List of usage examples for java.io PushbackInputStream read
public int read(byte b[]) throws IOException
b.length bytes of data from this input stream into an array of bytes. From source file:Main.java
/** * Creates a reader from the given input stream and encoding. * This method assumes the input stream working buffer is at least * 128 byte long. The input stream is restored before this method * returns. The 4 first bytes are skipped before creating the reader. *//*from www. j a v a 2 s. c o m*/ protected static Reader createXMLDeclarationReader(PushbackInputStream pbis, String enc) throws IOException { byte[] buf = new byte[128]; int len = pbis.read(buf); if (len > 0) { pbis.unread(buf, 0, len); } return new InputStreamReader(new ByteArrayInputStream(buf, 4, len), enc); }
From source file:Main.java
/** * Checks the InputStream if it contains GZIP compressed data * * @param inputStream InputStream to be checked * @return true or false if the stream contains GZIP compressed data * @throws java.io.IOException/*w w w .j a va2s . c om*/ */ public static boolean isInputStreamGZIPCompressed(final PushbackInputStream inputStream) throws IOException { if (inputStream == null) return false; byte[] signature = new byte[2]; int readStatus = inputStream.read(signature); inputStream.unread(signature); int streamHeader = ((int) signature[0] & 0xff) | ((signature[1] << 8) & 0xff00); return readStatus == 2 && GZIPInputStream.GZIP_MAGIC == streamHeader; }
From source file:eu.europeana.uim.sugarcrmclient.internal.ExtendedSaajSoapMessageFactory.java
/** * Detects if the incoming stream is Gzip encoded * /*from w w w. j a v a2 s. co m*/ * @param pb * @return an InputStream/GZIPInputStream * @throws IOException */ public static InputStream decompressStream(PushbackInputStream pb) throws IOException { byte[] signature = new byte[2]; pb.read(signature); pb.unread(signature); if (signature[0] == 31 && signature[1] == -117) return new GZIPInputStream(pb); else return pb; }
From source file:it.unifi.rcl.chess.traceanalysis.Trace.java
/** * Checks if the input stream is compressed, and in case it returns a GZIPInputStream * @param stream/*from w w w. ja va 2s . c om*/ * @return * @throws IOException */ private static boolean checkGZIP(InputStream stream) throws IOException { PushbackInputStream pb = new PushbackInputStream(stream, 2); //we need a pushbackstream to look ahead byte[] signature = new byte[2]; pb.read(signature); //read the signature pb.unread(signature); //push back the signature to the stream if (signature[0] == (byte) 0x1f && signature[1] == (byte) 0x8b) //check if matches standard gzip magic number return true; else return false; }
From source file:eu.scape_project.up2ti.container.ArcContainer.java
@Override public void init(String containerFileName, InputStream containerFileStream) throws IOException { this.containerFileName = containerFileName; // Read first two bytes to check if we have a gzipped input stream PushbackInputStream pb = new PushbackInputStream(containerFileStream, 2); byte[] signature = new byte[2]; pb.read(signature); pb.unread(signature);/* ww w . j a v a2 s .c om*/ // use compressed reader if gzip magic number is matched if (signature[0] == (byte) 0x1f && signature[1] == (byte) 0x8b) { reader = ArcReaderFactory.getReaderCompressed(pb); } else { reader = ArcReaderFactory.getReaderUncompressed(pb); } archiveRecords = new ArrayList<ArcRecordBase>(); // initialise object by create temporary files and the bidirectional // file-record map. arcRecContentsToTempFiles(); }
From source file:eu.europeana.uim.sugarcrmclient.internal.ExtendedSaajSoapMessageFactory.java
