PATCH: pws-24: Stop the unbearable beeping of magic-space

PATCH: pws-24: Stop the unbearable beeping of magic-space

[Bart Schaefer]

I should have noticed this when 6814 was posted, but I was preoccupied with
testing job control in 3.0.6 ...

If you have

	bindkey " " magic-space

then every time you press the spacebar you get a beep, unless there happens
to be an expandable history item in the word.

Now, the really good thing would be for doexpandhist() to return failure
only if there is a history reference but it can't be expanded, rather than
failing whenever no expansion took place.  It's beyond my comprehension of
the interaction between zle and the history mechanism to pull that off, so
I settled for making magic-space fail only if selfinsert() does [the old
code didn't make sense anyway -- why store the return from selfinsert and
then do nothing with it?].

Line numbers below may be off a bit, because I also got annoyed by Oliver
Kiddle's patch from 6772.  Gosh, I actually went one whole revision of zsh
without having to either put in or maintain a local change ...

Index: Src/Zle/zle_tricky.c
@@ -8249,8 +8243,9 @@
 {
     int ret;
     c = ' ';
-    ret = selfinsert(args);
-    return !doexpandhist();
+    if (!(ret = selfinsert(args)))
+	doexpandhist();
+    return ret;
 }
 
 /**/

-- 
Bart Schaefer                                 Brass Lantern Enterprises
http://www.well.com/user/barts              http://www.brasslantern.com

Re: PATCH: pws-24: Stop the unbearable beeping of magic-space

[Peter Stephenson]

"Bart Schaefer" wrote:
> Line numbers below may be off a bit, because I also got annoyed by Oliver
> Kiddle's patch from 6772.  Gosh, I actually went one whole revision of zsh
> without having to either put in or maintain a local change ...

Is there a suggestion for resolving this?  I put it in because there were
no complaints.  I use a different job completion function anyway.  That's
the solution for somebody, but what should be the default?  Anyone want to
write a patch to return the shortest unique string of complete words?

-- 
Peter Stephenson <pws@ibmth.df.unipi.it>       Tel: +39 050 844536
WWW:  http://www.ifh.de/~pws/
Dipartimento di Fisica, Via Buonarroti 2, 56127 Pisa, Italy

Re: PATCH: pws-24: Stop the unbearable beeping of magic-space

[Bart Schaefer]

On Jun 27,  3:25pm, Peter Stephenson wrote:
} Subject: Re: PATCH: pws-24: Stop the unbearable beeping of magic-space
}
} "Bart Schaefer" wrote:
} > I also got annoyed by Oliver Kiddle's patch from 6772.
} 
} Is there a suggestion for resolving this?

Oh, just do what Zefram and Zoltan always did and ignore me.  I'm probably
being excessively picky; and I probably wouldn't even have noticed if I
hadn't been debugging job control stuff, because most of the time I don't
have more than one stopped job anyway (it's just too easy to open and
close another xterm).

-- 
Bart Schaefer                                 Brass Lantern Enterprises
http://www.well.com/user/barts              http://www.brasslantern.com

job completion (was Re: PATCH: pws-24: Stop the unbearable beeping of magic-space)

[Oliver Kiddle]

Peter Stephenson wrote:
> 
> "Bart Schaefer" wrote:
> > Line numbers below may be off a bit, because I also got annoyed by Oliver
> > Kiddle's patch from 6772.  Gosh, I actually went one whole revision of zsh
> > without having to either put in or maintain a local change ...

I'm guessing that you don't like it because the full command is long and
looks messy when quoted and listed?

> Is there a suggestion for resolving this?  I put it in because there were
> no complaints.  I use a different job completion function anyway.  That's
> the solution for somebody, but what should be the default?  Anyone want to
> write a patch to return the shortest unique string of complete words?

I used it all last week and it was really useful to me so I'd prefer if
it stays. I'm not convinced that using the shortest unique string would
be the best approach. What might be better would be to treat completion
of the first word (i.e. the command name) as separate to completion of
the remaining arguments so initially only the commands are listed but
they are completed without a suffix and you get the full arguments if
you press tab again.

I can probably implement this for the new style completion and this
would be helped by 6772 remaining so I wouldn't have to use $(jobs). I
don't have much time at the moment so I'll wait for any comments or
alternative suggestions. If many people don't like 6772 then we probably
want a solution that isn't dependant on new style completion.

Oliver Kiddle