List of usage examples for java.io BufferedInputStream reset
public synchronized void reset() throws IOException
reset method of InputStream. From source file:edu.harvard.iq.dataverse.ingest.tabulardata.impl.plugins.sav.SAVFileReader.java
void decodeRecordType999(BufferedInputStream stream) throws IOException { dbgLog.fine("decodeRecordType999(): start"); try {/*from w w w. jav a 2s . c o m*/ if (stream == null) { throw new IllegalArgumentException("RT999: stream == null!"); } // first check the 4-byte header value //if (stream.markSupported()){ stream.mark(1000); //} // 999.0 check the first 4 bytes byte[] headerCodeRt999 = new byte[LENGTH_RECORD_TYPE999_CODE]; //dbgLog.fine("RT999: stream position="+stream.pos); int nbytes_rt999 = stream.read(headerCodeRt999, 0, LENGTH_RECORD_TYPE999_CODE); // to-do check against nbytes //printHexDump(headerCodeRt999, "RT999 header test"); ByteBuffer bb_header_code_rt999 = ByteBuffer.wrap(headerCodeRt999, 0, LENGTH_RECORD_TYPE999_CODE); if (isLittleEndian) { bb_header_code_rt999.order(ByteOrder.LITTLE_ENDIAN); } int intRT999test = bb_header_code_rt999.getInt(); dbgLog.fine("header test value: RT999=" + intRT999test); if (intRT999test != 999) { //if (stream.markSupported()){ dbgLog.fine("intRT999test failed=" + intRT999test); stream.reset(); throw new IOException("RT999:Header value(999) was not correctly detected:" + intRT999test); //} } // 999.1 check 4-byte integer Filler block byte[] length_filler = new byte[LENGTH_RT999_FILLER]; int nbytes_rt999_1 = stream.read(length_filler, 0, LENGTH_RT999_FILLER); // to-do check against nbytes //printHexDump(length_how_many_line_bytes, "RT999 how_many_line_bytes"); ByteBuffer bb_filler = ByteBuffer.wrap(length_filler, 0, LENGTH_RT999_FILLER); if (isLittleEndian) { bb_filler.order(ByteOrder.LITTLE_ENDIAN); } int rt999filler = bb_filler.getInt(); dbgLog.fine("rt999filler=" + rt999filler); if (rt999filler == 0) { dbgLog.fine("the end of the dictionary section"); } else { throw new IOException("RT999: failed to detect the end mark(0): value=" + rt999filler); } // missing value processing concerning HIGHEST/LOWEST values Set<Map.Entry<String, InvalidData>> msvlc = invalidDataTable.entrySet(); for (Iterator<Map.Entry<String, InvalidData>> itc = msvlc.iterator(); itc.hasNext();) { Map.Entry<String, InvalidData> et = itc.next(); String variable = et.getKey(); dbgLog.fine("variable=" + variable); InvalidData invalidDataInfo = et.getValue(); if (invalidDataInfo.getInvalidRange() != null && !invalidDataInfo.getInvalidRange().isEmpty()) { if (invalidDataInfo.getInvalidRange().get(0).equals(OBSTypeHexValue.get("LOWEST"))) { dbgLog.fine("1st value is LOWEST"); invalidDataInfo.getInvalidRange().set(0, "LOWEST"); } else if (invalidDataInfo.getInvalidRange().get(1).equals(OBSTypeHexValue.get("HIGHEST"))) { dbgLog.fine("2nd value is HIGHEST"); invalidDataInfo.getInvalidRange().set(1, "HIGHEST"); } } } dbgLog.fine("invalidDataTable:\n" + invalidDataTable); // TODO: take care of the invalid data! - add the appropriate // value labels (?) // should it be done here, or at the end of ingest? // -- L.A. 4.0 alpha ///smd.setInvalidDataTable(invalidDataTable); } catch (IOException ex) { //ex.printStackTrace(); //exit(1); throw ex; } dbgLog.fine("decodeRecordType999(): end"); }
From source file:edu.harvard.iq.dataverse.ingest.tabulardata.impl.plugins.sav.SAVFileReader.java
void decodeRecordType6(BufferedInputStream stream) throws IOException { dbgLog.fine("***** decodeRecordType6(): start *****"); try {/*from w ww. j a v a2 s. c o m*/ if (stream == null) { throw new IllegalArgumentException("stream == null!"); } // this section is optional; so let's first check the 4-byte header // value and see what type it is. //if (stream.markSupported()){ // -- ? L.A. 4.0 alpha stream.mark(1000); //} // 6.0 check the first 4 bytes byte[] headerCodeRt6 = new byte[LENGTH_RECORD_TYPE6_CODE]; int nbytes_rt6 = stream.read(headerCodeRt6, 0, LENGTH_RECORD_TYPE6_CODE); // to-do check against nbytes //printHexDump(headerCodeRt6, "RT6 header test"); ByteBuffer bb_header_code_rt6 = ByteBuffer.wrap(headerCodeRt6, 0, LENGTH_RECORD_TYPE6_CODE); if (isLittleEndian) { bb_header_code_rt6.order(ByteOrder.LITTLE_ENDIAN); } int intRT6test = bb_header_code_rt6.getInt(); dbgLog.fine("RT6: header test value=" + intRT6test); if (intRT6test != 6) { //if (stream.markSupported()){ //out.print("iteration="+safteyCounter); //dbgLog.fine("iteration="+safteyCounter); dbgLog.fine("intRT6test failed=" + intRT6test); stream.reset(); return; //} } // 6.1 check 4-byte integer that tells how many lines follow byte[] length_how_many_line_bytes = new byte[LENGTH_RT6_HOW_MANY_LINES]; int nbytes_rt6_1 = stream.read(length_how_many_line_bytes, 0, LENGTH_RT6_HOW_MANY_LINES); // to-do check against nbytes //printHexDump(length_how_many_line_bytes, "RT6 how_many_line_bytes"); ByteBuffer bb_how_many_lines = ByteBuffer.wrap(length_how_many_line_bytes, 0, LENGTH_RT6_HOW_MANY_LINES); if (isLittleEndian) { bb_how_many_lines.order(ByteOrder.LITTLE_ENDIAN); } int howManyLinesRt6 = bb_how_many_lines.getInt(); dbgLog.fine("how Many lines follow=" + howManyLinesRt6); // 6.2 read 80-char-long lines String[] documentRecord = new String[howManyLinesRt6]; for (int i = 0; i < howManyLinesRt6; i++) { byte[] line = new byte[80]; int nbytes_rt6_line = stream.read(line); documentRecord[i] = StringUtils.stripEnd( new String(Arrays.copyOfRange(line, 0, LENGTH_RT6_DOCUMENT_LINE), defaultCharSet), " "); dbgLog.fine(i + "-th line =" + documentRecord[i] + "<-"); } dbgLog.fine("documentRecord:\n" + StringUtils.join(documentRecord, "\n")); } catch (IOException ex) { //ex.printStackTrace(); throw ex; } dbgLog.fine("decodeRecordType6(): end"); }
From source file:edu.harvard.iq.dvn.ingest.statdataio.impl.plugins.sav.SAVFileReader.java