/** * Checks for the UTF-8 Byte Order Mark, and removes it if present. The SAAJ RI cannot cope with these BOMs. * * @see <a href="http://jira.springframework.org/browse/SWS-393">SWS-393</a> * @see <a href="http://unicode.org/faq/utf_bom.html#22">UTF-8 BOMs</a> *///from w w w. j av a 2 s .c o m private InputStream checkForUtf8ByteOrderMark(InputStream inputStream) throws IOException { PushbackInputStream pushbackInputStream = new PushbackInputStream(new BufferedInputStream(inputStream), 3); byte[] bom = new byte[3]; if (pushbackInputStream.read(bom) != -1) { // check for the UTF-8 BOM, and remove it if there. See SWS-393 if (!(bom[0] == (byte) 0xEF && bom[1] == (byte) 0xBB && bom[2] == (byte) 0xBF)) { pushbackInputStream.unread(bom); } } return pushbackInputStream; }
From source file:z.hol.net.http.entity.DeflateDecompressingEntity.java
/** * Returns the non-null InputStream that should be returned to by all requests to * {@link #getContent()}.//from www . ja v a 2 s. c o m * * @return a non-null InputStream * @throws IOException if there was a problem */ @Override InputStream getDecompressingInputStream(final InputStream wrapped) throws IOException { /* * A zlib stream will have a header. * * CMF | FLG [| DICTID ] | ...compressed data | ADLER32 | * * * CMF is one byte. * * * FLG is one byte. * * * DICTID is four bytes, and only present if FLG.FDICT is set. * * Sniff the content. Does it look like a zlib stream, with a CMF, etc? c.f. RFC1950, * section 2.2. http://tools.ietf.org/html/rfc1950#page-4 * * We need to see if it looks like a proper zlib stream, or whether it is just a deflate * stream. RFC2616 calls zlib streams deflate. Confusing, isn't it? That's why some servers * implement deflate Content-Encoding using deflate streams, rather than zlib streams. * * We could start looking at the bytes, but to be honest, someone else has already read * the RFCs and implemented that for us. So we'll just use the JDK libraries and exception * handling to do this. If that proves slow, then we could potentially change this to check * the first byte - does it look like a CMF? What about the second byte - does it look like * a FLG, etc. */ /* We read a small buffer to sniff the content. */ byte[] peeked = new byte[6]; PushbackInputStream pushback = new PushbackInputStream(wrapped, peeked.length); int headerLength = pushback.read(peeked); if (headerLength == -1) { throw new IOException("Unable to read the response"); } /* We try to read the first uncompressed byte. */ byte[] dummy = new byte[1]; Inflater inf = new Inflater(); try { int n; while ((n = inf.inflate(dummy)) == 0) { if (inf.finished()) { /* Not expecting this, so fail loudly. */ throw new IOException("Unable to read the response"); } if (inf.needsDictionary()) { /* Need dictionary - then it must be zlib stream with DICTID part? */ break; } if (inf.needsInput()) { inf.setInput(peeked); } } if (n == -1) { throw new IOException("Unable to read the response"); } /* * We read something without a problem, so it's a valid zlib stream. Just need to reset * and return an unused InputStream now. */ pushback.unread(peeked, 0, headerLength); return new InflaterInputStream(pushback); } catch (DataFormatException e) { /* Presume that it's an RFC1951 deflate stream rather than RFC1950 zlib stream and try * again. */ pushback.unread(peeked, 0, headerLength); return new InflaterInputStream(pushback, new Inflater(true)); } }
From source file:eu.scape_project.tpid.ContainerProcessing.java
/** * Prepare input// w ww . j a v a 2 s . com * * @param pt * @throws IOException IO Error * @throws java.lang.InterruptedException */ public void prepareInput(Path pt) throws InterruptedException, IOException { FileSystem fs = FileSystem.get(context.getConfiguration()); InputStream containerFileStream = fs.open(pt); String containerFileName = pt.getName(); ArcReader reader; // Read first two bytes to check if we have a gzipped input stream PushbackInputStream pb = new PushbackInputStream(containerFileStream, 2); byte[] signature = new byte[2]; pb.read(signature); pb.unread(signature); // use compressed reader if gzip magic number is matched if (signature[0] == (byte) 0x1f && signature[1] == (byte) 0x8b) { reader = ArcReaderFactory.getReaderCompressed(pb); } else { reader = ArcReaderFactory.getReaderUncompressed(pb); } long currTM = System.currentTimeMillis(); String