opencsv HeaderColumnNameAndOrderMappingStrategy

README.md

HeaderColumnNameAndOrderMappingStrategy

This creates a MappingStrategy for use with OpenCSV (specifically tested for generating a CSV from beans) which does the following:

1. Preserves the column name casing in the @CsvBindByName annotation
2. Adds a @CsvBindByNameOrder annotation you can apply to the bean class to define the order of the columns.

• Any field not included in the order, but is still annotated with @CsvBindName will still be included AFTER all the columns that have a defined order. Those remaining columns will be added in alphabetical order (this is the default behavior of the HeaderColumnNameMappingStrategy)

3. Overrides the converter used with @CsvDate to use a custom converter that adds support for the java time api (LocalDate, LocalTime, and LocalDateTime are tested)

Usage

Annotate your bean with something like...

@CsvBindByNameOrder({"Foo","Bar"})
public class MyBean {
    @CsvBindByName(column = "Foo")
    private String foo;
    
    @CsvBindByName(column = "Bar")
    private String bar;
    
    // getter/setters omitted for brevity
}

Setup your writer...

List<MyBean> beans = new ArrayList();
MyBean bean = new MyBean();
bean.setFoo("fooit");
bean.setBar("barit");
beans.add(bean);

StringWriter writer = new StringWriter();
StatefulBeanToCsv<MyBean> csvWriter = new StatefulBeanToCsvBuilder<MyBean>(writer)
    .withApplyQuotesToAll(false)
    .withMappingStrategy(new HeaderColumnNameAndOrderMappingStrategy<>(MyBean.class))
    .build();
csvWriter.write(beans);
return writer.toString();

Results

With the above you should get something like...

Foo,Bar
fooit,barit

ConverterDateAndJavaTime.java

package com.example.csv;

import com.opencsv.bean.ConverterDate;
import com.opencsv.exceptions.CsvDataTypeMismatchException;
import java.lang.reflect.InvocationTargetException;
import java.time.LocalDate;
import java.time.LocalTime;
import java.time.format.DateTimeFormatter;
import java.time.temporal.Temporal;
import java.util.Locale;
import org.apache.commons.lang3.StringUtils;

public class ConverterDateAndJavaTime extends ConverterDate {
    private static String DEFAULT_FORMAT = "yyyyMMdd'T'HHmmss";
    private static String DEFAULT_DATE_ONLY_FORMAT = "yyyyMMdd";
    private static String DEFAULT_TIME_ONLY_FORMAT = "HHmmss";
    private DateTimeFormatter format;
    /**
     * @param type         The type of the field being populated
     * @param formatString The string to use for formatting the date. See
     *                     {@link com.opencsv.bean.CsvDate#value()}
     * @param locale       If not null or empty, specifies the locale used for
     *                     converting locale-specific data types
     * @param errorLocale The locale to use for error messages.
     */
    public ConverterDateAndJavaTime(Class<?> type, String locale, Locale errorLocale, String formatString) {
        super(type, locale, errorLocale, formatString);

        // if the type is LocalDate and using the default format, we know it will fail. Lets use just the date portion of the default date format
        if (DEFAULT_FORMAT.equals(formatString) && LocalDate.class.isAssignableFrom(type)) {
            formatString = DEFAULT_DATE_ONLY_FORMAT;
        } else if (DEFAULT_FORMAT.equals(formatString) && LocalTime.class.isAssignableFrom(type)) {
            formatString = DEFAULT_TIME_ONLY_FORMAT;
        }
        if (this.locale != null) {
            format = DateTimeFormatter.ofPattern(formatString, this.locale);
        } else {
            format = DateTimeFormatter.ofPattern(formatString);
        }
    }

    @Override
    public Object convertToRead(String value) throws CsvDataTypeMismatchException {
        if (StringUtils.isNotBlank(value) && Temporal.class.isAssignableFrom(type)) {
            return convertToTemporal(value);
        }
        return super.convertToRead(value);
    }

    @Override
    public String convertToWrite(Object value) throws CsvDataTypeMismatchException {
        if (value != null && Temporal.class.isAssignableFrom(value.getClass())) {
            return convertFromTemporal((Temporal) value);
        }
        return super.convertToWrite(value);
    }

    private Temporal convertToTemporal(String value) {
        try {
            return (Temporal) type.getMethod("parse", CharSequence.class, DateTimeFormatter.class).invoke(null, value, format);
        } catch (NoSuchMethodException|IllegalAccessException|InvocationTargetException e) {
            throw new RuntimeException("Failed to invoke the parse method of " + type.getName(), e);
        }
    }

    private String convertFromTemporal(Temporal value) {
        return format.format(value);
    }
}

ConverterDateAndJavaTimeTest.java

package com.example.csv;

import static org.junit.Assert.assertEquals;
import static org.junit.Assert.assertNotNull;
import static org.junit.Assert.assertNull;

import com.opencsv.bean.CsvConverter;
import com.opencsv.bean.CsvDate;
import java.time.LocalDate;
import java.time.LocalDateTime;
import java.time.LocalTime;
import java.util.Locale;
import org.junit.Test;

public class ConverterDateAndJavaTimeTest {
    private static final Locale ERROR_LOCALE = Locale.getDefault();
    private static final String LOCALE = ERROR_LOCALE.toString();
    private static String DEFAULT_FORMAT;

    static {
        try {
            DEFAULT_FORMAT = (String) CsvDate.class.getMethod("value").getDefaultValue();
        } catch (NoSuchMethodException e) {
            e.printStackTrace();
        }
    }

    @Test
    public void convertToRead_When_ValueIsBlank_Expect_ToReceiveNull() throws Exception {
        assertNull(converter(LocalDate.class, DEFAULT_FORMAT).convertToRead(""));
    }

    @Test
    public void convertToRead_When_ValueIsValidAndTypeIsLocalDate_Expect_ToReceiveLocalDate() throws Exception {
        LocalDate actual = (LocalDate) converter(LocalDate.class, DEFAULT_FORMAT).convertToRead("20190214");
        assertNotNull(actual);
        assertEquals(2019, actual.getYear());
        assertEquals(2, actual.getMonthValue());
        assertEquals(14, actual.getDayOfMonth());
    }

    @Test
    public void convertToRead_When_ValueIsValidAndTypeIsLocalTime_Expect_ToReceiveLocalTime() throws Exception {
        LocalTime actual = (LocalTime) converter(LocalTime.class, DEFAULT_FORMAT).convertToRead("221546");
        assertNotNull(actual);
        assertEquals(22, actual.getHour());
        assertEquals(15, actual.getMinute());
        assertEquals(46, actual.getSecond());
    }

    @Test
    public void convertToRead_When_ValueIsValidAndTypeIsLocalDateTime_Expect_ToReceiveLocalDateTime() throws Exception {
        LocalDateTime actual = (LocalDateTime) converter(LocalDateTime.class, DEFAULT_FORMAT).convertToRead("20190214T221546");
        assertNotNull(actual);
        assertEquals(2019, actual.getYear());
        assertEquals(2, actual.getMonthValue());
        assertEquals(14, actual.getDayOfMonth());
        assertEquals(22, actual.getHour());
        assertEquals(15, actual.getMinute());
        assertEquals(46, actual.getSecond());
    }

    @Test
    public void convertToWrite_When_ValueIsLocalDate_Expect_ToReceiveFormattedDate() throws Exception {
        String actual = converter(LocalDate.class, DEFAULT_FORMAT).convertToWrite(LocalDate.of(2019, 2, 14));
        assertEquals("20190214", actual);
    }

    @Test
    public void convertToWrite_When_ValueIsLocalTime_Expect_ToReceiveFormattedTime() throws Exception {
        String actual = converter(LocalTime.class, DEFAULT_FORMAT).convertToWrite(LocalTime.of(22, 15, 46));
        assertEquals("221546", actual);
    }

    @Test
    public void convertToWrite_When_ValueIsLocalDateTime_Expect_ToReceiveFormattedDateTime() throws Exception {
        String actual = converter(LocalDateTime.class, DEFAULT_FORMAT).convertToWrite(LocalDateTime.of(2019, 2, 14,22, 15, 46));
        assertEquals("20190214T221546", actual);
    }

    private CsvConverter converter(Class type, String format) {
        return new ConverterDateAndJavaTime(type, LOCALE, ERROR_LOCALE, format);
    }
}

