stanfordnlp/CoreNLP · error · IllegalArgumentException

Cannot read from null object!

Error message

Cannot read from null object!

What it means

The DocumentPreprocessor constructor requires a non-null Reader to read the document from. Passing null input makes it impossible to produce any tokenized sentences, so the constructor immediately throws this IllegalArgumentException as a fail-fast guard.

Solutions

1. Check why the Reader is null before constructing — log/verify the source that produced it
2. Fix resource loading: use IOUtils.readerFromString/readerFromFile and handle IOException instead of a possibly-null stream
3. Pass a valid reader, e.g. new DocumentPreprocessor(new BufferedReader(new FileReader(path)))
4. Add an explicit null check on your input before calling the constructor

Example fix

// before
Reader r = MyClass.class.getResourceAsStream(path) != null
    ? new InputStreamReader(MyClass.class.getResourceAsStream(path)) : null;
DocumentPreprocessor dp = new DocumentPreprocessor(r);
// after
InputStream in = MyClass.class.getResourceAsStream(path);
if (in == null) throw new FileNotFoundException("Missing resource: " + path);
DocumentPreprocessor dp = new DocumentPreprocessor(new InputStreamReader(in, StandardCharsets.UTF_8));

if (reader == null) {
  throw new IllegalArgumentException("Document reader must not be null");
}

static Reader requireNonNullReader(Reader r) {
  if (r == null) throw new IllegalArgumentException("Reader is null: source failed to load");
  return r;
}

try {
  DocumentPreprocessor dp = new DocumentPreprocessor(reader);
} catch (IllegalArgumentException e) {
  if ("Cannot read from null object!".equals(e.getMessage())) {
    throw new DocumentLoadException("Input reader was null — check resource loading", e);
  }
  throw e;
}

When it happens

Trigger: Calling new DocumentPreprocessor((Reader) null) or new DocumentPreprocessor(reader, DocType.Plain) with a null reader — commonly when a method that should have produced a Reader (e.g. IOUtils / getResourceAsStream / FileReader) returned null and the result is passed along unchecked.

Common situations: Classpath resource lookup returning null (getResourceAsStream on a missing resource); a file that failed to open silently mapped to null; a config field for the document source left unset.

Related errors

• You can't make a Tokenizer out of a null Lexer!
• Cannot open null document path!
• Attempt to open file with null name
• WordsToSentencesAnnotator: unable to find words/tokens in:
• argsToProperties could not read properties file: " + file

AI-assisted analysis of stanfordnlp/CoreNLP@1b7edd19c4 (2026-09-10). Data as JSON: /api/errors/ae1761c7e3619398. Report an issue: GitHub.

Appendix: source

Thrown at src/edu/stanford/nlp/process/DocumentPreprocessor.java:109

  //From PTB conventions
  private final String[] sentenceFinalFollowers = {")", "]", "}", "\"", "'", "''", "-RRB-", "-RSB-", "-RCB-"};

  private boolean keepEmptySentences; // = false;


  /**
   * Constructs a preprocessor from an existing input stream.
   *
   * @param input An existing reader
   */
  public DocumentPreprocessor(Reader input) {
    this(input,DocType.Plain);
  }

  public DocumentPreprocessor(Reader input, DocType t) {
    if (input == null) {
      throw new IllegalArgumentException("Cannot read from null object!");
    }
    docType = t;
    inputReader = input;
  }

  public DocumentPreprocessor(String docPath) {
    this(docPath, DocType.Plain, "UTF-8");
  }

  public DocumentPreprocessor(String docPath, DocType t) {
    this(docPath, t, "UTF-8");
  }


  /**
   * Constructs a preprocessor from a file at a path, which can be either
   * a filesystem location, a classpath entry, or a URL.
   *

View on GitHub (pinned to 1b7edd19c4)