Startup Time

Startup Time

[Andy Wick]

I noticed that the startup time of zsh-2.6-beta11-test12-hzoli11
is much slower then zsh-2.6-beta11-test9-hzoli11.  It used
to take about 3 seconds to start a zsh and process my MANY 
named dirs.  Now it takes about 7 seconds.  (I ran time
several times on both versions.)  Anything I can
do to improve the speed.  I have the autonamedirs
option turned on, which I bet is what is slowing it down.  But
I can't turn it off because the "%~" expansion stops working
correctly (doesn't always get the shortest path).  

Platform: Solaris 2.4, Sparc 10.

2 Bugs:

1)  ^ by itself crashes zsh.  (present since at least beta6)

2) My favorite: Bolding in watch fmts doesn't work.  RC says my
fix (the one putshout in zle_misc.c should be putraw) has side effects.
I have been using it for half a year now, with no problems.  
What kind of side effects should I be looking for?

Thanks!!

-- 
awick@vt.edu                                    Andy Wick
awick@yertle.fp.trw.com                     Virginia Tech/TRW

Re: Startup Time

[Zoltan Hidvegi]

Andy Wick wrote:
> 
> I noticed that the startup time of zsh-2.6-beta11-test12-hzoli11
> is much slower then zsh-2.6-beta11-test9-hzoli11.  It used
> to take about 3 seconds to start a zsh and process my MANY 
> named dirs.  Now it takes about 7 seconds.  (I ran time
> several times on both versions.)  Anything I can
> do to improve the speed.  I have the autonamedirs
> option turned on, which I bet is what is slowing it down.  But
> I can't turn it off because the "%~" expansion stops working
> correctly (doesn't always get the shortest path).  

Richard made some changes here between test9 and test12.  I'm sure I did not
changed anything related to this.

> 1)  ^ by itself crashes zsh.  (present since at least beta6)

I'll have a look at it if I have some time.

Bye,

  Zoltan

Re: Startup Time

[P.Stephenson]

awick@barney.purple.org wrote:
> 1)  ^ by itself crashes zsh.  (present since at least beta6)

This seems to be the minmal fix.  Although I'm not sure this is
documented, what ^ on it's own does is attempt to replace the last
pattern substituted with nothing in the previous history line.  This
was failing if there was no previous substitution.  This bug crept in
when I tidied up the history code: before, the code hacked the input
and the history code status so that it thought a real substitution
like !!^ was coming up, which handles this case.  If your dead set on
the message "no previous subsitution with &" here too, I can add that.

*** Src/hist.c~	Tue Nov  7 09:48:34 1995
--- Src/hist.c	Wed Nov  8 09:25:47 1995
***************
*** 152,158 ****
  	inungetc(hatchar);
  	if (!(ehist = gethist(defev))
  	    || !(sline = getargs(ehist, 0, ehist->nwords-1))
! 	    || getsubsargs(sline))
  	    return -1;
  	subst(&sline, hsubl, hsubr, 0);
      } else {
--- 152,158 ----
  	inungetc(hatchar);
  	if (!(ehist = gethist(defev))
  	    || !(sline = getargs(ehist, 0, ehist->nwords-1))
! 	    || getsubsargs(sline) || !hsubl)
  	    return -1;
  	subst(&sline, hsubl, hsubr, 0);
      } else {


-- 
Peter Stephenson <pws@ifh.de>       Tel: +49 33762 77366
WWW:  http://www.ifh.de/~pws/       Fax: +49 33762 77330
Deutches Electronen-Synchrotron --- Institut fuer Hochenergiephysik Zeuthen
DESY-IfH, 15735 Zeuthen, Germany.