lynx-dev
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [2.8.1rel.2] lineedit cleanup/enhancement - take 3 (was: Re: lynx-de


From: David Combs
Subject: Re: [2.8.1rel.2] lineedit cleanup/enhancement - take 3 (was: Re: lynx-dev [lynx2.8.1rel.2] lineedit bug fix (and enhancement))
Date: Fri, 6 Nov 1998 05:21:26 -0800 (PST)

> From address@hidden Tue Nov  3 17:10:09 1998
> Date: Tue, 3 Nov 1998 17:06:45 -0800
> From: Kim DeVaughn <address@hidden>
> 
> On Tue, Nov 03, 1998, Bela Lubkin (address@hidden) said:
> |
> | Kim DeVaughn wrote:
> | >
> | > The attached patch now removes the redundent LYE_DELC case code from
> | > LYStrings.c, and add the new LYE_DELEOL case (which deletes characters
> | > from the current cursor position, thru the end-of-line).
> | >
> | > Default binding for DELEOL is ^\ (formerly a NOP).
> |
> | On almost all Unix systems, ^\ is, by default, the `quit' character.  It
> | sends SIGQUIT, which causes most programs to dump core (that's its
> | purpose).  This can be blocked by ignoring SIGQUIT.
> |
> | I find no references to SIGQUIT in the Lynx source, including in your
> | patch.  Hitting ^\ in my (un-patched) Lynx causes it to dump core, as
> | expected.
> |
> | This isn't a good default binding, as it will cause problems on a large
> | percentage of the systems that use Lynx.  Fixing it by blocking SIGQUIT
> | wouldn't be good either -- it exists for good reason (e.g. to be able to
> | generate a dump from a hung Lynx, for analysis).
> |
> | I don't object to the feature, just the default binding...
> 
> Heh.  Forgot about that usage of C-\.
> 
> FWIW, stty -a shows it bound to "quit" on this FreeBSD 2.2.7-STABLE system,
> but it seems to have no effect on lynx (nor any other app, for that matter).
> Wonder where it's getting snarfed up ...?...
> 
> 
> As may be, I rebound the default for LYE_DELEOL to ^_ (which I *hope* is a
> safe choice).
> 
> Also fixed a couple more doc references to emacs/vi keys to reflect that
> they have no effect on the line-edit bindings.
> 
> /kim
<snipped the code>

OH NO!  Not "control-backslash"!  Please!

On my suns (before, sun3/160, sunos; now sparc5, solaris) that
char does NOTHING except echo.

But not KERMIT.  I dare not even try it.

At least, "control-\ C" is kermit's "get to kermit command-level"
signal-type-thing.   Maybe control-\ itself is ok with kermit.
Surely someone knows the answer.

David

reply via email to

[Prev in Thread] Current Thread [Next in Thread]