List of usage examples for java.io BufferedInputStream reset
public synchronized void reset() throws IOException
reset method of InputStream. From source file:org.apache.hadoop.tools.rumen.TestRumenJobTraces.java
@Test public void testHadoop20JHParser() throws Exception { // Disabled// w ww . jav a 2 s . c om if (true) return; final Configuration conf = new Configuration(); final FileSystem lfs = FileSystem.getLocal(conf); boolean success = false; final Path rootInputDir = new Path(System.getProperty("test.tools.input.dir", "")).makeQualified(lfs); final Path rootTempDir = new Path(System.getProperty("test.build.data", "/tmp")).makeQualified(lfs); final Path rootInputPath = new Path(rootInputDir, "rumen/small-trace-test"); final Path tempDir = new Path(rootTempDir, "TestHadoop20JHParser"); lfs.delete(tempDir, true); final Path inputPath = new Path(rootInputPath, "v20-single-input-log.gz"); final Path goldPath = new Path(rootInputPath, "v20-single-input-log-event-classes.text.gz"); InputStream inputLogStream = new PossiblyDecompressedInputStream(inputPath, conf); InputStream inputGoldStream = new PossiblyDecompressedInputStream(goldPath, conf); BufferedInputStream bis = new BufferedInputStream(inputLogStream); bis.mark(10000); Hadoop20JHParser parser = new Hadoop20JHParser(bis); final Path resultPath = new Path(tempDir, "result.text"); System.out.println("testHadoop20JHParser sent its output to " + resultPath); Compressor compressor; FileSystem fs = resultPath.getFileSystem(conf); CompressionCodec codec = new CompressionCodecFactory(conf).getCodec(resultPath); OutputStream output; if (codec != null) { compressor = CodecPool.getCompressor(codec); output = codec.createOutputStream(fs.create(resultPath), compressor); } else { output = fs.create(resultPath); } PrintStream printStream = new PrintStream(output); try { assertEquals("Hadoop20JHParser can't parse the test file", true, Hadoop20JHParser.canParse(inputLogStream)); bis.reset(); HistoryEvent event = parser.nextEvent(); while (event != null) { printStream.println(event.getClass().getCanonicalName()); event = parser.nextEvent(); } printStream.close(); LineReader goldLines = new LineReader(inputGoldStream); LineReader resultLines = new LineReader(new PossiblyDecompressedInputStream(resultPath, conf)); int lineNumber = 1; try { Text goldLine = new Text(); Text resultLine = new Text(); int goldRead = goldLines.readLine(goldLine); int resultRead = resultLines.readLine(resultLine); while (goldRead * resultRead != 0) { if (!goldLine.equals(resultLine)) { assertEquals("Type mismatch detected", goldLine, resultLine); break; } goldRead = goldLines.readLine(goldLine); resultRead = resultLines.readLine(resultLine); ++lineNumber; } if (goldRead != resultRead) { assertEquals("the " + (goldRead > resultRead ? "gold" : resultRead) + " file contains more text at line " + lineNumber, goldRead, resultRead); } success = true; } finally { goldLines.close(); resultLines.close(); if (success) { lfs.delete(resultPath, false); } } } finally { if (parser == null) { inputLogStream.close(); } else { if (parser != null) { parser.close(); } } if (inputGoldStream != null) { inputGoldStream.close(); } // it's okay to do this twice [if we get an error on input] printStream.close(); } }
From source file:org.apereo.portal.io.xml.JaxbPortalDataHandlerService.java
protected MediaType getMediaType(BufferedInputStream inputStream, String fileName) throws IOException { final TikaInputStream tikaInputStreamStream = TikaInputStream.get(new CloseShieldInputStream(inputStream)); try {/*w ww. j a v a 2s .c om*/ final Detector detector = new DefaultDetector(); final Metadata metadata = new Metadata(); metadata.set(Metadata.RESOURCE_NAME_KEY, fileName); final MediaType type = detector.detect(tikaInputStreamStream, metadata); logger.debug("Determined '{}' for '{}'", type, fileName); return type; } catch (IOException e) { logger.warn("Failed to determine media type for '" + fileName + "' assuming XML", e); return null; } finally { IOUtils.closeQuietly(tikaInputStreamStream); //Reset the buffered stream to make up for anything read by the detector inputStream.reset(); } }
