回答自己一年前的一个问题，python如何动态拼接sql

首先谈谈应用场景吧，前提是针对查询接口做接口自动化，接口校验的脚本中，一般以响应报文作为预期值，通过sql查出的数据库值作为实际值，二者对比通过则认为接口输出正确。而sql从何而来呢，对于查询接口一般是通过入参作为筛选条件+程序内部逻辑共同决定了我们脚本中查询sql该怎么写。下面将举例工作中遇到的几种场景，带入比较简单粗暴的场景，不一定形象，主打一个意会~。>.<

假设有一个水果店水果统计的查询接口，它的入参有店名、水果名称、进货日期，店名必填，后两者如果传则作为筛选条件，不传则全量查。那分析下来可能就存在四种情况:

1.店名='水果店' 
2.店名='水果店' and 水果名称='西瓜'
3.店名='水果店' and 进货日期='20240316'
4.店名='水果店' and 水果名称='西瓜' and 进货日期='20240316'

这时候就到了我之前疑问的地方，这么多场景，我要怎么决定我的sql怎么写呢？首先可以明确的，肯定是要根据入参是否传，来决定sql中要不要筛选，那么就有了下面的实现方式：

def sgdDemo(reqmsg, resmsg,):#重点关注where后面查询条件的处理，对于与整个sql拼接的部分，后面有时间再写啦~ser = []if(reqmsg["店名"]) != "":sql = "店名='" + reqmsg["店名"]+"'"ser.append(sql)if(reqmsg["水果名称"]) != "":sql = "水果名称='" + reqmsg["水果名称"]+"'"ser.append(sql)if(reqmsg["进货日期"]) != "":   sql= "进货日期='" + reqmsg["进货日期"]+"'"ser.append(sql)ret = ' and '.join(ser)return retif __name__ == '__main__':ret1 = sgdDemo({"店名":"水果店","水果名称":"","进货日期":""},"")ret2 = sgdDemo({"店名":"水果店","水果名称":"西瓜","进货日期":""},"")ret3 = sgdDemo({"店名":"水果店","水果名称":"","进货日期":"20240316"},"")ret4 = sgdDemo({"店名":"水果店","水果名称":"西瓜","进货日期":"20240316"},"")print(ret1)print(ret2)print(ret3)print(ret4)

运行效果如下：

问题看似好像解决了，那么又引入了新的问题，示例只有3个入参，如果入参扩大到成百上千个，那还要写几百几千个if来一一判断吗，好像不太现实呢，效率也很低。这个时候就考虑引入for循环，一旦发现为空则跳出循环，否则进行赋值处理

def sgdDemo2(reqmsg, resmsg):ser = []for i in reqmsg:if(reqmsg[i] != ""):sql = i + "='" +reqmsg[i]+"'"ser.append(sql)ret = ' and '.join(ser)return retif __name__ == '__main__':ret1 = sgdDemo2({"店名": "水果店", "水果名称": "", "进货日期": ""}, "")ret2 = sgdDemo2({"店名": "水果店", "水果名称": "西瓜", "进货日期": ""}, "")ret3 = sgdDemo2({"店名": "水果店", "水果名称": "", "进货日期": "20240316"}, "")ret4 = sgdDemo2({"店名": "水果店", "水果名称": "西瓜", "进货日期": "20240316"}, "")ret5 = sgdDemo2({"店名": "水果店", "水果名称": "西瓜", "进货日期": "20240316","1":"1","2":"2","3":"3","4":"4","5":"5","6":"6","7":"7","8":"8","9":"9","10":"10","11":"11","12":"12",}, "")print(ret1)print(ret2)print(ret3)print(ret4)print(ret5)