Subject | Re: [ib-support] bug ? |
---|---|
Author | Helen Borrie |
Post date | 2002-02-12T14:14:43Z |
At 03:01 PM 12-02-02 +0100, you wrote:
It's not a "bug" in a sardine can if the user tries to open it with a chain saw and clues up with an inedible snack.
H.
All for Open and Open for All
Firebird Open SQL Database ยท http://firebirdsql.org
_______________________________________________________
>Daniel,Not really; it seems common sense if you are going to hit the system tables directly instead of by means of the facilities and checks provided by the engine and language layers, to acquaint yourself with the table structures more closely and map their dependencies.
>
>>The proper way to change it is to create a temp field with the desired
>>properties...
>
>I am puzzled.
>
>There is a politically correct way to change a field attribute ?
>
>When doing it, can one dress causally or is a necktie de-rigour ? :)
>
>If fields attributes are stored in RDB$RELATION_FIELDS, I expect to get
>logic results when I change the table.
>
>Can you quote public documents that advice against touching system tables ?
>For what I know, Interbase WorkBench, Marathon and surely QuickDesk: allModifying the system tables directly can be a back-door way to do some things, that experienced tool developers have mapped out for themselves and which the database language and tools do under the hood, with full comprehension of the dependencies.
>these tools routinely change fields in system tables in order to get quick
>results.
>
>Of course if you want to suggest a method that will work when compared to a
>method that does not work, I take that you are confirming a bug in FB ...
It's not a "bug" in a sardine can if the user tries to open it with a chain saw and clues up with an inedible snack.
H.
All for Open and Open for All
Firebird Open SQL Database ยท http://firebirdsql.org
_______________________________________________________