PATCH: 3.1.6: promptexpand (Re: Weird behaviour with certain PS1-sequences)

["Bart Schaefer" <schaefer@candle.brasslantern.com>]

On Sep 28, 10:54am, Peter Stephenson wrote:
} Subject: Re: Weird behaviour with certain PS1-sequences
}
} Dan Nelson wrote:
} > % zsh -f
} > % PROMPT="%m %n %/ %%"
} > dan dan /home/dan % TERM=dumb
} > dan dan /home/dan % 
} >                     ^ blanks start here and don't stop
} 
} Yes, I can see this, and what's going wrong, in the refresh code.
} Unfortunately I'm not sure I understand the logic well enough to fix it,
} though if Geoff doesn't get around to it I will have to try.

I was looking at this because 3.0.6 has the same bug, and I can't figure
out what Geoff thinks is happening either.  There doesn't seem to ever be
any reason to output more than `ct' characters unless you're reprinting
from the beginning of the line (which is what the block lower down does,
but after sending a '\r').

However, while experimenting with forcing zsh to take all three branches
of that code, I discovered that there's a bug in promptexpand() when the
%{...%} construct is used -- it leaves garbage characters in the string
beyond the end of the prompt.  The following should take care of it.

This reminds me, though, that a while back somebody noticed that zalloc()
called calloc() rather than malloc() and "fixed" it.  I wonder if there
are other bugs lurking around because callers of zalloc() assume that they
are getting a zero'd buffer?

Index: Src/prompt.c
===================================================================
@@ -161,7 +161,7 @@
     rstring = rs;
     Rstring = Rs;
     fm = s;
-    bp = bufline = buf = zalloc(bufspc = 256);
+    bp = bufline = buf = zcalloc(bufspc = 256);
     bp1 = NULL;
     trunclen = 0;
     putpromptchar(1, '\0');

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