List of usage examples for org.dom4j ElementPath getCurrent
Element getCurrent();
From source file:StreamFlusher.java
License:Apache License
private Fst xml2fst(String filepath) throws Exception { // read an XML file representing a network, return the network final Fst fst = lib.EmptyLanguageFst(); final HashSet<Integer> sigma = fst.getSigma(); SAXReader reader = new SAXReader(); // SAXReader from dom4j // each SAXReader handler must define onStart() and onEnd() methods // when the kleeneFst element is first found reader.addHandler("/kleeneFst", new ElementHandler() { public void onStart(ElementPath path) { Element current = path.getCurrent(); // semiring is an attribute on the kleeneFst node String semiring = current.attribute("semiring").getValue(); }/*from ww w . j av a2 s . c om*/ public void onEnd(ElementPath path) { } }); reader.addHandler("/kleeneFst/sigma", new ElementHandler() { public void onStart(ElementPath path) { if (path.getCurrent().attribute("containsOther").getValue().equals("true")) { fst.setContainsOther(true); } else { fst.setContainsOther(false); } ; } public void onEnd(ElementPath path) { } }); reader.addHandler("/kleeneFst/sigma/sym", new ElementHandler() { public void onStart(ElementPath path) { } public void onEnd(ElementPath path) { Element sym = path.getCurrent(); sigma.add(symmap.putsym(sym.getText())); sym.detach(); } }); // when the arcs element is first found reader.addHandler("/kleeneFst/arcs", new ElementHandler() { public void onStart(ElementPath path) { // grab the two attrs and convert to int int startState = Integer.parseInt(path.getCurrent().attribute("start").getValue()); int numStates = Integer.parseInt(path.getCurrent().attribute("numStates").getValue()); lib.AddStates(fst, numStates); // native function, add this many // states to the new Fst lib.SetStart(fst, startState); // set the start state } public void onEnd(ElementPath path) { } }); // handle each whole arc element reader.addHandler("/kleeneFst/arcs/arc", new ElementHandler() { // in an ElementHandler, need to supply .onStart(), // called when the start tag is found, and // .onEnd(), which is called when the end tag is found. public void onStart(ElementPath path) { } public void onEnd(ElementPath path) { // retrieve the entire arc element Element arc = path.getCurrent(); // these two are always present int src_id = Integer.parseInt(arc.attribute("s").getValue()); int dest_id = Integer.parseInt(arc.attribute("d").getValue()); // there will be either one io attr xor separate i and o attrs // (keep the io option to facilitate hand-written XML files) String input; String output; Attribute io = arc.attribute("io"); if (io != null) { input = io.getValue(); output = io.getValue(); } else { input = arc.attribute("i").getValue(); output = arc.attribute("o").getValue(); } if (!symmap.containsKey(input)) { // symbol name in XML file not in the // current internal symtab symmap.putsym(input); } if (!symmap.containsKey(output)) { symmap.putsym(output); } // the w attr is optional in the arc elmt Attribute w = arc.attribute("w"); if (w != null) { // call AddArc to add an arc to the fst // being built from the XML file description // semiring generalization point lib.AddArc(fst, src_id, symmap.getint(input), symmap.getint(output), Float.parseFloat(w.getValue()), dest_id); } else { // semiring generalization point lib.AddArcNeutralWeight(fst, src_id, symmap.getint(input), symmap.getint(output), dest_id); } arc.detach(); } }); // for each full final element reader.addHandler("/kleeneFst/arcs/final", new ElementHandler() { public void onStart(ElementPath path) { } public void onEnd(ElementPath path) { Element arc = path.getCurrent(); // s attr is always present int src_id = Integer.parseInt(arc.attribute("s").getValue()); // the w attr is optional Attribute w = arc.attribute("w"); if (w != null) { lib.SetFinal(fst, src_id, Float.parseFloat(w.getValue())); } else { lib.SetFinalNeutralWeight(fst, src_id); } arc.detach(); } }); Document document = null; // the actual XML reading/parsing is done here try { // XmlReader detects the encoding of the XML document and // handles BOMs, including the UTF-8 BOMs that Java usually // chokes on document = reader.read(new XmlReader(new FileInputStream(filepath))); // Old, pre-XmlReader code //if (encoding.equals("UTF-8")) { // // then need to work around SUN's irresponsible decision not to // // handle the optional UTF-8 BOM correctly // document = reader.read(new InputStreamReader( // new UTF8BOMStripperInputStream(new FileInputStream(filepath)), // "UTF-8") // ) ; //} else { // document = reader.read(new InputStreamReader( // new FileInputStream(filepath), // encoding) // ) ; //} } catch (DocumentException de) { // dom4j DocumentException extends Exception de.printStackTrace(); throw de; } catch (FileNotFoundException fnfe) { fnfe.printStackTrace(); throw fnfe; } catch (Exception e) { e.printStackTrace(); throw e; } correctSigmaOther(fst); return fst; }
From source file:com.globalsight.everest.projecthandler.importer.XmlReader.java
License:Apache License
