Search Postgresql Archives

Re: Enforce primary key on every table during dev?

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



On 03/01/2018 02:32 PM, marcelo wrote:


On 01/03/2018 17:21 , Ron Johnson wrote:
On 03/01/2018 02:08 PM, marcelo wrote:


On 01/03/2018 16:42 , Ron Johnson wrote:
On 03/01/2018 01:11 PM, marcelo wrote:

On 01/03/2018 16:00 , Ron Johnson wrote:
[snip]
If your only unique index is a synthetic key, then you can insert the same "business data" multiple times with different synthetic keys.


--
Angular momentum makes the world go 'round.
IMHO, business logic can and must preclude "garbage insertion". Except you are inserting data directly to database using SQL, any n-tier architecture will be checking data validity.
bl
Any n-tier architecture that's bug-free.

Do you know about unit testing?

Way Back When Dinosaurs Still Roamed The Earth and I first learned the trade, the focus was on proper design instead of throwing crud against the wall and hoping tests caught any bugs.  Because, of course, unit tests are only as good as you imagination in devising tests.


So, you are fully convinced that there´s no bug free software... Same as I (and you) can code following the business rules, you (and me) can design unit tests not from "imagination" but from same rules. Moreover: you can have a surrogate key (to speedup foreign keys) and simultaneously put a unique constraint on the columns requiring it. What´s the question?

Implementing tests to cover edge cases is much harder than implementing business rules in natural (and foreign) keys.


--
Angular momentum makes the world go 'round.




[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
[Index of Archives]     [Postgresql Jobs]     [Postgresql Admin]     [Postgresql Performance]     [Linux Clusters]     [PHP Home]     [PHP on Windows]     [Kernel Newbies]     [PHP Classes]     [PHP Books]     [PHP Databases]     [Postgresql & PHP]     [Yosemite]

  Powered by Linux