本節要點
掌握API開發中的要點
掌握JSONPath的概念
瞭解內網穿透的概念
課前準備
無
課程內容
API開發要點
預覽推播的格式
擴展功能 - 資料推播 - 欄位對照表及JSON樣例
1. 判斷推播類型*
op：推播事件；包括data_create(資料送出)、data_update(資料修改)、data_remove(資料刪除)、data_recover(資料恢復)
透過op來區分具體的推播事件
from flask import Flask, request import json app = Flask(__name__) @app.route('/test/', methods=['POST']) #因為Jodoo只接收POST請求 def hello_world(): print("params: ",request.args) print("headers: ",request.headers) print("未編碼data: ",request.data) print("已編碼data: ",json.loads(request.data)) return 'success',200 if __name__ == '__main__': app.run(host='0.0.0.0',port=10000)url: http://101.132.103.228：10000/test/?nonce=5932d0×tamp=1585724440 params: ImmutableMultiDict([('nonce', '5932d0'), ('timestamp', '1585724440')]) headers: Content-Type: application/json X-Jdy-Deliverid: 2b85594d-454a-4063-ac1d-91a2b3318b63 Host: 101.132.103.228:10000 Content-Length: 765 Connection: close 原始data: b'{"data":{"_id":"5dd6740646357c0006e6eb6e","_widget_1574134687834":"\xe6\xb6\xa1\xe8\xbd\xae\xe5\x87\x8f\xe9\x80\x9f\xe6\x9c\xba1","_widget_1574134687849":"DJSM9K2000","_widget_1574134687894":"2019-11-21T11:24:54.000Z","_widget_1574134688078":345999,"_widget_1574134688093":101,"_widget_1574134688108":3459990,"_widget_1574134688185":"\xe5\x8d\x97\xe4\xba\xac\xe6\x82\x9f\xe5\xb8\x86\xe7\xa7\x91\xe6\x8a\x80\xe8\x82\xa1\xe4\xbb\xbd\xe6\x9c\x89\xe9\x99\x90\xe5\x85\xac\xe5\x8f\xb8","_widget_1574142293244":"TSCJ001574335494","appId":"5dce13f43087860006c70e7a","createTime":"2019-11-21T11:24:54.213Z","creator":{"_id":"5cecd40dd23e194ab0867aab","name":"\xe6\x9f\xa5\xe7\x90\x86","username":"cxt7777"},"deleteTime":null,"deleter":null,"entryId":"5dce145c26aecf00062e7db0","formName":"\xe8\xae\xa2\xe5\x8d\x95\xe6\xb1\x87\xe6\x80\xbb","updateTime":"2020-04-01T07:00:40.876Z","updater":{"_id":"5cecd40dd23e194ab0867aab","name":"\xe6\x9f\xa5\xe7\x90\x86","username":"cxt7777"}},"op":"data_update"}' 轉換成Python字典後的data: {'data': {'_id': '5dd6740646357c0006e6eb6e', '_widget_1574134687834': '渦輪減速機1', '_widget_1574134687849': 'DJSM9K2000', '_widget_1574134687894': '2019-11-21T11:24:54.000Z', '_widget_1574134688078': 345999, '_widget_1574134688093': 101, '_widget_1574134688108': 3459990, '_widget_1574134688185': '南京悟帆科技股份有限公司', '_widget_1574142293244': 'TSCJ001574335494', 'appId': '5dce13f43087860006c70e7a', 'createTime': '2019-11-21T11:24:54.213Z', 'creator': {'_id': '5cecd40dd23e194ab0867aab', 'name': '查理', 'username': 'cxt7777'}, 'deleteTime': None, 'deleter': None, 'entryId': '5dce145c26aecf00062e7db0', 'formName': '訂單彙總', 'updateTime': '2020-04-01T07:00:40.876Z', 'updater': {'_id': '5cecd40dd23e194ab0867aab', 'name': '查理', 'username': 'cxt7777'}}, 'op': 'data_update'}2. 多線程處理
Threading ：Python 實現多線程編程需要藉助於 threading 模塊，我們使用多線程來防止處理時間長導致回應超時；
多線程類似於同時執行多個不同程式
使用線程可以把佔據長時間的程式中的任務放到後台去處理；
使用者界面可以更加吸引人，比如使用者點擊了一個按鈕去觸發某些事件的處理，可以彈出一個進度條來顯示處理的進度；
程式的運行速度可能加快；
在一些等待的任務實現上如使用者輸入、文件讀寫和網絡收發資料等，線程就比較有用了。在這種情況下我們可以釋放一些珍貴的資源如內存佔用等等。
threading.Thread(target=<函數名>).start() # 多線程賦值 threading.Thread(target=<函數名>, args=(<變數>,).start()