From source file:edu.harvard.iq.dvn.ingest.statdataio.impl.plugins.por.PORFileReader.java
private File decodeHeader(BufferedInputStream stream) throws IOException { File tempPORfile = null;/*from w ww .j a va 2s . c o m*/ if (stream == null) { throw new IllegalArgumentException("file == null!"); } byte[] headerByes = new byte[POR_HEADER_SIZE]; if (stream.markSupported()) { stream.mark(1000); } int nbytes = stream.read(headerByes, 0, POR_HEADER_SIZE); //printHexDump(headerByes, "hex dump of the byte-array"); if (nbytes == 0) { throw new IOException("decodeHeader: reading failure"); } else if (nbytes < 491) { // Size test: by defnition, it must have at least // 491-byte header, i.e., the file size less than this threshold // is not a POR file dbgLog.fine("this file is NOT spss-por type"); throw new IllegalArgumentException("file is not spss-por type"); } // rewind the current reading position back to the beginning if (stream.markSupported()) { stream.reset(); } // line-terminating characters are usually one or two by defnition // however, a POR file saved by a genuine SPSS for Windows // had a three-character line terminator, i.e., failed to remove the // original file's one-character terminator when it was opened, and // saved it with the default two-character terminator without // removing original terminators. So we have to expect such a rare // case // // terminator // windows [0D0A]=> [1310] = [CR/LF] // unix [0A] => [10] // mac [0D] => [13] // 3char [0D0D0A]=> [131310] spss for windows rel 15 // // terminating characters should be found at the following // column positions[counting from 0]: // unix case: [0A] : [80], [161], [242], [323], [404], [485] // windows case: [0D0A] : [81], [163], [245], [327], [409], [491] // : [0D0D0A] : [82], [165], [248], [331], [414], [495] // convert b into a ByteBuffer ByteBuffer buff = ByteBuffer.wrap(headerByes); byte[] nlch = new byte[36]; int pos1; int pos2; int pos3; int ucase = 0; int wcase = 0; int mcase = 0; int three = 0; int nolines = 6; int nocols = 80; for (int i = 0; i < nolines; ++i) { int baseBias = nocols * (i + 1); // 1-char case pos1 = baseBias + i; buff.position(pos1); dbgLog.finer("\tposition(1)=" + buff.position()); int j = 6 * i; nlch[j] = buff.get(); if (nlch[j] == 10) { ucase++; } else if (nlch[j] == 13) { mcase++; } // 2-char case pos2 = baseBias + 2 * i; buff.position(pos2); dbgLog.finer("\tposition(2)=" + buff.position()); nlch[j + 1] = buff.get(); nlch[j + 2] = buff.get(); // 3-char case pos3 = baseBias + 3 * i; buff.position(pos3); dbgLog.finer("\tposition(3)=" + buff.position()); nlch[j + 3] = buff.get(); nlch[j + 4] = buff.get(); nlch[j + 5] = buff.get(); dbgLog.finer(i + "-th iteration position =" + nlch[j] + "\t" + nlch[j + 1] + "\t" + nlch[j + 2]); dbgLog.finer(i + "-th iteration position =" + nlch[j + 3] + "\t" + nlch[j + 4] + "\t" + nlch[j + 5]); if ((nlch[j + 3] == 13) && (nlch[j + 4] == 13) && (nlch[j + 5] == 10)) { three++; } else if ((nlch[j + 1] == 13) && (nlch[j + 2] == 10)) { wcase++; } buff.rewind(); } boolean windowsNewLine = true; if (three == nolines) { windowsNewLine = false; // lineTerminator = "0D0D0A" } else if ((ucase == nolines) && (wcase < nolines)) { windowsNewLine = false; // lineTerminator = "0A" } else if ((ucase < nolines) && (wcase == nolines)) { windowsNewLine = true; //lineTerminator = "0D0A" } else if ((mcase == nolines) && (wcase < nolines)) { windowsNewLine = false; //lineTerminator = "0D" } buff.rewind(); int PORmarkPosition = POR_MARK_POSITION_DEFAULT; if (windowsNewLine) { PORmarkPosition = PORmarkPosition + 5; } else if (three == nolines) { PORmarkPosition = PORmarkPosition + 10; } byte[] pormark = new byte[8]; buff.position(PORmarkPosition); buff.get(pormark, 0, 8); String pormarks = new String(pormark); //dbgLog.fine("pormark =>" + pormarks + "<-"); dbgLog.fine( "pormark[hex: 53 50 53 53 50 4F 52 54 == SPSSPORT] =>" + new String(Hex.encodeHex(pormark)) + "<-"); if (pormarks.equals(POR_MARK)) { dbgLog.fine("POR ID toke test: Passed"); init(); smd.getFileInformation().put("mimeType", MIME_TYPE); smd.getFileInformation().put("fileFormat", MIME_TYPE); } else { dbgLog.fine("this file is NOT spss-por type"); throw new IllegalArgumentException("decodeHeader: POR ID token was not found"); } // save the POR file without new line characters FileOutputStream fileOutPOR = null; Writer fileWriter = null; // Scanner class can handle three-character line-terminator Scanner porScanner = null; try { tempPORfile = File.createTempFile("tempPORfile.", ".por"); fileOutPOR = new FileOutputStream(tempPORfile); fileWriter = new BufferedWriter(new OutputStreamWriter(fileOutPOR, "utf8")); porScanner = new Scanner(stream); // Because 64-bit and 32-bit machines decode POR's first 40-byte // sequence differently, the first 5 leader lines are skipped from // the new-line-stripped file int lineCounter = 0; while (porScanner.hasNextLine()) { lineCounter++; if (lineCounter <= 5) { String line = porScanner.nextLine().toString(); dbgLog.fine("line=" + lineCounter + ":" + line.length() + ":" + line); } else { fileWriter.write(porScanner.nextLine().toString()); } } } finally { try { if (fileWriter != null) { fileWriter.close(); } } catch (IOException ex) { ex.printStackTrace(); } if (porScanner != null) { porScanner.close(); } } return tempPORfile; }
From source file:nl.minbzk.dwr.zoeken.enricher.util.XmlHtmlReader.java
/** * Returns the encoding declared in the <meta http-equiv="content-type" ..>, or NULL if none. * /*from ww w .j av a 2 s . com*/ * @param is * @param guessedEnc * @return String * @throws IOException */ private String getHtmlMeta(final BufferedInputStream is, final String guessedEnc) throws IOException { String encoding = null; if (guessedEnc != null) { // Mark the stream up until the maximum bytes we'd like to read (4096 * 2), which should be enough to account for double-byte encodings like UTF-16 is.mark(BUFFER_SIZE * 2); // Rather than read until we run into the meta tag, we just read out 4096 * 4 bytes byte[] buffer = new byte[BUFFER_SIZE * 2]; int count = is.read(buffer); if (count > 0) { String content = new String(buffer, 0, count, guessedEnc); if (content.indexOf("<") != -1 && content.indexOf(">") != -1) { Matcher contentTypeMatcher = META_CONTENT_TYPE_PATTERN.matcher(content); Matcher contentLanguageMatcher = META_CONTENT_LANGUAGE_PATTERN.matcher(content); // Either the first or second group contains the match, the format of should be equivalent to the Content-Type header if (contentTypeMatcher.find()) encoding = getContentTypeEncoding( (StringUtils.isEmpty(contentTypeMatcher.group(1)) ? contentTypeMatcher.group(2) : contentTypeMatcher.group(1))); // Likewise for the language, except store it directly within the class if (contentLanguageMatcher.find()) { String[] languages = (StringUtils.isEmpty(contentLanguageMatcher.group(1)) ? contentLanguageMatcher.group(2) : contentLanguageMatcher.group(1)).split(","); _languages = new ArrayList<String>(); for (String language : languages) _languages.add(language.split("-")[0].trim()); } } is.reset(); } } return encoding; }
