Question

我还有另一个问题：以下代码执行以下操作

对于文件夹中的每个文件
打开文件并阅读其内容
将每一行划分为令牌
将每个标记（单词）保存在hasMap
准备数据库查询（选择表单字词......）
对于令牌与数据库Write 1.0中包含的单词之间的每个匹配，如果为true，否则为0.0;

此时出现问题：

try{
    while (rs_mail.next()) {
        if(result_m.contains(rs_mail.getString("voc_w").toString()))  //HERE I GET THE ERROR! java.lang.NullPointerException
            out_final.print("1.0;"); 
        else
            out_final.print("0.0;"); 
    }//Close While
}       //Close TRY
finally{
    rs_mail.close();
    //result_m.clear();
    mail.clear(); //Clear MAP
}

完整代码下方：

String path ="C:/Users/.../file";
File currentDIR = new File("C:/Users/.../file");
File files_mail[]=currentDIR.listFiles();
String tmp_mail="";

// prepares the file tmpTraning.txt to receive value 1.0, 0.0 obtained by comparison with database
PrintWriter out_final=null;
File ff=new File("C:/Users/.../tmpTraning.txt");

//Seach for File in DIR
for( File fX : files_mail ){
    String name_Filex = fX.getName();
    FileReader fr = null;
    BufferedReader fINx = null;
    String sx;
    //Create MAP 
    Map<String, Set<String>> mail = new HashMap<String, Set<String>>();
    //Open File
    try{
        Set<String> sq = new HashSet<String>();

        fr = new FileReader(path+"/"+name_Filex);
        fINx = new BufferedReader(fr);
        sx = fINx.readLine();

        //scroll the file
        while(sx != null) {
            StringTokenizer stq = new StringTokenizer(sx);
            while(stq.hasMoreTokens()) { //Extract form line the single word
                tmp_mail = stq.nextToken();

                sq.add(tmp_mail.toString().toLowerCase()); //add the word to sq -> HashMap
                mail.put(nome_Filex, sq);

            }// Close st.hasMoreTokens()

            sx = fINx.readLine();
        } //Close  while for scroll File
        fr.close(); //Close fileReader
        sq.clear(); //Clear HasSet

    } //Close il TRAY
    catch (FileNotFoundException ex) {
        ex.printStackTrace();
    } catch (IOException ex) {
        ex.printStackTrace();
    }

    Set<String> result_m = mail.get(name_Filex);
    ResultSet rs_mail = stmt.executeQuery("SELECT DISTINCT voc.words as voc_w FROM voc_words as voc");

    //Prepare for writing on the file " tmpTraning.txt " 
    OutputStreamWriter fout_f = new OutputStreamWriter(new FileOutputStream(ff,true));
    out_final = new PrintWriter(fout_f);

    try{
        while (rs_mail.next()) {
            //If the word extract from the database is in MAP (name_Filex) then print 1.0; on the file tmpTraning.txt
            if(result_m.contains(rs_mail.getString("voc_w").toString()))  //HERE I GET THE ERROR! java.lang.NullPointerException
                out_final.print("1.0;"); 
            else
                //else print 0.0;
                out_final.print("0.0;"); 
        }
    }       //Close TRY
    finally{
        rs_mail.close();
        //result_m.clear();
        mail.clear(); //Clear MAP
    }

    out_final.println(""); //Send CR char ASCII to set the coursor for the next file on the new line
    out_final.close();
    out_final.flush();
} // End SCAN DIR

感谢您的任何建议！

代码更改 - 打印result_m的内容：

String path ="...";
File currentDIR = new File("...");
File files_mail[]=currentDIR.listFiles();
String tmp_mail="";

// prepares the file tmpTraning.txt to receive value 1.0, 0.0 obtained by comparison with database
PrintWriter out_final=null;
File ff=new File("...");

//Seach for File in DIR
for( File fX : currentDIR.listFiles() ){
    String name_Filex = fX.getName();

    String sx;
    //Create MAP 
    Map<String, Set<String>> mail = new HashMap<String, Set<String>>();
    //Open File
    try{
        Set<String> sq = new HashSet<String>();
        BufferedReader fINx = new BufferedReader(new FileReader(fX));
        sx = fINx.readLine();
        //scroll the file
        while(sx != null) {
            StringTokenizer stq = new StringTokenizer(sx);
            while(stq.hasMoreTokens()) { //Extract form line the single word
                tmp_mail = stq.nextToken();

                sq.add(tmp_mail.toString().toLowerCase()); //add the word to sq -> HashMap
                mail.put(name_Filex, sq);

            }// Close st.hasMoreTokens()

            sx = fINx.readLine();
        } //Close  while for scroll File
        fr.close(); //Close fileReader
        sq.clear(); //Clear HasSet

    } //Close il TRAY
    catch (FileNotFoundException ex) {
        ex.printStackTrace();
    } catch (IOException ex) {
        ex.printStackTrace();
    }

    /*
     * print the contents of  result_m
     */

    System.out.println("----- START FILE -----");
    Set<String> result_m = mail.get(name_Filex);
    Object[] toArray_m = mail.get(name_Filex).toArray();
    for (int ncc=0; ncc<result_m.size();ncc++){
        System.out.println(toArray_m[ncc]);
    }
    System.out.println("----- END  FILE  -----");


} // End SCAN DIR

如果程序读取的文件包含空行（没有字符，没有字符串），则保存空值

Answer 1

你是否有充分理由在字符串上调用toString()？如果getString("voc_w")为null，则会导致您的例外。

正如manji在评论中指出的那样，它是我提到的电话，或result_m Set null。

Answer 2

遗憾的是，您的代码存在很多问题。这是一个，立刻引起了我的注意。

您正在使用字符串操作来创建要传递给FileReader构造函数的文件名，尽管据我所知，正在打开的文件是从目录列表返回的文件。那只是在惹麻烦。相反，代码会更好地写得更像......

// Lots of things are omitted; this is just a sketch...
String path = "...";
File currentDIR = new File(path);
for (File fX : currentDIR.listFiles()) {
    try {
        BufferedReader fINx = new BufferedReader(new FileReader(fX));
        // ...
    } catch (IOException e) {
        e.printErrorStack();
    }
}

然而，你有更多的问题。例如，sq.clear()的使用具有错误的代码气味。您刚刚在Set中存储了对Map的引用;为什么要再次删除其内容？变量超出范围;你可以简单地留下那些代码。 clear()之后的result_m的下游结果将是一个空集，因此其contains测试将始终返回false。我不能随便告诉你这是否是你的流氓null的原因，但根据你声称要做的事情，它必须是错误的。

尝试将该代码重构为几个更容易验证的小块。我建议作为第一个剪切：一个私有方法从File获取一组单词（作为参数提供），一个私有方法来比较一组单词与数据库，以及一个方法，将这两个与一些循环结合起来实现你的总体目标。

MAP JAVA和Sqlite数据库的问题

2 个答案: