寫給自己的技術筆記 — 作為程式開發者我們絕對不能忽略的JSON — Python 如何處理JSON文件

如果覺得我的圖不清楚都可以直接在我的Github上看喔，Github連結

哈囉，由於JSON系列的第一篇(寫給自己的技術筆記 — 作為程式開發者我們絕對不能忽略的JSON — 教學和用法 — 以JavaScript為例(一))已經針對JSON做過介紹了，所以這篇就不會再詳細介紹JSON囉，這篇會把重點放在如何使用Python來處理JSON檔案

1. JSON 介紹

• JSON 全名為 JavaScript Object Notation，即 JavaScript 對象標記法
• 2001年提出
• 特性:
• 輕量型(Light-Weight): 表示產出的檔案非常小
• 基於文本的(Text-Based): 可以使用文本編譯器(ex. 記事本)開啟
• 可讀的(Human-Readable)格式: 可以輕鬆被人閱讀
• 穩定性(Stability): JSON創始人說 JSON 格式永遠不會改變，也就是說很久以前寫的 JSON 文件，很久以後也能使用，就不會有兼容性的問題
• 目前 REST API 的主流形式，可見它在數據交換格式的重要性

2. Python中的JSON套件

• json套件: Python中用來編碼和解碼JSON文件的套件
• json套件為Python中自帶的套件，所以不用額外安裝，只需要直接載入即可

JSON 和 Python 資料型態轉換表

Python 轉 JSON

JSON 轉 Python

3. Python 如何處理 JSON 套件

處理過程分成:

• Marshalling / Encoder 序列化: 代表將 Python 的物件資料轉換成為 JSON 物件，使用方法 — json.dumps 或 json.dump
• Unmarshalling / Decoder 反序列化: 代表將 JSON 物件轉換成為 Python 的資料物件，使用方法 — json.loads 或 json.load

使用方法

完整參數

• json.dumps

dumps(obj, *, skipkeys=False, ensure_ascii=True, check_circular=True, allow_nan=True, cls=None, indent=None, separators=None, default=None, sort_keys=False, **kw)

• json.dump

dump(obj, fp, *, skipkeys=False, ensure_ascii=True, check_circular=True, allow_nan=True, cls=None, indent=None, separators=None, default=None, sort_keys=False, **kw)

• json.loads

loads(s, *, encoding=None, cls=None, object_hook=None, parse_float=None, parse_int=None, parse_constant=None, object_pairs_hook=None, **kw)

• json.load

load(fp, *, cls=None, object_hook=None, parse_float=None, parse_int=None, parse_constant=None, object_pairs_hook=None, **kw)

4. 實作

JSON 檔案或字符串 轉 Python字典

1. 從JSON檔中導入

• 我們有一份JSON擋 — test.json

{
  "Student 1": {
    "Student_id": 126,
    "Name": "Jack",
    "Score": 99,
    "Friends": ["Tim", "Jen", "Ken"]
 },
​
  "Student 2": {
    "Student_id": 128,
    "Name": "Ken",
    "Score": 99,
    "Friends": ["Jack", "Jessy", "Cathy"]
 }
}
​

• 方法一: json.load

import json
​
## 讀取JSON檔
with open('test.json', 'r') as f:
  ## 轉成Python Dict
  python_dict = json.load(fp = f)
print("Python Dict: ", python_dict)
print("Type: ", type(python_dict))

執行結果

Python Dict: {'Student 1': {'Student_id': 126, 'Name': 'Jack', 'Score': 99, 'Friends': ['Tim', 'Jen', 'Ken']}, 'Student 2': {'Student_id': 128, 'Name': 'Ken', 'Score': 99, 'Friends': ['Jack', 'Jessy', 'Cathy']}}
Type: <class 'dict'>