void decodeRecordType999(BufferedInputStream stream) throws IOException { dbgLog.fine("***** decodeRecordType999(): start *****"); try {// w w w.ja va 2 s .co m if (stream == null) { throw new IllegalArgumentException("RT999: stream == null!"); } // first check the 4-byte header value //if (stream.markSupported()){ stream.mark(1000); //} // 999.0 check the first 4 bytes byte[] headerCodeRt999 = new byte[LENGTH_RECORD_TYPE999_CODE]; //dbgLog.fine("RT999: stream position="+stream.pos); int nbytes_rt999 = stream.read(headerCodeRt999, 0, LENGTH_RECORD_TYPE999_CODE); // to-do check against nbytes //printHexDump(headerCodeRt999, "RT999 header test"); ByteBuffer bb_header_code_rt999 = ByteBuffer.wrap(headerCodeRt999, 0, LENGTH_RECORD_TYPE999_CODE); if (isLittleEndian) { bb_header_code_rt999.order(ByteOrder.LITTLE_ENDIAN); } int intRT999test = bb_header_code_rt999.getInt(); dbgLog.fine("header test value: RT999=" + intRT999test); if (intRT999test != 999) { //if (stream.markSupported()){ dbgLog.fine("intRT999test failed=" + intRT999test); stream.reset(); throw new IOException("RT999:Header value(999) was not correctly detected:" + intRT999test); //} } // 999.1 check 4-byte integer Filler block byte[] length_filler = new byte[LENGTH_RT999_FILLER]; int nbytes_rt999_1 = stream.read(length_filler, 0, LENGTH_RT999_FILLER); // to-do check against nbytes //printHexDump(length_how_many_line_bytes, "RT999 how_many_line_bytes"); ByteBuffer bb_filler = ByteBuffer.wrap(length_filler, 0, LENGTH_RT999_FILLER); if (isLittleEndian) { bb_filler.order(ByteOrder.LITTLE_ENDIAN); } int rt999filler = bb_filler.getInt(); dbgLog.fine("rt999filler=" + rt999filler); if (rt999filler == 0) { dbgLog.fine("the end of the dictionary section"); } else { throw new IOException("RT999: failed to detect the end mark(0): value=" + rt999filler); } // missing value processing concerning HIGHEST/LOWEST values Set<Map.Entry<String, InvalidData>> msvlc = invalidDataTable.entrySet(); for (Iterator<Map.Entry<String, InvalidData>> itc = msvlc.iterator(); itc.hasNext();) { Map.Entry<String, InvalidData> et = itc.next(); String variable = et.getKey(); dbgLog.fine("variable=" + variable); InvalidData invalidDataInfo = et.getValue(); if (invalidDataInfo.getInvalidRange() != null && !invalidDataInfo.getInvalidRange().isEmpty()) { if (invalidDataInfo.getInvalidRange().get(0).equals(OBSTypeHexValue.get("LOWEST"))) { dbgLog.fine("1st value is LOWEST"); invalidDataInfo.getInvalidRange().set(0, "LOWEST"); } else if (invalidDataInfo.getInvalidRange().get(1).equals(OBSTypeHexValue.get("HIGHEST"))) { dbgLog.fine("2nd value is HIGHEST"); invalidDataInfo.getInvalidRange().set(1, "HIGHEST"); } } } dbgLog.fine("invalidDataTable:\n" + invalidDataTable); smd.setInvalidDataTable(invalidDataTable); } catch (IOException ex) { //ex.printStackTrace(); //exit(1); throw ex; } dbgLog.fine("***** decodeRecordType999(): end *****"); }
From source file:edu.harvard.iq.dvn.ingest.statdataio.impl.plugins.sav.SAVFileReader.java
void decodeRecordType6(BufferedInputStream stream) throws IOException { dbgLog.fine("***** decodeRecordType6(): start *****"); try {//from ww w . j a v a 2 s. c om if (stream == null) { throw new IllegalArgumentException("stream == null!"); } // this secton may not exit so first check the 4-byte header value //if (stream.markSupported()){ stream.mark(1000); //} // 6.0 check the first 4 bytes byte[] headerCodeRt6 = new byte[LENGTH_RECORD_TYPE6_CODE]; int nbytes_rt6 = stream.read(headerCodeRt6, 0, LENGTH_RECORD_TYPE6_CODE); // to-do check against nbytes //printHexDump(headerCodeRt6, "RT6 header test"); ByteBuffer bb_header_code_rt6 = ByteBuffer.wrap(headerCodeRt6, 0, LENGTH_RECORD_TYPE6_CODE); if (isLittleEndian) { bb_header_code_rt6.order(ByteOrder.LITTLE_ENDIAN); } int intRT6test = bb_header_code_rt6.getInt(); dbgLog.fine("RT6: header test value=" + intRT6test); if (intRT6test != 6) { //if (stream.markSupported()){ //out.print("iteration="+safteyCounter); //dbgLog.fine("iteration="+safteyCounter); dbgLog.fine("intRT6test failed=" + intRT6test); stream.reset(); return; //} } // 6.1 check 4-byte integer that tells how many lines follow byte[] length_how_many_line_bytes = new byte[LENGTH_RT6_HOW_MANY_LINES]; int nbytes_rt6_1 = stream.read(length_how_many_line_bytes, 0, LENGTH_RT6_HOW_MANY_LINES); // to-do check against nbytes //printHexDump(length_how_many_line_bytes, "RT6 how_many_line_bytes"); ByteBuffer bb_how_many_lines = ByteBuffer.wrap(length_how_many_line_bytes, 0, LENGTH_RT6_HOW_MANY_LINES); if (isLittleEndian) { bb_how_many_lines.order(ByteOrder.LITTLE_ENDIAN); } int howManyLinesRt6 = bb_how_many_lines.getInt(); dbgLog.fine("how Many lines follow=" + howManyLinesRt6); // 6.2 read 80-char-long lines String[] documentRecord = new String[howManyLinesRt6]; for (int i = 0; i < howManyLinesRt6; i++) { byte[] line = new byte[80]; int nbytes_rt6_line = stream.read(line); documentRecord[i] = StringUtils.stripEnd( new String(Arrays.copyOfRange(line, 0, LENGTH_RT6_DOCUMENT_LINE), defaultCharSet), " "); dbgLog.fine(i + "-th line =" + documentRecord[i] + "<-"); } dbgLog.fine("documentRecord:\n" + StringUtils.join(documentRecord, "\n")); } catch (IOException ex) { //ex.printStackTrace(); throw ex; } dbgLog.fine("***** decodeRecordType6(): end *****"); }
From source file:edu.harvard.iq.dataverse.ingest.tabulardata.impl.plugins.sav.SAVFileReader.java