From source file:com.seajas.search.contender.scripting.XmlHtmlReader.java
/** * Returns the encoding declared in the <meta http-equiv="content-type" ..>, or NULL if none. * //w w w . j a v a 2s .c o m * @param is * @param guessedEnc * @return String * @throws IOException */ private String getHtmlMeta(final BufferedInputStream is, final String guessedEnc) throws IOException { String encoding = null; if (guessedEnc != null) { // Mark the stream up until the maximum bytes we'd like to read (4096 * 2), which should be enough to account for double-byte encodings like UTF-16 is.mark(BUFFER_SIZE * 2); // Rather than read until we run into the meta tag, we just read out 4096 * 4 bytes byte[] buffer = new byte[BUFFER_SIZE * 2]; int count = is.read(buffer); if (count > 0) { String content = new String(buffer, 0, count, guessedEnc); if (content.indexOf("<") != -1 && content.indexOf(">") != -1) { Matcher contentTypeMatcher = META_CONTENT_TYPE_PATTERN.matcher(content); Matcher contentLanguageMatcher = META_CONTENT_LANGUAGE_PATTERN.matcher(content); // Either the first or second group contains the match, the format of should be equivalent to the Content-Type header if (contentTypeMatcher.find()) encoding = getContentTypeEncoding( (StringUtils.isEmpty(contentTypeMatcher.group(1)) ? contentTypeMatcher.group(2) : contentTypeMatcher.group(1))); // Likewise for the language, except store it directly within the class if (contentLanguageMatcher.find()) { String languageList = StringUtils.isEmpty(contentLanguageMatcher.group(1)) ? contentLanguageMatcher.group(2) : contentLanguageMatcher.group(1); if (!StringUtils.isEmpty(languageList)) { _languages = new ArrayList<String>(); for (String language : languageList.split(",")) _languages.add(language.trim()); } } } is.reset(); } } return encoding; }
From source file:edu.harvard.iq.dataverse.ingest.tabulardata.impl.plugins.por.PORFileReader.java
private File decodeHeader(BufferedInputStream stream) throws IOException { dbgLog.fine("decodeHeader(): start"); File tempPORfile = null;/* ww w . j a v a2s .c om*/ if (stream == null) { throw new IllegalArgumentException("file == null!"); } byte[] headerByes = new byte[POR_HEADER_SIZE]; if (stream.markSupported()) { stream.mark(1000); } int nbytes = stream.read(headerByes, 0, POR_HEADER_SIZE); //printHexDump(headerByes, "hex dump of the byte-array"); if (nbytes == 0) { throw new IOException("decodeHeader: reading failure"); } else if (nbytes < 491) { // Size test: by defnition, it must have at least // 491-byte header, i.e., the file size less than this threshold // is not a POR file dbgLog.fine("this file is NOT spss-por type"); throw new IllegalArgumentException("file is not spss-por type"); } // rewind the current reading position back to the beginning if (stream.markSupported()) { stream.reset(); } // line-terminating characters are usually one or two by defnition // however, a POR file saved by a genuine SPSS for Windows // had a three-character line terminator, i.e., failed to remove the // original file's one-character terminator when it was opened, and // saved it with the default two-character terminator without // removing original terminators. So we have to expect such a rare // case // // terminator // windows [0D0A]=> [1310] = [CR/LF] // unix [0A] => [10] // mac [0D] => [13] // 3char [0D0D0A]=> [131310] spss for windows rel 15 // // terminating characters should be found at the following // column positions[counting from 0]: // unix case: [0A] : [80], [161], [242], [323], [404], [485] // windows case: [0D0A] : [81], [163], [245], [327], [409], [491] // : [0D0D0A] : [82], [165], [248], [331], [414], [495] // convert b into a ByteBuffer ByteBuffer buff = ByteBuffer.wrap(headerByes); byte[] nlch = new byte[36]; int pos1; int pos2; int pos3; int ucase = 0; int wcase = 0; int mcase = 0; int three = 0; int nolines = 6; int nocols = 80; for (int i = 0; i < nolines; ++i) { int baseBias = nocols * (i + 1); // 1-char case pos1 = baseBias + i; buff.position(pos1); dbgLog.finer("\tposition(1)=" + buff.position()); int j = 6 * i; nlch[j] = buff.get(); if (nlch[j] == 10) { ucase++; } else if (nlch[j] == 13) { mcase++; } // 2-char case pos2 = baseBias + 2 * i; buff.position(pos2); dbgLog.finer("\tposition(2)=" + buff.position()); nlch[j + 1] = buff.get(); nlch[j + 2] = buff.get(); // 3-char case pos3 = baseBias + 3 * i; buff.position(pos3); dbgLog.finer("\tposition(3)=" + buff.position()); nlch[j + 3] = buff.get(); nlch[j + 4] = buff.get(); nlch[j + 5] = buff.get(); dbgLog.finer(i + "-th iteration position =" + nlch[j] + "\t" + nlch[j + 1] + "\t" + nlch[j + 2]); dbgLog.finer(i + "-th iteration position =" + nlch[j + 3] + "\t" + nlch[j + 4] + "\t" + nlch[j + 5]); if ((nlch[j + 3] == 13) && (nlch[j + 4] == 13) && (nlch[j + 5] == 10)) { three++; } else if ((nlch[j + 1] == 13) && (nlch[j + 2] == 10)) { wcase++; } buff.rewind(); } boolean windowsNewLine = true; if (three == nolines) { windowsNewLine = false; // lineTerminator = "0D0D0A" } else if ((ucase == nolines) && (wcase < nolines)) { windowsNewLine = false; // lineTerminator = "0A" } else if ((ucase < nolines) && (wcase == nolines)) { windowsNewLine = true; //lineTerminator = "0D0A" } else if ((mcase == nolines) && (wcase < nolines)) { windowsNewLine = false; //lineTerminator = "0D" } buff.rewind(); int PORmarkPosition = POR_MARK_POSITION_DEFAULT; if (windowsNewLine) { PORmarkPosition = PORmarkPosition + 5; } else if (three == nolines) { PORmarkPosition = PORmarkPosition + 10; } byte[] pormark = new byte[8]; buff.position(PORmarkPosition); buff.get(pormark, 0, 8); String pormarks = new String(pormark); //dbgLog.fine("pormark =>" + pormarks + "<-"); dbgLog.fine( "pormark[hex: 53 50 53 53 50 4F 52 54 == SPSSPORT] =>" + new String(Hex.encodeHex(pormark)) + "<-"); if (pormarks.equals(POR_MARK)) { dbgLog.fine("POR ID toke test: Passed"); init(); dataTable.setOriginalFileFormat(MIME_TYPE); dataTable.setUnf("UNF:6:NOTCALCULATED"); } else { dbgLog.fine("this file is NOT spss-por type"); throw new IllegalArgumentException("decodeHeader: POR ID token was not found"); } // save the POR file without new line characters FileOutputStream fileOutPOR = null; Writer fileWriter = null; // Scanner class can handle three-character line-terminator Scanner porScanner = null; try { tempPORfile = File.createTempFile("tempPORfile.", ".por"); fileOutPOR = new FileOutputStream(tempPORfile); fileWriter = new BufferedWriter(new OutputStreamWriter(fileOutPOR, "utf8")); porScanner = new Scanner(stream); // Because 64-bit and 32-bit machines decode POR's first 40-byte // sequence differently, the first 5 leader lines are skipped from // the new-line-stripped file int lineCounter = 0; while (porScanner.hasNextLine()) { lineCounter++; if (lineCounter <= 5) { String line = porScanner.nextLine(); dbgLog.fine("line=" + lineCounter + ":" + line.length() + ":" + line); } else { fileWriter.write(porScanner.nextLine()); } } } finally { try { if (fileWriter != null) { fileWriter.close(); } } catch (IOException ex) { ex.printStackTrace(); } if (porScanner != null) { porScanner.close(); } } return tempPORfile; }
From source file:org.jzkit.a2j.codec.runtime.ChunkingBERInputStream.java