CsvBindByNameOrder.java

package com.example.csv;

import java.lang.annotation.Documented;
import java.lang.annotation.ElementType;
import java.lang.annotation.Retention;
import java.lang.annotation.RetentionPolicy;
import java.lang.annotation.Target;

@Documented
@Retention(RetentionPolicy.RUNTIME)
@Target(ElementType.TYPE)
public @interface CsvBindByNameOrder {
    String[] value() default {};
}

HeaderColumnNameAndOrderMappingStrategy.java

package com.example.csv;

import com.opencsv.bean.BeanField;
import com.opencsv.bean.CsvBindByName;
import com.opencsv.bean.CsvCustomBindByName;
import com.opencsv.bean.HeaderColumnNameMappingStrategy;
import com.opencsv.bean.comparator.LiteralComparator;
import com.opencsv.exceptions.CsvBadConverterException;
import com.opencsv.exceptions.CsvRequiredFieldEmptyException;
import java.util.Arrays;
import org.apache.commons.lang3.StringUtils;

public class HeaderColumnNameAndOrderMappingStrategy<T> extends HeaderColumnNameMappingStrategy<T> {

    public HeaderColumnNameAndOrderMappingStrategy(Class<T> type) {
        setType(type);
    }

    @Override
    public String[] generateHeader(T bean) throws CsvRequiredFieldEmptyException {
        // overriding this method to allow us to preserve the header column name casing

        String[] header = super.generateHeader(bean);
        final int numColumns = findMaxFieldIndex();
        if (!isAnnotationDriven() || numColumns == -1) {
            return header;
        }

        header = new String[numColumns + 1];

        BeanField beanField;
        for (int i = 0; i <= numColumns; i++) {
            beanField = findField(i);
            String columnHeaderName = extractHeaderName(beanField);
            header[i] = columnHeaderName;
        }
        return header;
    }


    @Override
    protected void loadFieldMap() throws CsvBadConverterException {
        // overriding this method to support setting column order by the custom `CsvBindByNameOrder` annotation
        if (writeOrder == null && type.isAnnotationPresent(CsvBindByNameOrder.class)) {
            setColumnOrderOnWrite(
                    new LiteralComparator<>(Arrays.stream(type.getAnnotation(CsvBindByNameOrder.class).value())
                            .map(String::toUpperCase).toArray(String[]::new)));
        }
        super.loadFieldMap();
    }

    private String extractHeaderName(final BeanField beanField) {
        if (beanField == null || beanField.getField() == null
                || beanField.getField().getDeclaredAnnotationsByType(CsvBindByName.class).length == 0) {
            return StringUtils.EMPTY;
        }

        if (beanField.getField().isAnnotationPresent(CsvBindByName.class)) {
            return beanField.getField().getDeclaredAnnotationsByType(CsvBindByName.class)[0].column();
        } else if (beanField.getField().isAnnotationPresent(CsvCustomBindByName.class)) {
            return beanField.getField().getDeclaredAnnotationsByType(CsvCustomBindByName.class)[0].column();
        }
        return StringUtils.EMPTY;

    }
}

HeaderColumnNameAndOrderMappingStrategyTest.java

package com.example.csv;

import com.opencsv.bean.AbstractCsvConverter;
import com.opencsv.bean.BeanField;
import com.opencsv.bean.CsvBindByName;
import com.opencsv.bean.CsvConverter;
import com.opencsv.bean.CsvCustomBindByName;
import com.opencsv.bean.CsvDate;
import com.opencsv.bean.HeaderColumnNameMappingStrategy;
import com.opencsv.bean.comparator.LiteralComparator;
import com.opencsv.exceptions.CsvBadConverterException;
import com.opencsv.exceptions.CsvRequiredFieldEmptyException;
import java.lang.reflect.Field;
import java.util.Arrays;
import org.apache.commons.lang3.StringUtils;

public class HeaderColumnNameAndOrderMappingStrategy<T> extends HeaderColumnNameMappingStrategy<T> {

    public HeaderColumnNameAndOrderMappingStrategy(Class<T> type) {
        setType(type);
    }

    @Override
    public String[] generateHeader(T bean) throws CsvRequiredFieldEmptyException {
        // overriding this method to allow us to preserve the header column name casing

        String[] header = super.generateHeader(bean);
        final int numColumns = findMaxFieldIndex();
        if (!isAnnotationDriven() || numColumns == -1) {
            return header;
        }

        header = new String[numColumns + 1];

