Re: Column naming convention good or bad?

From: <MSherrill_at_compuserve.com>
Date: Wed, 22 Nov 2000 17:44:55 GMT
Message-ID: <3a1bf2cb.5724401_at_news.compuserve.com>


On Sun, 19 Nov 2000 06:18:14 GMT, genew_at_shuswap.net (Gene Wirchenko) wrote:

>MSherrill_at_compuserve.com wrote:
>
>>On Thu, 16 Nov 2000 20:34:39 GMT, spamfilter_at_rosinowski.de (Jan-Erik
>>Rosinowski) wrote:
>>
>>>if you change a type from int to long (i.e. i32->i64) your clientsw
>>>will break anyway if you don't change it
>
> Yes.

It's probably more accurate to say that client software that's designed <cough> to break when an underlying data type changes from integer to long will break when an underlying data type changes from integer to long. Client software that's designed to withstand such a change will withstand such a change.

>>"Select EmployeeID From Employees;" breaks if you change EmployeeID
>>from int to long?
>
> Possible. C has different format specifications for these two.

The SQL statement won't break if you change the data type. (There's a lesson there. There there.) Is your point that no client software written in C can survive changing a column type from integer to long integer on a server database?

Think about abstract data types. Think about a browser. Think about perl's DBI module. Think about data independence.

-- 
Mike Sherrill
Information Management Systems
Received on Wed Nov 22 2000 - 18:44:55 CET

Original text of this message