Parsing Nested Text in C#

Question

If I have a series of strings that have this base format:

"[id value]"//id and value are space delimited.  id will never have spaces

They can then be nested like this:

[a]
[a [b value]]
[a [b [c [value]]]

So every item can have 0 or 1 value entries.

What is the best approach to go about parsing this format? Do I just use stuff like string.Split() or string.IndexOf() or are there better methods?

Answer 1

A little recursion and split would work, the main point is use recursion, it'll make it so much easier. Your input syntax looks kind of like LISP :)

Parsing a, split, no second part. done.
Parsing a [b value]. has second part, go to the beginning.
...

You get the idea.

Answer 2

Simple split should work For every id,there is one bracket [
So when you split that string you have n-brackets so n-1 id(s) where the last element contains the value.

Answer 3

Regex is alway a nice solution.

string test = "[a [b [c [value]]]";
Regex r = new Regex("\\[(?<id>[A-Za-z]*) (?<value>.*)\\]");
var res = r.Match(test);

Then you can get the value (which is [b [c [value]] after the first iteration) and apply the same again until the match fails.

string id = res.Groups[1].Value;
string value = res.Groups[2].Value;

Answer 4

there is nothing wrong with split and indexof methods, they exist for string parsing. Here is a sample for your case:

        string str = "[a [b [c [d value]]]]";

        while (str.Trim().Length > 0)
        {
            int start = str.LastIndexOf('[');
            int end = str.IndexOf(']');

            string s = str.Substring(start +1, end - (start+1)).Trim();
            string[] pair = s.Split(' ');// this is what you are looking for. its length will be 2 if it has a value

            str = str.Remove(start, (end + 1)- start);
        }