stanfordnlp/CoreNLP · error · RuntimeException

expecting BEGIN block; got

Error message

expecting BEGIN block; got 

What it means

confirmBeginBlock validates that the current line of a text-format parser grammar file begins a BEGIN block. If the file ends before any line is read (line == null) or the line does not start with "BEGIN", a RuntimeException is thrown naming the grammar file and the offending line. This means the grammar text file is malformed or is not a text grammar at all.

Solutions

1. Check the file at the path printed in the exception: it must be a text grammar whose sections start with BEGIN lines (e.g. BEGIN_GENERIC, BEGIN_STATE)
2. If you actually have a serialized model, load it with getParserFromSerializedFile / LexicalizedParser.loadModel instead of the text-file path
3. Re-download or restore the original grammar file — it may be truncated or corrupted
4. Ensure the file is plain UTF-8 with no BOM and no stray leading blank/garbage line before the first BEGIN
5. Enable parser verbosity (-vv or set op.testOptions.verbose) and compare with a known-good shipped grammar like englishPCFG.ser.gz format

Example fix

// before
LexicalizedParser lp = LexicalizedParser.loadModel("myGrammar.txt"); // binary file read as text grammar
// after
LexicalizedParser lp = LexicalizedParser.loadModel("englishPCFG.ser.gz"); // correct serialized model

// Java: sanity-check the file looks like a text grammar before loading
BufferedReader r = Files.newBufferedReader(Paths.get(path), StandardCharsets.UTF_8);
String first = r.readLine();
r.close();
if (first == null || !first.startsWith("BEGIN")) throw new IllegalArgumentException("not a text grammar: " + path);

When it happens

Trigger: Calling LexicalizedParser.getParserFromTextFile (via loadParserFromTextFile / the main pd path after a serialized load fails) on a file whose structure is not the expected BEGIN/END sectioned grammar text format.

Common situations: Passing a serialized (binary) parser model path or an unrelated text file as a text grammar; a grammar file that was truncated during download/copy; editing a grammar file and deleting or renaming the BEGIN header; CRLF or BOM corruption changing the first characters of the header line.

Understand the failure class

Background: "Invalid ... format", "must be in format X", "does not look like a ..." — invalid argument format errors across CLI tools and libraries — this error's family across 17 libraries.

Related errors

• Unexpected number of field , expected >= for line (,):
• First line of input file should be header definition
• format error in embeddings
• format error unexpected featureFactory line:
• weights format error

AI-assisted analysis of stanfordnlp/CoreNLP@1b7edd19c4 (2026-09-10). Data as JSON: /api/errors/162c2fd667f90e76. Report an issue: GitHub.

Appendix: source

Thrown at src/edu/stanford/nlp/parser/lexparser/LexicalizedParser.java:508

        dg.writeData(out);
      }
      out.println();
      log.info(".");

      out.flush();
      out.close();
      log.info("done.");
    } catch (IOException e) {
      log.info("Trouble saving parser data to ASCII format.");
      throw new RuntimeIOException(e);
    }
  }

  private static void confirmBeginBlock(String file, String line) {
    if (line == null) {
      throw new RuntimeException(file + ": expecting BEGIN block; got end of file.");
    } else if (! line.startsWith("BEGIN")) {
      throw new RuntimeException(file + ": expecting BEGIN block; got " + line);
    }
  }

  protected static LexicalizedParser getParserFromTextFile(String textFileOrUrl, Options op) {
    try (BufferedReader in = IOUtils.readerFromString(textFileOrUrl)) {
      Timing tim = new Timing();

      String line = in.readLine();
      confirmBeginBlock(textFileOrUrl, line);
      op.readData(in);

      line = in.readLine();
      confirmBeginBlock(textFileOrUrl, line);
      Index<String> stateIndex = HashIndex.loadFromReader(in);

      line = in.readLine();
      confirmBeginBlock(textFileOrUrl, line);
      Index<String> wordIndex = HashIndex.loadFromReader(in);

View on GitHub (pinned to 1b7edd19c4)