stanfordnlp/CoreNLP · error · RuntimeException

invalid line

Error message

invalid line '%s'

What it means

findAnnotation scans gigaword document lines looking for the '</DOC>' terminator. A line that merely CONTAINS '</DOC>' but is not exactly that string means the document markup is malformed (trailing content on the closing tag), so it throws a RuntimeException with the offending line rather than silently mis-parsing the document.

Solutions

1. Normalize the input file: strip trailing whitespace/CR characters after '</DOC>' and re-save.
2. Pre-process each line with line.trim() comparison logic or fix the file so the closing tag stands alone.
3. Re-download or re-extract the gigaword archive to rule out extraction corruption.

Example fix

// before
if (line.contains("</DOC>")) {
  throw new RuntimeException(String.format("invalid line '%s'", line));
}
// after
if (line.trim().equals("</DOC>")) {
  break;
}

if (line.contains("</DOC>") && !line.trim().equals("</DOC>")) {
  logger.warning("Malformed </DOC> line: " + line);
  line = line.trim(); // or sanitize the corpus beforehand
}

try {
  return it.next();
} catch (RuntimeException e) {
  if (e.getMessage() != null && e.getMessage().startsWith("invalid line")) {
    logger.warning("Skipping malformed document boundary");
  }
  throw e;
}

When it happens

Trigger: Reading a gigaword file whose '</DOC>' closing tag has trailing whitespace, characters, or CRLF remnants making line.equals("</DOC>") false while line.contains("</DOC>") is true.

Common situations: Corrupted or re-encoded gigaword archives; files edited by tools that appended spaces or converted line endings; concatenated documents leaving stray markup after the closing tag.

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

• Unparsable sentence:
• error:\n \ninput:\n
• Error (line ): 10 fields expected but are present
• Bad data format:
• Cannot find matching labelled span for

AI-assisted analysis of stanfordnlp/CoreNLP@1b7edd19c4 (2026-09-10). Data as JSON: /api/errors/2019d84997b9d976. Report an issue: GitHub.

Appendix: source

Thrown at src/edu/stanford/nlp/time/ParsedGigawordReader.java:94

      private Annotation findAnnotation() {
        if (this.reader == null) {
          return null;
        }
        try {
          String line;
          StringBuilder doc = new StringBuilder();
          while ((line = this.reader.readLine()) != null) {
            doc.append(line);
            doc.append('\n');
//            if(line.contains("<DOC id")){
//              log.info(line);
//            }
            if (line.equals("</DOC>")) {
              break;
            }
            if (line.contains("</DOC>")) {
              throw new RuntimeException(String.format("invalid line '%s'", line));
            }
          }
          if (line == null) {
            this.reader.close();
            this.reader = findReader();
          }
          String xml = doc.toString().replaceAll("&", "&amp;");
          if(xml == null || xml.equals("")) {
            return findAnnotation();
          }

          xml = xml.replaceAll("num=([0-9]+) (.*)", "num=\"$1\" $2");
          xml = xml.replaceAll("sid=(.*)>", "sid=\"$1\">");
          xml = xml.replaceAll("</SENT>\n</DOC>", "</SENT>\n</TEXT>\n</DOC>");
          xml = new String(xml.getBytes(), "UTF8");
          //log.info("This is what goes in:\n" + xml);
          return toAnnotation(xml);
        } catch (IOException e) {

View on GitHub (pinned to 1b7edd19c4)