使用SQLAlchemy Integer字段创建timedelta对象以进行过滤 [英] Using a SQLAlchemy Integer field to create a timedelta object for filtering

问题描述

我有一个名为Item的对象.它具有2个字段:日期时间行"created_on"和整数行"days".

I have an object called Item. It has 2 fields: a datetime row called "created_on," and an integer row called "days."

我想查询许多天前创建的所有对象.

I want to query for all objects that were created "days" many days ago.

这是我认为应该完成的方式:

Here's how I thought it should be done:

now = utcnow()
session.query(Item).filter(Item.created_on + Interval(timedelta(days=Item.days)) <= now)

但是我无法创建这样的时间增量.我收到此错误:

But I'm unable to create a timedelta like this. I'm getting this error:

TypeError: unsupported type for timedelta minutes component: InstrumentedAttribute

更新:

多亏了van，我应该使用内置函数.我正在使用Mysql 5.1，因此它是timestampadd.我的新查询是这样:

Thanks to van, I should be using a built-in function. I'm using Mysql 5.1, so it would be timestampadd. My new query is this:

now = utcnow()
session.query(Item).filter(func.timestampadd('DAY', Item.days, Item.created_on) <= now)

但是，我遇到的新错误是:

However, the new error I'm getting is:

You have an error in your SQL syntax; check the manual that corresponds to your MySQL server version for the right syntax to use near ''DAY', items.days, items.created_on) <= '2015-07-12 ' at line 3

推荐答案

我认为Interval不能为您提供帮助，因为它无法强制使用特定于DB的date [time]函数.因此，您可能应该使用平台特定的date [time]函数来解决此问题.

I do not think that Interval will help you there, as it would not be able to coerce into DB-specific date[time] functions. So you should probably use the platform specific date[time] functions to solve this.

如果您使用的是 postgresql ，则下面的代码应该可以工作(假设Items.days是整数):

If you are using postgresql, the code below should work (assuming Items.days is integer):

q = (session.query(Item)
     .filter(func.age(now, Item.created_on) <=
             func.make_interval(0, 0, 0, Item.days)
             )
     )

请参见postgres' 日期/时间函数和运算符了解更多信息.

See postgres' Date/Time Functions and Operators for more information.

这篇关于使用SQLAlchemy Integer字段创建timedelta对象以进行过滤的文章就介绍到这了，希望我们推荐的答案对大家有所帮助，也希望大家多多支持IT屋！