Möhring, René wrote:
>>Turns out, this is extremely unlikely to occur.
>
> But isn't that what exceptions are made for? If it wasn't unlikely
> to happen it would not be an exception.
You're arguing theory against reality. Are you seriously asking me to
break the ABI, just to cover a case where the MySQL database is -- I
would argue -- deliberately misconfigured?
Maybe we should mandate that wheels be welded onto car axles because
someone could remove all the lug nuts but one. It could happen!
> There could also be breakdown of the server just in that tiny moment.
Yes, and there could also be a microcomet that comes through my roof,
crushing my 'e' key, so I can't type my password.
If you want to continue that line, let's talk probabilities, and balance
that against return.
>>I've withdrawn that change because it breaks the ABI, pointlessly.
>
> Pointless depends onyour viewpoint, wether you want to get your library
> correct or just nearly correct.
The feature is slated for a future revision, where breaking the ABI is
more acceptable. If that's not soon enough, give me a rational reason
why not.
| Thread |
|---|
| • Re: AW: Minor problem with the resetdb | Warren Young | 12 Mar |