在创建“精度浮动"类型的表列时列被创建为“真实的"键入 SQL 服务器 [英] On creating a table column of type "float with precision" column gets created as "real" type in SQL server

问题描述

我创建了一个简单的表格，如下所示:

I created a simple table as below:

create table table1
(
col1 float(1),
col2 float
)

当我使用 sp_help 或脚本表作为创建检查表定义时，我得到的 float(1) 是真实的

When I check the table definition using sp_help or script table as create, I get the float(1) as real

CREATE TABLE [dbo].[table1](
[col1] [real] NULL,
[col2] [float] NULL
)

根据我对 msdn 和 stackoverflow 引用，float 的精度 n 可以从 1 到 53.

As per my understanding from msdn and stackoverflow references, float can have a precision n ranging from 1 to 53.

为什么当 float 指定了精度时，默认情况下它会转换为 real?

Why is it getting converted to real by default whenever float has a precision specified?

推荐答案

SQL Server 不记得您创建表时使用的语言 - 它只是查询当前表定义并呈现一段文本可以用于创建相同的表.

SQL Server doesn't remember what language you used when you create the table - it just consults the current table definition and renders a piece of text that could have been used to create an identical table.

正如 文档 指出的那样，SQL Server 没有支持任意精度 - 它只支持两个精度 - 24 和 53:

As the documentation points out, SQL Server doesn't support arbitrary precisions - it only supports two precisions - 24 and 53:

SQL Server 将 n 视为两个可能值之一.如果 1<=n<=24，n 被视为 24.如果 25<=n<=53，n 被视为 53.

SQL Server treats n as one of two possible values. If 1<=n<=24, n is treated as 24. If 25<=n<=53, n is treated as 53.

正如它还指出的那样，real 被视为 float(24) 的同义词:

As it also points out, real is treated as a synonym for float(24):

real 的 ISO 同义词是 float(24).

The ISO synonym for real is float(24).

因此，任何指定精度为 24 或更低的浮点列实际上都将创建为 float(24).float(24) 的同义词是 real.系统不记得它最初是指定为 float(1)、float(24) 还是 real.

So, any float column specified with a precision of 24 or lower will in fact be created as a float(24). And a synonym for float(24) is real. The system doesn't remember whether it was originally specified as float(1), float(24) or real.

为什么当 float 指定了精度时，默认情况下它会转换为 real?

Why is it getting converted to real by default whenever float has a precision specified?

它没有.如上所述，如果您创建一个类型为 float(25)(或任何更高精度)的列，您会发现它以普通的 float 形式返回，而不是real，因为它是作为 float(53) 创建的，当没有提供时，53 是默认精度.

It doesn't. As per the above, if you create a column with type float(25) (or any higher precision), you'll find it gets returned as plain float, rather than real, because it was created as a float(53) and 53 is the default precision when none is supplied.

这篇关于在创建“精度浮动"类型的表列时列被创建为“真实的"键入 SQL 服务器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！