我知道这有点晚了，这个问题已经得到了很好的回答，但我想就#3关于列名前缀的问题提出我的看法。

所有列都应该使用一个对定义它们的表唯一的前缀命名。

例如，给定表“customer”和“address”，让我们分别使用前缀“cust”和“addr”。"customer"中会有"cust_id"， "cust_name"等。“address”将包含“addr_id”，“addr_cust_id”(FK返回给客户)，“addr_street”等。

当我第一次看到这个标准时，我坚决反对它;我讨厌这个主意。我无法忍受所有额外的输入和冗余。现在我已经有了足够的经验，我再也不会回去了。

这样做的结果是数据库模式中的所有列都是唯一的。这有一个主要的好处，它压倒了所有反对它的论点(当然，在我看来):

您可以搜索整个代码库，并可靠地找到涉及特定列的每一行代码。

The benefit from #1 is incredibly huge. I can deprecate a column and know exactly what files need to be updated before the column can safely be removed from the schema. I can change the meaning of a column and know exactly what code needs to be refactored. Or I can simply tell if data from a column is even being used in a particular portion of the system. I can't count the number of times this has turned a potentially huge project into a simple one, nor the amount of hours we've saved in development work.

另一个相对较小的好处是，当你进行自连接时，你只需要使用表别名:

SELECT cust_id, cust_name, addr_street, addr_city, addr_state
    FROM customer
        INNER JOIN address ON addr_cust_id = cust_id
    WHERE cust_name LIKE 'J%';

我知道这有点晚了，这个问题已经得到了很好的回答，但我想就#3关于列名前缀的问题提出我的看法。

所有列都应该使用一个对定义它们的表唯一的前缀命名。

例如，给定表“customer”和“address”，让我们分别使用前缀“cust”和“addr”。"customer"中会有"cust_id"， "cust_name"等。“address”将包含“addr_id”，“addr_cust_id”(FK返回给客户)，“addr_street”等。

当我第一次看到这个标准时，我坚决反对它;我讨厌这个主意。我无法忍受所有额外的输入和冗余。现在我已经有了足够的经验，我再也不会回去了。

这样做的结果是数据库模式中的所有列都是唯一的。这有一个主要的好处，它压倒了所有反对它的论点(当然，在我看来):

您可以搜索整个代码库，并可靠地找到涉及特定列的每一行代码。

The benefit from #1 is incredibly huge. I can deprecate a column and know exactly what files need to be updated before the column can safely be removed from the schema. I can change the meaning of a column and know exactly what code needs to be refactored. Or I can simply tell if data from a column is even being used in a particular portion of the system. I can't count the number of times this has turned a potentially huge project into a simple one, nor the amount of hours we've saved in development work.

另一个相对较小的好处是，当你进行自连接时，你只需要使用表别名:

SELECT cust_id, cust_name, addr_street, addr_city, addr_state
    FROM customer
        INNER JOIN address ON addr_cust_id = cust_id
    WHERE cust_name LIKE 'J%';

这里的回答有点晚，但简而言之:

复数表名:我的偏好是复数 单个列名:是的 前缀表或列:

表:*通常*没有前缀是最好的。 列:没有。

在命名项时使用任何大小写:表和列都使用PascalCase。

细化:

(1)你必须做什么。很少有事情是你每次都必须以某种方式去做的，但还是有一些。

Name your primary keys using "[singularOfTableName]ID" format. That is, whether your table name is Customer or Customers, the primary key should be CustomerID. Further, foreign keys must be named consistently in different tables. It should be legal to beat up someone who does not do this. I would submit that while defined foreign key constraints are often important, consistent foreign key naming is always important You database must have internal conventions. Even though in later sections you'll see me being very flexible, within a database naming must be very consistent . Whether your table for customers is called Customers or Customer is less important than that you do it the same way throughout the same database. And you can flip a coin to determine how to use underscores, but then you must keep using them the same way. If you don't do this, you are a bad person who should have low self-esteem.

你可能应该做的事。

Fields representing the same kind of data on different tables should be named the same. Don't have Zip on one table and ZipCode on another. To separate words in your table or column names, use PascalCasing. Using camelCasing would not be intrinsically problematic, but that's not the convention and it would look funny. I'll address underscores in a moment. (You may not use ALLCAPS as in the olden days. OBNOXIOUSTABLE.ANNOYING_COLUMN was okay in DB2 20 years ago, but not now.) Don't artifically shorten or abbreviate words. It is better for a name to be long and clear than short and confusing. Ultra-short names is a holdover from darker, more savage times. Cus_AddRef. What on earth is that? Custodial Addressee Reference? Customer Additional Refund? Custom Address Referral?

