Question

Example of string I am working with:

s = "{new {value1 value2 value3}} {old {value2 value1 value1}} {{old school} {value2 value3 value1}}"

The {}’s are affected by spaces, which is why “old school” is surrounded while “new” and “old” are not.

Parsing the first two (new and old) are easily done using s.split[1] to access “new” and s.split[3..5] for the values. The problem comes when “new” or “old” has a space, in this case “old school”. In the database I am accessing, these names with spaces occur randomly.

How can I alter my parsing to account for these occurrences?

Answer 1

You can do it with this one line:

s.split("}} {").map{|x| x.split(" {")}.map{|x| x.map{|y| y.gsub("{","").gsub("}","")}}

Kind of ugly but works with your example, returns:

[["new", "value1 value2 value3"], ["old", "value2 value1 value1"], ["old school", "value2 value3 value1"]]

You can then parse if further by breaking values into their own objects etc. If you want it as hash, you can get it like this:

Hash[s.split("}} {").map{|x| x.split(" {")}.map{|x| x.map{|y| y.gsub("{","").gsub("}","")}}]

This will return:

{"new"=>"value1 value2 value3", "old"=>"value2 value1 value1", "old school"=>"value2 value3 value1"}