如何在iBatis中返回NULL值? [英] How to Return NULL-values in iBatis?

问题描述

假设我有一个Oracle数据库和一个类似这样的接口:

Let's say I have an Oracle database and an interface like this:

public interface DaoMapper {

  @Select({
      "SELECT col1, col2, col3",
        "FROM my_table" })
  List<Map<String, Object>> getUntyped();

}

如果我调用getUntyped()，并且所有列都有一个值，则映射包含三个条目.但是，如果col2为NULL，则映射仅具有两个条目.在很多情况下，这不是问题，但是在我们代码的通用部分中，我实际上想在该地图上调用.values()，并希望包含三个条目的列表.任何条目都可以为null(或与Oracle中相同的空字符串).

If I call getUntyped() and all columns have a value the map contains three entries. However, if col2 is NULL, the map has only two entries. In many cases this isn't a problem, but in a generic part of our code I actually want to call .values() on that map and want a list consisting of three entries. Any entry may be null (or an empty string as that's the same in Oracle).

实际上，我真的很高兴这样的事情，其中​​每个外部列表都由具有三个条目的列表组成:

Actually, what I would be really happy about is something like this where each outer list consists of lists with three entries:

  @Select({
      "SELECT col1, col2, col3",
        "FROM my_table" })
  List<List<Object>> getUntypedList();

但是，iBatis告诉我这是一项不受支持的操作.

However, iBatis tells me that this is an unsupported operation.

因此，我在这里问我如何告诉iBatis包含NULL或空字符串的列.

Therefore, I'm here to ask how I can tell iBatis to include columns that are NULL or an empty string.

推荐答案

感谢乔瓦尼的答案我注意到类型处理程序的示例，然后从那里去:

Thanks to Giovanni's answer I noticed the example for type handlers and went from there:

public class EmptyStringTypeHandler extends StringTypeHandler {

  @Override
  public String getResult(ResultSet rs, String columnName) throws SQLException {
    return unnulledString(super.getResult(rs, columnName));
  }

  @Override
  public String getResult(ResultSet rs, int columnIndex) throws SQLException {
    return unnulledString(super.getResult(rs, columnIndex));
  }

  @Override
  public String getResult(CallableStatement cs, int columnIndex) throws SQLException {
    return unnulledString(super.getResult(cs, columnIndex));
  }

  private String unnulledString(String value) {
    return StringUtils.defaultString(value, "");
  }

}

界面现在为:

public interface DaoMapper {

  @Select({
      "SELECT col1, col2, col3",
        "FROM my_table" })
  @Results(value = {
      @Result(column = "col1", property = "col1", typeHandler = EmptyStringTypeHandler.class),
      @Result(column = "col2", property = "col2", typeHandler = EmptyStringTypeHandler.class),
      @Result(column = "col3", property = "col3", typeHandler = EmptyStringTypeHandler.class)
  })
  List<LinkedHashMap<String, ?>> getUntyped();

}

我应该补充一点，最大的好处是我可以在每条语句的每一列中指定它.对于更通用的用法，最好在每个语句中指定它.也许是将来的版本?

I should add that the big advantage is that I can specify this per column per statement. For more generic use it would be better to specify this per statement. Maybe in some future version?

这篇关于如何在iBatis中返回NULL值?的文章就介绍到这了，希望我们推荐的答案对大家有所帮助，也希望大家多多支持IT屋！