GNU bug report logs - #6800
23.1; EOT / ^D inserted into comint input string

Previous Next

Full log

View this message in rfc822 format

From: help-debbugs <at> gnu.org (GNU bug Tracking System)
To: David Fox <ddssff <at> gmail.com>
Cc: tracker <at> debbugs.gnu.org
Subject: bug#6800: closed (23.1; EOT / ^D inserted into comint input string)
Date: Fri, 12 Aug 2016 19:53:01 +0000

[Message part 1 (text/plain, inline)]

Your message dated Fri, 12 Aug 2016 11:40:03 -0700
with message-id <CAJN_RF4RQtWdP_E-yycZC4NQXzNU4zPKocg10ZYrKXcYm_snEw <at> mail.gmail.com>
and subject line Re: bug#6800: 23.1; EOT / ^D inserted into comint input string
has caused the debbugs.gnu.org bug report #6800,
regarding 23.1; EOT / ^D inserted into comint input string
to be marked as done.

(If you believe you have received this mail in error, please contact
help-debbugs <at> gnu.org.)


-- 
6800: http://debbugs.gnu.org/cgi/bugreport.cgi?bug=6800
GNU Bug Tracking System
Contact help-debbugs <at> gnu.org with problems

[Message part 2 (message/rfc822, inline)]

From: David Fox <ddssff <at> gmail.com>
To: bug-gnu-emacs <at> gnu.org
Subject: 23.1; EOT / ^D inserted into comint input string
Date: Thu, 5 Aug 2010 07:58:13 -0700

[Message part 3 (text/plain, inline)]

When an input longer than 255 characters is typed into comint (or
shell) an EOT character (ascii 4, ^D) is inserted into the string.
This can cause an error depending how the sub process handles these
extra characters.  For example, GHC doesn't like it when an EOT
appears inside of a string:

 ghci
 > Prelude> putStrLn "<a 241 character string>"
works fine, but
 > Prelude> putStrLn "<a 242 character string>"
<interactive>:1:255: lexical error at character '\EOT'

I inserted a function to break up the input into comint-send-string to work
around the problem:

(require 'comint)

(defun comint-send-string (process string)
  "Like `process-send-string', but also does extra bookkeeping for Comint
mode."
  (if process
      (with-current-buffer (if (processp process)
                   (process-buffer process)
                 (get-buffer process))
    (comint-snapshot-last-prompt))
    (comint-snapshot-last-prompt))
  (my-process-send-string process string))

;; Break up the string so that we don't get EOT characters in our input
stream.
(defun my-process-send-string (process string)
  (if (> (length string) 200)
      (progn (process-send-string process (substring string 0 200))
(my-process-send-string process (substring string 200)))
    (process-send-string process string)))

[Message part 4 (text/html, inline)]

[Message part 5 (message/rfc822, inline)]

From: David Fox <ddssff <at> gmail.com>
To: Alan Third <alan <at> idiocy.org>
Cc: 6800-done <at> debbugs.gnu.org, Stefan Monnier <monnier <at> iro.umontreal.ca>
Subject: Re: bug#6800: 23.1; EOT / ^D inserted into comint input string
Date: Fri, 12 Aug 2016 11:40:03 -0700

[Message part 6 (text/plain, inline)]

On Fri, Aug 12, 2016 at 11:36 AM, Alan Third <alan <at> idiocy.org> wrote:

> Stefan Monnier <monnier <at> iro.umontreal.ca> writes:
>
> >> When an input longer than 255 characters is typed into comint (or
> >> shell) an EOT character (ascii 4, ^D) is inserted into the string.
> >> This can cause an error depending how the sub process handles these
> >> extra characters.  For example, GHC doesn't like it when an EOT
> >> appears inside of a string:
> >
> > I believe this bug is fixed on the Emacs trunk (to become Emacs-24), but
> > the fix may introduce other problems.  Apparently nobody knows how to
> > fix it right and there is about as much as no documentation at all on
> > this part of the behavior of ttys.
>
> IIRC this code is no longer in Emacs. I'll close this bug.
>
> --
> Alan Third
>

I do know that commands longer than 4096 characters are truncated in emacs
24.3.1

[Message part 7 (text/html, inline)]

Previous Next

GNU bug tracking system
Copyright (C) 1999 Darren O. Benham, 1997,2003 nCipher Corporation Ltd, 1994-97 Ian Jackson.