Subj : binktrim in ruby
To   : All
From : tony summerfelt
Date : Wed Feb 04 2004 04:34 pm

this is the binktrim program in ruby. i'm not happy with it. this took me WAY 
longer to write than the perl version. ruby's 'principle of least suprise' 
turned out to be the 'principle of most suprise'

ruby does not have perl's magic. it could be that i was trying to write it too 
perlish, but for a quick and dirty script i should have had to fiddle with 
classes, etc.

the output isn't quite what i wanted (you'll see if you compare the perl 
trimmed file to the ruby trimmed file.)

the only advantage i can see with the ruby version is the parse method for 
Time. and the program is definitely slower.

i think for larger oo projects i would stick with itcl or ruby, probably not 
perl, and definitely not python.

==========================================================================
require 'time'
require 'tempfile'
require 'ftools'

if ARGV.length!=2
   raise  "usage: binktrim <path/to/binkd.log> <days>"
end

year=Time.now.year
t2=Time.now

t=Tempfile.new("trimmed")
l=File.open(ARGV[0])

l.each_line {|line|
 logline="#{line}" if line=~/\[\d+\]/
 if line=~/\[\d+\]/
    td1=logline.split(/\[\d+\]/)
    td2=td1.first.split(/\s+/)
    td2.delete_at(0)
    td2.push(year)
    td3=td2.join(s=" ")
    t1=Time.parse(td3)
    t3 = t2 - 24 * 3600 * ARGV[1].to_i
    t.print("#{logline}") if t1>t3
 end
 t.print "#{line}" if line=~/^\s*$/
}
l.close
File::copy(t.path, ARGV[0])
t.close(true)
==========================================================================
--- GoldED/spleen 3.0.1-os1+
 * Origin: ventedspleen.dyndns.org (1:261/38.5)

.