Question

我正在制作一个Android字典越南语 - 英语（超过20,000个条目）。在我的应用程序中包含edittext和Listview。要在我的列表视图中搜索项目，我使用二进制搜索。但问题是搜索方法和越南语，它搜索不好它可以找到一些词，有些则不能。

以下是我的搜索代码，用于在编辑文本更改时按字词前缀查找。

public void searchWords(String[] words, String prefix)
    {
        int first = 0, last = words.length - 1;
        int mid = 0;

        while (first <= last)
        {
            mid = (first + last) / 2;
           int c = prefix.compareTo(words[mid]);
            if (c == 0)
            {
                first = mid;    // first indicates the beginning
                break;
            }
            if (c > 0)
                first = mid + 1;
            else
                last = mid - 1;
        }

        int i;
        for (i = first; i < words.length; i++)
        {
            if (words[i].startsWith(prefix))
            {
             pos=i;
             break;
            }
        }

        lv.setSelection(pos);
 //  Toast.makeText(getApplicationContext(), ""+pos,Toast.LENGTH_SHORT).show();
    }

和onTextchange代码我调用我的方法：

public void onTextChanged(CharSequence s, int start, int before, int count){
        // TODO Auto-generated method stub


         searchWords(w,s.toString());

       }

这就是我将条目从数据库加载到数组的方式：

d=handle.retrieve();

    if(d.moveToFirst())
     {
         do
         {
             w[ii++]=d.getString(1);

         }while(d.moveToNext());

     }

那么，我该怎么做才能使我的搜索工作正常？

Answer 1

使用“compareTo（）”的二进制搜索仅适用于SORTED单词（或排序的字符串，即按字母顺序排序 - 无论语言如何）。含义：数组以A，AA，AAA，......开头，以z，zz，zzz结尾... Btw，小写前大写。例如：

String[] seq = {"Ape", "Bird", "Donkey", "Eagle", "Fish", "Gnu", "Horse", "Koala"};
String[] ran = {"Gnu", "Koala", "Horse", "Fish", "Bird", "Donkey", "Eagle", "Ape"};
String[] queries = {"Eagle", "Bird", "Donkey", "Fish", "Ape", "Horse", "Eagle", "Gnu"};
// binarySearch with random[] will produce unpredictable results when prefix starts with
// a word that is either after "Bird" or "Ape" or "Gnu" 
for (String s : queries) {
   System.out.println("Random: BS for " + s + " = " + binarySearch(ran, s));
}
//
// binarySearch with sequence[] will produce correct results whatever prefix is.
for (String s : queries) {
        System.out.println("Sequence: BS for " + s + " = " + binarySearch(seq, s));
}
//-------------------------------------------------------------------------
public static int binarySearch(String[] words, String value) {
    return binarySearch(words, value, 0, words.length - 1);
}
//
public static int binarySearch(String[] words, String value, int min, int max) {
    if (min > max) {
        return -1;
    }

    int mid = (max + min) / 2;
    if (words[mid].equals(value)) return mid;
    if(words[mid].compareTo(value) > 0)
        return binarySearch(words, value, min, mid - 1);
    return binarySearch(words, value, mid + 1, max);
}

二进制搜索越南词从java android中的数组字符串

1 个答案: