OpenCSV单列映射多字段的挑战与自定义策略解析

public class MyDto {
    @CsvBindByName(column = "AFBP")
    String placeholderA;
    @CsvBindByNames({
            @CsvBindByName(column = "ABCD"),
            @CsvBindByName(column = "AFEL")
    })
    String placeholderB;

    @CsvBindByNames({
            @CsvBindByName(column = "ABCD"),
            @CsvBindByName(column = "ALTM")
    })
    String placeholderC;

    @Override
    public String toString() {
        return "placeholder A = " + placeholderA + ", placeholderB = " + placeholderB + ", placeholderC = " + placeholderC;
    }
}

AFBP,ABCD
this is A,this is B and C

placeholder A = this is A, placeholderB = null, placeholderC = this is B and C

import com.opencsv.bean.HeaderNameBaseMappingStrategy;
import com.opencsv.bean.CsvBindByName;
import com.opencsv.bean.CsvBindByNames;
import com.opencsv.bean.CsvToBeanBuilder;
import com.opencsv.exceptions.CsvBadConverterException;
import com.opencsv.exceptions.CsvRequiredFieldException;
import java.beans.IntrospectionException;
import java.lang.reflect.Field;
import java.util.ArrayList;
import java.util.HashMap;
import java.util.List;
import java.util.Map;
import java.util.stream.Collectors;

// 假设这是您的自定义映射策略
public class MultiFieldColumnMappingStrategy<T> extends HeaderNameBaseMappingStrategy<T> {

    // 存储一个CSV列名到多个Java字段的映射
    private final Map<String, List<Field>> multiFieldMap = new HashMap<>();

    @Override
    public void captureHeader(String[] header) throws CsvRequiredFieldException, CsvBadConverterException {
        // 在这里，您需要遍历所有字段，并根据注解构建multiFieldMap
        // 这是一个简化示例，实际情况需要更复杂的逻辑来处理所有注解类型
        // 和确保字段可访问性等
        for (Field field : loadFields(getType())) { // loadFields是一个假设的方法，用于获取所有字段
            if (field.isAnnotationPresent(CsvBindByName.class)) {
                CsvBindByName bindByName = field.getAnnotation(CsvBindByName.class);
                multiFieldMap.computeIfAbsent(bindByName.column(), k -> new ArrayList<>()).add(field);
            } else if (field.isAnnotationPresent(CsvBindByNames.class)) {
                CsvBindByNames bindByNames = field.getAnnotation(CsvBindByNames.class);
                for (CsvBindByName bindByName : bindByNames.value()) {
                    multiFieldMap.computeIfAbsent(bindByName.column(), k -> new ArrayList<>()).add(field);
                }
            }
        }
        // 调用父类的captureHeader以确保其他默认逻辑也被执行
        super.captureHeader(header);
    }

    // 重写setter，以处理多字段映射
    @Override
    protected void set \(T bean, String headerName, String value, int colPos) throws CsvBadConverterException {
        List<Field> fieldsToSet = multiFieldMap.get(headerName);
        if (fieldsToSet != null) {
            for (Field field : fieldsToSet) {
                try {
                    field.setAccessible(true); // 确保字段可访问
                    field.set(bean, value); // 假设值类型匹配
                } catch (IllegalAccessException e) {
                    // 处理异常
                    e.printStackTrace();
                }
            }
        } else {
            // 如果没有自定义映射，则使用父类（默认）逻辑
            super.set(bean, headerName, value, colPos);
        }
    }

    // 辅助方法：获取所有字段 (实际OpenCSV内部有更完善的机制)
    private List<Field> loadFields(Class<T> type) {
        List<Field> fields = new ArrayList<>();
        Class<?> currentClass = type;
        while (currentClass != null && currentClass != Object.class) {
            for (Field field : currentClass.getDeclaredFields()) {
                fields.add(field);
            }
            currentClass = currentClass.getSuperclass();
        }
        return fields;
    }
}

import com.opencsv.bean.CsvToBean;
import com.opencsv.bean.CsvToBeanBuilder;
import java.io.StringReader;
import java.util.List;

public class CsvProcessor {
    public static void main(String[] args) {
        var csv = "AFBP,ABCD
this is A,this is B and C";

        // 使用自定义映射策略
        CsvToBean<MyDto> csvToBean = new CsvToBeanBuilder<MyDto>(new StringReader(csv))
                .withType(MyDto.class)
                .withMappingStrategy(new MultiFieldColumnMappingStrategy<>(MyDto.class)) // 传入自定义策略实例
                .build();

        List<MyDto> dtos = csvToBean.parse();
        dtos.forEach(System.out::println);
    }
}