| Subject | Re: Re[6]: [ib-support] Recomended component for FireBird & Delphi ? |
|---|---|
| Author | Woody (TMW) |
| Post date | 2003-02-26T16:30:13Z |
Sorry, HeLen, this is becoming quite off-topic but I don't subscribe to the
other groups so I'll answer one last time here.
something that is needed versus something that is desired. In place
inserting is not a practicality of CS development since records are not
maintained in any particular physical order. I remember that discussion and
IIRC, Jeff's answer was that implementing this is
a.) detrimental to prior code which has always counted on inserted records
being placed at the end of the current set of data as it always has.
b.) more of a personal opinion of how it should work.
c.) has a simple workaraound involving rereading (closing/opening) the data
set.
You can argue about the merits of suggestions like these, but ultimately,
they are convenience type modifications that are almost always left up to
the developer. I have many suggestions for my own components but not all of
them appeal to me or are worth the effort to implement. That's just the way
it is.
And now, sadly, I relinquish control to the very lovely hat lady, HeLen, who
is surely to be ready to press the re-direction key at any time. <g>
Woody (TMW)
"Never trust a computer you can't throw out a window."
-Steve Wozniak
other groups so I'll answer one last time here.
>the
> WT> I shouldn't really be commenting but I feel that I must. <g> One of
> WT> issues about getting changes into IBX is that Borland now owns itsomething
> Yes, I absolutely agree with you. Jeff can not make a decision
> himself for most cases. However, sometimes if he does not make
> it is not only a Borland point of view :).Agreed. I'm not sure how much "control" Jeff has over them, though.
>always
> WT> that only help a select few. Bug fixes, errors and the like, are
> WT> considered and mostly implemented.Ah, you happen to point out a suggestion that is a prime example of
> I do not know all cases but for example the bug with Insert method
> (which works as Append) still exists in IBX. And this bug exists
> in IBX since FreeIBComponents :). Jeff does not want to fix it at all.
> I think some of ex-IBX-users can give you much more examples.
something that is needed versus something that is desired. In place
inserting is not a practicality of CS development since records are not
maintained in any particular physical order. I remember that discussion and
IIRC, Jeff's answer was that implementing this is
a.) detrimental to prior code which has always counted on inserted records
being placed at the end of the current set of data as it always has.
b.) more of a personal opinion of how it should work.
c.) has a simple workaraound involving rereading (closing/opening) the data
set.
You can argue about the merits of suggestions like these, but ultimately,
they are convenience type modifications that are almost always left up to
the developer. I have many suggestions for my own components but not all of
them appeal to me or are worth the effort to implement. That's just the way
it is.
And now, sadly, I relinquish control to the very lovely hat lady, HeLen, who
is surely to be ready to press the re-direction key at any time. <g>
Woody (TMW)
"Never trust a computer you can't throw out a window."
-Steve Wozniak