Path: nntp.eternal-september.org!news.eternal-september.org!eternal-september.org!feeder3.eternal-september.org!weretis.net!feeder8.news.weretis.net!usenet.goja.nl.eu.org!dotsrc.org!filter.dotsrc.org!news.dotsrc.org!not-for-mail
Date: Thu, 10 Jul 2025 19:05:42 -0400
MIME-Version: 1.0
User-Agent: Mozilla Thunderbird
Subject: Re: VMS x86-64 database server
Newsgroups: comp.os.vms
References: <104ejo8$2cobv$1@dont-email.me>
 <686af3b4$0$686$14726298@news.sunsite.dk> <104f0a6$2gn2r$5@dont-email.me>
 <104f2ic$2h75q$2@dont-email.me> <104fc66$2n4ir$2@dont-email.me>
 <104h2d1$31cae$2@dont-email.me> <104hgeh$3474l$11@dont-email.me>
 <104hl76$3595d$2@dont-email.me> <104hojt$360c1$1@dont-email.me>
 <686d12d9$0$694$14726298@news.sunsite.dk> <104k487$3p7d8$8@dont-email.me>
 <104k6p0$3pmuk$1@dont-email.me> <104ka6h$3qkfr$5@dont-email.me>
 <104ki4c$3r3fl$1@dont-email.me> <104l5i4$4bv9$2@dont-email.me>
 <686ec41e$0$686$14726298@news.sunsite.dk> <104msna$fd02$9@dont-email.me>
 <686f0082$0$686$14726298@news.sunsite.dk> <104n4jb$gs8r$2@dont-email.me>
 <686f2472$0$686$14726298@news.sunsite.dk> <104nfgt$mfqt$1@dont-email.me>
Content-Language: en-US
From: =?UTF-8?Q?Arne_Vajh=C3=B8j?= <arne@vajhoej.dk>
In-Reply-To: <104nfgt$mfqt$1@dont-email.me>
Content-Type: text/plain; charset=UTF-8; format=flowed
Content-Transfer-Encoding: 8bit
Lines: 32
Message-ID: <68704746$0$690$14726298@news.sunsite.dk>
Organization: SunSITE.dk - Supporting Open source
NNTP-Posting-Host: a1b9736a.news.sunsite.dk
X-Trace: 1752188742 news.sunsite.dk 690 arne@vajhoej.dk/70.172.216.99:51349
X-Complaints-To: staff@sunsite.dk

On 7/10/2025 12:28 AM, Lawrence D'Oliveiro wrote:
> On Wed, 9 Jul 2025 22:24:50 -0400, Arne Vajhøj wrote:
>> There are fundamentally two different ways of handling errors:
>> * return status
>> * exception
> 
> There’s a difference between different statuses that might be returned as
> part of the normal execution of an operation, and unexpected conditions
> indicating program bugs.
> 
> Syntax errors in the embedded language would almost always belong in the
> latter category.

Yes.

Java operate with checked and unchecked exceptions based on
that logic.

But I don't see any point for the problem at hand.

If the PHP developer configure the database connection
to not throw exceptions because the developer want to
check return status, then that must be respected. It
does not make any sense of the database extension is
is coded like "SQL syntax is so serious a bug, so even
though the developer asked me not to throw exceptions, then
I will throw one anyway". The code is not prepared to
handle exceptions, because the developer expect the
driver to throw one when told not to.

Arne