apache/druid · error · UnsupportedOperationException

CharSequence not supported

Error message

CharSequence not supported

What it means

When pulling a local segment, LocalDataSegmentPuller exposes the segment archive as a JavaCompiler FileObject for tools that inspect it. getCharContent is deliberately unimplemented for this FileObject, so any caller requesting the file's character content receives this UnsupportedOperationException.

Source

Thrown at server/src/main/java/org/apache/druid/segment/loading/LocalDataSegmentPuller.java:94

        return new FileInputStream(file);
      }

      @Override
      public OutputStream openOutputStream() throws IOException
      {
        return new FileOutputStream(file);
      }

      @Override
      public Reader openReader(boolean ignoreEncodingErrors) throws IOException
      {
        return Files.newReader(file, Charset.defaultCharset());
      }

      @Override
      public CharSequence getCharContent(boolean ignoreEncodingErrors)
      {
        throw new UOE("CharSequence not supported");
      }

      @Override
      public Writer openWriter() throws IOException
      {
        return Files.newWriter(file, Charset.defaultCharset());
      }

      @Override
      public long getLastModified()
      {
        return file.lastModified();
      }

      @Override
      public boolean delete()
      {
        return file.delete();

View on GitHub (pinned to 9b90983fd2)

Solutions

1. Don't call getCharContent on this FileObject; open it as bytes instead (openInputStream) and read the zip binary.
2. Copy the file to a regular File and read text from there if a CharSequence is required.
3. Guard calls: check the FileObject implementation and fall back to a byte-stream reader.

Example fix

// before
CharSequence text = fileObject.getCharContent(true);
// after
try (InputStream in = fileObject.openInputStream()) {
  byte[] data = ByteStreams.toByteArray(in); // read segment as bytes
}

try {
  CharSequence cs = fileObject.getCharContent(true);
} catch (UnsupportedOperationException e) {
  try (InputStream in = fileObject.openInputStream()) {
    byte[] bytes = in.readAllBytes(); // handle segment as binary
  }
}

When it happens

Trigger: Calling getCharContent(ignoreEncodingErrors) on the FileObject returned by LocalDataSegmentPuller.get() — typically from tooling or extension code that tries to read the zip as text.

Common situations: Compiler/annotation-processing style tooling (javax.tools.FileObject consumers) applied to segment archives; custom code assuming all FileObjects support text reads.

Related errors

• HDFS Reader not supported
• HDFS CharSequence not supported
• HDFS Writer not supported
• Casting to float type is not supported
• Casting to long type is not supported

AI-assisted analysis of apache/druid@9b90983fd2 (2026-09-07). Data as JSON: /api/errors/344eb3f7b22dfa13. Report an issue: GitHub.