Detect if file contains XML

Detect whether a File seems to contain XML, taking encoding and whitespace into account

462

With the help of EncodingDetectingInputStream we can make an assumption whether a File is supposed to contain XML or not, without actually parsing the file (for performance reasons). Note that we do NOT make any assumptions of whether the XML is valid.

  public static boolean isXml(File file) {
    FileInputStream fis = null;
    InputStreamReader isr = null;
    try {
      fis = new FileInputStream(file);
      EncodingDetectingInputStream encodingDetectingInputStream = new EncodingDetectingInputStream(fis);

      final Charset charset = encodingDetectingInputStream.getCharset();

      isr = (charset != null) ? // UTF encoding detected by BOM 
        new InputStreamReader(encodingDetectingInputStream, charset) :
        new InputStreamReader(encodingDetectingInputStream);

      char c;

      // Skip leading whitespace
      do {
        c = (char) isr.read();
      } while(c != -1 && Character.isWhitespace(c));

      // If first real character is <, then assume XML. Your case may require reading more characters
      return c == '<';
    }
    catch(IOException ioex) {
      return false; // Probably not an XML file
    }
    finally {
      try {
        if(isr != null)
          isr.close();
        if(fis != null)
          fis.close();
      }
      catch (IOException ioex) {
        // TODO: Handle error
      }
    }