客户名单附加索引
我正在 Access 2003 中构建客户和案例数据库。父客户表包含客户姓名、公司（如果有）、状态（值列表）、附加注释（文本或备忘录类型）和 Y/ 的字段。 N 我们的邮件列表的切换框。 PK 是递增的自动编号。根据其他地方的线程的标准化提示，地址和电话号码将获得自己的表格，如下所示：
tbl_Client
tbl_Address
tbl_AddressType
tbl_Phone
tbl_PhoneType
（tbl_Client将链接到tbl_Address和 tbl_Phone 通过联结表，但我不知道如何表示。）
我的研究表明表应该至少有两个索引的、非重复的列——一个 PK 和一个额外的唯一列场地。 tbl_Client 中唯一能保证唯一的列是备注字段——姓氏可能会再次出现，而且同一家公司可能有多个客户。
所以这是我的问题：对 Note 字段建立索引、不允许重复，还是只接受 ClientID AutoNumber 将是唯一的唯一值，这两种方法是否更好？
（当然，这个问题是建立在上述整个结构没有无可挽回的缺陷的假设之上的。如果是这样，请随意表达你的恐惧。）
I'm building a client and case database in Access 2003. The parent client table has fields for the clients' names, company (if any), status (value list) an additional note (text or memo type), and a Y/N toggle box for our mailing list. The PK is in incrementing AutoNumber. Following a normalization tip from a thread elsewhere, addresses and phone numbers will get their own tables, like so:
tbl_Client
tbl_Address
tbl_AddressType
tbl_Phone
tbl_PhoneType
(tbl_Client will link to tbl_Address and tbl_Phone via junction tables, but I don't know how to represent that.)
My research suggests that tables should have at least two indexed, non-duplicated columns -- one PK, and one additional unique field. The only other guaranteed unique column in tbl_Client is the memo field -- last names may reappear and we may have more than one client at a given company.
So here's my question: Is it better to index the Note field, disallowing duplicates, or to just accept that the ClientID AutoNumber will be the only unique value?
(Of course, this question is predicated on the assumption that the entire above structure is not irredeemably flawed. If that's the case, please feel free to express your horror.)
如果你对这篇内容有疑问，欢迎到本站社区发帖提问 参与讨论，获取更多帮助，或者扫码二维码加入 Web 技术交流群。
绑定邮箱获取回复消息
由于您还没有绑定你的真实邮箱，如果其他用户或者作者回复了您的评论，将不能在第一时间通知您！
发布评论
评论（1）
我的研究表明表应该至少有两个索引的、非重复的列 - 一个 PK 和一个附加的唯一字段。
您的 PK 应该唯一地标识每一行。我不明白第二个独特字段的要求。
* tbl_Client 中唯一有保证的唯一列是备注字段——姓氏可能会再次出现，并且我们在给定公司可能有多个客户。*
不要仅仅为了拥有第二个唯一字段而对备注字段建立索引。
My research suggests that tables should have at least two indexed, non-duplicated columns -- one PK, and one additional unique field.
Your PK should uniquely identify each row. I don't understand the requirement for a second unique field.
*The only other guaranteed unique column in tbl_Client is the memo field -- last names may reappear and we may have more than one client at a given company.*
Don't index a memo field just so you can have a second unique field.