Path: utzoo!utgpu!news-server.csri.toronto.edu!rpi!zaphod.mps.ohio-state.edu!swrinde!elroy.jpl.nasa.gov!jpl-devvax!lwall
From: lwall@jpl-devvax.jpl.nasa.gov (Larry Wall)
Newsgroups: comp.lang.perl
Subject: Re: Bug with $` in s///; in 4.000,4.003
Message-ID: <1991Apr22.175438.8564@jpl-devvax.jpl.nasa.gov>
Date: 22 Apr 91 17:54:38 GMT
References: <1991Apr20.193701.17201@uvaarpa.Virginia.EDU>
Reply-To: lwall@jpl-devvax.JPL.NASA.GOV (Larry Wall)
Organization: Jet Propulsion Laboratory, Pasadena, CA
Lines: 25

In article <1991Apr20.193701.17201@uvaarpa.Virginia.EDU> ckclark@athena.mit.edu writes:
: A problem occurs when trying to replace tabs with spaces using the
: invocation:
: 
: while (<>) {
:     1 while s/\t+/' 'x (length($&) * 8 - length($`) % 8)/e;
:     print;
: }
: 
: (taken from the Gospel According to lwall, Chapter 3, pg. 105, footnote 31)
: 
: When I run it in 3.044, it works fine, but when I try it in 4.000 or
: 4.003, it acts as if length($`) is 0; i.e., as if $` is not bring set.
: 
: I have not seen anything in the documentation which indictates that this
: behavior change is intentional.  Am I missing something?

No, you're right.  There was a bug in 3.044 having to do with recursion
if the RHS called the subroutine containing the current substitution.  In
fixing this, I managed to "unfix" $`.  It will be fixed in patch 4.

Sorry.  My testing organization is either too small, or too large, depending
on how you look at it.  :-)

Larry