Question

So, I have

puts "test\\nstring".gsub(/\\n/, "\n")

and that works.

But how do I write one statement that replaces \n, \r, and \t with their correctly escaped counterparts?

Answer 1

Those aren’t escaped characters, those are literal characters that are only represented as being escaped so they’re human readable. What you need to do is this:

escapes = {
  'n' => "\n",
  'r' => "\r",
  't' => "\t"
}

"test\\nstring".gsub(/\\([nrt])/) { escapes[$1] }
# => "test\nstring"

You will have to add other escape characters as required, and this still won’t accommodate some of the more obscure ones if you really need to interpret them all. A potentially dangerous but really simple solution is to just eval it:

eval("test\\nstring")

So long as you can be assured that your input stream doesn’t contain things like #{ ... } that would allow injecting arbitrary Ruby, which is possible if this is a one shot repair to fix some damaged encoding, this would be fine.

Update

There might be a mis-understanding as to what these backslashes are. Here’s an example:

"\n".bytes.to_a
# => [10]

"\\n".bytes.to_a
# => [92, 110]

You can see these are two entirely different things. \n is a representation of ASCII character 10, a linefeed.