void decodeRecordType3and4(BufferedInputStream stream) throws IOException { dbgLog.fine("decodeRecordType3and4(): start"); Map<String, Map<String, String>> valueLabelTable = new LinkedHashMap<String, Map<String, String>>(); int safteyCounter = 0; while (true) { try {//from www . j a v a2 s .co m if (stream == null) { throw new IllegalArgumentException("stream == null!"); } // this secton may not exit so first check the 4-byte header value //if (stream.markSupported()){ stream.mark(1000); //} // 3.0 check the first 4 bytes byte[] headerCode = new byte[LENGTH_RECORD_TYPE3_CODE]; int nbytes_rt3 = stream.read(headerCode, 0, LENGTH_RECORD_TYPE3_CODE); // to-do check against nbytes //printHexDump(headerCode, "RT3 header test"); ByteBuffer bb_header_code = ByteBuffer.wrap(headerCode, 0, LENGTH_RECORD_TYPE3_CODE); if (isLittleEndian) { bb_header_code.order(ByteOrder.LITTLE_ENDIAN); } int intRT3test = bb_header_code.getInt(); dbgLog.fine("header test value: RT3=" + intRT3test); if (intRT3test != 3) { //if (stream.markSupported()){ dbgLog.fine("iteration=" + safteyCounter); // We have encountered a record that's not type 3. This means we've // processed all the type 3/4 record pairs. So we want to rewind // the stream and return -- so that the appropriate record type // reader can be called on it. // But before we return, we need to save all the value labels // we have found: //smd.setValueLabelTable(valueLabelTable); assignValueLabels(valueLabelTable); stream.reset(); return; //} } // 3.1 how many value-label pairs follow byte[] number_of_labels = new byte[LENGTH_RT3_HOW_MANY_LABELS]; int nbytes_3_1 = stream.read(number_of_labels); if (nbytes_3_1 == 0) { throw new IOException("RT 3: reading recordType3.1: no byte was read"); } ByteBuffer bb_number_of_labels = ByteBuffer.wrap(number_of_labels, 0, LENGTH_RT3_HOW_MANY_LABELS); if (isLittleEndian) { bb_number_of_labels.order(ByteOrder.LITTLE_ENDIAN); } int numberOfValueLabels = bb_number_of_labels.getInt(); dbgLog.fine("number of value-label pairs=" + numberOfValueLabels); ByteBuffer[] tempBB = new ByteBuffer[numberOfValueLabels]; String valueLabel[] = new String[numberOfValueLabels]; for (int i = 0; i < numberOfValueLabels; i++) { // read 8-byte as value byte[] value = new byte[LENGTH_RT3_VALUE]; int nbytes_3_value = stream.read(value); if (nbytes_3_value == 0) { throw new IOException("RT 3: reading recordType3 value: no byte was read"); } // note these 8 bytes are interpreted later // currently no information about which variable's (=> type unknown) ByteBuffer bb_value = ByteBuffer.wrap(value, 0, LENGTH_RT3_VALUE); if (isLittleEndian) { bb_value.order(ByteOrder.LITTLE_ENDIAN); } tempBB[i] = bb_value; dbgLog.fine("bb_value=" + Hex.encodeHex(bb_value.array())); /* double valueD = bb_value.getDouble(); dbgLog.fine("value="+valueD); */ // read 1st byte as unsigned integer = label_length // read label_length byte as label byte[] labelLengthByte = new byte[LENGTH_RT3_LABEL_LENGTH]; int nbytes_3_label_length = stream.read(labelLengthByte); // add check-routine here dbgLog.fine("labelLengthByte" + Hex.encodeHex(labelLengthByte)); dbgLog.fine("label length = " + labelLengthByte[0]); // the net-length of a value label is saved as // unsigned byte; however, the length is less than 127 // byte should be ok int rawLabelLength = labelLengthByte[0] & 0xFF; dbgLog.fine("rawLabelLength=" + rawLabelLength); // -1 =>1-byte already read int labelLength = getSAVobsAdjustedBlockLength(rawLabelLength + 1) - 1; byte[] valueLabelBytes = new byte[labelLength]; int nbytes_3_value_label = stream.read(valueLabelBytes); // ByteBuffer bb_label = ByteBuffer.wrap(valueLabel,0,labelLength); valueLabel[i] = StringUtils.stripEnd( new String(Arrays.copyOfRange(valueLabelBytes, 0, rawLabelLength), defaultCharSet), " "); dbgLog.fine(i + "-th valueLabel=" + valueLabel[i] + "<-"); } // iter rt3 dbgLog.fine("end of RT3 block"); dbgLog.fine("start of RT4 block"); // 4.0 check the first 4 bytes byte[] headerCode4 = new byte[LENGTH_RECORD_TYPE4_CODE]; int nbytes_rt4 = stream.read(headerCode4, 0, LENGTH_RECORD_TYPE4_CODE); if (nbytes_rt4 == 0) { throw new IOException("RT4: reading recordType4 value: no byte was read"); } //printHexDump(headerCode4, "RT4 header test"); ByteBuffer bb_header_code_4 = ByteBuffer.wrap(headerCode4, 0, LENGTH_RECORD_TYPE4_CODE); if (isLittleEndian) { bb_header_code_4.order(ByteOrder.LITTLE_ENDIAN); } int intRT4test = bb_header_code_4.getInt(); dbgLog.fine("header test value: RT4=" + intRT4test); if (intRT4test != 4) { throw new IOException("RT 4: reading recordType4 header: no byte was read"); } // 4.1 read the how-many-variables bytes byte[] howManyVariablesfollow = new byte[LENGTH_RT4_HOW_MANY_VARIABLES]; int nbytes_rt4_1 = stream.read(howManyVariablesfollow, 0, LENGTH_RT4_HOW_MANY_VARIABLES); ByteBuffer bb_howManyVariablesfollow = ByteBuffer.wrap(howManyVariablesfollow, 0, LENGTH_RT4_HOW_MANY_VARIABLES); if (isLittleEndian) { bb_howManyVariablesfollow.order(ByteOrder.LITTLE_ENDIAN); } int howManyVariablesRT4 = bb_howManyVariablesfollow.getInt(); dbgLog.fine("how many variables follow: RT4=" + howManyVariablesRT4); int length_indicies = LENGTH_RT4_VARIABLE_INDEX * howManyVariablesRT4; byte[] variableIdicesBytes = new byte[length_indicies]; int nbytes_rt4_2 = stream.read(variableIdicesBytes, 0, length_indicies); // !!!!! Caution: variableIndex in RT4 starts from 1 NOT ** 0 ** int[] variableIndex = new int[howManyVariablesRT4]; int offset = 0; for (int i = 0; i < howManyVariablesRT4; i++) { ByteBuffer bb_variable_index = ByteBuffer.wrap(variableIdicesBytes, offset, LENGTH_RT4_VARIABLE_INDEX); offset += LENGTH_RT4_VARIABLE_INDEX; if (isLittleEndian) { bb_variable_index.order(ByteOrder.LITTLE_ENDIAN); } variableIndex[i] = bb_variable_index.getInt(); dbgLog.fine(i + "-th variable index number=" + variableIndex[i]); } dbgLog.fine("variable index set=" + ArrayUtils.toString(variableIndex)); dbgLog.fine("subtract 1 from variableIndex for getting a variable info"); boolean isNumeric = OBSwiseTypelList.get(variableIndex[0] - 1) == 0 ? true : false; Map<String, String> valueLabelPair = new LinkedHashMap<String, String>(); if (isNumeric) { // numeric variable dbgLog.fine("processing of a numeric value-label table"); for (int j = 0; j < numberOfValueLabels; j++) { valueLabelPair.put(doubleNumberFormatter.format(tempBB[j].getDouble()), valueLabel[j]); } } else { // String variable dbgLog.fine("processing of a string value-label table"); for (int j = 0; j < numberOfValueLabels; j++) { valueLabelPair.put( StringUtils.stripEnd(new String((tempBB[j].array()), defaultCharSet), " "), valueLabel[j]); } } dbgLog.fine("valueLabePair=" + valueLabelPair); dbgLog.fine("key variable's (raw) index =" + variableIndex[0]); valueLabelTable.put(OBSIndexToVariableName.get(variableIndex[0] - 1), valueLabelPair); dbgLog.fine("valueLabelTable=" + valueLabelTable); // create a mapping table that finds the key variable for this mapping table String keyVariableName = OBSIndexToVariableName.get(variableIndex[0] - 1); for (int vn : variableIndex) { valueVariableMappingTable.put(OBSIndexToVariableName.get(vn - 1), keyVariableName); } dbgLog.fine("valueVariableMappingTable:\n" + valueVariableMappingTable); } catch (IOException ex) { //ex.printStackTrace(); throw ex; } safteyCounter++; if (safteyCounter >= 1000000) { break; } } //while ///smd.setValueLabelTable(valueLabelTable); assignValueLabels(valueLabelTable); dbgLog.fine("***** decodeRecordType3and4(): end *****"); }
From source file:edu.harvard.iq.dvn.ingest.statdataio.impl.plugins.sav.SAVFileReader.java
void decodeRecordType3and4(BufferedInputStream stream) throws IOException { dbgLog.fine("***** decodeRecordType3and4(): start *****"); Map<String, Map<String, String>> valueLabelTable = new LinkedHashMap<String, Map<String, String>>(); int safteyCounter = 0; while (true) { try {/*from ww w . j a v a 2 s . c o m*/ if (stream == null) { throw new IllegalArgumentException("stream == null!"); } // this secton may not exit so first check the 4-byte header value //if (stream.markSupported()){ stream.mark(1000); //} // 3.0 check the first 4 bytes byte[] headerCode = new byte[LENGTH_RECORD_TYPE3_CODE]; int nbytes_rt3 = stream.read(headerCode, 0, LENGTH_RECORD_TYPE3_CODE); // to-do check against nbytes //printHexDump(headerCode, "RT3 header test"); ByteBuffer bb_header_code = ByteBuffer.wrap(headerCode, 0, LENGTH_RECORD_TYPE3_CODE); if (isLittleEndian) { bb_header_code.order(ByteOrder.LITTLE_ENDIAN); } int intRT3test = bb_header_code.getInt(); dbgLog.fine("header test value: RT3=" + intRT3test); if (intRT3test != 3) { //if (stream.markSupported()){ dbgLog.fine("iteration=" + safteyCounter); // We have encountered a record that's not type 3. This means we've // processed all the type 3/4 record pairs. So we want to rewind // the stream and return -- so that the appropriate record type // reader can be called on it. // But before we return, we need to save all the value labels // we have found: smd.setValueLabelTable(valueLabelTable); stream.reset(); return; //} } // 3.1 how many value-label pairs follow byte[] number_of_labels = new byte[LENGTH_RT3_HOW_MANY_LABELS]; int nbytes_3_1 = stream.read(number_of_labels); if (nbytes_3_1 == 0) { throw new IOException("RT 3: reading recordType3.1: no byte was read"); } ByteBuffer bb_number_of_labels = ByteBuffer.wrap(number_of_labels, 0, LENGTH_RT3_HOW_MANY_LABELS); if (isLittleEndian) { bb_number_of_labels.order(ByteOrder.LITTLE_ENDIAN); } int numberOfValueLabels = bb_number_of_labels.getInt(); dbgLog.fine("number of value-label pairs=" + numberOfValueLabels); ByteBuffer[] tempBB = new ByteBuffer[numberOfValueLabels]; String valueLabel[] = new String[numberOfValueLabels]; for (int i = 0; i < numberOfValueLabels; i++) { // read 8-byte as value byte[] value = new byte[LENGTH_RT3_VALUE]; int nbytes_3_value = stream.read(value); if (nbytes_3_value == 0) { throw new IOException("RT 3: reading recordType3 value: no byte was read"); } // note these 8 bytes are interpreted later // currently no information about which variable's (=> type unknown) ByteBuffer bb_value = ByteBuffer.wrap(value, 0, LENGTH_RT3_VALUE); if (isLittleEndian) { bb_value.order(ByteOrder.LITTLE_ENDIAN); } tempBB[i] = bb_value; dbgLog.fine("bb_value=" + Hex.encodeHex(bb_value.array())); /* double valueD = bb_value.getDouble(); dbgLog.fine("value="+valueD); */ // read 1st byte as unsigned integer = label_length // read label_length byte as label byte[] labelLengthByte = new byte[LENGTH_RT3_LABEL_LENGTH]; int nbytes_3_label_length = stream.read(labelLengthByte); // add check-routine here dbgLog.fine("labelLengthByte" + Hex.encodeHex(labelLengthByte)); dbgLog.fine("label length = " + labelLengthByte[0]); // the net-length of a value label is saved as // unsigned byte; however, the length is less than 127 // byte should be ok int rawLabelLength = labelLengthByte[0] & 0xFF; dbgLog.fine("rawLabelLength=" + rawLabelLength); // -1 =>1-byte already read int labelLength = getSAVobsAdjustedBlockLength(rawLabelLength + 1) - 1; byte[] valueLabelBytes = new byte[labelLength]; int nbytes_3_value_label = stream.read(valueLabelBytes); // ByteBuffer bb_label = ByteBuffer.wrap(valueLabel,0,labelLength); valueLabel[i] = StringUtils.stripEnd( new String(Arrays.copyOfRange(valueLabelBytes, 0, rawLabelLength), defaultCharSet), " "); dbgLog.fine(i + "-th valueLabel=" + valueLabel[i] + "<-"); } // iter rt3 dbgLog.fine("end of RT3 block"); dbgLog.fine("start of RT4 block"); // 4.0 check the first 4 bytes byte[] headerCode4 = new byte[LENGTH_RECORD_TYPE4_CODE]; int nbytes_rt4 = stream.read(headerCode4, 0, LENGTH_RECORD_TYPE4_CODE); if (nbytes_rt4 == 0) { throw new IOException("RT4: reading recordType4 value: no byte was read"); } //printHexDump(headerCode4, "RT4 header test"); ByteBuffer bb_header_code_4 = ByteBuffer.wrap(headerCode4, 0, LENGTH_RECORD_TYPE4_CODE); if (isLittleEndian) { bb_header_code_4.order(ByteOrder.LITTLE_ENDIAN); } int intRT4test = bb_header_code_4.getInt(); dbgLog.fine("header test value: RT4=" + intRT4test); if (intRT4test != 4) { throw new IOException("RT 4: reading recordType4 header: no byte was read"); } // 4.1 read the how-many-variables bytes byte[] howManyVariablesfollow = new byte[LENGTH_RT4_HOW_MANY_VARIABLES]; int nbytes_rt4_1 = stream.read(howManyVariablesfollow, 0, LENGTH_RT4_HOW_MANY_VARIABLES); ByteBuffer bb_howManyVariablesfollow = ByteBuffer.wrap(howManyVariablesfollow, 0, LENGTH_RT4_HOW_MANY_VARIABLES); if (isLittleEndian) { bb_howManyVariablesfollow.order(ByteOrder.LITTLE_ENDIAN); } int howManyVariablesRT4 = bb_howManyVariablesfollow.getInt(); dbgLog.fine("how many variables follow: RT4=" + howManyVariablesRT4); int length_indicies = LENGTH_RT4_VARIABLE_INDEX * howManyVariablesRT4; byte[] variableIdicesBytes = new byte[length_indicies]; int nbytes_rt4_2 = stream.read(variableIdicesBytes, 0, length_indicies); // !!!!! Caution: variableIndex in RT4 starts from 1 NOT ** 0 ** int[] variableIndex = new int[howManyVariablesRT4]; int offset = 0; for (int i = 0; i < howManyVariablesRT4; i++) { ByteBuffer bb_variable_index = ByteBuffer.wrap(variableIdicesBytes, offset, LENGTH_RT4_VARIABLE_INDEX); offset += LENGTH_RT4_VARIABLE_INDEX; if (isLittleEndian) { bb_variable_index.order(ByteOrder.LITTLE_ENDIAN); } variableIndex[i] = bb_variable_index.getInt(); dbgLog.fine(i + "-th variable index number=" + variableIndex[i]); } dbgLog.fine("variable index set=" + ArrayUtils.toString(variableIndex)); dbgLog.fine("subtract 1 from variableIndex for getting a variable info"); boolean isNumeric = OBSwiseTypelList.get(variableIndex[0] - 1) == 0 ? true : false; Map<String, String> valueLabelPair = new LinkedHashMap<String, String>(); if (isNumeric) { // numeric variable dbgLog.fine("processing of a numeric value-label table"); for (int j = 0; j < numberOfValueLabels; j++) { valueLabelPair.put(doubleNumberFormatter.format(tempBB[j].getDouble()), valueLabel[j]); } } else { // String variable dbgLog.fine("processing of a string value-label table"); for (int j = 0; j < numberOfValueLabels; j++) { valueLabelPair.put( StringUtils.stripEnd(new String((tempBB[j].array()), defaultCharSet), " "), valueLabel[j]); } } dbgLog.fine("valueLabePair=" + valueLabelPair); dbgLog.fine("key variable's (raw) index =" + variableIndex[0]); valueLabelTable.put(OBSIndexToVariableName.get(variableIndex[0] - 1), valueLabelPair); dbgLog.fine("valueLabelTable=" + valueLabelTable); // create a mapping table that finds the key variable for this mapping table String keyVariableName = OBSIndexToVariableName.get(variableIndex[0] - 1); for (int vn : variableIndex) { valueVariableMappingTable.put(OBSIndexToVariableName.get(vn - 1), keyVariableName); } dbgLog.fine("valueVariableMappingTable:\n" + valueVariableMappingTable); } catch (IOException ex) { //ex.printStackTrace(); throw ex; } safteyCounter++; if (safteyCounter >= 1000000) { break; } } //while smd.setValueLabelTable(valueLabelTable); dbgLog.fine("***** decodeRecordType3and4(): end *****"); }
From source file:edu.harvard.iq.dataverse.ingest.tabulardata.impl.plugins.sav.SAVFileReader.java
void decodeRecordType7(BufferedInputStream stream) throws IOException { dbgLog.fine("decodeRecordType7(): start"); int counter = 0; int[] headerSection = new int[2]; // the variables below may no longer needed; // but they may be useful for debugging/logging purposes. /// // RecordType 7 /// // Subtype 3 /// List<Integer> releaseMachineSpecificInfo = new ArrayList<Integer>(); /// List<String> releaseMachineSpecificInfoHex = new ArrayList<String>(); /// // Subytpe 4 /// Map<String, Double> OBSTypeValue = new LinkedHashMap<String, Double>(); /// Map<String, String> OBSTypeHexValue = new LinkedHashMap<String, String>(); //Subtype 11/* www. jav a 2s. c o m*/ /// List<Integer> measurementLevel = new ArrayList<Integer>(); /// List<Integer> columnWidth = new ArrayList<Integer>(); /// List<Integer> alignment = new ArrayList<Integer>(); while (true) { try { if (stream == null) { throw new IllegalArgumentException("RT7: stream == null!"); } // first check the 4-byte header value //if (stream.markSupported()){ stream.mark(1000); //} // 7.0 check the first 4 bytes byte[] headerCodeRt7 = new byte[LENGTH_RECORD_TYPE7_CODE]; int nbytes_rt7 = stream.read(headerCodeRt7, 0, LENGTH_RECORD_TYPE7_CODE); // to-do check against nbytes //printHexDump(headerCodeRt7, "RT7 header test"); ByteBuffer bb_header_code_rt7 = ByteBuffer.wrap(headerCodeRt7, 0, LENGTH_RECORD_TYPE7_CODE); if (isLittleEndian) { bb_header_code_rt7.order(ByteOrder.LITTLE_ENDIAN); } int intRT7test = bb_header_code_rt7.getInt(); dbgLog.fine("RT7: header test value=" + intRT7test); if (intRT7test != 7) { //if (stream.markSupported()){ //out.print("iteration="+safteyCounter); //dbgLog.fine("iteration="+safteyCounter); dbgLog.fine("intRT7test failed=" + intRT7test); dbgLog.fine("counter=" + counter); stream.reset(); return; //} } // 7.1 check 4-byte integer Sub-Type Code byte[] length_sub_type_code = new byte[LENGTH_RT7_SUB_TYPE_CODE]; int nbytes_rt7_1 = stream.read(length_sub_type_code, 0, LENGTH_RT7_SUB_TYPE_CODE); // to-do check against nbytes //printHexDump(length_how_many_line_bytes, "RT7 how_many_line_bytes"); ByteBuffer bb_sub_type_code = ByteBuffer.wrap(length_sub_type_code, 0, LENGTH_RT7_SUB_TYPE_CODE); if (isLittleEndian) { bb_sub_type_code.order(ByteOrder.LITTLE_ENDIAN); } int subTypeCode = bb_sub_type_code.getInt(); dbgLog.fine("RT7: subTypeCode=" + subTypeCode); switch (subTypeCode) { case 3: // 3: Release andMachine-Specific Integer Information //parseRT7SubTypefield(stream); headerSection = parseRT7SubTypefieldHeader(stream); if (headerSection != null) { int unitLength = headerSection[0]; int numberOfUnits = headerSection[1]; for (int i = 0; i < numberOfUnits; i++) { dbgLog.finer(i + "-th fieldData"); byte[] work = new byte[unitLength]; int nb = stream.read(work); dbgLog.finer("raw bytes in Hex:" + new String(Hex.encodeHex(work))); ByteBuffer bb_field = ByteBuffer.wrap(work); if (isLittleEndian) { bb_field.order(ByteOrder.LITTLE_ENDIAN); } String dataInHex = new String(Hex.encodeHex(bb_field.array())); /// releaseMachineSpecificInfoHex.add(dataInHex); dbgLog.finer("raw bytes in Hex:" + dataInHex); if (unitLength == 4) { int fieldData = bb_field.getInt(); dbgLog.finer("fieldData(int)=" + fieldData); dbgLog.finer("fieldData