PATCH: 3.1.5: bad handling of 8 bit character

PATCH: 3.1.5: bad handling of 8 bit character

[Peter Stephenson]

Marc Baudoin wrote:
> Hi,
> 
> I've upgraded from zsh 3.0.5 to try zsh 3.1.5 and I noticed two
> problems.
> 
> The new zsh seems to have problems with files whose name contains
> non ASCII characters:

I suspect this was my fault last time I fiddled with input.c.  It only
shows up on machines where char is signed, so I didn't notice it.  At
least, this seems to do the trick for SUNOS, where I was seeing the
problem.

*** Src/input.c.lastc	Sat Sep 26 19:02:15 1998
--- Src/input.c	Thu Nov 12 14:18:23 1998
***************
*** 176,182 ****
  int
  ingetc(void)
  {
!     char lastc;
  
      if (lexstop)
  	return ' ';
--- 176,182 ----
  int
  ingetc(void)
  {
!     int lastc;
  
      if (lexstop)
  	return ' ';

-- 
Peter Stephenson <pws@ibmth.df.unipi.it>       Tel: +39 050 844536
WWW:  http://www.ifh.de/~pws/
Dipartimento di Fisica, Via Buonarotti 2, 56100 Pisa, Italy

Re: PATCH: 3.1.5: bad handling of 8 bit character

[Marc Baudoin]

Peter Stephenson <pws@ibmth.df.unipi.it> écrit :
> Marc Baudoin wrote:
> > 
> > I've upgraded from zsh 3.0.5 to try zsh 3.1.5 and I noticed two
> > problems.
> > 
> > The new zsh seems to have problems with files whose name contains
> > non ASCII characters:
> 
> I suspect this was my fault last time I fiddled with input.c.  It only
> shows up on machines where char is signed, so I didn't notice it.  At
> least, this seems to do the trick for SUNOS, where I was seeing the
> problem.

Looks like it!

Thanks.

-- 
Marc Baudoin   -=-   <babafou@babafou.eu.org>

Re: PATCH: 3.1.5: bad handling of 8 bit character

[Bart Schaefer]

On Nov 12,  2:12pm, Peter Stephenson wrote:
} Subject: PATCH: 3.1.5: bad handling of 8 bit character
}
} I suspect this was my fault last time I fiddled with input.c.  It only
} shows up on machines where char is signed, so I didn't notice it.  At
} least, this seems to do the trick for SUNOS, where I was seeing the
} problem.
} 
}   int
}   ingetc(void)
}   {
} !     int lastc;
}   
}       if (lexstop)
}   	return ' ';

I'm confused.  I have this patch applied to 3.0.5:

 PWS's patch from zsh-workers 4172 to eliminate the `lastc' global
 and thereby clean up some goofy history management and a couple of
 unexpected exits.

That means that my 3.0.5-extended has `char lastc;' in ingetc() as well.
Yet 3.0.5-extended still manages to get the 8-bit characters right, even
though 3.1.5 built on the same machine gets it wrong.

What *else* changed in 3.1.5 to make it sensitive to this?

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

Re: PATCH: 3.1.5: bad handling of 8 bit character

[Bart Schaefer]

On Nov 12,  8:55am, Bart Schaefer wrote:
} Subject: Re: PATCH: 3.1.5: bad handling of 8 bit character
}
} On Nov 12,  2:12pm, Peter Stephenson wrote:
} } Subject: PATCH: 3.1.5: bad handling of 8 bit character
} }
} } I suspect this was my fault last time I fiddled with input.c.
} 
} I'm confused.  I have this patch applied to 3.0.5:
} 
}  PWS's patch from zsh-workers 4172 to eliminate the `lastc' global

My mistake:  The version of 3.0.5 where I tried to reproduce the bug
was not the same one that had the above-mentioned patch.

The char --> int change in input.c is indeed necessary and sufficient.

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