953N/Apackage org.opensolaris.opengrok.analysis;

953N/A

953N/Aimport java.io.IOException;

953N/Aimport java.io.InputStream;

953N/Aimport java.io.InputStreamReader;

953N/Aimport java.nio.charset.Charset;

953N/A

953N/Aimport org.apache.lucene.document.Document;

953N/A

953N/Apublic abstract class TextAnalyzer extends FileAnalyzer {

953N/A public TextAnalyzer(FileAnalyzerFactory factory) {

953N/A super(factory);

953N/A }

953N/A

953N/A public final void analyze(Document doc, InputStream in) throws IOException {

953N/A String charset = null;

953N/A

953N/A in.mark(3);

953N/A

953N/A byte[] head = new byte[3];

953N/A int br = in.read(head, 0, 3);

953N/A

953N/A if (br >= 2) {

953N/A if ((head[0] == (byte)0xFE && head[1] == (byte)0xFF) || (head[0] == (byte)0xFF && (byte)head[1] == (byte)0xFE)) {

953N/A charset = "UTF16";

953N/A in.reset();

953N/A }

953N/A }

953N/A if (br >= 3) {

953N/A if (head[0] == (byte)0xEF && head[1] == (byte)0xBB && head[2] == (byte)0xBF) {

953N/A /* InputStreamReader does not properly discard BOM on UTF8 streams,

953N/A charset = "UTF8";

953N/A }

953N/A }

953N/A

953N/A if (charset == null) {

953N/A in.reset();

953N/A charset = Charset.defaultCharset().name();

953N/A }

953N/A

953N/A analyze(doc, new InputStreamReader(in, charset));

953N/A }

953N/A

953N/A protected abstract void analyze(Document doc, InputStreamReader reader) throws IOException;

953N/A}