RetroBBS - DOVE-Net.Synchronet_Discussion - Re: atexit compilation errors?

Re: atexit compilation errors?

<64F6C2B5.50355.sync@vert.synchro.net>

https://www.rocksolidbbs.com/dovenet/article-flat.php?id=1428&group=DOVE-Net.Synchronet_Discussion#1428

copy link Newsgroups: DOVE-Net.Synchronet_Discussion

From: fusion@VERT/CFBBS (fusion)
To: Ryan Fantus
Subject: Re: atexit compilation errors?
Message-ID: <64F6C2B5.50355.sync@vert.synchro.net>
Date: Tue, 5 Sep 2023 01:54:00 +0000
X-Comment-To: Ryan Fantus
Path: rocksolidbbs.com!not-for-mail
Newsgroups: DOVE-Net.Synchronet_Discussion
X-FTN-PID: Synchronet 3.20a-Linux master/3e1aa12ec Sep 2 2023 GCC 12.2.0
X-FTN-MSGID: 50355.sync@1:103/705 295cbd68
X-FTN-CHRS: ASCII 1
WhenImported: 20230904225501-0700 c1e0
WhenExported: 20230905021451-0700 c1e0
ExportedFrom: VERT sync 50355
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 8bit

by: fusion - Tue, 5 Sep 2023 01:54 UTC

On 04 Sep 2023, Ryan Fantus said the following...
RF> BTW, how necessary is calling 'atexit' in OpenDoors? I commented out the
RF> line and compiled without `-shared` and it compiled fine. Wondering if
RF> it's truly needed, especially for linux x64?

atexit is a way to create a stack of functions that are called when a program
exits.

my guess is since it's a door library (and they don't necessarily control the program flow, the door programmed with it does) they want to ensure if the door game just abruptly decides to do exit(), odoors can at least clean up after itself.

the other option would be to make the door programmer responsible for it.. they'd call the atexit()-ed functions themselves when they're finishing up.

--- Mystic BBS v1.12 A47 2021/12/25 (Windows/32)
* Origin: cold fusion - cfbbs.net - grand rapids, mi

Subject	Author
atexit compilation errors?	fusion

micro: Thinker toys.

dovenet / Synchronet Discussion / Re: atexit compilation errors?