🚀 go-pugleaf

RetroBBS NetNews Server

Inspired by RockSolid Light RIP Retro Guy

Article View: gnu.emacs.bug
Article #98399

Re: shell mode bug

#98399
From: Wolfgang S. Rupprecht
Date: Tue, 30 Jul 1991 17:04
53 lines
2807 bytes
snicoud@bcsaic.UUCP (Stephen Nicoud) writes:
>In article <9107231836.AA15126@vega.Berkeley.EDU> haygood@vega.berkeley.edu (Ralph Haygood) writes:
>>When I insert (via mouse stuff, yank or process-send-region) more than 895
>>characters at a time into a shell mode buffer, the following error message
>>appears in the minibuffer
>>writing to process: no more processes, shell
>>and the shell dies.  Seemingly, 895 is a magic number; one less, and all  
>>goes well.
>>My version of GNU Emacs is marked
>>GNU Emacs 18.57.35 of Wed Jul 10 1991 on minestrone.Berkeley.EDU (berkeley-unix)
>>and my Sun 3/60 runs
>>SunOS Release 4.1.1 (GENERIC) #2 
>>I've just switched to this from  
>>Sun UNIX 4.2 Release 3.5 (GENERIC) #5 
>>and an older version of GNU Emacs marked
>>GNU Emacs 18.51.2 of Mon May  9 1988 on priam.Berkeley.EDU (berkeley-unix)
>>which doesn't have this problem. 
>I've run into a similar situation.  The main difference is that rather
>than a shell buffer, I copy stuff over to a buffer running REFINE,
>which is a system built with Franz's Allegro Common Lisp.
>I haven't verified or confirmed whether 895 is a magic number, but
>that looks about right.
>My configuration is Sun4/IPC (Sunos 4.1.1), GNU Emacs 18.57 (with
>HAVE_X_WINDOWS X11 HAVE_X_MENU).

This appears to be the same bug I used to see here.  It if is, this
patch will fix it.

*** process.c-dist	Tue Jan 15 20:28:48 1991
--- process.c	Thu Jul  4 15:18:57 1991
***************
*** 1908,1914 ****
  	if (rv < 0)
  	  {
  #ifdef EWOULDBLOCK
! 	    if (errno == EWOULDBLOCK)
  	      {
  		/* It would be nice to accept process output here,
  		   but that is difficult.  For example, it could
--- 1908,1914 ----
  	if (rv < 0)
  	  {
  #ifdef EWOULDBLOCK
! 	    if ((errno == EWOULDBLOCK) || errno == EAGAIN)
  	      {
  		/* It would be nice to accept process output here,
  		   but that is difficult.  For example, it could
***************

-wolfgang
-- 
Wolfgang Rupprecht    wolfgang@wsrcc.com (or) uunet!wsrcc!wolfgang
Snail Mail Address:   Box 6524, Alexandria, VA 22306-0524
Technical Details
Message-ID: <1991Jul30.170440.2624@wsrcc.com>
Path: arc1-ca.pugleaf.net!gmdzi!unido!mcsun!uunet!cis.ohio-state.edu!wsrcc.COM!wolfgang
References: <9107231836.AA15126@vega.Berkeley.EDU>

Show Full Headers
Complete Article Headers
X-Google-Language: ENGLISH,ASCII-7-bit
X-Google-Thread: fb2f1,65734ef34f3f2668
X-Google-Attributes: gidfb2f1,public
X-Google-ArrivalTime: 1991-07-30 10:04:40 PST
Path: arc1-ca.pugleaf.net!gmdzi!unido!mcsun!uunet!cis.ohio-state.edu!wsrcc.COM!wolfgang
From: wolfgang@wsrcc.COM (Wolfgang S. Rupprecht)
Newsgroups: gnu.emacs.bug
Subject: Re: shell mode bug
Keywords: refine, allegro common lisp, process
Message-ID: <1991Jul30.170440.2624@wsrcc.com>
Date: 30 Jul 91 17:04:40 GMT
References: <9107231836.AA15126@vega.Berkeley.EDU>
Sender: gnulists@ai.mit.edu
Distribution: gnu
Organization: Wolfgang S Rupprecht Computer Consulting, Washington DC.
Lines: 53
Approved: bug-gnu-emacs@prep.ai.mit.edu