import threading import time def handle_a(): for i in range(5): print("a",time.strftime("%Y-%m-%d %H:%M:%S", time.localtime()) ) time.sleep(1) def handle_b(): for i in range(5): print("b",time.strftime("%Y-%m-%d %H:%M:%S", time.localtime()) ) time.sleep(1) #使用多線程 threading.Thread(target=handle_a).start() time.sleep(0.1) handle_b() #正常按順序執行函數 handle_a() handle_b() #傳入參數 def printtime(a): print(a, time.time()) x = 123 threading.Thread(target=print,args=(x,)).start()3. API 加密
加密：防止 API 被攻擊；
import hashlib def get_signature(nonce, payload, secret, timestamp): content = ':'.join([nonce, payload, secret, timestamp]).encode('utf-8') m = hashlib.sha1() m.update(content) return m.hexdigest() @app.route('/callback', methods=['POST']) def callback(): payload = request.data.decode('utf-8') nonce = request.args['nonce'] timestamp = request.args['timestamp'] if request.headers['x-jdy-signature'] != get_signature(nonce, payload, 'test-secret', timestamp): return 'fail', 401 threading.Thread(target=handle, args=(json.loads(payload), )).start() return 'success' JSONPath
測試網址：點擊查看
{ "data": [ { "creator": { "_id": "5cecd40dd23e194ab0867aab", "name": "查理", "username": "cxt7777" }, "updater": { "_id": "5cecd40dd23e194ab0867aab", "name": "查理", "username": "cxt7777" }, "deleter": null, "createTime": "2020-03-26T02:41:06.491Z", "updateTime": "2020-03-26T02:46:27.825Z", "deleteTime": null, "_widget_1557886562320": "iPhone 11", "_widget_1557886562335": "5998", "_widget_1557886562350": "17", "_id": "5e7c164229e01a00063be284", "appId": "5e798363b587cc0006b40445", "entryId": "5cdb765b5a6ae613aeed0f84" }, { "creator": { "_id": "5cecd40dd23e194ab0867aab", "name": "查理", "username": "cxt7777" }, "updater": { "_id": "5cecd40dd23e194ab0867aab", "name": "查理", "username": "cxt7777" }, "deleter": null, "createTime": "2020-03-26T02:47:02.037Z", "updateTime": "2020-03-26T02:47:02.037Z", "deleteTime": null, "_widget_1557886562320": "iPhone X", "_widget_1557886562335": "4998", "_widget_1557886562350": "5", "_id": "5e7c17a650bccb0006441778", "appId": "5e798363b587cc0006b40445", "entryId": "5cdb765b5a6ae613aeed0f84" }, { "creator": { "_id": "5cecd40dd23e194ab0867aab", "name": "查理", "username": "cxt7777" }, "updater": { "_id": "5cecd40dd23e194ab0867aab", "name": "查理", "username": "cxt7777" }, "deleter": null, "createTime": "2020-03-26T02:47:43.059Z", "updateTime": "2020-03-26T02:47:43.059Z", "deleteTime": null, "_widget_1557886562320": "iPhone 8", "_widget_1557886562335": "3998", "_widget_1557886562350": "32", "_id": "5e7c17cfcd87510006cf8189", "appId": "5e798363b587cc0006b40445", "entryId": "5cdb765b5a6ae613aeed0f84" } ] } 提取規則類似Python：
取出iPhone 11的商品數量（“_widget_1557886562350”）
$.data[0]._widget_1557886562350
$.data[0][‘_widget_1557886562350’]
內網穿透（演示）
現在我們編寫的API，僅能在本地（你自己的主機，localhost 或 127.0.0.1）造訪，我們需要將本地端口映射至公網環境，才能被Jodoo或是其他人造訪（這裡不要求掌握如何進行內網穿透，瞭解概念即可）；
當然也可以直接把服務部署在具有公網地址的雲伺服器（ECS）上，這是我們下節課介紹的內容！
這裡編寫了一個API，請求與返回格式均為JSON
from flask import Flask, request import json app = Flask(__name__) @app.route('/test/', methods=['POST']) #因為Jodoo只接收POST請求 def hello_world(): print("params: ",request.args) print("headers: ",request.headers) print("未編碼data: ",request.data) print("已編碼data: ",json.loads(request.data)) return 'success',200 if __name__ == '__main__': app.run(host='0.0.0.0',port=3100)