• 方法二: json.loads

​
import json
​
## 讀取JSON檔
with open('test.json', 'r') as f:
  ## 將內容裝進text
  text = f.read()
  
print("JSON Format: ", text)
print("Type: ", type(text))
print("From JSON String Transform To Python Dict")
print("Python Dict: ", json.loads(text))
print("Type: ", type(json.loads(text)))
​

執行結果

​JSON Format: {
 "Student 1": {
   "Student_id": 126,
   "Name": "Jack",
   "Score": 99,
   "Friends": ["Tim", "Jen", "Ken"]
 },
​
 "Student 2": {
   "Student_id": 128,
   "Name": "Ken",
   "Score": 99,
   "Friends": ["Jack", "Jessy", "Cathy"]
 }
}
​
Type: <class 'str'>
From JSON String Transform To Python Dict
Python Dict: {'Student 1': {'Student_id': 126, 'Name': 'Jack', 'Score': 99, 'Friends': ['Tim', 'Jen', 'Ken']}, 'Student 2': {'Student_id': 128, 'Name': 'Ken', 'Score': 99, 'Friends': ['Jack', 'Jessy', 'Cathy']}}
Type: <class 'dict'>

2. 有一筆JSON Format的字符串 轉 Python Dict

• 只能使用json.loads

import json
​
## JSON Format的字符串
text = '{"Name":"Jack", "Score": 99}'
​
print("JSON Format: ", text)
print("Type:, "type(text))
print("From JSON String Transform To Python Dict")
print("Python Dict: ", json.loads(text))
print("Type: ", type(json.loads(text)))
​

執行結果

JSON Format: {"Name":"Jack", "Score": 99}
Type: <class 'str'>
From JSON String Transform To Python Dict
Python Dict: {'Name': 'Jack', 'Score': 99}
Type: <class 'dict'>

Python 字典(Dict) 轉 JSON字符串

1. 寫入到JSON檔 — 會自動保存成JSON檔

• 方法一: json.dump

import json
​
## Python 字典
python_dict = {
  'Name': "Jack",
  'Score': 99
}
​
## 存成JSON File
with open("python2json.json", 'w') as f:
  json.dump(python_dict, f)

• 方法二: JSON.dumps

import json
​
## Python 字典
python_dict = {
  'Name': "Jack",
  "Score": 99,
}
print("Python Dict: ", python_dict)
print("Type: ", type(python_dict))
print("From Python Dict Transform To JSON Format")
print("JSON Format: ", json.dumps(python_dict))
print("Type: ", type(json.dumps(python_dict)))
​
## 存成JSON File
with open('python2json1.json', 'w') as f:
  f.write(json.dumps(python_dict))

執行結果

Python Dict: {'Name': 'Jack', 'Score': 99}
Type: <class 'dict'>
From Python Dict Transform To JSON Format
JSON Format: {"Name": "Jack", "Score": 99}
Type: <class 'str'>

2. 轉成 JSON Format的字符串

• 只能使用函數: json.dumps

import json
​
## Python 字典
python_dict = {
  'Name': "Jack",
  "Score": 99,
}
print("Python Dict: ", python_dict)
print("Type: ", type(python_dict))
print("From Python Dict Transform To JSON Format")
print("JSON Format: ", json.dumps(python_dict))
print("Type: ", type(json.dumps(python_dict)))

執行結果

Python Dict: {'Name': 'Jack', 'Score': 99}
Type: <class 'dict'>
From Python Dict Transform To JSON Format
JSON Format: {"Name": "Jack", "Score": 99}
Type: <class 'str'>

Python 字典(Dict)中包含list 轉 JSON 字符串

• 範例: 這邊是使用dumps來存進檔案

import json
​
## Python 字典
python_dict = {
  "Student A": {
    "Student_id": 126,
    "Name": "Jack",
    "Score": 99,
    "Friends": ["Tim", "Jen", "Cindy"]
 },
  