public int completeConstructedType(BufferedInputStream in, ByteArrayOutputStream baos, int level) throws java.io.IOException { int bytes_written = 0; int first_byte = in.read(); if (first_byte == -1) throw new java.io.IOException("EOF"); byte c = (byte) first_byte; // byte c = (byte)in.read(); baos.write(c);/*from w ww . j a va2s .c om*/ bytes_written++; c &= 0xFF; int next_tag_class = c & 0xC0; boolean next_is_constructed = (c & 0x20) != 0; boolean next_is_indefinite = false; int next_tag_number = c & 0x1F; // If there are multiple octets to encode the tag if (next_tag_number == 0x1F) { next_tag_number = 0; do { c = (byte) in.read(); baos.write(c); bytes_written++; // Shift value 7 bits left next_tag_number = next_tag_number << 7; // Merge with the octets we just got next_tag_number = (next_tag_number | (c & 0x7F)); } while ((c & 0x80) != 0); } // dbg("tag: "+next_tag_number+" class:"+next_tag_class, level); int datalen; byte lenpart = (byte) in.read(); baos.write(lenpart); bytes_written++; // System.err.println("First len octet is "+lenpart); if ((lenpart & 0x80) == 0) // If bit 8 is 0 { // Single octet length encoding // System.err.println("Single octet length encoding"); datalen = lenpart; next_is_indefinite = false; } else if ((lenpart & 0x7F) == 0) // Otherwise we are multiple octets (Maybe 0, which = indefinite) { // System.err.println("Indefinite length encoding"); next_is_indefinite = true; datalen = 0; } else { next_is_indefinite = false; // System.err.println("Multiple octet length encoding ("+(lenpart & 0x7F )+"octets)"); lenpart &= 0x7F; datalen = 0; while (lenpart-- > 0) { byte lenbyte = (byte) in.read(); datalen = (datalen << 8) | (lenbyte & 0xFF); baos.write(lenbyte); bytes_written++; } } // System.err.print(" indefinite: "+next_is_indefinite+" cons:"+next_is_constructed+" len:"+datalen); // OK, len now contains the size of the octets. // If it's definite length encoding, just copy that many bytes if (next_is_indefinite) { if (next_is_constructed) { // System.err.print(" {\n"); // Peek ahead looking for terminating octets. boolean more_data = true; in.mark(5); byte i1 = (byte) in.read(); byte i2 = (byte) in.read(); in.reset(); if ((i1 == 0) && (i2 == 0)) { more_data = false; } while (more_data) { completeConstructedType(in, baos, level + 1); in.mark(5); i1 = (byte) in.read(); i2 = (byte) in.read(); in.reset(); if ((i1 == 0) && (i2 == 0)) { more_data = false; } } // Better consume the terminating octets. in.read(); in.read(); baos.write(0); bytes_written++; baos.write(0); bytes_written++; // dbg("} ("+bytes_written+")\n",level); } else { // Indefinite length primitive type // System.err.print(" Indefinite length primitive type"); byte b1 = (byte) in.read(); baos.write(b1); bytes_written++; byte b2 = (byte) in.read(); baos.write(b2); bytes_written++; while (!((b1 == 0) && (b2 == 0))) { b1 = b2; b2 = (byte) in.read(); baos.write(b2); bytes_written++; } // System.err.println("("+bytes_written+")"); } } else { // System.err.println("copy definite length encoding, remain="+datalen); if (next_is_constructed) { // System.err.println(" {"); while (datalen > 0) { int child_len = completeConstructedType(in, baos, level + 1); datalen -= child_len; bytes_written += child_len; } // dbg("} ("+bytes_written+")\n",level); } else { // System.err.print(" Definite length primitive type"); byte[] buff = new byte[4096]; while (datalen > 0) { int bytes_read = in.read(buff, 0, datalen > 4096 ? 4096 : datalen); // System.err.println("processed "+bytes_read); baos.write(buff, 0, bytes_read); datalen -= bytes_read; bytes_written += bytes_read; // System.err.println("("+bytes_written+")"); } } } return bytes_written; }
From source file:org.lockss.remote.RemoteApi.java
