Skip to content

[format] Enable blob write/read null #7055

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
leaves12138 wants to merge 1 commit into
base: master
Choose a base branch
from
+26 −18
Open

[format] Enable blob write/read null #7055

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
4 changes: 3 additions & 1 deletion paimon-format/src/main/java/org/apache/paimon/format/blob/BlobFormatReader.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -83,7 +83,9 @@ public InternalRow next() {
Blob blob;
long offset = fileMeta.blobOffset(currentPosition) + 4;
long length = fileMeta.blobLength(currentPosition) - 16;
if (in != null) {
if (length == 0) {
blob = null;
} else if (in != null) {
blob = Blob.fromData(readInlineBlob(in, offset, length));
} else {
blob = Blob.fromFile(fileIO, filePath.toString(), offset, length);
Expand Down
20 changes: 10 additions & 10 deletions paimon-format/src/main/java/org/apache/paimon/format/blob/BlobFormatWriter.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -55,21 +55,21 @@ public BlobFormatWriter(PositionOutputStream out) {
@Override
public void addElement(InternalRow element) throws IOException {
checkArgument(element.getFieldCount() == 1, "BlobFormatWriter only support one field.");
checkArgument(!element.isNullAt(0), "BlobFormatWriter only support non-null blob.");
Blob blob = element.getBlob(0);

long previousPos = out.getPos();
crc32.reset();

write(MAGIC_NUMBER_BYTES);
Copy link
Contributor

@JingsongLi JingsongLi Jan 17, 2026

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Why we need to write magic number for null blob?

try (SeekableInputStream in = blob.newInputStream()) {
int bytesRead = in.read(tmpBuffer);
while (bytesRead >= 0) {
write(tmpBuffer, bytesRead);
bytesRead = in.read(tmpBuffer);

if (!element.isNullAt(0)) {
Blob blob = element.getBlob(0);
try (SeekableInputStream in = blob.newInputStream()) {
int bytesRead = in.read(tmpBuffer);
while (bytesRead >= 0) {
write(tmpBuffer, bytesRead);
bytesRead = in.read(tmpBuffer);
}
}
}

crc32.reset();
long binLength = out.getPos() - previousPos + 12;
lengths.add(binLength);
byte[] lenBytes = longToLittleEndian(binLength);
Expand Down
20 changes: 13 additions & 7 deletions paimon-format/src/test/java/org/apache/paimon/format/blob/BlobFileFormatTest.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -78,11 +78,13 @@ private void innerTest(boolean blobAsDescriptor) throws IOException {

// write
FormatWriterFactory writerFactory = format.createWriterFactory(rowType);
List<byte[]> blobs = Arrays.asList("hello".getBytes(), "world".getBytes());
List<byte[]> blobs =
Arrays.asList(
"hello".getBytes(), "world".getBytes(), null, "asldfjasldkfjas".getBytes());
try (PositionOutputStream out = fileIO.newOutputStream(file, false)) {
FormatWriter formatWriter = writerFactory.create(out, null);
for (byte[] bytes : blobs) {
formatWriter.addElement(GenericRow.of(new BlobData(bytes)));
formatWriter.addElement(GenericRow.of(bytes == null ? null : new BlobData(bytes)));
}
formatWriter.close();
}
Expand All @@ -96,13 +98,17 @@ private void innerTest(boolean blobAsDescriptor) throws IOException {
.createReader(context)
.forEachRemaining(
row -> {
Blob blob = row.getBlob(0);
if (blobAsDescriptor) {
assertThat(blob).isInstanceOf(BlobRef.class);
if (row.isNullAt(0)) {
result.add(null);
} else {
assertThat(blob).isInstanceOf(BlobData.class);
Blob blob = row.getBlob(0);
if (blobAsDescriptor) {
assertThat(blob).isInstanceOf(BlobRef.class);
} else {
assertThat(blob).isInstanceOf(BlobData.class);
}
result.add(blob.toData());
}
result.add(blob.toData());
});

// assert
Expand Down