List of usage examples for org.jdom2 Document Document
public Document(List<? extends Content> content)
Document, with the supplied list of content, and a DocType declaration only if the content contains a DocType instance. From source file:edu.wisc.ssec.mcidasv.data.hydra.SuomiNPPDataSource.java
License:Open Source License
public void setup() throws VisADException { // which format, NASA or NOAA? isNOAA = false;//from w ww . j a v a 2 s. co m // store filenames for possible bundle unpersistence for (Object o : sources) { oldSources.add((String) o); } // time zone for product labels SimpleTimeZone stz = new SimpleTimeZone(0, "GMT"); sdf.setTimeZone(stz); sdfOut.setTimeZone(stz); // looking to populate 3 things - path to lat, path to lon, path to relevant products String pathToLat = null; String pathToLon = null; Set<String> pathToProducts = new LinkedHashSet<>(); Map<String, String> prodToDesc = new HashMap<>(); // flag to differentiate VIIRS from one of the other Suomi sensors boolean isVIIRS = true; // check source filenames to see if this is a combined product. everything // from last file separator to first underscore should be product info int lastSeparator = filename.lastIndexOf(File.separatorChar); int firstUnderscore = filename.indexOf("_", lastSeparator + 1); String prodStr = filename.substring(lastSeparator + 1, firstUnderscore); // only do this check for NOAA data if (filename.endsWith(".h5")) { isNOAA = true; StringTokenizer st = new StringTokenizer(prodStr, "-"); logger.debug("SNPPDS check for embedded GEO, tokenizing: " + prodStr); while (st.hasMoreTokens()) { String singleProd = st.nextToken(); for (int i = 0; i < JPSSUtilities.geoProductIDs.length; i++) { if (singleProd.equals(JPSSUtilities.geoProductIDs[i])) { logger.debug("Setting isCombinedProduct true, Found embedded GEO: " + singleProd); isCombinedProduct = true; break; } } } } // various metatdata we'll need to gather on a per-product basis Map<String, String> unsignedFlags = new LinkedHashMap<>(); Map<String, String> unpackFlags = new LinkedHashMap<>(); // geo product IDs for each granule Set<String> geoProductIDs = new LinkedHashSet<>(); // aggregations will use sets of NetCDFFile readers List<NetCDFFile> ncdfal = new ArrayList<>(); // we should be able to find an XML Product Profile for each data/product type SuomiNPPProductProfile nppPP = null; // and also Profile metadata for geolocation variables boolean haveGeoMetaData = false; // number of source granules which make up the data source int granuleCount = 1; try { nppPP = new SuomiNPPProductProfile(); // for each source file provided, find the appropriate geolocation, // get the nominal time and various other granule-level metadata Iterator keyIterator = filenameMap.keySet().iterator(); while (keyIterator.hasNext()) { String keyStr = (String) keyIterator.next(); List fileNames = (List) filenameMap.get(keyStr); granuleCount = fileNames.size(); setProperty(Constants.PROP_GRANULE_COUNT, granuleCount + " Granule"); for (int fileCount = 0; fileCount < granuleCount; fileCount++) { // need to open the main NetCDF file to determine the geolocation product NetcdfFile ncfile = null; String fileAbsPath = null; try { fileAbsPath = (String) fileNames.get(fileCount); logger.debug("Trying to open file: " + fileAbsPath); ncfile = NetcdfFile.open(fileAbsPath); if (!isCombinedProduct) { if (isNOAA) { Attribute a = ncfile.findGlobalAttribute("N_GEO_Ref"); logger.debug("Value of GEO global attribute: " + a.getStringValue()); String tmpGeoProductID = a.getStringValue(); geoProductIDs.add(tmpGeoProductID); } else { geoProductIDs.add(keyStr.replace("L1B", "GEO")); } } Group rg = ncfile.getRootGroup(); List<Group> gl = rg.getGroups(); if (gl != null) { for (Group g : gl) { logger.trace("Group name: " + g.getFullName()); if (isNOAA) { // when we find the Data_Products group, go down another group level and pull out // what we will use for nominal day and time (for now anyway). // XXX TJJ fileCount check is so we don't count the GEO file in time array! if (g.getFullName().contains("Data_Products") && (fileCount != fileNames.size())) { List<Group> dpg = g.getGroups(); // cycle through once looking for XML Product Profiles for (Group subG : dpg) { String subName = subG.getFullName(); // use actual product, not geolocation, to id XML Product Profile if (!subName.contains("-GEO")) { // determine the instrument name (VIIRS, ATMS, CrIS, OMPS) instrumentName = subG.findAttribute("Instrument_Short_Name"); // note any EDR products, will need to check for and remove // fill scans later Attribute adtt = subG.findAttribute("N_Dataset_Type_Tag"); if (adtt != null) { String baseName = adtt.getStringValue(); if ((baseName != null) && (baseName.equals("EDR"))) { // have to loop through sub groups variables to determine band List<Variable> tmpVar = subG.getVariables(); for (Variable v : tmpVar) { // if Imagery EDR attribute for band is specified, save it Attribute mBand = v.findAttribute("Band_ID"); if (mBand != null) { whichEDR = mBand.getStringValue(); } } } } // This is also where we find the attribute which tells us which // XML Product Profile to use! Attribute axpp = subG.findAttribute("N_Collection_Short_Name"); if (axpp != null) { String baseName = axpp.getStringValue(); productName = baseName; // TJJ Apr 2018 // Hack so we can look at CrIS Full Spectrum, until we can // track down existence of an official Product Profile for it. // http://mcidas.ssec.wisc.edu/inquiry-v/?inquiry=2634 // The regular SDR profile lets us visualize it. if (productName.equals("CrIS-FS-SDR")) productName = "CrIS-SDR"; String productProfileFileName = nppPP .getProfileFileName(productName); logger.trace("Found profile: " + productProfileFileName); if (productProfileFileName == null) { throw new Exception( "XML Product Profile not found in catalog"); } try { nppPP.addMetaDataFromFile(productProfileFileName); } catch (Exception nppppe) { logger.error("Error parsing XML Product Profile: " + productProfileFileName); throw new Exception("XML Product Profile Error", nppppe); } } } } // 2nd pass through sub-group to extract date/time for aggregation for (Group subG : dpg) { List<Variable> vl = subG.getVariables(); for (Variable v : vl) { Attribute aDate = v.findAttribute("AggregateBeginningDate"); Attribute aTime = v.findAttribute("AggregateBeginningTime"); // did we find the attributes we are looking for? if ((aDate != null) && (aTime != null)) { // set time for display to day/time of 1st granule examined if (!nameHasBeenSet) { String sDate = aDate.getStringValue(); String sTime = aTime.getStringValue(); logger.debug("For day/time, using: " + sDate + sTime.substring(0, sTime.indexOf('Z') - 3)); Date d = sdf.parse( sDate + sTime.substring(0, sTime.indexOf('Z') - 3)); theDate = d; setName(instrumentName.getStringValue() + " " + sdfOut.format(d)); nameHasBeenSet = true; } break; } } } if (!nameHasBeenSet) { throw new VisADException("No date time found in Suomi NPP granule"); } } } else { // NASA data - date/time from global attribute // set time for display to day/time of 1st granule examined Attribute timeStartNASA = ncfile.findGlobalAttribute("time_coverage_start"); Date d = sdfNASA.parse(timeStartNASA.getStringValue()); theDate = d; if (!nameHasBeenSet) { instrumentName = ncfile.findGlobalAttribute("instrument"); setName(instrumentName.getStringValue() + " " + sdfOut.format(d)); nameHasBeenSet = true; } } } } } catch (Exception e) { logger.warn("Exception during processing of file: " + fileAbsPath); throw (e); } finally { ncfile.close(); } } } // build each union aggregation element Iterator<String> iterator = geoProductIDs.iterator(); for (int elementNum = 0; elementNum < granuleCount; elementNum++) { String s = null; // build an XML (NCML actually) representation of the union aggregation of these two files Namespace ns = Namespace.getNamespace("http://www.unidata.ucar.edu/namespaces/netcdf/ncml-2.2"); Element root = new Element("netcdf", ns); Document document = new Document(root); Element agg = new Element("aggregation", ns); agg.setAttribute("type", "union"); // TJJ - Loop over filename map, could be several products that need to be aggregated Set set = filenameMap.keySet(); Iterator mapIter = set.iterator(); while (mapIter.hasNext()) { String key = (String) mapIter.next(); List l = (List) filenameMap.get(key); Element fData = new Element("netcdf", ns); fData.setAttribute("location", (String) l.get(elementNum)); agg.addContent(fData); s = (String) l.get(elementNum); } String geoFilename = null; Element fGeo = new Element("netcdf", ns); ; if (!isCombinedProduct) { if (isNOAA) { geoFilename = s.substring(0, s.lastIndexOf(File.separatorChar) + 1); // check if we have the whole file name or just the prefix String geoProductID = iterator.next(); if (geoProductID.endsWith("h5")) { geoFilename += geoProductID; } else { geoFilename += geoProductID; geoFilename += s.substring(s.lastIndexOf(File.separatorChar) + 6); } // Be sure file as specified by N_GEO_Ref global attribute really is there. File tmpGeo = new File(geoFilename); if (!tmpGeo.exists()) { // Ok, the expected file defined (supposedly) exactly by a global att is not there... // We need to check for similar geo files with different creation dates String geoFileRelative = geoFilename .substring(geoFilename.lastIndexOf(File.separatorChar) + 1); // also check for Terrain Corrected version of geo String geoTerrainCorrected = geoFileRelative; geoTerrainCorrected = geoTerrainCorrected.replace("OD", "TC"); geoTerrainCorrected = geoTerrainCorrected.replace("MG", "TC"); // now we make a file filter, and see if a matching geo file is present File fList = new File( geoFilename.substring(0, geoFilename.lastIndexOf(File.separatorChar) + 1)); // current directory FilenameFilter geoFilter = new FilenameFilter() { public boolean accept(File dir, String name) { if (name.matches(JPSSUtilities.SUOMI_GEO_REGEX_NOAA)) { return true; } else { return false; } } }; File[] files = fList.listFiles(geoFilter); for (File file : files) { if (file.isDirectory()) { continue; } // get the file name for convenience String fName = file.getName(); // is it one of the standard Ellipsoid geo types we are looking for? if (fName.substring(0, 5).equals(geoFileRelative.substring(0, 5))) { int geoStartIdx = geoFileRelative.indexOf("_d"); int prdStartIdx = fName.indexOf("_d"); String s1 = geoFileRelative.substring(geoStartIdx, geoStartIdx + JPSSUtilities.NOAA_CREATION_DATE_INDEX); String s2 = fName.substring(prdStartIdx, prdStartIdx + JPSSUtilities.NOAA_CREATION_DATE_INDEX); if (s1.equals(s2)) { geoFilename = s.substring(0, s.lastIndexOf(File.separatorChar) + 1) + fName; break; } } // same check, but for Terrain Corrected version if (fName.substring(0, 5).equals(geoTerrainCorrected.substring(0, 5))) { int geoStartIdx = geoTerrainCorrected.indexOf("_d"); int prdStartIdx = fName.indexOf("_d"); String s1 = geoTerrainCorrected.substring(geoStartIdx, geoStartIdx + JPSSUtilities.NOAA_CREATION_DATE_INDEX); String s2 = fName.substring(prdStartIdx, prdStartIdx + JPSSUtilities.NOAA_CREATION_DATE_INDEX); if (s1.equals(s2)) { geoFilename = s.substring(0, s.lastIndexOf(File.separatorChar) + 1) + fName; break; } } } } } else { // NASA format geoFilename = JPSSUtilities.replaceLast(s, "L1B", "GEO"); // get list of files in current directory File fList = new File( geoFilename.substring(0, geoFilename.lastIndexOf(File.separatorChar) + 1)); // make a NASA style file filter, and see if a matching geo file is present FilenameFilter geoFilter = new FilenameFilter() { public boolean accept(File dir, String name) { if (name.matches(JPSSUtilities.SUOMI_GEO_REGEX_NASA)) { return true; } else { return false; } } }; File[] files = fList.listFiles(geoFilter); for (File file : files) { if (file.isDirectory()) { continue; } // get the file name for convenience String fName = file.getName(); String tmpStr = geoFilename.substring(s.lastIndexOf(File.separatorChar) + 1, s.lastIndexOf(File.separatorChar) + (JPSSUtilities.NASA_CREATION_DATE_INDEX + 1)); if (fName.substring(0, JPSSUtilities.NASA_CREATION_DATE_INDEX) .equals(tmpStr.substring(0, JPSSUtilities.NASA_CREATION_DATE_INDEX))) { geoFilename = s.substring(0, s.lastIndexOf(File.separatorChar) + 1) + fName; break; } } } logger.debug("Determined GEO file name should be: " + geoFilename); fGeo.setAttribute("location", geoFilename); // add this to list used if we create a zipped bundle geoSources.add(geoFilename); agg.addContent(fGeo); } root.addContent(agg); XMLOutputter xmlOut = new XMLOutputter(); String ncmlStr = xmlOut.outputString(document); ByteArrayInputStream is = new ByteArrayInputStream(ncmlStr.getBytes()); MultiDimensionReader netCDFReader = new NetCDFFile(is); // let's try and look through the NetCDF reader and see what we can learn... NetcdfFile ncdff = ((NetCDFFile) netCDFReader).getNetCDFFile(); Group rg = ncdff.getRootGroup(); // this is a list filled with unpacked qflag products, if any ArrayList<VariableDS> qfProds = new ArrayList<VariableDS>(); // this is a list filled with pseudo Brightness Temp variables converted from Radiance ArrayList<VariableDS> btProds = new ArrayList<VariableDS>(); List<Group> gl = rg.getGroups(); if (gl != null) { int xDimNASA = -1; int yDimNASA = -1; // Make a first pass to determine the shape of the geolocation data for (Group g : gl) { if (g.getFullName().contains("geolocation_data")) { List<Variable> vl = g.getVariables(); for (Variable v : vl) { if (v.getShortName().equals("latitude")) { // XXX TJJ Nov 2015 // Hack because fill value in attribute does not match // what I am seeing in the data. Attribute fillAtt = new Attribute("_FillValue", -999.0); v.addAttribute(fillAtt); pathToLat = v.getFullName(); pathToProducts.add(v.getFullName()); prodToDesc.put(v.getFullName(), v.getDescription()); xDimNASA = v.getDimension(0).getLength(); yDimNASA = v.getDimension(1).getLength(); } if (v.getShortName().equals("longitude")) { // XXX TJJ Nov 2015 // Hack because fill value in attribute does not match // what I am seeing in the data. Attribute fillAtt = new Attribute("_FillValue", -999.0); v.addAttribute(fillAtt); pathToLon = v.getFullName(); pathToProducts.add(v.getFullName()); prodToDesc.put(v.getFullName(), v.getDescription()); } } } } for (Group g : gl) { logger.debug("Group name: " + g.getFullName()); // NASA only - looking through observation_data and geolocation_data if (g.getFullName().contains("observation_data")) { List<Variable> vl = g.getVariables(); for (Variable v : vl) { // keep any data which matches geolocation dimensions if (v.getDimension(0).getLength() == xDimNASA && v.getDimension(1).getLength() == yDimNASA) { logger.debug("Adding product: " + v.getFullName()); pathToProducts.add(v.getFullName()); prodToDesc.put(v.getFullName(), v.getDescription()); Attribute aUnsigned = v.findAttribute("_Unsigned"); if (aUnsigned != null) { unsignedFlags.put(v.getFullName(), aUnsigned.getStringValue()); } else { unsignedFlags.put(v.getFullName(), "false"); } // store units in a map for later Attribute unitAtt = v.findAttribute("units"); if (unitAtt != null) { unitsNASA.put(v.getShortName(), unitAtt.getStringValue()); } else { unitsNASA.put(v.getShortName(), "Unknown"); } // TJJ Nov 2018 - SIPS V2+ mods // Regridding with bow-tie interpolation wasn't working since there are // now multiple fill value categories and we need to look specifically // for the bowtie deletion flag Attribute longNameAtt = v.findAttribute("long_name"); String longName = "empty"; if (longNameAtt != null) longName = longNameAtt.getStringValue(); if (longName.contains("reflectance") || longName.contains("radiance")) { Attribute flagMeanings = v .findAttribute(JPSSUtilities.SIPS_FLAG_MEANINGS_ATTRIBUTE); // If this is not null, we must be v2.0.0 or higher if (flagMeanings != null) { String meanings = flagMeanings.getStringValue(); // Tokenize meanings string, multiple flags defined there StringTokenizer st = new StringTokenizer(meanings); int bowtieIdx = -1; boolean foundBowTieAttribute = false; String tokStr = null; while (st.hasMoreTokens()) { tokStr = st.nextToken(); bowtieIdx++; if (tokStr.equals(JPSSUtilities.SIPS_BOWTIE_DELETED_FLAG)) { foundBowTieAttribute = true; break; } } if (foundBowTieAttribute) { Attribute flagValues = v .findAttribute(JPSSUtilities.SIPS_FLAG_VALUES_ATTRIBUTE); Array flagValsArr = flagValues.getValues(); int bowTieVal = (int) flagValsArr.getInt(bowtieIdx); Attribute a1 = new Attribute("_FillValue", bowTieVal); v.addAttribute(a1); } } } // TJJ Feb 2016 - Create BT variables where applicable if ((v.getShortName().matches("M12|M13|M14|M15|M16")) || (v.getShortName().matches("I04|I05"))) { // Get the LUT variable, load into primitive array Variable lut = g .findVariable(v.getShortName() + "_brightness_temperature_lut"); int[] lutShape = lut.getShape(); logger.debug("Handling NASA LUT Variable, LUT size: " + lutShape[0]); // pull out valid min, max - these will be used for our new VariableDS Attribute aVMin = lut.findAttribute("valid_min"); Attribute aVMax = lut.findAttribute("valid_max"); Attribute fillAtt = lut.findAttribute("_FillValue"); logger.debug("valid_min from LUT: " + aVMin.getNumericValue()); logger.debug("valid_max from LUT: " + aVMax.getNumericValue()); // A little hacky, but at this point the class is such a mess // that what's a little more, right? Load M12-M16, I4-I5 LUTS if (v.getShortName().matches("M12")) { m12LUT = new float[lutShape[0]]; ArrayFloat.D1 lutArray = (ArrayFloat.D1) lut.read(); for (int lutIdx = 0; lutIdx < lutShape[0]; lutIdx++) { m12LUT[lutIdx] = lutArray.get(lutIdx); } } if (v.getShortName().matches("M13")) { m13LUT = new float[lutShape[0]]; ArrayFloat.D1 lutArray = (ArrayFloat.D1) lut.read(); for (int lutIdx = 0; lutIdx < lutShape[0]; lutIdx++) { m13LUT[lutIdx] = lutArray.get(lutIdx); } } if (v.getShortName().matches("M14")) { m14LUT = new float[lutShape[0]]; ArrayFloat.D1 lutArray = (ArrayFloat.D1) lut.read(); for (int lutIdx = 0; lutIdx < lutShape[0]; lutIdx++) { m14LUT[lutIdx] = lutArray.get(lutIdx); } } if (v.getShortName().matches("M15")) { m15LUT = new float[lutShape[0]]; ArrayFloat.D1 lutArray = (ArrayFloat.D1) lut.read(); for (int lutIdx = 0; lutIdx < lutShape[0]; lutIdx++) { m15LUT[lutIdx] = lutArray.get(lutIdx); } } if (v.getShortName().matches("M16")) { m16LUT = new float[lutShape[0]]; ArrayFloat.D1 lutArray = (ArrayFloat.D1) lut.read(); for (int lutIdx = 0; lutIdx < lutShape[0]; lutIdx++) { m16LUT[lutIdx] = lutArray.get(lutIdx); } } if (v.getShortName().matches("I04")) { i04LUT = new float[lutShape[0]]; ArrayFloat.D1 lutArray = (ArrayFloat.D1) lut.read(); for (int lutIdx = 0; lutIdx < lutShape[0]; lutIdx++) { i04LUT[lutIdx] = lutArray.get(lutIdx); } } if (v.getShortName().matches("I05")) { i05LUT = new float[lutShape[0]]; ArrayFloat.D1 lutArray = (ArrayFloat.D1) lut.read(); for (int lutIdx = 0; lutIdx < lutShape[0]; lutIdx++) { i05LUT[lutIdx] = lutArray.get(lutIdx); } } // Create a pseudo-variable, fill using LUT // make a copy of the source variable // NOTE: by using a VariableDS here, the original // variable is used for the I/O, this matters! VariableDS vBT = new VariableDS(g, v, false); // Name is orig name plus suffix vBT.setShortName(v.getShortName() + "_BT"); vBT.addAttribute(fillAtt); vBT.addAttribute(aVMin); vBT.addAttribute(aVMax); if (v.getShortName().matches("M12")) { lutMap.put(vBT.getFullName(), m12LUT); } if (v.getShortName().matches("M13")) { lutMap.put(vBT.getFullName(), m13LUT); } if (v.getShortName().matches("M14")) { lutMap.put(vBT.getFullName(), m14LUT); } if (v.getShortName().matches("M15")) { lutMap.put(vBT.getFullName(), m15LUT); } if (v.getShortName().matches("M16")) { lutMap.put(vBT.getFullName(), m16LUT); } if (v.getShortName().matches("I04")) { lutMap.put(vBT.getFullName(), i04LUT); } if (v.getShortName().matches("I05")) { lutMap.put(vBT.getFullName(), i05LUT); } pathToProducts.add(vBT.getFullName()); String newName = vBT.getDescription().replace("radiance", "brightness temperature"); prodToDesc.put(vBT.getFullName(), newName); btProds.add(vBT); } } } } if (g.getFullName().contains("geolocation_data")) { List<Variable> vl = g.getVariables(); for (Variable v : vl) { // keep any data which matches geolocation dimensions if (v.getDimension(0).getLength() == xDimNASA && v.getDimension(1).getLength() == yDimNASA) { // except we already found Lat and Lon, skip those if ((v.getShortName().equals("latitude")) || (v.getShortName().equals("latitude"))) continue; logger.debug("Adding product: " + v.getFullName()); pathToProducts.add(v.getFullName()); prodToDesc.put(v.getFullName(), v.getDescription()); } } } // NOAA only - we are looking through All_Data, finding displayable data if (g.getFullName().contains("All_Data")) { List<Group> adg = g.getGroups(); int xDim = -1; int yDim = -1; // two sub-iterations, first one to find geolocation and product dimensions for (Group subG : adg) { logger.debug("Sub group name: " + subG.getFullName()); String subName = subG.getFullName(); if (subName.contains("-GEO")) { // this is the geolocation data String geoBaseName = subG.getShortName(); geoBaseName = geoBaseName.substring(0, geoBaseName.indexOf('_')); if (!haveGeoMetaData) { String geoProfileFileName = nppPP.getProfileFileName(geoBaseName); // also add meta data from geolocation profile nppPP.addMetaDataFromFile(geoProfileFileName); haveGeoMetaData = true; } List<Variable> vl = subG.getVariables(); for (Variable v : vl) { if (v.getFullName().endsWith(SEPARATOR_CHAR + "Latitude")) { pathToLat = v.getFullName(); logger.debug("Ellipsoid Lat/Lon Variable: " + v.getFullName()); // get the dimensions of the lat variable Dimension dAlongTrack = v.getDimension(0); yDim = dAlongTrack.getLength(); Dimension dAcrossTrack = v.getDimension(1); xDim = dAcrossTrack.getLength(); logger.debug("Lat across track dim: " + dAcrossTrack.getLength()); } if (v.getFullName().endsWith(SEPARATOR_CHAR + "Longitude")) { // we got dimensions from lat, don't need 'em twice, but need path pathToLon = v.getFullName(); } } // one more pass in case there is terrain-corrected Lat/Lon for (Variable v : vl) { if (v.getFullName().endsWith(SEPARATOR_CHAR + "Latitude_TC")) { pathToLat = v.getFullName(); logger.debug("Switched Lat/Lon Variable to TC: " + v.getFullName()); // get the dimensions of the lat variable Dimension dAlongTrack = v.getDimension(0); yDim = dAlongTrack.getLength(); Dimension dAcrossTrack = v.getDimension(1); xDim = dAcrossTrack.getLength(); logger.debug("Lat across track dim: " + dAcrossTrack.getLength()); } if (v.getFullName().endsWith(SEPARATOR_CHAR + "Longitude_TC")) { // we got dimensions from lat, don't need 'em twice, but need path pathToLon = v.getFullName(); } } } } // second to identify displayable products for (Group subG : adg) { // this is the product data List<Variable> vl = subG.getVariables(); for (Variable v : vl) { boolean useThis = false; String vName = v.getFullName(); logger.trace("Variable: " + vName); String varShortName = vName.substring(vName.lastIndexOf(SEPARATOR_CHAR) + 1); // Special code to handle quality flags. We throw out anything // that does not match bounds of the geolocation data if (varShortName.startsWith("QF")) { logger.trace("Handling Quality Flag: " + varShortName); // this check is done later for ALL variables, but we need // it early here to weed out those quality flags that are // simply a small set of data w/no granule geo nbounds boolean xScanOk = false; boolean yScanOk = false; List<Dimension> dl = v.getDimensions(); // toss out > 2D Quality Flags if (dl.size() > 2) { logger.trace("SKIPPING QF, > 2D: " + varShortName); continue; } for (Dimension d : dl) { // in order to consider this a displayable product, make sure // both scan direction dimensions are present and look like a granule if (d.getLength() == xDim) { xScanOk = true; } if (d.getLength() == yDim) { yScanOk = true; } } if (!(xScanOk && yScanOk)) { logger.trace("SKIPPING QF, does not match geo bounds: " + varShortName); continue; } ArrayList<QualityFlag> qfal = nppPP.getQualityFlags(varShortName); if (qfal != null) { for (QualityFlag qf : qfal) { qf.setPackedName(vName); // make a copy of the qflag variable // NOTE: by using a VariableDS here, the original // variable is used for the I/O, this matters! VariableDS vqf = new VariableDS(subG, v, false); // prefix with QF num to help guarantee uniqueness across groups // this will cover most cases, but could still be dupe names // within a single QF. This is handled when fetching XMLPP metadata vqf.setShortName(varShortName.substring(0, 3) + "_" + qf.getName()); logger.debug("New QF var full name: " + vqf.getFullName()); qfProds.add(vqf); qfMap.put(vqf.getFullName(), qf); } } } // for CrIS instrument, first find dimensions of var matching // CrIS filter, then throw out all variables which don't match // those dimensions if (instrumentName.getStringValue().equals("CrIS")) { if (!vName.contains("GEO")) { if (!varShortName.startsWith(crisFilter)) { logger.trace("Skipping variable: " + varShortName); continue; } } else { // these variables are all GEO-related // if they match lat/lon bounds, keep them List<Dimension> dl = v.getDimensions(); if (dl.size() == 3) { boolean isDisplayableCrIS = true; for (Dimension d : dl) { if ((d.getLength() != xDim) && (d.getLength() != yDim) && (d.getLength() != 9)) { isDisplayableCrIS = false; } } if (!isDisplayableCrIS) { continue; } } } } DataType dt = v.getDataType(); if ((dt.getSize() != 4) && (dt.getSize() != 2) && (dt.getSize() != 1)) { continue; } List<Dimension> dl = v.getDimensions(); if (dl.size() > 4) { continue; } // for now, skip any 3D VIIRS data if (instrumentName.getStringValue().equals("VIIRS")) { if (dl.size() == 3) { continue; } } boolean xScanOk = false; boolean yScanOk = false; for (Dimension d : dl) { // in order to consider this a displayable product, make sure // both scan direction dimensions are present and look like a granule if (d.getLength() == xDim) { xScanOk = true; } if (d.getLength() == yDim) { yScanOk = true; } } if (xScanOk && yScanOk) { useThis = true; } // For ATMS, only 3-D variable we pass through is BrightnessTemperature // Dimensions for BT are (lon, lat, channel) if (instrumentName.getStringValue().equals("ATMS")) { if (dl.size() == 3) { boolean isDisplayableATMS = false; for (Dimension d : dl) { if (d.getLength() == JPSSUtilities.ATMSChannelCenterFrequencies.length) { isDisplayableATMS = true; logger.trace( "This variable has a dimension matching num ATMS channels"); break; } } if (!isDisplayableATMS) useThis = false; } } // sensor data with a channel dimension if (useThis) { if ((instrumentName.getStringValue().equals("CrIS")) || (instrumentName.getStringValue().equals("ATMS")) || (instrumentName.getStringValue().contains("OMPS"))) { isVIIRS = false; logger.debug("Handling non-VIIRS data source..."); } } if (useThis) { // loop through the variable list again, looking for a corresponding "Factors" float scaleVal = 1f; float offsetVal = 0f; boolean unpackFlag = false; // if the granule has an entry for this variable name // get the data, data1 = scale, data2 = offset // create and poke attributes with this data // endif String factorsVarName = nppPP.getScaleFactorName(varShortName); if (factorsVarName != null) { logger.debug("Mapping: " + varShortName + " to: " + factorsVarName); for (Variable fV : vl) { if (fV.getShortName().equals(factorsVarName)) { logger.trace("Pulling scale and offset values from variable: " + fV.getShortName()); ucar.ma2.Array a = fV.read(); float[] so = (float[]) a.copyTo1DJavaArray(); scaleVal = so[0]; offsetVal = so[1]; logger.trace("Scale value: " + scaleVal + ", Offset value: " + offsetVal); unpackFlag = true; break; } } } // poke in scale/offset attributes for now Attribute a1 = new Attribute("scale_factor", scaleVal); v.addAttribute(a1); Attribute a2 = new Attribute("add_offset", offsetVal); v.addAttribute(a2); // add valid range and fill value attributes here // try to fill in valid range if (nppPP.hasNameAndMetaData(varShortName)) { String rangeMin = nppPP.getRangeMin(varShortName); String rangeMax = nppPP.getRangeMax(varShortName); logger.trace("range min: " + rangeMin + ", range max: " + rangeMax); // only store range attribute if VALID range found if ((rangeMin != null) && (rangeMax != null)) { int[] shapeArr = new int[] { 2 }; ArrayFloat af = new ArrayFloat(shapeArr); try { af.setFloat(0, Float.parseFloat(rangeMin)); } catch (NumberFormatException nfe) { af.setFloat(0, new Float(Integer.MIN_VALUE)); } try { af.setFloat(1, Float.parseFloat(rangeMax)); } catch (NumberFormatException nfe) { af.setFloat(1, new Float(Integer.MAX_VALUE)); } Attribute rangeAtt = new Attribute("valid_range", af); v.addAttribute(rangeAtt); } // check for and load fill values too... // we need to check two places, first, the XML product profile ArrayList<Float> fval = nppPP.getFillValues(varShortName); // 2nd, does the variable already have one defined? // if there was already a fill value associated with this variable, make // sure we bring that along for the ride too... Attribute aFill = v.findAttribute("_FillValue"); // determine size of our fill value array int fvArraySize = 0; if (aFill != null) fvArraySize++; if (!fval.isEmpty()) fvArraySize += fval.size(); int[] fillShape = new int[] { fvArraySize }; // allocate the array ArrayFloat afFill = new ArrayFloat(fillShape); // and FINALLY, fill it! if (!fval.isEmpty()) { for (int fillIdx = 0; fillIdx < fval.size(); fillIdx++) { afFill.setFloat(fillIdx, fval.get(fillIdx)); logger.trace( "Adding fill value (from XML): " + fval.get(fillIdx)); } } if (aFill != null) { Number n = aFill.getNumericValue(); // is the data unsigned? Attribute aUnsigned = v.findAttribute("_Unsigned"); float fillValAsFloat = Float.NaN; if (aUnsigned != null) { if (aUnsigned.getStringValue().equals("true")) { DataType fvdt = aFill.getDataType(); logger.trace("Data String: " + aFill.toString()); logger.trace("DataType primitive type: " + fvdt.getPrimitiveClassType()); // signed byte that needs conversion? if (fvdt.getPrimitiveClassType() == byte.class) { fillValAsFloat = (float) Util .unsignedByteToInt(n.byteValue()); } else if (fvdt.getPrimitiveClassType() == short.class) { fillValAsFloat = (float) Util .unsignedShortToInt(n.shortValue()); } else { fillValAsFloat = n.floatValue(); } } } afFill.setFloat(fvArraySize - 1, fillValAsFloat); logger.trace( "Adding fill value (from variable): " + fillValAsFloat); } Attribute fillAtt = new Attribute("_FillValue", afFill); v.addAttribute(fillAtt); } Attribute aUnsigned = v.findAttribute("_Unsigned"); if (aUnsigned != null) { unsignedFlags.put(v.getFullName(), aUnsigned.getStringValue()); } else { unsignedFlags.put(v.getFullName(), "false"); } if (unpackFlag) { unpackFlags.put(v.getFullName(), "true"); } else { unpackFlags.put(v.getFullName(), "false"); } logger.debug("Adding product: " + v.getFullName()); pathToProducts.add(v.getFullName()); prodToDesc.put(v.getFullName(), v.getDescription()); } } } } } } // add in any unpacked qflag products for (VariableDS qfV : qfProds) { // skip the spares - they are reserved for future use if (qfV.getFullName().endsWith("Spare")) { continue; } // String.endsWith is case sensitive so gotta check both cases if (qfV.getFullName().endsWith("spare")) { continue; } ncdff.addVariable(qfV.getGroup(), qfV); logger.trace("Adding QF product: " + qfV.getFullName()); pathToProducts.add(qfV.getFullName()); prodToDesc.put(qfV.getFullName(), qfV.getDescription()); unsignedFlags.put(qfV.getFullName(), "true"); unpackFlags.put(qfV.getFullName(), "false"); } // add in any pseudo BT products from NASA data for (Variable vBT : btProds) { logger.trace("Adding BT product: " + vBT.getFullName()); ncdff.addVariable(vBT.getGroup(), vBT); unsignedFlags.put(vBT.getFullName(), "true"); unpackFlags.put(vBT.getFullName(), "false"); } ncdfal.add((NetCDFFile) netCDFReader); } } catch (Exception e) { logger.error("cannot create NetCDF reader for files selected", e); if (e.getMessage() != null && e.getMessage().equals("XML Product Profile Error")) { throw new VisADException("Unable to extract metadata from required XML Product Profile", e); } } // TJJ Feb 2018 // Doing a reorder of variable names here, as per HP's request from // http://mcidas.ssec.wisc.edu/inquiry-v/?inquiry=2613 if (isVIIRS) { // Copy the variable Set to a sortable List List<String> sortedList = new ArrayList(pathToProducts); Collections.sort(sortedList, new VIIRSSort()); // Clear the original data structure which retains insert order // (it's a LinkedHashSet) pathToProducts.clear(); // Re-add the variables in corrected order for (String s : sortedList) { pathToProducts.add(s); } } // initialize the aggregation reader object try { if (isNOAA) { nppAggReader = new GranuleAggregation(ncdfal, pathToProducts, "Track", "XTrack", isVIIRS); ((GranuleAggregation) nppAggReader).setQfMap(qfMap); } else { nppAggReader = new GranuleAggregation(ncdfal, pathToProducts, "number_of_lines", "number_of_pixels", isVIIRS); ((GranuleAggregation) nppAggReader).setLUTMap(lutMap); } } catch (Exception e) { throw new VisADException("Unable to initialize aggregation reader", e); } // make sure we found valid data if (pathToProducts.size() == 0) { throw new VisADException("No data found in files selected"); } logger.debug("Number of adapters needed: " + pathToProducts.size()); adapters = new MultiDimensionAdapter[pathToProducts.size()]; Hashtable<String, String[]> properties = new Hashtable<>(); Iterator<String> iterator = pathToProducts.iterator(); int pIdx = 0; boolean adapterCreated = false; while (iterator.hasNext()) { String pStr = iterator.next(); logger.debug("Working on adapter number " + (pIdx + 1) + ": " + pStr); Map<String, Object> swathTable = SwathAdapter.getEmptyMetadataTable(); Map<String, Object> spectTable = SpectrumAdapter.getEmptyMetadataTable(); swathTable.put("array_name", pStr); swathTable.put("lon_array_name", pathToLon); swathTable.put("lat_array_name", pathToLat); swathTable.put("XTrack", "XTrack"); swathTable.put("Track", "Track"); swathTable.put("geo_Track", "Track"); swathTable.put("geo_XTrack", "XTrack"); // TJJ is this even needed? Is product_name used anywhere? if (productName == null) productName = pStr.substring(pStr.indexOf(SEPARATOR_CHAR) + 1); swathTable.put("product_name", productName); swathTable.put("_mapping", prodToDesc); // array_name common to spectrum table spectTable.put("array_name", pStr); spectTable.put("product_name", productName); spectTable.put("_mapping", prodToDesc); if (!isVIIRS) { // 3D data is either ATMS, OMPS, or CrIS if ((instrumentName.getShortName() != null) && (instrumentName.getStringValue().equals("ATMS"))) { spectTable.put(SpectrumAdapter.channelIndex_name, "Channel"); swathTable.put(SpectrumAdapter.channelIndex_name, "Channel"); swathTable.put("array_dimension_names", new String[] { "Track", "XTrack", "Channel" }); swathTable.put("lon_array_dimension_names", new String[] { "Track", "XTrack" }); swathTable.put("lat_array_dimension_names", new String[] { "Track", "XTrack" }); spectTable.put("array_dimension_names", new String[] { "Track", "XTrack", "Channel" }); spectTable.put("lon_array_dimension_names", new String[] { "Track", "XTrack" }); spectTable.put("lat_array_dimension_names", new String[] { "Track", "XTrack" }); spectTable.put(SpectrumAdapter.channelType, "wavelength"); spectTable.put(SpectrumAdapter.channels_name, "Channel"); spectTable.put(SpectrumAdapter.x_dim_name, "XTrack"); spectTable.put(SpectrumAdapter.y_dim_name, "Track"); int numChannels = JPSSUtilities.ATMSChannelCenterFrequencies.length; float[] bandArray = new float[numChannels]; String[] bandNames = new String[numChannels]; for (int bIdx = 0; bIdx < numChannels; bIdx++) { bandArray[bIdx] = JPSSUtilities.ATMSChannelCenterFrequencies[bIdx]; bandNames[bIdx] = "Channel " + (bIdx + 1); } spectTable.put(SpectrumAdapter.channelValues, bandArray); spectTable.put(SpectrumAdapter.bandNames, bandNames); } else { if (instrumentName.getStringValue().equals("CrIS")) { swathTable.put("XTrack", "dim1"); swathTable.put("Track", "dim0"); swathTable.put("geo_XTrack", "dim1"); swathTable.put("geo_Track", "dim0"); swathTable.put("product_name", "CrIS_SDR"); swathTable.put(SpectrumAdapter.channelIndex_name, "dim3"); swathTable.put(SpectrumAdapter.FOVindex_name, "dim2"); spectTable.put(SpectrumAdapter.channelIndex_name, "dim3"); spectTable.put(SpectrumAdapter.FOVindex_name, "dim2"); spectTable.put(SpectrumAdapter.x_dim_name, "dim1"); spectTable.put(SpectrumAdapter.y_dim_name, "dim0"); } else if (instrumentName.getStringValue().contains("OMPS")) { spectTable.put(SpectrumAdapter.channelIndex_name, "Channel"); swathTable.put(SpectrumAdapter.channelIndex_name, "Channel"); swathTable.put("array_dimension_names", new String[] { "Track", "XTrack", "Channel" }); swathTable.put("lon_array_dimension_names", new String[] { "Track", "XTrack" }); swathTable.put("lat_array_dimension_names", new String[] { "Track", "XTrack" }); spectTable.put("array_dimension_names", new String[] { "Track", "XTrack", "Channel" }); spectTable.put("lon_array_dimension_names", new String[] { "Track", "XTrack" }); spectTable.put("lat_array_dimension_names", new String[] { "Track", "XTrack" }); spectTable.put(SpectrumAdapter.channelType, "wavelength"); spectTable.put(SpectrumAdapter.channels_name, "Channel"); spectTable.put(SpectrumAdapter.x_dim_name, "XTrack"); spectTable.put(SpectrumAdapter.y_dim_name, "Track"); int numChannels = 200; if (instrumentName.getStringValue().equals("OMPS-TC")) { numChannels = 260; } logger.debug("Setting up OMPS adapter, num channels: " + numChannels); float[] bandArray = new float[numChannels]; String[] bandNames = new String[numChannels]; for (int bIdx = 0; bIdx < numChannels; bIdx++) { bandArray[bIdx] = bIdx; bandNames[bIdx] = "Channel " + (bIdx + 1); } spectTable.put(SpectrumAdapter.channelValues, bandArray); spectTable.put(SpectrumAdapter.bandNames, bandNames); } else { // sorry, if we can't id the instrument, we can't display the data! throw new VisADException("Unable to determine instrument