Writing Arrays using grep - ruby

Question

I'm trying to search through a specified string and assign the results to an array.
Opening and writing to "input.txt" and "ms3.txt" files works fine. putting a normal string like reassign << "hello" in works fine its just when i use line.grep and the regex following it prints nothing to the console or the ms3 file it doesn't even throw up any errors
i've also tried a search and replace: reassign << line.gsub(/[abc]/, '£')
Here's the code
# encoding: utf-8
#!/usr/bin/ruby
file = File.open("input.txt", "w+")
reassign = []
file.each_line do |line|
reassign << line.grep(/[abc]/)
end
new_file = File.open("ms3.txt", "w+")
new_file.puts(reassign)
new_file.close

Answer 1

Your code can be streamlined a lot to make it more Ruby-like, and to make it behave better:
# encoding: utf-8
#!/usr/bin/ruby
file = File.open("input.txt", "w+")
reassign = []
file.each_line do |line|
reassign << line.grep(/[abc]/)
end
new_file = File.open("ms3.txt", "w+")
new_file.puts(reassign)
new_file.close
#! lines have to be first, so reverse the encoding and "slash-bang" lines.
open takes a block, which will allow Ruby to automatically close the file when the block exits. This is a very powerful, and smart thing to do, as it keeps your file I/O environment clean. It's possible, and often a problem as in your code, to open files but never close them, which can exhaust all the file handles on a machine if it's done in a loop, causing all apps to fail. Using the block form will avoid this.
The IO class has foreach, which makes it simple to iterate over the lines of a file. Take advantage of it instead of opening the file then using each_line, because it simplifies your code.
Here's how I'd initially write your code:
#!/usr/bin/ruby
# encoding: utf-8
reassign = []
File.foreach("input.txt") do |line|
reassign << line[/[abc]/]
end
File.write("ms3.txt", reassign.join("\n"))
But, after refactoring it I'd end up with:
#!/usr/bin/ruby
# encoding: utf-8
File.open('ms3.txt', 'w') do |fo|
fo.puts File.foreach('input.txt').grep(/[abc/])
end
The open opens the output file using a block to take advantage of automatically closing the file when the block exits.
foreach is an iterator, and normally is used with a block to pass each line read into the block. Instead, I'm letting grep read all the lines found and search for the pattern.
Any lines found by grep that match the pattern are returned as an array to puts which will iterate over them, appending "\n" to the end of each.
fo.puts directs the output of puts to the output file.
end causes the block to exit, which causes open to close the file.
That's untested but looks correct.

Answer 2

There are several issues with your code:
You open "input.txt" with open mode "w+". According to the documentation, this truncates your file to zero length. An empty file doesn't contain any lines and therefore, file.each_line doesn't invoke the block.
If you want to read from the file, use "r", which is the default:
file = File.open("input.txt")
You don't close file. Use the block form which closes the file automatically:
File.open("input.txt") do |file|
# ...
end
line is a String and there's no String#grep method. But since File includes Enumerable, you can use Enumerable#grep instead:
reassign = file.grep(/[abc]/)
A complete example:
File.open("input.txt") do |file|
reassign = file.grep(/[abc]/)
File.open("ms3.txt", "w+") do |new_file|
new_file.puts(reassign)
end
end