糾正JSON格式，並以星火

Question

落實我與格式線JSON文件：

{"user1" : "a", 
"mobile": "b", 
"address": "c" 

{"user2" : "aa", 
"mobile": "bb", 
"address": "cc" 

此日誌是不正確的，沒有}對各行的末尾

我試過

text_file = open('abc.txt', "r", encoding='utf-8') 
read = text_file.read() 
a = read.split("\n") 
for i in a: 
    print(i+"}") 

這給了我一個額外的}所有行結束。我該如何避免它？

另外，我需要在spark中實現相同的邏輯。請讓我知道需要什麼樣的修改，或者如果有火花更好的邏輯

Answer 1

與Python

text_file = open('abc.txt', 'r') 
read = text_file.read() 
a = read.split('{') # split with '{' 
del a[0] # remove the first line contains space 
for i in a: 
    print("{"+i.strip()+"}") # if you want to remove '\n' add .replace('\n','') 

結果：

{"user1" : "a", 
"mobile": "b", 
"address": "c"} 
{"user2" : "aa", 
"mobile": "bb", 
"address": "cc"} 

星火

# use wholeTextFiles to read all lines, textFile split lines with '\n'  
text_file = sc.wholeTextFiles("abc.txt") 
a = text_file.map(lambda (pathFile , lines) : lines). # select only lines 
       flatMap(lambda text : text.split('{')). # split with '{' 
       filter(lambda line : len(line) > 0). # remove the first line 
       map(lambda line : '{'+line.strip()+'}') 
for i in a.collect() : 
    print i 

結果：

{"user1" : "a", 
"mobile": "b", 
"address": "c"} 
{"user2" : "aa", 
"mobile": "bb", 
"address": "cc"}