name"); } } } else { swathTable.put("array_dimension_names", new String[] { "Track", "XTrack" }); swathTable.put("lon_array_dimension_names", new String[] { "Track", "XTrack" }); swathTable.put("lat_array_dimension_names", new String[] { "Track", "XTrack" }); } swathTable.put("scale_name", "scale_factor"); swathTable.put("offset_name", "add_offset"); swathTable.put("fill_value_name", "_FillValue"); swathTable.put("range_name", pStr.substring(pStr.indexOf(SEPARATOR_CHAR) + 1)); spectTable.put("range_name", pStr.substring(pStr.indexOf(SEPARATOR_CHAR) + 1)); // set the valid range hash if data is available if (nppPP != null) { if (nppPP.getRangeMin(pStr.substring(pStr.lastIndexOf(SEPARATOR_CHAR) + 1)) != null) { swathTable.put("valid_range", "valid_range"); } } String unsignedAttributeStr = unsignedFlags.get(pStr); if ((unsignedAttributeStr != null) && (unsignedAttributeStr.equals("true"))) { swathTable.put("unsigned", unsignedAttributeStr); } String unpackFlagStr = unpackFlags.get(pStr); if ((unpackFlagStr != null) && (unpackFlagStr.equals("true"))) { swathTable.put("unpack", "true"); } // For Suomi NPP data, do valid range check AFTER applying scale/offset swathTable.put("range_check_after_scaling", "true"); // pass in a GranuleAggregation reader... if (!isVIIRS) { if (instrumentName.getStringValue().equals("ATMS")) { adapters[pIdx] = new SwathAdapter(nppAggReader, swathTable); adapterCreated = true; SpectrumAdapter sa = new SpectrumAdapter(nppAggReader, spectTable); DataCategory.createCategory("MultiSpectral"); categories = DataCategory.parseCategories("MultiSpectral;MultiSpectral;IMAGE"); MultiSpectralData msd = new MultiSpectralData((SwathAdapter) adapters[pIdx], sa, "BrightnessTemperature", "BrightnessTemperature", "SuomiNPP", "ATMS"); msd.setInitialWavenumber(JPSSUtilities.ATMSChannelCenterFrequencies[0]); multiSpectralData.add(msd); } if (instrumentName.getStringValue().equals("CrIS")) { if (pStr.contains(crisFilter)) { adapters[pIdx] = new CrIS_SDR_SwathAdapter(nppAggReader, swathTable); adapterCreated = true; CrIS_SDR_Spectrum csa = new CrIS_SDR_Spectrum(nppAggReader, spectTable); DataCategory.createCategory("MultiSpectral"); categories = DataCategory.parseCategories("MultiSpectral;MultiSpectral;IMAGE"); MultiSpectralData msd = new CrIS_SDR_MultiSpectralData( (CrIS_SDR_SwathAdapter) adapters[pIdx], csa); msd.setInitialWavenumber(csa.getInitialWavenumber()); msd_CrIS.add(msd); } } if (instrumentName.getStringValue().contains("OMPS")) { adapters[pIdx] = new SwathAdapter(nppAggReader, swathTable); adapterCreated = true; SpectrumAdapter sa = new SpectrumAdapter(nppAggReader, spectTable); DataCategory.createCategory("MultiSpectral"); categories = DataCategory.parseCategories("MultiSpectral;MultiSpectral;IMAGE"); MultiSpectralData msd = new MultiSpectralData((SwathAdapter) adapters[pIdx], sa, "RadianceEarth", "RadianceEarth", "SuomiNPP", "OMPS"); msd.setInitialWavenumber(0); multiSpectralData.add(msd); } if (pIdx == 0) { // generate default subset for ATMS and OMPS if (!instrumentName.getStringValue().equals("CrIS")) { defaultSubset = multiSpectralData.get(pIdx).getDefaultSubset(); } } } else { // setting NOAA-format units String varName = pStr.substring(pStr.indexOf(SEPARATOR_CHAR) + 1); String varShortName = pStr.substring(pStr.lastIndexOf(SEPARATOR_CHAR) + 1); String units = nppPP.getUnits(varShortName); // setting NASA-format units if (!isNOAA) { units = unitsNASA.get(varShortName); // Need to set _BT variables manually, since they are created on the fly if (varShortName.endsWith("_BT")) units = "Kelvin"; } if (units == null) units = "Unknown"; Unit u = null; try { u = Parser.parse(units); } catch (NoSuchUnitException e) { u = new DerivedUnit(units); logger.debug("Unknown units: " + units); } catch (ParseException e) { u = new DerivedUnit(units); logger.debug("Unparseable units: " + units); } // associate this variable with these units, if not done already RealType.getRealType(varName, u); adapters[pIdx] = new SwathAdapter(nppAggReader, swathTable); adapterCreated = true; if (pIdx == 0) { defaultSubset = adapters[pIdx].getDefaultSubset(); } categories = DataCategory.parseCategories("IMAGE"); } // only increment count if we created an adapter, some products are skipped if (adapterCreated) pIdx++; adapterCreated = false; } if (msd_CrIS.size() > 0) { try { MultiSpectralAggr aggr = new MultiSpectralAggr( msd_CrIS.toArray(new MultiSpectralData[msd_CrIS.size()])); aggr.setInitialWavenumber(902.25f); multiSpectralData.add(aggr); defaultSubset = ((MultiSpectralData) msd_CrIS.get(0)).getDefaultSubset(); } catch (Exception e) { logger.error("Exception: ", e); } } // Merge with pre-set properties Hashtable tmpHt = getProperties(); tmpHt.putAll(properties); setProperties(tmpHt); }
From source file:Ejercicios.punto5d.CreatorEntity.java
/** * * @return documento jdom xml con el contenido de la entidad * @throws IOException si hay un error al guardar el archivo en disco *///from w ww.j a v a2 s . c o m public Document crearXML() throws IOException { if ("".equals(nombre) || "".equals(raiz) || 0 == atributos.size()) { return null; } root = new Element(raiz); doc = new Document(root); for (String s : atributos.keySet()) { Element e = new Element(s).addContent(atributos.get(s)); getDoc().getRootElement().addContent(e); } try { // new XMLOutputter().output(doc, System.out); XMLOutputter xmlOutput = new XMLOutputter(); // display nice nice xmlOutput.setFormat(Format.getPrettyFormat()); xmlOutput.output(getDoc(), new FileWriter(nombre + ".xml")); System.out.println("Archivo Guardado!"); return doc; } catch (IOException e) { System.out.println(e.getMessage()); return null; } }
From source file:es.ucm.fdi.ac.Analysis.java
License:Open Source License
/** * Saves this analysis to a file/*from w ww. j a v a 2 s .co m*/ * * @param f the file to write to * @throws java.io.IOException */ public void saveToFile(File f) throws IOException { FileOutputStream fos = null; try { XMLOutputter outputter = new XMLOutputter(Format.getPrettyFormat()); fos = new FileOutputStream(f); outputter.output(new Document(saveToXML()), fos); } finally { if (fos != null) { fos.close(); } } }
From source file:es.ucm.fdi.clover.gui.CloverSave.java
License:Open Source License
/** * Saves the CloverSave to an XML file//from w ww . ja va 2 s .c o m */ public static void save(Collection<ClusterView> views, File f) throws IOException { Element root = new Element("clover"); root.setAttribute("version", saveVersion); root.setAttribute("requiresVersion", compatibleWith); root.setAttribute("date", new Date().toString()); Element shared = new Element("shared"); root.addContent(shared); HashMap<Filter, String> filterToId = new HashMap<Filter, String>(); HashMap<ClusteringEngine, String> engineToId = new HashMap<ClusteringEngine, String>(); HashMap<ClusterHierarchy, String> hierarchyToId = new HashMap<ClusterHierarchy, String>(); for (ClusterView v : views) { Element view = new Element("view"); v.save(view); Element layoutCache = new Element("layoutCache"); v.getAnimator().getLayoutCache().save(layoutCache, v.getBase()); view.addContent(layoutCache); Element animatorProps = new Element("animatorProps"); v.getAnimator().save(animatorProps); view.addContent(animatorProps); ClusteredGraph cg = (ClusteredGraph) v.getBase(); ClusterHierarchy h = cg.getHierarchy(); BaseGraph bg = cg.getBase(); // create the hierarchy element (if necessary) if (!hierarchyToId.containsKey(h)) { String hierarchyId = "" + generateId(); hierarchyToId.put(h, hierarchyId); Element hierarchy = new Element("hierarchy"); hierarchy.setAttribute("id", hierarchyId); h.save(hierarchy); // save the filtering used in the hierarchy (if necessary) if (bg instanceof FilteredGraph) { Filter filter = ((FilteredGraph) bg).getFilter(); if (!filterToId.containsKey(filter)) { String filterId = "" + generateId(); filterToId.put(filter, filterId); Element fe = new Element("filter"); fe.setAttribute("id", filterId); filter.save(fe); shared.addContent(fe); } hierarchy.setAttribute("filterId", filterToId.get(filter)); } // save the hierarchy itself: clustering and update-engine ClusteringEngine e = h.getEngine(); Element engine = new Element("engine"); engine.setAttribute("engineClass", e.getClass().getName()); e.save(engine); hierarchy.addContent(engine); shared.addContent(hierarchy); } view.setAttribute("hierarchyId", hierarchyToId.get(h)); root.addContent(view); } XMLOutputter outputter = new XMLOutputter(Format.getPrettyFormat()); outputter.output(new Document(root), new FileOutputStream(f)); }
From source file:es.upm.dit.xsdinferencer.generation.generatorimpl.schemageneration.XMLSchemaDocumentGenerator.java
License:Apache License
