Re^4: reading a line from file two times

There's definitely someting screwy with using a negative, relative seek (on Win32). To back up over a line consisting of 8 chars + NL, requires seeking back -6.

I'd expect to have to seek back more than the line length, not less to account for the newline?

#! perl -slw
use strict;

while( <DATA> ) {
    chomp;
    print;
    if( /the line/ ) {
        my $backup = -length() + 2;
        print "Backing up $backup places";
        seek DATA, $backup, 1;
        chomp( $_ = <DATA> );
        print "?$_?";
    }
}

=output
c:\test>junk
junk
junk
more junk
the line
Backing up -6 places
?the line?
other junk
yet more junk
=cut
__DATA__
junk
junk
more junk
the line
other junk
yet more junk
[download]

Examine what is said, not who speaks -- Silence betokens consent -- Love the truth but pardon error.

Lingua non convalesco, consenesco et abolesco. -- Rule 1 has a caveat! -- Who broke the cabal?

"Science is about questioning the status quo. Questioning authority".

In the absence of evidence, opinion is indistinguishable from prejudice.

Comment on Re^4: reading a line from file two times Download Code

Replies are listed 'Best First'.
Re^5: reading a line from file two times by jettero (Monsignor) on Dec 21, 2006 at 05:36 UTC
Should be on systems where `seek($fh, $amount, SEEK_CUR)` works correctly, you'd have to backup over all the characters read. So if the `length($entire_string)` is 10 (including LF or CRLF), that'd be: `seek($fh, -10, SEEK_CUR)`. I do seem to recall that it behaves unexpectedly (for a unix nerd) in windows, but my experience was in foxpro, not perl — that definitely shouldn't count. -Paul	[reply] [d/l] [select]