in Hex=0x" + Integer.toHexString(fieldData)); /// releaseMachineSpecificInfo.add(fieldData); } } /// dbgLog.fine("releaseMachineSpecificInfo="+releaseMachineSpecificInfo); /// dbgLog.fine("releaseMachineSpecificInfoHex="+releaseMachineSpecificInfoHex); } else { // throw new IOException } dbgLog.fine("***** end of subType 3 ***** \n"); break; case 4: // Release andMachine-SpecificOBS-Type Information headerSection = parseRT7SubTypefieldHeader(stream); if (headerSection != null) { int unitLength = headerSection[0]; int numberOfUnits = headerSection[1]; for (int i = 0; i < numberOfUnits; i++) { dbgLog.finer(i + "-th fieldData:" + RecordType7SubType4Fields.get(i)); byte[] work = new byte[unitLength]; int nb = stream.read(work); dbgLog.finer("raw bytes in Hex:" + new String(Hex.encodeHex(work))); ByteBuffer bb_field = ByteBuffer.wrap(work); dbgLog.finer("byte order=" + bb_field.order().toString()); if (isLittleEndian) { bb_field.order(ByteOrder.LITTLE_ENDIAN); } ByteBuffer bb_field_dup = bb_field.duplicate(); OBSTypeHexValue.put(RecordType7SubType4Fields.get(i), new String(Hex.encodeHex(bb_field.array()))); // dbgLog.finer("raw bytes in Hex:"+ // OBSTypeHexValue.get(RecordType7SubType4Fields.get(i))); if (unitLength == 8) { double fieldData = bb_field.getDouble(); /// OBSTypeValue.put(RecordType7SubType4Fields.get(i), fieldData); dbgLog.finer("fieldData(double)=" + fieldData); OBSTypeHexValue.put(RecordType7SubType4Fields.get(i), Double.toHexString(fieldData)); dbgLog.fine("fieldData in Hex=" + Double.toHexString(fieldData)); } } /// dbgLog.fine("OBSTypeValue="+OBSTypeValue); /// dbgLog.fine("OBSTypeHexValue="+OBSTypeHexValue); } else { // throw new IOException } dbgLog.fine("***** end of subType 4 ***** \n"); break; case 5: // Variable Sets Information parseRT7SubTypefield(stream); break; case 6: // Trends date information parseRT7SubTypefield(stream); break; case 7: // Multiple response groups parseRT7SubTypefield(stream); break; case 8: // Windows Data Entry data parseRT7SubTypefield(stream); break; case 9: // parseRT7SubTypefield(stream); break; case 10: // TextSmart data parseRT7SubTypefield(stream); break; case 11: // Msmt level, col width, & alignment //parseRT7SubTypefield(stream); headerSection = parseRT7SubTypefieldHeader(stream); if (headerSection != null) { int unitLength = headerSection[0]; int numberOfUnits = headerSection[1]; for (int i = 0; i < numberOfUnits; i++) { dbgLog.finer(i + "-th fieldData"); byte[] work = new byte[unitLength]; int nb = stream.read(work); dbgLog.finer("raw bytes in Hex:" + new String(Hex.encodeHex(work))); ByteBuffer bb_field = ByteBuffer.wrap(work); if (isLittleEndian) { bb_field.order(ByteOrder.LITTLE_ENDIAN); } dbgLog.finer("raw bytes in Hex:" + new String(Hex.encodeHex(bb_field.array()))); if (unitLength == 4) { int fieldData = bb_field.getInt(); dbgLog.finer("fieldData(int)=" + fieldData); dbgLog.finer("fieldData in Hex=0x" + Integer.toHexString(fieldData)); int remainder = i % 3; dbgLog.finer("remainder=" + remainder); if (remainder == 0) { /// measurementLevel.add(fieldData); } else if (remainder == 1) { /// columnWidth.add(fieldData); } else if (remainder == 2) { /// alignment.add(fieldData); } } } } else { // throw new IOException } /// dbgLog.fine("measurementLevel="+measurementLevel); /// dbgLog.fine("columnWidth="+columnWidth); /// dbgLog.fine("alignment="+alignment); dbgLog.fine("end of subType 11\n"); break; case 12: // Windows Data Entry GUID parseRT7SubTypefield(stream); break; case 13: // Extended variable names // parseRT7SubTypefield(stream); headerSection = parseRT7SubTypefieldHeader(stream); if (headerSection != null) { int unitLength = headerSection[0]; dbgLog.fine("RT7: unitLength=" + unitLength); int numberOfUnits = headerSection[1]; dbgLog.fine("RT7: numberOfUnits=" + numberOfUnits); byte[] work = new byte[unitLength * numberOfUnits]; int nbtyes13 = stream.read(work); String[] variableShortLongNamePairs = new String(work, "US-ASCII").split("\t"); for (int i = 0; i < variableShortLongNamePairs.length; i++) { dbgLog.fine("RT7: " + i + "-th pair" + variableShortLongNamePairs[i]); String[] pair = variableShortLongNamePairs[i].split("="); shortToLongVariableNameTable.put(pair[0], pair[1]); } dbgLog.fine("RT7: shortToLongVarialbeNameTable" + shortToLongVariableNameTable); // We are saving the short-to-long name map; at the // end of ingest, we'll go through the data variables and // change the names accordingly. // smd.setShortToLongVarialbeNameTable(shortToLongVarialbeNameTable); } else { // throw new IOException } break; case 14: // Extended strings //parseRT7SubTypefield(stream); headerSection = parseRT7SubTypefieldHeader(stream); if (headerSection != null) { int unitLength = headerSection[0]; dbgLog.fine("RT7.14: unitLength=" + unitLength); int numberOfUnits = headerSection[1]; dbgLog.fine("RT7.14: numberOfUnits=" + numberOfUnits); byte[] work = new byte[unitLength * numberOfUnits]; int nbtyes13 = stream.read(work); String[] extendedVariablesSizePairs = new String(work, defaultCharSet).split("\000\t"); for (int i = 0; i < extendedVariablesSizePairs.length; i++) { dbgLog.fine("RT7.14: " + i + "-th pair" + extendedVariablesSizePairs[i]); if (extendedVariablesSizePairs[i].indexOf("=") > 0) { String[] pair = extendedVariablesSizePairs[i].split("="); extendedVariablesSizeTable.put(pair[0], Integer.valueOf(pair[1])); } } dbgLog.fine("RT7.14: extendedVariablesSizeTable" + extendedVariablesSizeTable); } else { // throw new IOException } break; case 15: // Clementine Metadata parseRT7SubTypefield(stream); break; case 16: // 64 bit N of cases parseRT7SubTypefield(stream); break; case 17: // File level attributes parseRT7SubTypefield(stream); break; case 18: // Variable attributes parseRT7SubTypefield(stream); break; case 19: // Extended multiple response groups parseRT7SubTypefield(stream); break; case 20: // Character encoding, aka code page. // Must be a version 16+ feature (?). // Starting v.16, the default character encoding for SAV // files is UTF-8; but then it is possible to specify an // alternative encoding here. // A typical use case would be people setting it to "ISO-Latin" // or "windows-1252", or a similar 8-bit encoding to store // text with standard Western European accents. // -- L.A. headerSection = parseRT7SubTypefieldHeader(stream); if (headerSection != null) { int unitLength = headerSection[0]; dbgLog.fine("RT7-20: unitLength=" + unitLength); int numberOfUnits = headerSection[1]; dbgLog.fine("RT7-20: numberOfUnits=" + numberOfUnits); byte[] rt7st20bytes = new byte[unitLength * numberOfUnits]; int nbytes20 = stream.read(rt7st20bytes); String dataCharSet = new String(rt7st20bytes, "US-ASCII"); if (dataCharSet != null && !