/** * It generates the XSD file of the targetNamespace given at the constructor, taking into account that * the main namespace is the one given at the constructor. * //from ww w . j av a 2s .c om * @param schema the schema object * @param configuration the inference configuration * * @return a JDOM2 {@link Document} object containing the XSD contents. * * @see SchemaDocumentGenerator#generateSchemaDocument(Schema, XSDInferenceConfiguration) */ @Override public Document generateSchemaDocument(Schema schema, XSDInferenceConfiguration configuration) { // if(!configuration.getElementsGlobal()==false || // !configuration.getComplexTypesGlobal()==true || // !configuration.getSimpleTypesGlobal()==true // ) // throw new UnsupportedOperationException("Not implemented yet."); // checkArgument(schema.getNamespacesToPossiblePrefixMappingModifiable().containsKey(mainNamespace), "The main namespace must be a known namespace"); checkArgument(schema.getNamespacesToPossiblePrefixMappingModifiable().containsKey(targetNamespace), "The target namespace must be a known namespace"); // checkArgument(!schema.getNamespacesToPossiblePrefixMappingModifiable().containsKey(XSD_NAMESPACE_URI),"The XSD namespace must not be a known namespace"); // checkArgument(!schema.getNamespacesToPossiblePrefixMappingModifiable().containsKey(XSI_NAMESPACE_URI),"The XSI namespace must not be a known namespace"); Map<String, String> namespaceURIToPrefixMappings = schema.getSolvedNamespaceMappings(); if (configuration.getSkipNamespaces().contains(targetNamespace)) { throw new IllegalArgumentException("This is an skipped namespace, so its XSD should not be generated"); } if (targetNamespace.equals(XSD_NAMESPACE_URI)) System.err.println( "The XML Schema namespace is being considered as a target namespace in your documents. Independing of the inferred schemas, the only valid XSD for an XSD would be the normative one present at its first RFC"); Namespace xsdNamespace = Namespace.getNamespace(XSD_NAMESPACE_PREFIX.replace(":", ""), XSD_NAMESPACE_URI); List<Namespace> namespaceDeclarations = getNamespaceDeclarations(namespaceURIToPrefixMappings, xsdNamespace); Element elementSchema = new Element("schema", xsdNamespace); for (int i = 0; i < namespaceDeclarations.size(); i++) { Namespace currentNamespace = namespaceDeclarations.get(i); elementSchema.addNamespaceDeclaration(currentNamespace); String currentNamespaceUri = currentNamespace.getURI(); if (!targetNamespace.equals(mainNamespace) && !currentNamespaceUri.equals(mainNamespace)) continue; if (currentNamespace.equals(Namespace.XML_NAMESPACE) && (!schema.getAttributes().containsRow(XSDInferenceConfiguration.XML_NAMESPACE_URI) && !schema.getElements().containsRow(XSDInferenceConfiguration.XML_NAMESPACE_URI))) { continue; } if (currentNamespaceUri.equals(XSD_NAMESPACE_URI) && !namespaceURIToPrefixMappings.containsKey(XSD_NAMESPACE_URI)) continue; if (targetNamespace.equals(currentNamespaceUri) || (currentNamespaceUri.equals("") && (fileNameGenerator == null))) continue; if (currentNamespaceUri.equals("") && !currentNamespaceUri.equals(mainNamespace) && !schema.getElements().containsRow("")) continue; Element importElement = new Element("import", xsdNamespace); if (!currentNamespaceUri.equals("")) { Attribute namespaceAttr = new Attribute("namespace", currentNamespaceUri); importElement.setAttribute(namespaceAttr); } if (fileNameGenerator != null && !configuration.getSkipNamespaces().contains(currentNamespaceUri)) { String fileName = fileNameGenerator.getSchemaDocumentFileName(currentNamespaceUri, namespaceURIToPrefixMappings); Attribute schemaLocationAttr = new Attribute("schemaLocation", fileName); importElement.setAttribute(schemaLocationAttr); } elementSchema.addContent(importElement); } if (!targetNamespace.equals("")) { Attribute targetNamespaceAttr = new Attribute("targetNamespace", targetNamespace); elementSchema.setAttribute(targetNamespaceAttr); } SortedSet<SimpleType> sortedSimpleTypes = new TreeSet<>(new SimpleTypeComparator()); sortedSimpleTypes.addAll(schema.getSimpleTypes().values()); SortedSet<ComplexType> sortedComplexTypes = new TreeSet<>(new ComplexTypeComparator()); sortedComplexTypes.addAll(schema.getComplexTypes().values()); //CONTINUE FROM HERE: Generate sorted sets for SchemaElement and SchemaAttribute objects and use them where needed. Attribute elementFormDefault = new Attribute("elementFormDefault", "qualified"); elementSchema.setAttribute(elementFormDefault); Document resultingDocument = new Document(elementSchema); if (targetNamespace.equals(mainNamespace)) { //First, we declare global SimpleTypes. //If simpleTypesGlobal is true, any enumeration will be declared as a global simple type. //if not, simple types of complex types which have attributes but not children will be declared globally //(due to limitations of XSD, they may not be declared locally together with the attributes info) if (configuration.getSimpleTypesGlobal()) { for (SimpleType simpleType : sortedSimpleTypes) { if (!simpleType.isEnum() || simpleType.isEmpty()) continue; Element simpleTypeElement = generateSimpleType(simpleType, false, configuration, xsdNamespace); elementSchema.addContent(simpleTypeElement); } } else { for (ComplexType complexType : sortedComplexTypes) { SimpleType simpleType = complexType.getTextSimpleType(); if (complexType.getAttributeList().isEmpty() || !(complexType.getAutomaton().nodeCount() == 0) || !simpleType.isEnum() || simpleType.isEmpty()) continue; Element simpleTypeElement = generateSimpleType(simpleType, false, configuration, xsdNamespace); elementSchema.addContent(simpleTypeElement); } } //Global complexType elements are only generated in the main schema (i.e. the one whose targetNamespace is equal to mainNamespace) if (configuration.getComplexTypesGlobal()) { for (ComplexType complexType : sortedComplexTypes) { boolean hasNoChildren = complexType.getRegularExpression().equals(new EmptyRegularExpression()); boolean hasNoAttributes = complexType.getAttributeList().size() == 0; boolean hasNoComments = complexType.getComments().size() == 0; // boolean simpleTypeIsNotEmpty = !complexType.getTextSimpleType().isEmpty(); boolean simpleTypeIsWhiteSpaceOnlyOrEmpty = !(complexType.getTextSimpleType().isEmpty() || complexType.getTextSimpleType().consistOnlyOfWhitespaceCharacters()); if (hasNoChildren && hasNoAttributes && simpleTypeIsWhiteSpaceOnlyOrEmpty && hasNoComments) continue; //Because the elements which are linked to this ComplexType at our internal model //will be linked to an XSD simple type elsewhere, either a builtin or a custom one. Element complexTypeElement = generateComplexType(configuration, complexType, false, targetNamespace, namespaceURIToPrefixMappings, mainNamespace, xsdNamespace); elementSchema.addContent(complexTypeElement); } } } //If there are many namespaces and the workaround is disabled, we must declare global attributes. //If the targetNamespace is not the mainNamespace, we must declare all the attributes. //if the target namespace is the main namespace, we do not need to declare anything, because the complex types which hold the attributes //are also in the main namespace. if ((namespaceURIToPrefixMappings.size() - configuration.getSkipNamespaces().size()) > 1) { SortedMap<String, SchemaAttribute> globalAttributeCandidates = new TreeMap<>( schema.getAttributes().row(targetNamespace)); if (!targetNamespace.equals(mainNamespace) && !targetNamespace.equals("")) { globalAttributesLoop: for (Map.Entry<String, SchemaAttribute> schemaAttributeEntry : globalAttributeCandidates .entrySet()) { SchemaAttribute schemaAttribute = schemaAttributeEntry.getValue(); //First, we check if the attribute has been already declared when the workaround is disabled. //If so, we update the "use" property. //The type should have been already merged. if (!configuration.getStrictValidRootDefinitionWorkaround()) { List<Element> alreadyGeneratedAttributeElements = elementSchema.getChildren("attribute", xsdNamespace); for (int i = 0; i < alreadyGeneratedAttributeElements.size(); i++) { Element currentAttributeElement = alreadyGeneratedAttributeElements.get(i); if (currentAttributeElement.getAttributeValue("name") .equals(schemaAttribute.getName())) { continue globalAttributesLoop; } } } Element attributeOrAttributeGroupElement = generateAttribute(schemaAttribute, true, configuration, namespaceURIToPrefixMappings, targetNamespace, mainNamespace, schemaAttributeEntry.getKey(), xsdNamespace); elementSchema.addContent(attributeOrAttributeGroupElement); } } } //Now, we declare global elements. //An element will be declared globally if and only if: //1-elementsGlobal is true in the configuration //2-The element is a valid root //3-The element is in a namespace other than the main namespace. Note that the element WILL be surrounded by the corresponding group if the workaround is enabled. //Another important remark: Iterating over a set copy implies iterating over DISTINCT SchemaElements, so if two keys pointed to equal SchemaElements, we would generate it only once- SortedSet<SchemaElement> schemaElementsAtTargetNamespace = new TreeSet<>(new SchemaElementComparator()); schemaElementsAtTargetNamespace.addAll(schema.getElements().row(targetNamespace).values()); globalSchemaElementsLoop: for (SchemaElement schemaElement : schemaElementsAtTargetNamespace) { // if(!configuration.getElementsGlobal()&& // !schemaElement.isValidRoot()&& // (targetNamespace.equals(mainNamespace)||configuration.getStrictValidRootDefinitionWorkaround())) if (!configuration.getElementsGlobal() && !schemaElement.isValidRoot() && (targetNamespace.equals(mainNamespace))) continue; // for(Element currentElement:elementSchema.getContent(Filters.element("element",xsdNamespace))){ // if(schemaElement.getName().equals(currentElement.getAttributeValue("name"))) // continue globalSchemaElementsLoop; // } String possibleGroupName = schemaElement.getName() + configuration.getTypeNamesAncestorsSeparator() + schemaElement.getType().getName(); for (Element currentElement : elementSchema.getContent(Filters.element("group", xsdNamespace))) { if (possibleGroupName.equals(currentElement.getAttributeValue("name"))) continue globalSchemaElementsLoop; } Element elementOrGroupElement = generateElement(schemaElement, true, configuration, targetNamespace, mainNamespace, null, namespaceURIToPrefixMappings, xsdNamespace); if (elementOrGroupElement.getName().equals("element")) { for (Element currentElement : elementSchema.getChildren("element", xsdNamespace)) { if (schemaElement.getName().equals(currentElement.getAttributeValue("name"))) continue globalSchemaElementsLoop; } } elementSchema.addContent(elementOrGroupElement); } return resultingDocument; }
