Re: removeheaders

> > I don't know if performance is at all a concern with this small 
> > script, but in its current state it is probably running
> > 3 times slower than it needs to be. It compares the line to three 
> > regular expressions, even though only one should ever match. The 
> > expressions should also be ordered such that the most 
> likely ones are 
> > tried first.
> > 
> > The diff below corrects these issues. I've only tested it with data 
> > that matches the first expression.
> > 
> > Also, in this line:
> > 
> >   while (defined($ThisLine = <STDIN>)) {
> > 
> > is defined() doing anything useful? Even a blank line, due to the 
> > newline, will still return true.
> > 
> >   -Tom
> > 
> 
> 

Tom,

I think the removeheaders optimization you posted can be further reduced,
potentially avoiding calling the RE generator twice for non-matching lines,
and avoids traversing the same sub-expression twice for matching lines.

# Removes the beginning of each line of a standard /var/log/messages-style
# logfile.
#
# Assumptions: OnlyService is called before RemoveHeaders

while (<>)) {
   if (s/^... .. ..:..:.. [^ ]* //) {       # Strip date, time, service
      if (s/^[^\[:]*(?:\[\d*\])?: //) {     # ":" after service name may not
exist
         s/^\[ID \d+(?: \w+\.\w+)?] //;     # Strip Solaris ID tag style
-mgt
      }
   }
   print;
}

MrC