_NEWLINE       
             16 LOAD_CONST               0 (None)
             19 RETURN_VALUE