(3)你应该考虑什么。

I really think you should have plural names for tables; some think singular. Read the arguments elsewhere. Column names should be singular however. Even if you use plural table names, tables that represent combinations of other tables might be in the singular. For example, if you have a Promotions and an Items table, a table representing an item being a part of a promotion could be Promotions_Items, but it could also legitimately be Promotion_Items I think (reflecting the one-to-many relationship). Use underscores consistently and for a particular purpose. Just general tables names should be clear enough with PascalCasing; you don't need underscores to separate words. Save underscores either (a) to indicate an associative table or (b) for prefixing, which I'll address in the next bullet. Prefixing is neither good or bad. It usually is not best. In your first db or two, I would not suggest using prefixes for general thematic grouping of tables. Tables end up not fitting your categories easily, and it can actually make it harder to find tables. With experience, you can plan and apply a prefixing scheme that does more good than harm. I worked in a db once where data tables began with tbl, config tables with ctbl, views with vew, proc's sp, and udf's fn, and a few others; it was meticulously, consistently applied so it worked out okay. The only time you NEED prefixes is when you have really separate solutions that for some reason reside in the same db; prefixing them can be very helpful in grouping the tables. Prefixing is also okay for special situations, like for temporary tables that you want to stand out. Very seldom (if ever) would you want to prefix columns.

虽然很晚了，但我仍然想对列前缀发表我的意见

对于使用table_column(或tableColumn)列命名标准，似乎有两个主要的论据，都是基于列名本身在整个数据库中是唯一的这一事实:

1)你不需要一直在你的查询中指定表名和/或列别名

2)你可以很容易地在整个代码中搜索列名

我认为这两种观点都有缺陷。不使用前缀解决这两个问题很简单。以下是我的建议:

在SQL中始终使用表名。例如，总是用table。列而不是列。

它显然解决了2)你现在只需要搜索表。而不是table_column。

But I can hear you scream, how does it solve 1)? It was exactly about avoiding this. Yes, it was, but the solution was horribly flawed. Why? Well, the prefix solution boils down to: To avoid having to specify table.column when there's ambiguity, you name all your columns table_column! But this means you will from now on ALWAYS have to write the column name every time you specify a column. But if you have to do that anyways, what's the benefit over always explicitly writing table.column? Exactly, there is no benefit, it's the exact same number of characters to type.

编辑:是的，我知道用前缀命名列可以强制使用正确的用法，而我的方法依赖于程序员

好吧，既然我们有意见:

我认为表名应该是复数。表是实体的集合(表)。每一行表示一个实体，表表示集合。因此，我将Person实体表称为People(或Persons，随您喜欢)。

对于那些喜欢在查询中看到单一“实体名称”的人来说，这就是我使用表别名的原因:

SELECT person.Name
FROM People person

有点像LINQ的“from person in people select person. name”。

至于2、3和4，我同意@Lars的观点。

--Example SQL

CREATE TABLE D001_Students
(
    StudentID INTEGER CONSTRAINT nnD001_STID NOT NULL,
    ChristianName NVARCHAR(255) CONSTRAINT nnD001_CHNA NOT NULL,
    Surname NVARCHAR(255) CONSTRAINT nnD001_SURN NOT NULL,
    CONSTRAINT pkD001 PRIMARY KEY(StudentID)
);

CREATE INDEX idxD001_STID on D001_Students;

CREATE TABLE D002_Classes
(
    ClassID INTEGER CONSTRAINT nnD002_CLID NOT NULL,
    StudentID INTEGER CONSTRAINT nnD002_STID NOT NULL,
    ClassName NVARCHAR(255) CONSTRAINT nnD002_CLNA NOT NULL,
    CONSTRAINT pkD001 PRIMARY KEY(ClassID, StudentID),
    CONSTRAINT fkD001_STID FOREIGN KEY(StudentID) 
        REFERENCES D001_Students(StudentID)
);

CREATE INDEX idxD002_CLID on D002_Classes;

CREATE VIEW V001_StudentClasses
(
    SELECT
        D001.ChristianName,
        D001.Surname,
        D002.ClassName
    FROM
        D001_Students D001
            INNER JOIN
        D002_Classes D002
            ON
        D001.StudentID = D002.StudentID
);

这些是我学到的惯例，但是您应该适应您的开发软管使用的任何东西。

复数。它是实体的集合。 是的。属性是一个实体的单一属性的表示。 是的，前缀表名允许轻松跟踪所有约束索引和表别名的命名。 表和列名用大小写，索引和约束用前缀+ ALL大写。