>>>> "Craig" == Craig Lanning
<CraigL(a)DyCon.com> writes:
Craig> At 01:44 PM 10/21/99 +0200, Adrian Aichner wrote:
>>>>>> "APA" == Adrian Aichner
<aichner(a)ecf.teradyne.com> writes:
>
> Hello All,
>
APA> It works nicely on NT, but 95/98 needs investigation.
APA> I'll try to work this out with Craig Lanning.
>
> could any of you with access to Windows95 or Windows98 please try
> following patch against 21.2-b19, please?
Craig> It still doesn't work.
> I'll try to make a patch in time for 21.1.8 then.
>
> Please send me back any output XEmacs produced on STDERR (it will go
> to the DOS box, if you start XEmacs from there).
Craig> That output is shown right after the (call-process ...) call.
> Please report output and errors for these test-cases:
>
> Here is what I get on Windows NT:
>
> (call-process "cmd" nil '(t t) nil)
> Microsoft(R) Windows NT(TM)
> (C) Copyright 1985-1996 Microsoft Corp.
Craig> Win98 doesn't have "cmd".
Sorry, I wasn't thinking.
> d:\Users\AichnerAd\tmp\J750>
> (call-process "command" nil '(t t) nil)
> Spawning child process: Exec format error
Craig> I get the same error.
This is documented in
nt\PROBLEMS
BTW: Is "minitar.exe" a 16bit application?
> (call-process "dir" nil '(t t) nil
"c:\\")
> AUTOEXEC.BAT READER5 cygnus notes
> Acrobat3 RECYCLER enscript ntldr
> CONFIG.SYS Real ffastun.ffa pagefile.sys
> I386 TEMP ffastun.ffl pali.pl
> IO.SYS WINNT ffastun.ffo perl
> MSDOS.SYS XEmacs ffastun0.ffx perl.old
> Multimedia\ Files Xemacs.zip gcisdn telnet.log
>
NTDETECT.COM autosave gnuserv-2_1p1 tmp
> NTRESKIT boot.ini gstools
> Program\ Files cvs mks
Craig> "dir" is a built-in command of the "command" shell.
Craig> Therefore, this gives the same error as "command" above.
Craig> I then tried an "ls" command that I have:
Craig> (call-process "ls" nil '(t t) nil "c:\\")
Craig> create_child: exe=c:\BIN32\ls.exe,cmdline=c:\BIN32\ls.exe
Craig> create_child.CreateProcess leads child PID -725391, changing to PID 725391
Craig> Fatal error: assertion failed, file sysdep.c, line 353, pid != NULL
Craig> saw no output and XEmacs locked up.
Ah, this is what I was looking for.
The problem must occur in the maipulation of cp->pid afterwards.
I'll work on another patch.
Regards,
Adrian
Craig> Craig