【搬砖】xml与json处理的一些备忘录（备忘录）

import xml.dom.minidom
import json

####获取有问题图片的电站ID与具体帧号
root_path = 'D:/zkys-drp/work/光伏/可见光训练数据'
img_dirs = os.listdir(root_path)


json_dict = {}
for img_dir in img_dirs:
    if '九江沙岭光伏电站' in img_dir:
        #print('img_dir:', img_dir)
        
        xml_root = os.path.join(root_path,img_dir) + '/H'
        xml_list = os.listdir(xml_root)
        for xml_file in xml_list:
            xml_path = os.path.join(xml_root, xml_file)
            #print(xml_path)
            DOMTree = xml.dom.minidom.parse(xml_path)
            doc = DOMTree.documentElement
            path = doc.getElementsByTagName("path")[0]
            fullpath = path.childNodes[0].data
            power_id = '/'.join(fullpath.split('/')[4:6])+'/' + xml_file
            #print(power_id)
            frame_ =  doc.getElementsByTagName("irfps")[0]
            frame = frame_.childNodes[0].data
            json_dict[power_id] = frame
            
            #print('frame:', frame)
#print('json_dict:', len(json_dict))  
with open("九江沙岭光伏电站.json", "w") as f:
    f.write(json.dumps(json_dict, ensure_ascii=False, indent=4, separators=(',', ':'))) #, indent=4, separators=(',', ':')         


rm_path = 'rm001.txt'

json_path = '九江沙岭光伏电站.json'
with open(json_path) as f:
     data = ''
     while True:
         line = f.readline()
         #print('1111:', line)
         if not line:
             break
         #if 'imageData' in line:
         #    continue
         data += line.rstrip()
     data = json.loads(data)
#print('data:',len(data))

num = 0
rm_dict = {} 
with open(rm_path, 'r', encoding='utf-8') as f:
    rm_list = f.readlines()
    for rm_info in rm_list:
        rm_ = rm_info.split('\\')
        power_ = os.path.splitext(rm_[-1].rstrip())[0]+'.xml'
        power_id = 'zklfh20t_20200924181730_九江沙岭光伏电站/' + rm_[0] + '/'+power_
        #print(power_id)
        #if power_id in data.keys():
        #    num+=1
        #    print(power_id)
        rm_dict[power_id] = data[power_id]
with open("九江沙岭光伏电站_fault.json", "w") as f:
   f.write(json.dumps(rm_dict, ensure_ascii=False, indent=4, separators=(',', ':'))) #, indent=4, separators=(',', ':') 
#print(rm_dict, len(rm_dict))