(dataCharSet.equals(""))) { dbgLog.fine("RT7-20: data charset: " + dataCharSet); defaultCharSet = dataCharSet; } } /*else { // TODO: // decide if the exception should actually be thrown here! // -- L.A. 4.0 beta // throw new IOException }*/ break; case 21: // Value labels for long strings parseRT7SubTypefield(stream); break; case 22: // Missing values for long strings parseRT7SubTypefield(stream); break; default: parseRT7SubTypefield(stream); } } catch (IOException ex) { //ex.printStackTrace(); throw ex; } counter++; if (counter > 20) { break; } } dbgLog.fine("RT7: counter=" + counter); dbgLog.fine("RT7: decodeRecordType7(): end"); }
From source file:edu.harvard.iq.dvn.ingest.statdataio.impl.plugins.sav.SAVFileReader.java
void decodeRecordType7(BufferedInputStream stream) throws IOException { dbgLog.fine("***** decodeRecordType7(): start *****"); int counter = 0; int[] headerSection = new int[2]; // the variables below may no longer needed; // but they may be useful for debugging/logging purposes. /// // RecordType 7 /// // Subtype 3 /// List<Integer> releaseMachineSpecificInfo = new ArrayList<Integer>(); /// List<String> releaseMachineSpecificInfoHex = new ArrayList<String>(); /// // Subytpe 4 /// Map<String, Double> OBSTypeValue = new LinkedHashMap<String, Double>(); /// Map<String, String> OBSTypeHexValue = new LinkedHashMap<String, String>(); //Subtype 11//from www . j av a2s . c o m /// List<Integer> measurementLevel = new ArrayList<Integer>(); /// List<Integer> columnWidth = new ArrayList<Integer>(); /// List<Integer> alignment = new ArrayList<Integer>(); Map<String, String> shortToLongVarialbeNameTable = new LinkedHashMap<String, String>(); while (true) { try { if (stream == null) { throw new IllegalArgumentException("RT7: stream == null!"); } // first check the 4-byte header value //if (stream.markSupported()){ stream.mark(1000); //} // 7.0 check the first 4 bytes byte[] headerCodeRt7 = new byte[LENGTH_RECORD_TYPE7_CODE]; int nbytes_rt7 = stream.read(headerCodeRt7, 0, LENGTH_RECORD_TYPE7_CODE); // to-do check against nbytes //printHexDump(headerCodeRt7, "RT7 header test"); ByteBuffer bb_header_code_rt7 = ByteBuffer.wrap(headerCodeRt7, 0, LENGTH_RECORD_TYPE7_CODE); if (isLittleEndian) { bb_header_code_rt7.order(ByteOrder.LITTLE_ENDIAN); } int intRT7test = bb_header_code_rt7.getInt(); dbgLog.fine("RT7: header test value=" + intRT7test); if (intRT7test != 7) { //if (stream.markSupported()){ //out.print("iteration="+safteyCounter); //dbgLog.fine("iteration="+safteyCounter); dbgLog.fine("intRT7test failed=" + intRT7test); dbgLog.fine("counter=" + counter); stream.reset(); return; //} } // 7.1 check 4-byte integer Sub-Type Code byte[] length_sub_type_code = new byte[LENGTH_RT7_SUB_TYPE_CODE]; int nbytes_rt7_1 = stream.read(length_sub_type_code, 0, LENGTH_RT7_SUB_TYPE_CODE); // to-do check against nbytes //printHexDump(length_how_many_line_bytes, "RT7 how_many_line_bytes"); ByteBuffer bb_sub_type_code = ByteBuffer.wrap(length_sub_type_code, 0, LENGTH_RT7_SUB_TYPE_CODE); if (isLittleEndian) { bb_sub_type_code.order(ByteOrder.LITTLE_ENDIAN); } int subTypeCode = bb_sub_type_code.getInt(); dbgLog.fine("RT7: subTypeCode=" + subTypeCode); switch (subTypeCode) { case 3: // 3: Release andMachine-Specific Integer Information //parseRT7SubTypefield(stream); headerSection = parseRT7SubTypefieldHeader(stream); if (headerSection != null) { int unitLength = headerSection[0]; int numberOfUnits = headerSection[1]; for (int i = 0; i < numberOfUnits; i++) { dbgLog.finer(i + "-th fieldData"); byte[] work = new byte[unitLength]; int nb = stream.read(work); dbgLog.finer("raw bytes in Hex:" + new String(Hex.encodeHex(work))); ByteBuffer bb_field = ByteBuffer.wrap(work); if (isLittleEndian) { bb_field.order(ByteOrder.LITTLE_ENDIAN); } String dataInHex = new String(Hex.encodeHex(bb_field.array())); /// releaseMachineSpecificInfoHex.add(dataInHex); dbgLog.finer("raw bytes in Hex:" + dataInHex); if (unitLength == 4) { int fieldData = bb_field.getInt(); dbgLog.finer("fieldData(int)=" + fieldData); dbgLog.finer("fieldData in Hex=0x" + Integer.toHexString(fieldData)); /// releaseMachineSpecificInfo.add(fieldData); } } /// dbgLog.fine("releaseMachineSpecificInfo="+releaseMachineSpecificInfo); /// dbgLog.fine("releaseMachineSpecificInfoHex="+releaseMachineSpecificInfoHex); } else { // throw new IOException } dbgLog.fine("***** end of subType 3 ***** \n"); break; case 4: // Release andMachine-SpecificOBS-Type Information headerSection = parseRT7SubTypefieldHeader(stream); if (headerSection != null) { int unitLength = headerSection[0]; int numberOfUnits = headerSection[1]; for (int i = 0; i < numberOfUnits; i++) { dbgLog.finer(i + "-th fieldData:" + RecordType7SubType4Fields.get(i)); byte[] work = new byte[unitLength]; int nb = stream.read(work); dbgLog.finer("raw bytes in Hex:" + new String(Hex.encodeHex(work))); ByteBuffer bb_field = ByteBuffer.wrap(work); dbgLog.finer("byte order=" + bb_field.order().toString()); if (isLittleEndian) { bb_field.order(ByteOrder.LITTLE_ENDIAN); } ByteBuffer