  "Student B":{
    "Student_id": 128,
    "Name": "Ken",
    "Score": 98,
    "Friends": ["Jack", "Jessy", "Cathy"]
 }
}
print("Python Dict: ", python_dict)
print("Type: ", type(python_dict))
print("From Python Dict Transform To JSON Format")
print("JSON Format: ", json.dumps(python_dict))
print("Type: ", type(json.dumps(python_dict)))
​
## 存成JSON File
with open('python2json2.json', 'w') as f:
  f.write(json.dumps(python_dict))

執行結果

Python Dict: {'Student A': {'Student_id': 126, 'Name': 'Jack', 'Score': 99, 'Friends': ['Tim', 'Jen', 'Cindy']}, 'Student B': {'Student_id': 128, 'Name': 'Ken', 'Score': 98, 'Friends': ['Jack', 'Jessy', 'Cathy']}}
Type: <class 'dict'>
From Python Dict Transform To JSON Format
JSON Format: {"Student A": {"Student_id": 126, "Name": "Jack", "Score": 99, "Friends": ["Tim", "Jen", "Cindy"]}, "Student B": {"Student_id": 128, "Name": "Ken", "Score": 98, "Friends": ["Jack", "Jessy", "Cathy"]}}
Type: <class 'str'>

大家也可以試試用json.dump的方式存成JSON檔喔

5. 重要補充: 如果寫進JSON檔，想要更易讀的格式來寫

• 前面的範例中，當我們把Python Dict轉成JSON Format的字符串時，都會變成一排(行)，但實際上的狀況是我們需要閱讀這些JSON檔，如果數據一多就會很難閱讀
• 在json.dumps()裡面填上一個參數indent，用來指定每行開頭要空幾格
• 範例一: 我將indent設4，代表空四格

import json
​
## Python 字典
python_dict = {
  'Name': "Jack",
  "Score": 99,
}
print("Python Dict: ", python_dict)
print("Type: ", type(python_dict))
print("From Python Dict Transform To JSON Format")
print("JSON Format: ", json.dumps(python_dict))
print("Type: ", type(json.dumps(python_dict)))
​
## 存成JSON File
with open('python2json3.json', 'w') as f:
  f.write(json.dumps(python_dict, indent = 4))

執行結果

Python Dict: {'Name': 'Jack', 'Score': 99}
Type: <class 'dict'>
From Python Dict Transform To JSON Format
JSON Format: {"Name": "Jack", "Score": 99}
Type: <class 'str'>

• json.dump也擁有參數indent喔
• 範例二:

import json
​
## Python 字典
python_dict = {
  "Student A": {
    "Student_id": 126,
    "Name": "Jack",
    "Score": 99,
    "Friends": ["Tim", "Jen", "Cindy"]
 },
  
  "Student B":{
    "Student_id": 128,
    "Name": "Ken",
    "Score": 98,
    "Friends": ["Jack", "Jessy", "Cathy"]
 }
}
print("Python Dict: ", python_dict)
print("Type: ", type(python_dict))
print("From Python Dict Transform To JSON Format")
print("JSON Format: ", json.dumps(python_dict))
print("Type: ", type(json.dumps(python_dict)))
​
## 存成JSON File
with open('python2json4.json', 'w') as f:
  f.write(json.dumps(python_dict, indent = 4))
  f.write('\n')

執行結果

Python Dict: {'Student A': {'Student_id': 126, 'Name': 'Jack', 'Score': 99, 'Friends': ['Tim', 'Jen', 'Cindy']}, 'Student B': {'Student_id': 128, 'Name': 'Ken', 'Score': 98, 'Friends': ['Jack', 'Jessy', 'Cathy']}}
Type: <class 'dict'>
From Python Dict Transform To JSON Format
JSON Format: {"Student A": {"Student_id": 126, "Name": "Jack", "Score": 99, "Friends": ["Tim", "Jen", "Cindy"]}, "Student B": {"Student_id": 128, "Name": "Ken", "Score": 98, "Friends": ["Jack", "Jessy", "Cathy"]}}
Type: <class 'str'>