Ruby for system administration

clean-data.rb

# Small example of the use of Ruby for system administration
# This sample script does the following:
# - Read a text file, "data.in". This is our main data
# - Read a list of data patterns, one per line. Each line is just a word or a sentence.
# - Suppress each line of the input file that contains one of the patterns
# - Output to "data.out"

# Version 1: simple syntax without blocks. But the opened files have to be
#  closed manually if necessary.
# Read file as an array of lines
data_in = File.readlines("data.in")
# Same thing, but strip newlines
rej = File.readlines("content-to-delete.txt").map { |l| l.strip }
# Build a simple regular expression
rejreg = Regexp.new(rej.join("|"))
# And output the result without the rejected lines
data_out = open("data.out", "w")
data_out.puts(data_in.reject { |l| rejreg.match(l) })
data_out.close

# Version 2: Feel the power of Ruby blocks
#  No leftover open file descriptors or local variables
#  have been hurt in this version.
File.open("data.in") do |infile|
  # Build a simple regular expression. I left this one on two lines
  # instead of writing an obscure one-liner..
  rej = File.readlines("content-to-delete.txt").map { |line| line.strip }
  rejreg = Regexp.new(rej.join("|"))
  File.open("data.out", "w") do |outfile|
    outfile.puts(infile.readlines.reject { |line| rejreg.match(line) })
  end
end