Re: PATCH: Re: History bug (Re: Completion debugging)

[Sven Wischnowsky <wischnow@xxxxxxxxxxxxxxxxxxxxxxx>]

Bart Schaefer wrote:

> On May 3,  9:07am, Sven Wischnowsky wrote:
> } Subject: PATCH: Re: History bug (Re: Completion debugging)
> }
> } > Why are there two of numbers 24, 29 and 32?  In each case, the first of the
> } > two was inserted by calling "print -s ..." during completion.  Apparently
> } > that doesn't work very well.
> } 
> } Of course this isn't only in completion, but in every widget.
> } 
> } The problem is that the history number for the currently edited line
> } is `reserved' and the `print -s' makes it be used. The patch below is
> } the simplest solution I can think of.
> 
> OOOooh, this is fun.
> 
> Change the "print -zR" in _complete_debug to "print -sR", and then:
> 
> zagzig[112] ls <C-x?>
> Debugging output left in /tmp/zsh13117ls3
> (listing omitted)
> zagzig[112] ls <C-n>
> zagzig[112] emacs /tmp/zsh13117ls3 ;: "ls -CF "<C-p>
> zagzig[112] ls <wait a few minutes, then RET>
> (listing omitted)
> zagzig[114] history -d
>   113  09:48  emacs /tmp/zsh13117ls3 ;: "ls -CF "
>   113  09:51  ls
> zagzig[115] 
> 
> So there are still two 113s in the history list, but the prompt has the
> right history number.  I'm expecting that particular shell to crash any
> time now ...

I get a SEGV reproducibly after C-p C-n. If I take out my patch for
this, I get it after the C-p.

Hm, maybe someone more knowledgeable with history stuff...?

Bye
 Sven


--
Sven Wischnowsky                         wischnow@xxxxxxxxxxxxxxxxxxxxxxx