public BatchAuStatus processSavedConfigProps(BufferedInputStream auTxtStream) throws IOException, InvalidAuConfigBackupFile { int commentLen = AU_BACKUP_FILE_COMMENT.length(); // There is apparently hidden buffering in the InputStreamReader's // StreamDecoder which throws off our calculation as to how much // auTxtStream needs to buffer, so use a large number auTxtStream.mark(paramBackupStreamMarkSize); BufferedReader rdr = new BufferedReader(new InputStreamReader(auTxtStream, Constants.DEFAULT_ENCODING), commentLen * 2);/*from www .j av a 2 s .c o m*/ // We really want rdr.readLine(), but we need to limit amount it reads // (in case it doesn't find newline) char[] buf = new char[commentLen]; int chars = StreamUtil.readChars(rdr, buf, commentLen); log.debug3("chars: " + chars); if (chars == 0) { throw new InvalidAuConfigBackupFile("Uploaded file is empty"); } String line1 = new String(buf); if (chars < commentLen || !line1.startsWith(AU_BACKUP_FILE_COMMENT)) { log.debug("line1: " + line1); throw new InvalidAuConfigBackupFile("Uploaded file does not appear to be a saved AU configuration"); } try { auTxtStream.reset(); } catch (IOException e) { throw new IOException("Internal error: please report \"Insufficient buffering for restore\"."); } Properties allAuProps = new Properties(); try { allAuProps.load(auTxtStream); } catch (Exception e) { log.warning("Loading AU config backup file", e); throw new InvalidAuConfigBackupFile("Uploaded file has illegal format: " + e.getMessage()); } Configuration allAuConfig = ConfigManager.fromPropertiesUnsealed(allAuProps); int ver = checkLegalAuConfigTree(allAuConfig); return batchProcessAus(false, BATCH_ADD_RESTORE, allAuConfig, null); }
From source file:org.sakaiproject.gradebook.gwt.server.ImportExportUtilityImpl.java
public Upload parseImportXLS(MultipartFile file, ImportSettings settings) throws InvalidInputException, FatalException, IOException { log.debug("parseImportXLS() called"); // Strip off extension String fileName = removeFileExenstion(file.getOriginalFilename().toLowerCase()); String realFileName = fileName; boolean isOriginalName; try {//from ww w . ja va2 s . c o m realFileName = getUniqueItemName(fileName, settings.getGradebookUid()); } catch (GradebookImportException e) { return emptyUploadFileWithNotes(e.getMessage()); } isOriginalName = realFileName.equals(fileName); log.debug("realFileName=" + realFileName); log.debug("isOriginalName=" + isOriginalName); org.apache.poi.ss.usermodel.Workbook inspread = null; BufferedInputStream bufStream = new BufferedInputStream(file.getInputStream()); //inspread = readPoiSpreadsheet(tempfi); ///TODO: this is broken buonly called by the *Test class for this class and not by default if (inspread != null) { log.debug("Found a POI readable spreadsheet"); bufStream.close(); return handlePoiSpreadSheet(inspread, realFileName, isOriginalName, settings); } // else log.debug("POI couldn't handle the spreadsheet, using jexcelapi"); bufStream.reset(); return handleJExcelAPISpreadSheet(bufStream, realFileName, isOriginalName, settings); }
From source file:tufts.vue.URLResource.java
private Properties scrapeHTMLmetaData(URLConnection connection, int maxSearchBytes) throws java.io.IOException { Properties metaData = new Properties(); InputStream byteStream = connection.getInputStream(); if (DEBUG.DND && DEBUG.META) { System.err.println("Getting headers from " + connection); System.err.println("Headers: " + connection.getHeaderFields()); }//from w ww . j a va2 s . co m // note: be sure to call getContentType and don't rely on getting it from the HeaderFields map, // as sometimes it's set by the OS for a file:/// URL when there are no header fields (no http server) // (actually, this is set by java via a mime type table based on file extension, or a guess based on the stream) if (DEBUG.DND) System.err.println("*** getting contentType & encoding..."); final String contentType = connection.getContentType(); final String contentEncoding = connection.getContentEncoding(); final int contentLength = connection.getContentLength(); if (DEBUG.DND) System.err.println("*** contentType [" + contentType + "]"); if (DEBUG.DND) System.err.println("*** contentEncoding [" + contentEncoding + "]"); if (DEBUG.DND) System.err.println("*** contentLength [" + contentLength + "]"); setProperty("url.contentType", contentType); setProperty("url.contentEncoding", contentEncoding); if (contentLength >= 0) setProperty("url.contentLength", contentLength); //if (contentType.toLowerCase().startsWith("text/html") == false) { if (!isHTML()) { // we only currently handle HTML if (DEBUG.Enabled) System.err.println("*** contentType [" + contentType + "] not HTML; skipping title extraction"); return metaData; } if (DEBUG.DND) System.err.println("*** scanning for HTML meta-data..."); try { final BufferedInputStream bufStream = new BufferedInputStream(byteStream, maxSearchBytes); bufStream.mark(maxSearchBytes); final byte[] byteBuffer = new byte[maxSearchBytes]; int bytesRead = 0; int len = 0; // BufferedInputStream still won't read thru a block, so we need to allow // a few reads here to get thru a couple of blocks, so we can get up to // our maxbytes (e.g., a common return chunk count is 1448 bytes, presumably related to the MTU) do { int max = maxSearchBytes - bytesRead; len = bufStream.read(byteBuffer, bytesRead, max); System.out.println("*** read " + len); if (len > 0) bytesRead += len; else if (len < 0) break; } while (len > 0 && bytesRead < maxSearchBytes); if (DEBUG.DND) System.out.println("*** Got total chars: " + bytesRead); String html = new String(byteBuffer, 0, bytesRead); if (DEBUG.DND && DEBUG.META) System.out.println("*** HTML-STRING[" + html + "]"); // first, look for a content encoding, so we can search for and get the title // on a properly encoded character stream String charset = null; Matcher cm = Content_Charset_Regex.matcher(html); if (cm.lookingAt()) { charset = cm.group(1); if (DEBUG.DND) System.err.println("*** found HTML specified charset [" + charset + "]"); setProperty("charset", charset); } if (charset == null && contentEncoding != null) { if (DEBUG.DND || true) System.err.println("*** no charset found: using contentEncoding charset " + contentEncoding); charset = contentEncoding; } final String decodedHTML; if (charset != null) { bufStream.reset(); InputStreamReader decodedStream = new InputStreamReader(bufStream, charset); //InputStreamReader decodedStream = new InputStreamReader(new ByteArrayInputStream(byteBuffer), charset); if (true || DEBUG.DND) System.out.println("*** decoding bytes into characters with official encoding " + decodedStream.getEncoding()); setProperty("contentEncoding", decodedStream.getEncoding()); char[] decoded = new char[bytesRead]; int decodedChars = decodedStream.read(decoded); decodedStream.close(); if (true || DEBUG.DND) System.err.println("*** " + decodedChars + " characters decoded using " + charset); decodedHTML = new String(decoded, 0, decodedChars); } else decodedHTML = html; // we'll just have to go with the default platform charset... // these needed to be left open till the decodedStream was done, which // although it should never need to read beyond what's already buffered, // some internal java code has checks that make sure the underlying stream // isn't closed, even it it isn't used. byteStream.close(); bufStream.close(); Matcher m = HTML_Title_Regex.matcher(decodedHTML); if (m.lookingAt()) { String title = m.group(1); if (true || DEBUG.DND) System.err.println("*** found title [" + title + "]"); metaData.put("title", title.trim()); } } catch (Throwable e) { System.err.println("scrapeHTMLmetaData: " + e); if (DEBUG.DND) e.printStackTrace(); } if (DEBUG.DND || DEBUG.Enabled) System.err.println("*** scrapeHTMLmetaData returning [" + metaData + "]"); return metaData; }