unpackHdfsPath = conf.get("unpack_hdfs_path", "tpid_unpacked"); String hdfsUnpackDirStr = StringUtils.normdir(unpackHdfsPath, Long.toString(currTM)); String hdfsJoboutputPath = conf.get("tooloutput_hdfs_path", "tpid_tooloutput"); String hdfsOutputDirStr = StringUtils.normdir(hdfsJoboutputPath, Long.toString(currTM)); Iterator<ArcRecordBase> arcIterator = reader.iterator(); // Number of files which should be processed per invokation int numItemsPerInvocation = conf.getInt("num_items_per_task", 50); int numItemCounter = numItemsPerInvocation; // List of input files to be processed String inliststr = ""; // List of output files to be generated String outliststr = ""; try { while (arcIterator.hasNext()) { ArcRecordBase arcRecord = arcIterator.next(); String recordKey = getRecordKey(arcRecord, containerFileName); String outFileName = RandomStringUtils.randomAlphabetic(25); String hdfsPathStr = hdfsUnpackDirStr + outFileName; Path hdfsPath = new Path(hdfsPathStr); String outputFileSuffix = conf.get("output_file_suffix", ".fits.xml"); String hdfsOutPathStr = hdfsOutputDirStr + outFileName + outputFileSuffix; FSDataOutputStream hdfsOutStream = fs.create(hdfsPath); ArcUtils.recordToOutputStream(arcRecord, hdfsOutStream); Text key = new Text(recordKey); Text value = new Text(fs.getHomeDirectory() + File.separator + hdfsOutPathStr); mos.write("keyfilmapping", key, value); String scapePlatformInvoke = conf.get("scape_platform_invoke", "fits dirxml"); Text ptmrkey = new Text(scapePlatformInvoke); // for the configured number of items per invokation, add the // files to the input and output list of the command. inliststr += "," + fs.getHomeDirectory() + File.separator + hdfsPathStr; outliststr += "," + fs.getHomeDirectory() + File.separator + hdfsOutPathStr; if (numItemCounter > 1 && arcIterator.hasNext()) { numItemCounter--; } else if (numItemCounter == 1 || !arcIterator.hasNext()) { inliststr = inliststr.substring(1); // cut off leading comma outliststr = outliststr.substring(1); // cut off leading comma String pattern = conf.get("tomar_param_pattern", "%1$s %2$s"); String ptMrStr = StringUtils.formatCommandOutput(pattern, inliststr, outliststr); Text ptmrvalue = new Text(ptMrStr); // emit tomar input line where the key is the tool invokation // (tool + operation) and the value is the parameter list // where input and output strings contain file lists. mos.write("tomarinput", ptmrkey, ptmrvalue); numItemCounter = numItemsPerInvocation; inliststr = ""; outliststr = ""; } } } catch (Exception ex) { mos.write("error", new Text("Error"), new Text(pt.toString())); } }
From source file:org.mcxiaoke.commons.http.impl.DeflateDecompressingEntity.java
/** * Returns the non-null InputStream that should be returned to by all * requests to {@link #getContent()}.// w ww. j a v a 2 s . c om * * @return a non-null InputStream * @throws IOException * if there was a problem */ @Override InputStream getDecompressingInputStream(final InputStream wrapped) throws IOException { /* * A zlib stream will have a header. * * CMF | FLG [| DICTID ] | ...compressed data | ADLER32 | * * * CMF is one byte. * * * FLG is one byte. * * * DICTID is four bytes, and only present if FLG.FDICT is set. * * Sniff the content. Does it look like a zlib stream, with a CMF, etc? * c.f. RFC1950, section 2.2. http://tools.ietf.org/html/rfc1950#page-4 * * We need to see if it looks like a proper zlib stream, or whether it * is just a deflate stream. RFC2616 calls zlib streams deflate. * Confusing, isn't it? That's why some servers implement deflate * Content-Encoding using deflate streams, rather than zlib streams. * * We could start looking at the bytes, but to be honest, someone else * has already read the RFCs and implemented that for us. So we'll just * use the JDK libraries and exception handling to do this. If that * proves slow, then we could potentially change this to check the first * byte - does it look like a CMF? What about the second byte - does it * look like a FLG, etc. */ /* We read a small buffer to sniff the content. */ byte[] peeked = new byte[6]; PushbackInputStream pushback = new PushbackInputStream(wrapped, peeked.length); int headerLength = pushback.read(peeked); if (headerLength == -1) { throw new IOException("Unable to read the response"); } /* We try to read the first uncompressed byte. */ byte[] dummy = new byte[1]; Inflater inf = new Inflater(); try { int n; while ((n = inf.inflate(dummy)) == 0) { if (inf.finished()) { /* Not expecting this, so fail loudly. */ throw new IOException("Unable to read the response"); } if (inf.needsDictionary()) { /* * Need dictionary - then it must be zlib stream with DICTID * part? */ break; } if (inf.needsInput()) { inf.setInput(peeked); } } if (n == -1) { throw new IOException("Unable to read the response"); } /* * We read something without a problem, so it's a valid zlib stream. * Just need to reset and return an unused InputStream now. */ pushback.unread(peeked, 0, headerLength); return new InflaterInputStream(pushback); } catch (DataFormatException e) { /* * Presume that it's an RFC1951 deflate stream rather than RFC1950 * zlib stream and try again. */ pushback.unread(peeked, 0, headerLength); return new InflaterInputStream(pushback, new Inflater(true)); } }
From source file:com.fanfou.app.opensource.http.support.DeflateDecompressingEntity.java
/** * Returns the non-null InputStream that should be returned to by all * requests to {@link #getContent()}.// w w w . j a v a2s .co m * * @return a non-null InputStream * @throws IOException * if there was a problem */ @Override InputStream getDecompressingInputStream(final InputStream wrapped) throws IOException { /* * A zlib stream will have a header. * * CMF | FLG [| DICTID ] | ...compressed data | ADLER32 | * * * CMF is one byte. * * * FLG is one byte. * * * DICTID is four bytes, and only present if FLG.FDICT is set. * * Sniff the content. Does it look like a zlib stream, with a CMF, etc? * c.f. RFC1950, section 2.2. http://tools.ietf.org/html/rfc1950#page-4 * * We need to see if it looks like a proper zlib stream, or whether it * is just a deflate stream. RFC2616 calls zlib streams deflate. * Confusing, isn't it? That's why some servers implement deflate * Content-Encoding using deflate streams, rather than zlib streams. * * We could start looking at the bytes, but to be honest, someone else * has already read the RFCs and implemented that for us. So we'll just * use the JDK libraries and exception handling to do this. If that * proves slow, then we could potentially change this to check the first * byte - does it look like a CMF? What about the second byte - does it * look like a FLG, etc. */ /* We read a small buffer to sniff the content. */ final byte[] peeked = new byte[6]; final PushbackInputStream pushback = new PushbackInputStream(wrapped, peeked.length); final int headerLength = pushback.read(peeked); if (headerLength == -1) { throw new IOException("Unable to read the response"); } /* We try to read the first uncompressed byte. */ final byte[] dummy = new byte[1]; final Inflater inf = new Inflater(); try { int n; while ((n = inf.inflate(dummy)) == 0) { if (inf.finished()) { /* Not expecting this, so fail loudly. */ throw new IOException("Unable to read the response"); } if (inf.needsDictionary()) { /* * Need dictionary - then it must be zlib stream with DICTID * part? */ break; } if (inf.needsInput()) { inf.setInput(peeked); } } if (n == -1) { throw new IOException("Unable to read the response"); } /* * We read something without a problem, so it's a valid zlib stream. * Just need to reset and return an unused InputStream now. */ pushback.unread(peeked, 0, headerLength); return new InflaterInputStream(pushback); } catch (final DataFormatException e) { /* * Presume that it's an RFC1951 deflate stream rather than RFC1950 * zlib stream and try again. */ pushback.unread(peeked, 0, headerLength); return new InflaterInputStream(pushback, new Inflater(true)); } }