用Ruby查找替换UTF-8文件中的中文字符

[b]Precondition:[/b]
1. The Source Java file is saved as UTF-8 type
2. The Source java file containts some chinese words as comments for some test data.
3. My ruby code file is saved as UTF-8 as well.

[b]Requirement:[/b]
1. Find all the '克' in the java files under a foder, and repleace it to '可'
2. User runs the replace.rb file through command line like:
replace <dir name>

[b]Code: replace.rb[/b]

##############################################################
dName = ARGV[0]     # Directory to process
srcStr = '克'  # srcStr = ARGV[1].dup.force_encoding('utf-8')
destStr = '可' # destStr = ARGV[2].dup.force_encoding('utf-8')
##############################################################
Dir.chdir(dName)
Dir["**/*.java"].each do |file|

  oldFileName = File.basename(file,".java") + ".org"
  File.rename(file, oldFileName)

  oldFile = File.open(oldFileName, "r")
  newFile = File.open(file, "w")

  puts "........................... #{File.basename(newFile)}"
  oldFile.each_line do |line|
    line.force_encoding('utf-8') 
    puts oldFile.lineno, line if line =~ /#{srcStr}/
    newFile.puts line.gsub(/#{srcStr}/, destStr)
  end

  oldFile.close
  newFile.close

end

[b]Issue:[/b]
I'm new to Ruby. The coding style and error handling is not enough

I failed to pass the chinese words as command arguments by above code like : replace <dir name> <sourceString>, <replaceString>
Example: replace <dir name> "克" "可"
This way doesn't work. I don't know the cause so far.

[b]Udpaded on 26/01/2011[/b]
I found the solution:
change the source code from

srcStr = ARGV[1].dup.force_encoding('utf-8')
destStr = ARGV[2].dup.force_encoding('utf-8')

to

srcStr = ARGV[1].dup.encode('utf-8')
destStr = ARGV[2].dup.encode('utf-8')

This can make the paramemter passed and convert to UTF-8 successfully.

Good article:
[url]http://blog.grayproductions.net/articles/ruby_19s_string[/url]

But later I tried to change the code

line.force_encoding('utf-8')

to

line.encode('utf-8')

It has error again. :(