如何用单词位置数组替换句子

Question

I have a sentence like this:

ジェーンは先週日本に来て、毎日４時間日本語のクラスで勉強しています

And data token like this:

[{"token":"ジェーン","type":"word","start_offset":0,"end_offset":4,"position":0},{"token":"は","type":"word","start_offset":4,"end_offset":5,"position":1},{"token":"先週","type":"word","start_offset":5,"end_offset":7,"position":2},{"token":"日本","type":"word","start_offset":7,"end_offset":9,"position":3},{"token":"に","type":"word","start_offset":9,"end_offset":10,"position":4},{"token":"来","type":"word","start_offset":10,"end_offset":11,"position":5},{"token":"て","type":"word","start_offset":11,"end_offset":12,"position":6},{"token":"毎日","type":"word","start_offset":13,"end_offset":15,"position":7},{"token":"４","type":"word","start_offset":15,"end_offset":16,"position":8},{"token":"時間","type":"word","start_offset":16,"end_offset":18,"position":9},{"token":"日本語","type":"word","start_offset":18,"end_offset":21,"position":10},{"token":"の","type":"word","start_offset":21,"end_offset":22,"position":11},{"token":"クラス","type":"word","start_offset":22,"e nd_offset":25,"position":12},{"token":"で","type":"word","start_offset":25,"end_offset":26,"position":13},{"token":"勉強","type":"word","start_offset":26,"end_offset":28,"position":14},{"token":"し","type":"word","start_offset":28,"end_offset":29,"position":15},{"token":"て","type":"word","start_offset":29,"end_offset":30,"position":16},{"token":"い","type":"word","start_offset":30,"end_offset":31,"position":17}]

how can i wrap text in sentence by start_offset and end_offset like this:

<span>ジェーン</span><span>は</span><span>先週</span>... 

I've tried StringBuilder to replace at position but index of words changed, so from token 2, it's wrong.

Answer 1

Inserting new elements moves the position of everything after that element. Therefore, try starting from the end of the string and working backwards. This means you don't have to recalculate the positions, because the positions that are affected are the ones you've already dealt with.

string result = sentence;

foreach (var token in dataTokens.OrderByDescending(x => x.position))
{
    result = result.Insert(token.end_offset, "</span>");
    result = result.Insert(token.start_offset, "<span>");
}

return result;

Testing this out yields the following string:

 <span>ジェーン</span><span>は</span><span>先週</span><span>日本</span><span>に</span><span>来</span><span>て</span>、<span>毎日</span><span>４</span><span>時間</span><span>日本語</span><span>の</span><span>クラス</span><span>で</span><span>勉強</span><span>し</span><span>て</span><span>い</span>ます

Answer 2

I would suggest to create a new string and then do something like this (not exact code):

string s = null;
foreach(string token in dataTokens)
{
   s+="<span>" + token + "</span>";
}

UPD: After comments I've tried to simulate your scenario with such token:

 class Token
{
    private int start_offset;
    private int end_offset;
    private int position;
    string type;
    string token;

    public Token(int so, int se, int pos, string type, string token)
    {
        start_offset = so;
        end_offset = se;
        position = pos;
        this.type = type;
        this.token = token;
    }

    public string TokenProp
    {
       get { return token; }
    }
}

The rolling with span I've done like this:

List<Token> tokens = new List<Token>();

        tokens.Add(new Token(0, 4, 0, "word", "abcd"));
        tokens.Add(new Token(4, 5, 1, "word", "e"));
        tokens.Add(new Token(6, 9, 2, "word", "fgh"));
        tokens.Add(new Token(9, 11, 3, "word", "ijk"));

        StringBuilder sb = new StringBuilder();
        foreach (Token t in tokens)
        {
            sb.Append("<span>");
            sb.Append(t.TokenProp);
            sb.Append("</span>");
        }

UPD2 The answer above is way better :)