From source file:es.upm.dit.xsdinferencer.generation.generatorimpl.statisticsgeneration.StatisticResultsDocGeneratorImpl.java
License:Apache License
/** * @see StatisticResultsDocGenerator#generateStatisticResultsDoc(Statistics) *///from www . ja v a 2 s.c o m @Override public Document generateStatisticResultsDoc(Statistics statistics) { Element statisticsElement = new Element("statistics", STATISTICS_NAMESPACE); Element generalStatisticsElement = generateGeneralStatistics(statistics); statisticsElement.addContent(generalStatisticsElement); Element rootElementsOccurrencesElement = generateRootElementsOccurrencesElement( statistics.getRootElementOccurrences()); statisticsElement.addContent(rootElementsOccurrencesElement); Element elementsAtPathElement = generateNodesAtPathElements("elementsAtPathOccurrences", "element", statistics.getElementAtPathInfo(), "valuesAtPath", "valueAtPath", statistics.getValuesAtPathInfo(), statistics.getStatisticsOfNumericValuesAtPath(), "numericValuesStatistics"); statisticsElement.addContent(elementsAtPathElement); Element attributesAtPathElement = generateNodesAtPathElements("attributesAtPathOccurrences", "attribute", statistics.getAttributeAtPathInfo(), "valuesAtPath", "valueAtPath", statistics.getValuesAtPathInfo(), statistics.getStatisticsOfNumericValuesAtPath(), "numericValuesStatistics"); statisticsElement.addContent(attributesAtPathElement); Element complexTypesInfoElement = generateComplexTypesInfo(statistics.getComplexTypeInfo()); statisticsElement.addContent(complexTypesInfoElement); Document resultingDocument = new Document(statisticsElement); return resultingDocument; }
From source file:eu.optimis.monitoring.amazoncollector.XMLHelper.java
License:Apache License
public static String createDocument(List<Measurement> measurements) { Element root = new Element(ROOT); Document doc = new Document(root); for (Measurement m : measurements) { root.addContent(createMonitoringResource(m)); }/* w ww . j a va2 s . c o m*/ XMLOutputter outputter = new XMLOutputter(); outputter.setFormat(Format.getPrettyFormat()); try { outputter.output(doc, new FileOutputStream("./aws.xml")); } catch (FileNotFoundException ex) { Logger.getLogger(XMLHelper.class.getName()).log(Level.SEVERE, null, ex); } catch (IOException ex) { Logger.getLogger(XMLHelper.class.getName()).log(Level.SEVERE, null, ex); } String xmlResult = outputter.outputString(doc); String result = xmlResult.replace("encoding=\"UTF-8\"", ""); //System.out.println (result); return result; }
From source file:eus.ixa.ixa.pipe.convert.AbsaSemEval.java
License:Apache License
public static String nafToAbsa2015(String inputNAF) throws IOException { Path kafPath = Paths.get(inputNAF); KAFDocument kaf = KAFDocument.createFromFile(kafPath.toFile()); Set<String> reviewIds = getReviewIdsFromXpathAttribute(kaf); // root element in ABSA 2015 and 2016 format Element reviewsElem = new Element("Reviews"); Document doc = new Document(reviewsElem); // creating Reviews children of Review for (String reviewId : reviewIds) { Element reviewElem = new Element("Review"); reviewElem.setAttribute("rid", reviewId); Element sentencesElem = new Element("sentences"); // getting the sentences in the review List<List<WF>> sentencesByReview = getSentencesByReview(kaf, reviewId); for (List<WF> sent : sentencesByReview) { String sentId = sent.get(0).getXpath(); Integer sentNumber = sent.get(0).getSent(); // getting text element from word forms in NAF String textString = NAFUtils.getSentenceStringFromWFs(sent); Element sentenceElem = new Element("sentence"); sentenceElem.setAttribute("id", sentId); Element textElem = new Element("text"); textElem.setText(textString); sentenceElem.addContent(textElem); // creating opinions element for sentence List<Opinion> opinionsBySentence = getOpinionsBySentence(kaf, sentNumber); Element opinionsElem = new Element("Opinions"); if (!opinionsBySentence.isEmpty()) { // getting opinion info from NAF Opinion layer for (Opinion opinion : opinionsBySentence) { Element opinionElem = new Element("Opinion"); // String polarity = opinion.getOpinionExpression().getPolarity(); String category = opinion.getOpinionExpression().getSentimentProductFeature(); String targetString = opinion.getStr(); int fromOffset = opinion.getOpinionTarget().getTerms().get(0).getWFs().get(0).getOffset(); List<WF> targetWFs = opinion.getOpinionTarget().getTerms() .get(opinion.getOpinionTarget().getTerms().size() - 1).getWFs(); int toOffset = targetWFs.get(targetWFs.size() - 1).getOffset() + targetWFs.get(targetWFs.size() - 1).getLength(); opinionElem.setAttribute("target", targetString); opinionElem.setAttribute("category", category); // TODO we still do not have polarity here opinionElem.setAttribute("polarity", "na"); opinionElem.setAttribute("from", Integer.toString(fromOffset)); opinionElem.setAttribute("to", Integer.toString(toOffset)); opinionsElem.addContent(opinionElem); }//from w w w . j a va 2 s .c o m } sentenceElem.addContent(opinionsElem); sentencesElem.addContent(sentenceElem); } reviewElem.addContent(sentencesElem); reviewsElem.addContent(reviewElem); } // end of review XMLOutputter xmlOutput = new XMLOutputter(); Format format = Format.getPrettyFormat(); xmlOutput.setFormat(format); return xmlOutput.outputString(doc); }
From source file:eus.ixa.ixa.pipe.convert.AbsaSemEval.java
License:Apache License
public static String nafToAbsa2014(String kafDocument) { KAFDocument kaf = null;// ww w .j a va2 s .co m try { Path kafPath = Paths.get(kafDocument); kaf = KAFDocument.createFromFile(kafPath.toFile()); } catch (IOException e) { e.printStackTrace(); } Element sentencesElem = new Element("sentences"); Document doc = new Document(sentencesElem); for (List<WF> sent : kaf.getSentences()) { String sentId = sent.get(0).getXpath(); Integer sentNumber = sent.get(0).getSent(); // getting text element from WFs in NAF String textString = NAFUtils.getSentenceStringFromWFs(sent); Element sentenceElem = new Element("sentence"); sentenceElem.setAttribute("id", sentId); Element textElem = new Element("text"); textElem.setText(textString); sentenceElem.addContent(textElem); // creating opinions element for sentence List<Opinion> opinionsBySentence = getOpinionsBySentence(kaf, sentNumber); if (!opinionsBySentence.isEmpty()) { Element aspectTerms = new Element("aspectTerms"); // getting opinion info from NAF Opinion layer for (Opinion opinion : opinionsBySentence) { String polarity = ""; String targetString = opinion.getStr(); int fromOffset = opinion.getOpinionTarget().getTerms().get(0).getWFs().get(0).getOffset(); List<WF> targetWFs = opinion.getOpinionTarget().getTerms() .get(opinion.getOpinionTarget().getTerms().size() - 1).getWFs(); int toOffset = targetWFs.get(targetWFs.size() - 1).getOffset() + targetWFs.get(targetWFs.size() - 1).getLength(); Element aspectTerm = new Element("aspectTerm"); aspectTerm.setAttribute("term", targetString); aspectTerm.setAttribute("polarity", polarity); aspectTerm.setAttribute("from", Integer.toString(fromOffset)); aspectTerm.setAttribute("to", Integer.toString(toOffset)); aspectTerms.addContent(aspectTerm); } sentenceElem.addContent(aspectTerms); } sentencesElem.addContent(sentenceElem); } XMLOutputter xmlOutput = new XMLOutputter(); Format format = Format.getPrettyFormat(); xmlOutput.setFormat(format); return xmlOutput.outputString(doc); }
From source file:eus.ixa.ixa.pipe.ml.document.features.DocumentFeatureDescriptor.java
License:Apache License
/** * Generate the XML feature descriptor from the docTrainer.properties file. * /* ww w. j av a 2 s . c om*/ * @param params * the properties file * @return the XML feature descriptor * @throws IOException * if input output fails */ public static String createDocumentFeatureDescriptor(final TrainingParameters params) throws IOException { // <generators> final Element generators = new Element("generators"); final Document doc = new Document(generators); // <custom bagofwords /. if (Flags.isBagOfWordsFeature(params)) { final String tokenFeatureRange = Flags.getBagOfWordsFeaturesRange(params); final Element tokenFeature = new Element("custom"); tokenFeature.setAttribute("class", BagOfWordsFeatureGenerator.class.getName()); tokenFeature.setAttribute("range", tokenFeatureRange); generators.addContent(tokenFeature); System.err.println("-> BOW features added!"); } if (Flags.isTokenClassFeature(params)) { final String tokenClassFeatureRange = Flags.getTokenClassFeaturesRange(params); final Element tokenClassFeature = new Element("custom"); tokenClassFeature.setAttribute("class", DocTokenClassFeatureGenerator.class.getName()); tokenClassFeature.setAttribute("range", tokenClassFeatureRange); generators.addContent(tokenClassFeature); System.err.println("-> Token Class Features added!"); } if (Flags.isOutcomePriorFeature(params)) { final Element outcomePriorFeature = new Element("custom"); outcomePriorFeature.setAttribute("class", DocOutcomePriorFeatureGenerator.class.getName()); generators.addContent(outcomePriorFeature); System.err.println("-> Outcome Prior Features added!"); } if (Flags.isSentenceFeature(params)) { final String beginSentence = Flags.getSentenceFeaturesBegin(params); final String endSentence = Flags.getSentenceFeaturesEnd(params); final Element sentenceFeature = new Element("custom"); sentenceFeature.setAttribute("class", DocSentenceFeatureGenerator.class.getName()); sentenceFeature.setAttribute("begin", beginSentence); sentenceFeature.setAttribute("end", endSentence); generators.addContent(sentenceFeature); System.err.println("-> Sentence Features added!"); } if (Flags.isPrefixFeature(params)) { final String beginPrefix = Flags.getPrefixFeaturesBegin(params); final String endPrefix = Flags.getPrefixFeaturesEnd(params); final Element prefixFeature = new Element("custom"); prefixFeature.setAttribute("class", DocPrefixFeatureGenerator.class.getName()); prefixFeature.setAttribute("begin", beginPrefix); prefixFeature.setAttribute("end", endPrefix); generators.addContent(prefixFeature); System.err.println("-> Prefix Features added!"); } if (Flags.isSuffixFeature(params)) { final String beginSuffix = Flags.getSuffixFeaturesBegin(params); final String endSuffix = Flags.getSuffixFeaturesEnd(params); final Element suffixFeature = new Element("custom"); suffixFeature.setAttribute("class", DocSuffixFeatureGenerator.class.getName()); suffixFeature.setAttribute("begin", beginSuffix); suffixFeature.setAttribute("end", endSuffix); generators.addContent(suffixFeature); System.err.println("-> Suffix Features added!"); } if (Flags.isNgramFeature(params)) { final String charngramRange = Flags.getNgramFeaturesRange(params); final String[] rangeArray = Flags.processNgramRange(charngramRange); final Element charngramFeature = new Element("custom"); charngramFeature.setAttribute("class", NGramFeatureGenerator.class.getName()); charngramFeature.setAttribute("minLength", rangeArray[0]); charngramFeature.setAttribute("maxLength", rangeArray[1]); generators.addContent(charngramFeature); System.err.println("-> Ngram Features added!"); } if (Flags.isCharNgramClassFeature(params)) { final String charngramRange = Flags.getCharNgramFeaturesRange(params); final String[] rangeArray = Flags.processNgramRange(charngramRange); final Element charngramFeature = new Element("custom"); charngramFeature.setAttribute("class", DocCharacterNgramFeatureGenerator.class.getName()); charngramFeature.setAttribute("minLength", rangeArray[0]); charngramFeature.setAttribute("maxLength", rangeArray[1]); generators.addContent(charngramFeature); System.err.println("-> CharNgram Class Features added!"); } // Polarity Dictionary Features if (Flags.isDictionaryPolarityFeatures(params)) { final String dictPath = Flags.getDictionaryPolarityFeatures(params); final List<File> fileList = StringUtils.getFilesInDir(new File(dictPath)); for (final File dictFile : fileList) { final Element dictFeatures = new Element("custom"); dictFeatures.setAttribute("class", DocPolarityDictionaryFeatureGenerator.class.getName()); dictFeatures.setAttribute("dict", IOUtils.normalizeLexiconName(dictFile.getName())); generators.addContent(dictFeatures); } System.err.println("-> Dictionary Features added!"); } // Frequent Word Features if (Flags.isFrequentWordFeatures(params)) { final String dictPath = Flags.getFrequentWordFeatures(params); final List<File> fileList = StringUtils.getFilesInDir(new File(dictPath)); for (final File dictFile : fileList) { final Element dictFeatures = new Element("custom"); dictFeatures.setAttribute("class", FrequentWordFeatureGenerator.class.getName()); dictFeatures.setAttribute("dict", IOUtils.normalizeLexiconName(dictFile.getName())); generators.addContent(dictFeatures); } System.err.println("-> Frequent Word Features added!"); } //Opinion Target Extraction Features if (Flags.isTargetFeatures(params)) { final String targetModelPath = Flags.getTargetFeatures(params); final String targetModelRange = Flags.getTargetFeaturesRange(params); final Element targetClassFeatureElement = new Element("custom"); targetClassFeatureElement.setAttribute("class", DocTargetFeatureGenerator.class.getName()); targetClassFeatureElement.setAttribute("model", IOUtils.normalizeLexiconName(new File(targetModelPath).getName())); targetClassFeatureElement.setAttribute("range", targetModelRange); generators.addContent(targetClassFeatureElement); System.err.println("-> Target Model Features added!"); } // Dictionary Features if (Flags.isDictionaryFeatures(params)) { final String dictPath = Flags.getDictionaryFeatures(params); final String seqCodec = Flags.getSequenceCodec(params); final List<File> fileList = StringUtils.getFilesInDir(new File(dictPath)); for (final File dictFile : fileList) { final Element dictFeatures = new Element("custom"); dictFeatures.setAttribute("class", DocDictionaryFeatureGenerator.class.getName()); dictFeatures.setAttribute("dict", IOUtils.normalizeLexiconName(dictFile.getName())); dictFeatures.setAttribute("seqCodec", seqCodec); generators.addContent(dictFeatures); } System.err.println("-> Dictionary Features added!"); } // Brown clustering features if (Flags.isBrownFeatures(params)) { final String brownClusterPath = Flags.getBrownFeatures(params); final List<File> brownClusterFiles = Flags.getClusterLexiconFiles(brownClusterPath); for (final File brownClusterFile : brownClusterFiles) { // brown bigram class features final Element brownBigramFeatures = new Element("custom"); brownBigramFeatures.setAttribute("class", DocBrownBigramFeatureGenerator.class.getName()); brownBigramFeatures.setAttribute("dict", IOUtils.normalizeLexiconName(brownClusterFile.getName())); //generators.addContent(brownBigramFeatures); // brown token feature final Element brownTokenFeature = new Element("custom"); brownTokenFeature.setAttribute("class", DocBrownTokenFeatureGenerator.class.getName()); brownTokenFeature.setAttribute("dict", IOUtils.normalizeLexiconName(brownClusterFile.getName())); generators.addContent(brownTokenFeature); // brown token class feature final Element brownTokenClassFeature = new Element("custom"); brownTokenClassFeature.setAttribute("class", DocBrownTokenClassFeatureGenerator.class.getName()); brownTokenClassFeature.setAttribute("dict", IOUtils.normalizeLexiconName(brownClusterFile.getName())); //generators.addContent(brownTokenClassFeature); } System.err.println("-> Brown Cluster Features added!"); } // Clark clustering features if (Flags.isClarkFeatures(params)) { final String clarkClusterPath = Flags.getClarkFeatures(params); final List<File> clarkClusterFiles = Flags.getClusterLexiconFiles(clarkClusterPath); for (final File clarkCluster : clarkClusterFiles) { final Element clarkFeatures = new Element("custom"); clarkFeatures.setAttribute("class", DocClarkFeatureGenerator.class.getName()); clarkFeatures.setAttribute("dict", IOUtils.normalizeLexiconName(clarkCluster.getName())); generators.addContent(clarkFeatures); } System.err.println("-> Clark Cluster Features added!"); } // word2vec clustering features if (Flags.isWord2VecClusterFeatures(params)) { final String word2vecClusterPath = Flags.getWord2VecClusterFeatures(params); final List<File> word2vecClusterFiles = Flags.getClusterLexiconFiles(word2vecClusterPath); for (final File word2vecFile : word2vecClusterFiles) { final Element word2vecClusterFeatures = new Element("custom"); word2vecClusterFeatures.setAttribute("class", DocWord2VecClusterFeatureGenerator.class.getName()); word2vecClusterFeatures.setAttribute("dict", IOUtils.normalizeLexiconName(word2vecFile.getName())); generators.addContent(word2vecClusterFeatures); } System.err.println("-> Word2Vec Clusters Features added!"); } // Morphological features if (Flags.isPOSTagModelFeatures(params)) { final String posModelPath = Flags.getPOSTagModelFeatures(params); final String posModelRange = Flags.getPOSTagModelFeaturesRange(params); final Element posTagClassFeatureElement = new Element("custom"); posTagClassFeatureElement.setAttribute("class", DocPOSTagModelFeatureGenerator.class.getName()); posTagClassFeatureElement.setAttribute("model", IOUtils.normalizeLexiconName(new File(posModelPath).getName())); posTagClassFeatureElement.setAttribute("range", posModelRange); generators.addContent(posTagClassFeatureElement); System.err.println("-> POSTagModel Features added!"); } if (Flags.isLemmaModelFeatures(params)) { final String lemmaModelPath = Flags.getLemmaModelFeatures(params); final Element lemmaClassFeatureElement = new Element("custom"); lemmaClassFeatureElement.setAttribute("class", DocLemmaModelFeatureGenerator.class.getName()); lemmaClassFeatureElement.setAttribute("model", IOUtils.normalizeLexiconName(new File(lemmaModelPath).getName())); generators.addContent(lemmaClassFeatureElement); System.err.println("-> LemmaModel Features added!"); } if (Flags.isLemmaDictionaryFeatures(params)) { final String lemmaDictPath = Flags.getLemmaDictionaryFeatures(params); final String[] lemmaDictResources = Flags.getLemmaDictionaryResources(lemmaDictPath); final Element lemmaClassFeatureElement = new Element("custom"); lemmaClassFeatureElement.setAttribute("class", DocLemmaDictionaryFeatureGenerator.class.getName()); lemmaClassFeatureElement.setAttribute("model", IOUtils.normalizeLexiconName(new File(lemmaDictResources[0]).getName())); lemmaClassFeatureElement.setAttribute("dict", IOUtils.normalizeLexiconName(new File(lemmaDictResources[1]).getName())); generators.addContent(lemmaClassFeatureElement); System.err.println("-> LemmaDictionary Features added!"); } final XMLOutputter xmlOutput = new XMLOutputter(); final Format format = Format.getPrettyFormat(); xmlOutput.setFormat(format); return xmlOutput.outputString(doc); }