Python: 如何串接Meta API ?如何將dict存成JSON檔? json.dump(dic_msg ,f , ensure_ascii=False ,indent=4)

by 儲蓄保險王 · 2024-01-14

Meta developer:
https://developers.facebook.com
我的應用程式 >
建立應用程式 >
其他 > 繼續
消費者>繼續
程式名稱:爬蟲>
建立應用程式

Python: 如何串接Meta API ?如何將dict存成JSON檔? json.dump(dic_msg ,f , ensure_ascii=False ,indent=4) - 儲蓄保險王

輸入自己FB的密碼>提交
工具>圖形API測試工具
fileds=posts > 提交
(權限至少要打開user_posts)

取得代碼>cURL

複製cURL中的網址:
“https://graph.facebook.com/v18.0/me?fields=posts&access_token=EAA…”
修改為:
https://graph.facebook.com/v18.0/me/posts?access_token=EAA…
#這樣可以少一層”posts” 的key
code:

# -*- coding: utf-8 -*-
"""
Created on Sun Jan 14 10:44:01 2024

@author: SavingKing
"""

import requests as req
import json

curl ="""
https://graph.facebook.com/v18.0/me/posts?access_token=EAA...
"""

curl = curl.strip()

response = req.get(curl)
#<Response [200]>
#requests.models.Response

json_data = json.loads(response.text)
#dict
# print(json_data)

lis_json_data = json_data['data']
lis_msg = []
for dic in lis_json_data:
    if 'message' in dic:
        msg=dic.get('message')
        print(msg)
        lis_msg.append(msg)

執行結果:

json_data:

一層一層撥進去json_data:

可將json_data[‘paging’][‘next’]賦值給url_next
url_next = json_data[‘paging’][‘next’]

封裝進去function中
return lis_msg, url_next
code:

# -*- coding: utf-8 -*-
"""
Created on Sun Jan 14 10:44:01 2024

@author: SavingKing
"""

import requests as req
import json

curl ="""
https://graph.facebook.com/v18.0/me/posts?access_token=EAAap...
"""

curl = curl.strip()
def url2lis_msg(curl):
    response = req.get(curl)
    #<Response [200]>
    #requests.models.Response
    
    json_data = json.loads(response.text)
    #dict
    # print(json_data)
    url_next = json_data['paging']['next']
    
    lis_json_data = json_data['data']
    lis_msg = []
    for dic in lis_json_data:
        if 'message' in dic:
            msg=dic.get('message')
            print(msg)
            lis_msg.append(msg)
    
    return lis_msg, url_next
lis_msg, url_next = url2lis_msg(curl)

執行結果:

Meta API的最後一頁:

Meta API的倒數第二頁(格式同其他頁):

為了從json_data作為迴圈終止的條件
將原本的function拆為兩個
code:

# -*- coding: utf-8 -*-
"""
Created on Sun Jan 14 10:44:01 2024

@author: SavingKing
"""

import requests as req
import json

# token = "nc..."
curl ="""
https://graph.facebook.com/v18.0/me/posts?access_token=EAA...
"""

curl = curl.strip()
def url2json_data(curl):
    response = req.get(curl)
    #<Response [200]>
    #requests.models.Response
    
    json_data = json.loads(response.text)
    #dict
    # print(json_data)
    return json_data


def json_data2lis(json_data):
    
    url_next = json_data['paging']['next']
    
    lis_json_data = json_data['data']
    lis_msg = []
    for dic in lis_json_data:
        if 'message' in dic:
            msg=dic.get('message')
            print(msg)
            lis_msg.append(msg)
    
    return lis_msg, url_next

lis_msg = []

while True:
    json_data=url2json_data(curl)
    if 'paging' not in json_data.keys():
        break
    lis_msg_tmp, curl = json_data2lis(json_data)
    lis_msg.extend(lis_msg_tmp)

dic_msg={"my lis_msg":lis_msg}
with open("my_lis_msg.json","w") as f: #應該要加encoding="UTF-8"
    json.dump(dic_msg,f, ensure_ascii=False,indent=4)

輸出的json有一部分資料,
但沒有全部成功
dic_msg={“my lis_msg”:lis_msg}
with open(“my_lis_msg.json”,”w”) as f:
json.dump(dic_msg,f, ensure_ascii=False,indent=4)
Traceback (most recent call last):

Cell In[123], line 3
json.dump(dic_msg,f, ensure_ascii=False,indent=4)

File C:\Python311\Lib\json__init__.py:180 in dump
fp.write(chunk)

UnicodeEncodeError: ‘cp950’ codec can’t encode character ‘\u654e’ in position 101: illegal multibyte sequence

with open(“my_lis_msg.json”,”w”) as f:
#應該要加encoding=”UTF-8″

執行結果: