Hi,
I can't get my bug report to the ILISP mailing list thru SF's filters, so
I'm sending it here.
I'm running CMUCL under detachtty. I connect to it using C-u M-x over ssh.
Clients and server are all FreeBSD 4.7. Emacs is 21.2 from ports.
ILISP 5.11 works fine. ILISP 5.12 eats "(", so I get error messages like
"unknown ORANDLOAD", which is actually "(or (and (load ..." somewhere inside
ILISP.
My transcript of the error is in the stuck mail and not easily accessible.
If anyone wants to see it, let me know.
Cheers.
--
Ng Pheng Siong <····@netmemetic.com> * http://www.netmemetic.com
····@netmemetic.com (Ng Pheng Siong) wrote in message news:<············@mawar.singnet.com.sg>...
> Hi,
>
> I can't get my bug report to the ILISP mailing list thru SF's filters, so
> I'm sending it here.
Gee, my mechanic is on vacation today, so I think I will take my car
to the drycleaners, because, well, I have to take it *somewhere*. Any
action is better than none!
According to Kaz Kylheku <···@ashi.footprints.net>:
> Gee, my mechanic is on vacation today, so I think I will take my car
> to the drycleaners, because, well, I have to take it *somewhere*. Any
> action is better than none!
ossifer, issh briitah herr aaan i ken seesh bettah hicc
--
Ng Pheng Siong <····@netmemetic.com> * http://www.netmemetic.com
According to Ng Pheng Siong <····@netmemetic.com>:
> ILISP 5.11 works fine. ILISP 5.12 eats "(", so I get error messages like
> "unknown ORANDLOAD", which is actually "(or (and (load ..." somewhere inside
> ILISP.
Yes, Ossifer, the bright place ain't the right place...
Turns out it was the interaction of detachtty with FreeBSD's erase2, which
defaults to "(".
So detachtty was eating "(", not ILISP.
When debugging, I was starting up detachtty in my shell, which has the
proper settings, but for the "deployment scenario", the server auto-pilots
under daemontools in a FreeBSD jail, where erase2 is the default "(". For a
while the whole thing appeared nondeterministic. Doh!
Here's a patch to detachtty.c:
--- detachtty.c.org Thu Jan 23 15:38:35 2003
+++ detachtty.c Thu Jan 23 15:27:17 2003
@@ -233,6 +233,7 @@
/* stermios.c_oflag &= ~(ONLCR); */
/* would also turn off NL to CR/NL mapping on output */
stermios.c_cc[VERASE]=0177;
+ stermios.c_cc[VERASE2]=0177;
if (tcsetattr(fd, TCSANOW, &stermios) < 0)
bail("tcsetattr error");
}
(Marco please forward this one as well. TIA! ;-)
--
Ng Pheng Siong <····@netmemetic.com> * http://www.netmemetic.com
····@netmemetic.com (Ng Pheng Siong) writes:
> Here's a patch to detachtty.c:
[snipped]
>
> (Marco please forward this one as well. TIA! ;-)
As I'm not an ilisp developer, and only read that list irregularly,
having Marco forward it would not be too productive. Is it really the
case that you also can't send email to ···@telent.net _or_ use the
Debian bug tracking system _or_ edit pages on CLiki?
Fortunately it happens that I read c.l.l rather more often. Thanks
for the patch, which will be in detachtty 7 as soon as that happens.
-dan
--
http://www.cliki.net/ - Link farm for free CL-on-Unix resources
According to Daniel Barlow <···@telent.net>:
> As I'm not an ilisp developer, and only read that list irregularly,
> having Marco forward it would not be too productive. Is it really the
> case that you also can't send email to ···@telent.net _or_ use the
> Debian bug tracking system _or_ edit pages on CLiki?
No, just that I noticed Marco forwarded the first one, and since I
misclaimed that the problem was with ILISP, I thought I should tell the
ILISP list, too.
I was going to edit the page on CLiki. I don't use Debian, so I didn't
think of the Debian bug tracker.
> Fortunately it happens that I read c.l.l rather more often. Thanks
> for the patch, which will be in detachtty 7 as soon as that happens.
My patch wasn't conditionalised for FreeBSD, which is why I didn't
send it to you directly.
Thanks for detachtty, btw.
Cheers.
--
Ng Pheng Siong <····@netmemetic.com> * http://www.netmemetic.com