仅在匹配阈值字节时才在地图中填充字符串值

Question

我有一个tasks列表对象,我正在迭代并附加每个任务对象,StringBuilder然后是新行,如下所示.现在我将继续在同一个字符串生成器中追加任务对象,直到它达到60000字节的大小限制.一旦达到限制,我将填充此字符串作为映射中的值,键将是带增量索引的文件名.然后我将重置字符串构建器和其他东西并再次重复此过程.

因此,如果我有一个大tasks对象,那么我将拆分成多个字符串对象,其大小应始终小于60000字节.

我得到了下面的代码,但我总是看到地图中的值大小超过60000字节.我做错了什么？此外,我正在填写HashMap两个不同的地方 - 一个达到限制,另一个是如果没有达到限制.

  public void populate(final List<Task> tasks) {
    Map<String, String> holder = new HashMap<>();
    int size = 0;
    int index = 0;
    StringBuilder sb = new StringBuilder();
    for (Task task : tasks) {
      sb.append(task).append(System.getProperty("line.separator"));
      size = sb.toString().getBytes(StandardCharsets.UTF_8).length;
      if (size > 60000) {
        String fileName = "tasks_info_" + index + ".txt";
        holder.put(fileName, sb.toString());
        index++;
        sb = new StringBuilder();
        size = 0;
      }
    }
    // for cases where we don't reach the limit
    if(sb.toString().length > 0) {
        String fileName = "tasks_info_" + index + ".txt";
        holder.put(fileName, sb.toString());        
    }
    System.out.println(holder);
  }

注意:如果每个Task对象都超过60000 bytes,那么我将立即删除该对象并移动到下一个条目.但实际上,它不会发生.

更新:

public void populate(final List<Task> tasks, final long timestamp) {
    Map<String, String> holder = new HashMap<>();
    int size = 0;
    int index = 0;
    int nl = System.getProperty("line.separator").getBytes(StandardCharsets.UTF_8).length;
    StringBuilder sb = new StringBuilder();
    // new change
    sb.append(timestamp).append(System.getProperty("line.separator"));
    for (Task task : tasks) {
        int ts = String.valueOf(task).getBytes(StandardCharsets.UTF_8).length;
        if (size + ts + nl > 60000) {
            String fileName = "tasks_info_" + index + ".txt";
            holder.put(fileName, sb.toString());
            index++;
            sb = new StringBuilder();
            // new change
            sb.append(timestamp).append(System.getProperty("line.separator"));          
            size = 0;
        }
        sb.append(task).append(System.getProperty("line.separator"));
        size += ts + nl;
    }
    // for cases where we don't reach the limit
    if (size > 0) { // size can only be 0 if you have 0 tasks
        String fileName = "tasks_info_" + index + ".txt";
        holder.put(fileName, sb.toString());        
    }
    System.out.println(holder);
}

Answer 1

其他答案已经提到了它不起作用的原因（您在超过限制后添加）。但我认为到目前为止，没有一个实现是正确的，不仅仅是因为省略了换行符的大小。

public Map<String, String> populate(final List<Task> tasks) {
    Map<String, String> holder = new HashMap<>();
    if (tasks.size() == 0)
        return holder;
    int index = 0;
    int nl = System.getProperty("line.separator").getBytes(StandardCharsets.UTF_8).length;
    StringBuilder sb = new StringBuilder();
    sb.append(System.currentTimeMillis()).append(System.getProperty("line.separator"));
    int size = sb.toString().getBytes(StandardCharsets.UTF_8).length;
    for (Task task : tasks) {
        int ts = String.valueOf(task).getBytes(StandardCharsets.UTF_8).length;
        if (size + ts + nl > 60000) {
            String fileName = "tasks_info_" + index + ".txt";
            holder.put(fileName, sb.toString());
            index++;
            sb = new StringBuilder();
            sb.append(System.currentTimeMillis()).append(System.getProperty("line.separator"));
            size = sb.toString().getBytes(StandardCharsets.UTF_8).length;
        }
        sb.append(task).append(System.getProperty("line.separator"));
        size += ts + nl;
    }
    String fileName = "tasks_info_" + index + ".txt";
    holder.put(fileName, sb.toString());        
    return holder;
}