        BeanField beanField;
        for (int i = 0; i <= numColumns; i++) {
            beanField = findField(i);
            String columnHeaderName = extractHeaderName(beanField);
            header[i] = columnHeaderName;
        }
        return header;
    }


    @Override
    protected void loadFieldMap() throws CsvBadConverterException {
        // overriding this method to support setting column order by the custom `CsvBindByNameOrder` annotation
        if (writeOrder == null && type.isAnnotationPresent(CsvBindByNameOrder.class)) {
            setColumnOrderOnWrite(
                    new LiteralComparator<>(Arrays.stream(type.getAnnotation(CsvBindByNameOrder.class).value())
                            .map(String::toUpperCase).toArray(String[]::new)));
        }
        super.loadFieldMap();
    }

    @Override
    protected CsvConverter determineConverter(Field field, Class<?> elementType, String locale,
            Class<? extends AbstractCsvConverter> customConverter) throws CsvBadConverterException {
        // overrides the converter for the `CsvDate` to use our custom converter that supports java.time api

        // A custom converter always takes precedence if specified.
        if(customConverter != null && !customConverter.equals(AbstractCsvConverter.class)) {
            return super.determineConverter(field, elementType, locale, customConverter);
        }

        // Perhaps a date instead
        else if(field.isAnnotationPresent(CsvDate.class)) {
            String formatString = field.getAnnotation(CsvDate.class).value();
            return new ConverterDateAndJavaTime(elementType, locale, errorLocale, formatString);
        }

        return super.determineConverter(field, elementType, locale, customConverter);
    }

    private String extractHeaderName(final BeanField beanField) {
        if (beanField == null || beanField.getField() == null) {
            return StringUtils.EMPTY;
        }

        if (beanField.getField().isAnnotationPresent(CsvBindByName.class)) {
            return beanField.getField().getDeclaredAnnotationsByType(CsvBindByName.class)[0].column();
        } else if (beanField.getField().isAnnotationPresent(CsvCustomBindByName.class)) {
            return beanField.getField().getDeclaredAnnotationsByType(CsvCustomBindByName.class)[0].column();
        }
        return StringUtils.EMPTY;

    }
}

what is the version of opencsv that you are using for?

With the version 4.6 is ok, but with the 5.0 there are some issues like deprecated method usages.

With the version 4.6 is ok, but with the 5.0 there are some issues like deprecated method usages.

Hi @ariielm,

I was able to get this working with opencsv version 5.2 by

• changing findMaxFieldIndex() to headerIndex.findMaxIndex() + 1
• removing !isAnnotationDriven() ||

from lines 27-31 of HeaderColumnNameAndOrderMappingStrategy.java.
Lines 27-31 now look like the following in my project and works perfectly:

String[] header = super.generateHeader(bean);
final int numColumns = headerIndex.findMaxIndex();
if (numColumns == -1) {
    return header;
}

Big thanks to @ammmze for creating this. It was exactly what I needed!

Hi @ariielm,
Just try it on 5.3, it works fine with some changes from @nerdmeeting.
Also in 5.3 LiteralComparator was deprecated, that's why loadFieldMap() method will look like this:

    @Override
    protected void loadFieldMap() throws CsvBadConverterException {
        // overriding this method to support setting column order by the custom `CsvBindByNameOrder` annotation
        if (writeOrder == null && type.isAnnotationPresent(CsvBindByNameOrder.class)) {
            List<String> predefinedList = Arrays.stream(type.getAnnotation(CsvBindByNameOrder.class).value())
                .map(String::toUpperCase).collect(Collectors.toList());
            FixedOrderComparator<String> fixedComparator = new FixedOrderComparator<>(predefinedList);
            fixedComparator.setUnknownObjectBehavior(FixedOrderComparator.UnknownObjectBehavior.AFTER);
            Comparator<String> comparator = new ComparatorChain<>(Arrays.asList(
                fixedComparator,
                new NullComparator<>(false),
                new ComparableComparator<>()));
            setColumnOrderOnWrite(comparator);
        }
        super.loadFieldMap();
    }

Thanks for this.

Nice. Are you releasing this under any particular license?

Thanks @ammmze and @nerdmeeting , exactly what I needed!