MuntashirAkon/AppManager · error · java.io.IOException

Unknown token with type

Error message

Unknown token ${token} with type ${type}

What it means

consumeToken() switches over the XmlPullParser token constants (START_DOCUMENT, START_TAG, TEXT, END_TAG, END_DOCUMENT). A token value outside the protocol's known set means the stream is desynchronized or corrupt, so it throws IOException "Unknown token <token> with type <type>".

Solutions

1. Restart parsing from a fresh stream positioned at the document start instead of continuing after a failure.
2. Verify the file's integrity and that it was produced by a compatible version of the binary XML writer.
3. Confirm the magic header was consumed exactly once (setInput does this); do not skip or re-read the header manually.
4. Log the offending token/type values to diagnose the misalignment.

Example fix

// before
parser.next(); // after an earlier IOException, stream is misaligned
parser.next(); // Unknown token ...
// after
parser = newBinaryXmlPullParser(); // fresh parser
parser.setInput(freshInputStream, null);

// Ensure a clean start for every parse attempt:
InputStream fresh = new FileInputStream(file);
byte[] magic = new byte[4];
if (fresh.read(magic) != 4 || !Arrays.equals(magic, PROTOCOL_MAGIC_VERSION_0)) {
    throw new IOException("Not binary XML");
}

try {
    parser.setInput(is, null);
    // parse
} catch (IOException e) {
    // stream is desynchronized or corrupt: re-create parser + stream, do not resume
}

When it happens

Trigger: Reading a token byte that maps to no known token — corrupted file, wrong stream offset, a stream produced by an incompatible writer/version, or byte misalignment after a prior parse failure.

Common situations: Continuing to parse after an earlier exception left the stream mid-token; passing non-binary-XML data that slipped past other checks; truncated files where data bytes are interpreted as tokens; protocol version mismatch.

Related errors

• Unexpected data type
• ${e}
• e.getMessage()
• e
• ${getPositionDescription()}

AI-assisted analysis of MuntashirAkon/AppManager@0152f468fc (2026-09-12). Data as JSON: /api/errors/707b8ae73ca4fe85. Report an issue: GitHub.

Appendix: source

Thrown at libcore/compat/src/main/java/io/github/muntashirakon/compat/xml/BinaryXmlPullParser.java:292

            case XmlPullParser.TEXT:
            case XmlPullParser.CDSECT:
            case XmlPullParser.PROCESSING_INSTRUCTION:
            case XmlPullParser.COMMENT:
            case XmlPullParser.DOCDECL:
            case XmlPullParser.IGNORABLE_WHITESPACE: {
                mCurrentName = null;
                mCurrentText = mIn.readUTF();
                if (mAttributeCount > 0) resetAttributes();
                break;
            }
            case XmlPullParser.ENTITY_REF: {
                mCurrentName = mIn.readUTF();
                mCurrentText = resolveEntity(mCurrentName);
                if (mAttributeCount > 0) resetAttributes();
                break;
            }
            default: {
                throw new IOException("Unknown token " + token + " with type " + type);
            }
        }
    }

    /**
     * When the current tag is {@link #TEXT}, consume all subsequent "text"
     * events, as described by {@link #next}. When finished, the current event
     * will still be {@link #TEXT}.
     */
    private void consumeAdditionalText() throws IOException, XmlPullParserException {
        String combinedText = mCurrentText;
        while (true) {
            final int token = peekNextExternalToken();
            switch (token) {
                case COMMENT:
                case PROCESSING_INSTRUCTION:
                    // Quietly consumed
                    consumeToken();

View on GitHub (pinned to 0152f468fc)