bb_field_dup = bb_field.duplicate(); OBSTypeHexValue.put(RecordType7SubType4Fields.get(i), new String(Hex.encodeHex(bb_field.array()))); // dbgLog.finer("raw bytes in Hex:"+ // OBSTypeHexValue.get(RecordType7SubType4Fields.get(i))); if (unitLength == 8) { double fieldData = bb_field.getDouble(); /// OBSTypeValue.put(RecordType7SubType4Fields.get(i), fieldData); dbgLog.finer("fieldData(double)=" + fieldData); OBSTypeHexValue.put(RecordType7SubType4Fields.get(i), Double.toHexString(fieldData)); dbgLog.fine("fieldData in Hex=" + Double.toHexString(fieldData)); } } /// dbgLog.fine("OBSTypeValue="+OBSTypeValue); /// dbgLog.fine("OBSTypeHexValue="+OBSTypeHexValue); } else { // throw new IOException } dbgLog.fine("***** end of subType 4 ***** \n"); break; case 5: // Variable Sets Information parseRT7SubTypefield(stream); break; case 6: // Trends date information parseRT7SubTypefield(stream); break; case 7: // Multiple response groups parseRT7SubTypefield(stream); break; case 8: // Windows Data Entry data parseRT7SubTypefield(stream); break; case 9: // parseRT7SubTypefield(stream); break; case 10: // TextSmart data parseRT7SubTypefield(stream); break; case 11: // Msmt level, col width, & alignment //parseRT7SubTypefield(stream); headerSection = parseRT7SubTypefieldHeader(stream); if (headerSection != null) { int unitLength = headerSection[0]; int numberOfUnits = headerSection[1]; for (int i = 0; i < numberOfUnits; i++) { dbgLog.finer(i + "-th fieldData"); byte[] work = new byte[unitLength]; int nb = stream.read(work); dbgLog.finer("raw bytes in Hex:" + new String(Hex.encodeHex(work))); ByteBuffer bb_field = ByteBuffer.wrap(work); if (isLittleEndian) { bb_field.order(ByteOrder.LITTLE_ENDIAN); } dbgLog.finer("raw bytes in Hex:" + new String(Hex.encodeHex(bb_field.array()))); if (unitLength == 4) { int fieldData = bb_field.getInt(); dbgLog.finer("fieldData(int)=" + fieldData); dbgLog.finer("fieldData in Hex=0x" + Integer.toHexString(fieldData)); int remainder = i % 3; dbgLog.finer("remainder=" + remainder); if (remainder == 0) { /// measurementLevel.add(fieldData); } else if (remainder == 1) { /// columnWidth.add(fieldData); } else if (remainder == 2) { /// alignment.add(fieldData); } } } } else { // throw new IOException } /// dbgLog.fine("measurementLevel="+measurementLevel); /// dbgLog.fine("columnWidth="+columnWidth); /// dbgLog.fine("alignment="+alignment); dbgLog.fine("***** end of subType 11 ***** \n"); break; case 12: // Windows Data Entry GUID parseRT7SubTypefield(stream); break; case 13: // Extended variable names // parseRT7SubTypefield(stream); headerSection = parseRT7SubTypefieldHeader(stream); if (headerSection != null) { int unitLength = headerSection[0]; dbgLog.fine("RT7: unitLength=" + unitLength); int numberOfUnits = headerSection[1]; dbgLog.fine("RT7: numberOfUnits=" + numberOfUnits); byte[] work = new byte[unitLength * numberOfUnits]; int nbtyes13 = stream.read(work); String[] variableShortLongNamePairs = new String(work, "US-ASCII").split("\t"); for (int i = 0; i < variableShortLongNamePairs.length; i++) { dbgLog.fine("RT7: " + i + "-th pair" + variableShortLongNamePairs[i]); String[] pair = variableShortLongNamePairs[i].split("="); shortToLongVarialbeNameTable.put(pair[0], pair[1]); } dbgLog.fine("RT7: shortToLongVarialbeNameTable" + shortToLongVarialbeNameTable); smd.setShortToLongVarialbeNameTable(shortToLongVarialbeNameTable); } else { // throw new IOException } break; case 14: // Extended strings //parseRT7SubTypefield(stream); headerSection = parseRT7SubTypefieldHeader(stream); if (headerSection != null) { int unitLength = headerSection[0]; dbgLog.fine("RT7.14: unitLength=" + unitLength); int numberOfUnits = headerSection[1]; dbgLog.fine("RT7.14: numberOfUnits=" + numberOfUnits); byte[] work = new byte[unitLength * numberOfUnits]; int nbtyes13 = stream.read(work); String[] extendedVariablesSizePairs = new String(work, defaultCharSet).split("\000\t"); for (int i = 0; i < extendedVariablesSizePairs.length; i++) { dbgLog.fine("RT7.14: " + i + "-th pair" + extendedVariablesSizePairs[i]); if (extendedVariablesSizePairs[i].indexOf("=") > 0) { String[] pair = extendedVariablesSizePairs[i].split("="); extendedVariablesSizeTable.put(pair[0], Integer.valueOf(pair[1])); } } dbgLog.fine("RT7.14: extendedVariablesSizeTable" + extendedVariablesSizeTable); } else { // throw new IOException } break; case 15: // Clementine Metadata parseRT7SubTypefield(stream); break; case 16: // 64 bit N of cases parseRT7SubTypefield(stream); break; case 17: // File level attributes parseRT7SubTypefield(stream); break; case 18: // Variable attributes parseRT7SubTypefield(stream); break; case 19: // Extended multiple response groups parseRT7SubTypefield(stream); break; case 20: // Encoding, aka code page parseRT7SubTypefield(stream); /* TODO: This needs to be researched; * Is this field really used, ever? headerSection = parseRT7SubTypefieldHeader(stream); if (headerSection != null){ int unitLength = headerSection[0]; dbgLog.fine("RT7-20: unitLength="+unitLength); int numberOfUnits = headerSection[1]; dbgLog.fine("RT7-20: numberOfUnits="+numberOfUnits); byte[] rt7st20bytes = new byte[unitLength*numberOfUnits]; int nbytes20 = stream.read(rt7st20bytes); String dataCharSet = new String(rt7st20bytes,"US-ASCII"); if (dataCharSet != null && !(dataCharSet.equals(""))) { dbgLog.fine("RT7-20: data charset: "+ dataCharSet); defaultCharSet = dataCharSet; } } else { // throw new IOException } * */ break; case 21: // Value labels for long strings parseRT7SubTypefield(stream); break; case 22: // Missing values for long strings parseRT7SubTypefield(stream); break; default: parseRT7SubTypefield(stream); } } catch (IOException ex) { //ex.printStackTrace(); throw ex; } counter++; if (counter > 20) { break; } } dbgLog.fine("RT7: counter=" + counter); dbgLog.fine("RT7: ***** decodeRecordType7(): end *****"); }