apache/flink · error · IllegalArgumentException

Type for included field {} should not be null.

Error message

Type for included field {} should not be null.

What it means

Thrown by GenericCsvInputFormat.setFieldsGeneric(boolean[], Class<?>[]) when an included field (mask[i]==true) is paired with a null type entry. While null entries are allowed in the simple setFieldTypesGeneric overload (meaning 'skip'), in the mask variant every included position must have a concrete type because the parser must be constructed for it.

Source

Thrown at flink-core/src/main/java/org/apache/flink/api/common/io/GenericCsvInputFormat.java:283

    protected void setFieldsGeneric(boolean[] includedMask, Class<?>[] fieldTypes) {
        checkNotNull(includedMask);
        checkNotNull(fieldTypes);

        ArrayList<Class<?>> types = new ArrayList<Class<?>>();

        // check if types are valid for included fields
        int typeIndex = 0;
        for (int i = 0; i < includedMask.length; i++) {

            if (includedMask[i]) {
                if (typeIndex > fieldTypes.length - 1) {
                    throw new IllegalArgumentException(
                            "Missing type for included field " + i + ".");
                }
                Class<?> type = fieldTypes[typeIndex++];

                if (type == null) {
                    throw new IllegalArgumentException(
                            "Type for included field " + i + " should not be null.");
                } else {
                    // check if we support parsers for this type
                    if (FieldParser.getParserForType(type) == null) {
                        throw new IllegalArgumentException(
                                "The type '"
                                        + type.getName()
                                        + "' is not supported for the CSV input format.");
                    }
                    types.add(type);
                }
            }
        }

        this.fieldTypes = types.toArray(new Class<?>[types.size()]);
        this.fieldIncluded = includedMask;
    }

View on GitHub (pinned to 2f3c205e92)

Solutions

1. Provide a concrete supported type for every included (mask true) position.
2. If a column should be skipped, set its mask entry to false rather than supplying a null type.
3. Validate with: if (mask[i]) Objects.requireNonNull(types[typeIndex]).

Example fix

// before
boolean[] mask = {true, true};
Class<?>[] types = {Integer.class, null}; // null for included field
format.setFieldsGeneric(mask, types);

// after
boolean[] mask = {true, false}; // skip second column instead
Class<?>[] types = {Integer.class};
format.setFieldsGeneric(mask, types);

int ti = 0;
for (int i = 0; i < includedMask.length; i++) {
    if (includedMask[i]) {
        Objects.requireNonNull(fieldTypes[ti], "type for included field " + i + " must not be null");
        ti++;
    }
}
format.setFieldsGeneric(includedMask, fieldTypes);

static boolean noNullTypeForIncluded(boolean[] mask, Class<?>[] types) {
    int ti = 0;
    for (boolean b : mask) {
        if (b) {
            if (types[ti] == null) return false;
            ti++;
        }
    }
    return true;
}

When it happens

Trigger: Building the fieldTypes array with a null slot that aligns with a true position in the mask; reusing a partially-null types array from another context.

Common situations: Refactoring that nulls out a type but leaves the mask position selected; mixing the 'skip via null' convention from the other overload into this mask-based API.

Related errors

• Field types must not be null.
• Delimiter must not be null
• Missing type for included field {}.
• The type '{}' is not supported for the CSV input format.
• Field indices must not be smaller than zero.

AI-assisted analysis of apache/flink@2f3c205e92 (2026-08-14). Data as JSON: /api/errors/a134f59699f3747d. Report an issue: GitHub.