/** * Reads an XML file and checks its correctness by validating * against the TMX DTD. If there's any error in the file, an * exception is thrown./*from w w w. jav a 2 s.c o m*/ */ private void analyzeXml(String p_url) throws Exception { CATEGORY.debug("Analyzing document: " + p_url); SAXReader reader = new SAXReader(); // TODO: Read the DTD and validate. // See com.globalsight.everest.tm.util.DtdResolver; // reader.setEntityResolver(DtdResolver.getInstance()); // reader.setValidation(true); // enable element complete notifications to conserve memory //TODO reader.addHandler("/projectdata/data", new ElementHandler() { public void onStart(ElementPath path) { ++m_entryCount; } public void onEnd(ElementPath path) { Element element = path.getCurrent(); // prune the current element to reduce memory element.detach(); } }); Document document = reader.read(p_url); // all done }
From source file:com.globalsight.everest.projecthandler.importer.XmlReaderThread.java
License:Apache License
public void run() { try {//from ww w . j a v a2 s. c o m SAXReader reader = new SAXReader(); // TODO: Read the DTD and validate. // See com.globalsight.everest.tm.util.DtdResolver; // reader.setEntityResolver(DtdResolver.getInstance()); // reader.setValidation(true); reader.addHandler("/projectdata", new ElementHandler() { public void onStart(ElementPath path) { Element element = path.getCurrent(); } public void onEnd(ElementPath path) { } }); // enable pruning to call me back as each Element is complete reader.addHandler("/projectdata/data", new ElementHandler() { public void onStart(ElementPath path) { m_count++; } public void onEnd(ElementPath path) { Element element = path.getCurrent(); // prune the current element to reduce memory element.detach(); m_result = m_results.hireResult(); try { // TODO: Create data objects Object o = /*createObject*/(element); if (CATEGORY.isDebugEnabled()) { CATEGORY.debug(o); } m_result.setResultObject(o); } catch (Throwable ex) { m_result.setError(ex.toString()); CATEGORY.warn("Error in object " + m_count, ex); } boolean done = m_results.put(m_result); m_result = null; // Stop reading the file. if (done) { throw new ThreadDeath(); } } }); String url = m_options.getFileName(); Document document = reader.read(url); } catch (ThreadDeath ignore) { CATEGORY.info("ReaderThread: interrupted"); } catch (Throwable ignore) { // Should never happen, and I don't know how to handle // this case other than passing the exception in // m_results, which I won't do for now. CATEGORY.error("unexpected error", ignore); } finally { if (m_result != null) { m_results.fireResult(m_result); } m_results.producerDone(); m_results = null; CATEGORY.debug("ReaderThread: done."); } }
From source file:com.globalsight.everest.tm.importer.TmxReader.java
License:Apache License
/** * Reads an XML file and checks its correctness by validating * against the TMX DTD. If there's any error in the file, an * exception is thrown./*from w w w . j a v a 2s. co m*/ * * As a side effect, this method builds a list of source and * target locales found in the file, including the declared source * locale from the header. */ private void analyzeXml(String p_url) throws Exception { if (m_tmxLevel == ImportUtil.TMX_LEVEL_TRADOS_RTF || m_tmxLevel == ImportUtil.TMX_LEVEL_TRADOS_HTML || m_tmxLevel == ImportUtil.TMX_LEVEL_TRADOS_FM || m_tmxLevel == ImportUtil.TMX_LEVEL_TRADOS_FM_SGML || m_tmxLevel == ImportUtil.TMX_LEVEL_TRADOS_IL || m_tmxLevel == ImportUtil.TMX_LEVEL_TRADOS_XPTAG) { // Convert the Trados codes to native System4 codes by // converting the file to RTF, saving it as HTML and // extracting the resulting TUVs. CATEGORY.info("Converting Trados TMX to native TMX: " + p_url); p_url = convertTradosTmx(p_url, m_tmxLevel); // Now we have a new file that contains native content. m_options.setFileName(p_url); m_options.setFileType(com.globalsight.everest.tm.importer.ImportOptions.TYPE_XML); m_tmxLevel = ImportUtil.TMX_LEVEL_NATIVE; } CATEGORY.debug("Analyzing document: " + p_url); // Reset list of locales found in the file. m_sourceLocales = new HashSet(); m_targetLocales = new HashSet(); SAXReader reader = new SAXReader(); reader.setXMLReaderClassName("org.apache.xerces.parsers.SAXParser"); // Read the DTD and validate. reader.setEntityResolver(DtdResolver.getInstance()); reader.setValidation(true); // enable element complete notifications to conserve memory reader.addHandler("/tmx", new ElementHandler() { public void onStart(ElementPath path) { Element element = path.getCurrent(); m_tmxVersion = element.attributeValue(Tmx.VERSION); } public void onEnd(ElementPath path) { } }); reader.addHandler("/tmx/header", new ElementHandler() { public void onStart(ElementPath path) { } public void onEnd(ElementPath path) { Element element = path.getCurrent(); m_header = new Tmx(element); m_header.setTmxVersion(m_tmxVersion); element.detach(); } }); reader.addHandler("/tmx/body/tu", new ElementHandler() { public void onStart(ElementPath path) { ++m_entryCount; } public void onEnd(ElementPath path) { Element element = path.getCurrent(); // Record optional source language declared on TU. String srclang = element.attributeValue(Tmx.SRCLANG); if (srclang != null) { m_sourceLocales.add(ImportUtil.normalizeLocale(srclang)); } // Find target languages HashSet langs = new HashSet(); List tuvs = element.selectNodes("./tuv"); for (int i = 0, max = tuvs.size(); i < max; i++) { Element tuv = (Element) tuvs.get(i); String lang = tuv.attributeValue(Tmx.LANG); // Collect TUV locales langs.add(ImportUtil.normalizeLocale(lang)); } langs.remove(srclang); m_targetLocales.addAll(langs); // prune the current element to reduce memory element.detach(); } }); Document document = reader.read(p_url); // Add declared source language from header. String sourceLocale = ImportUtil.normalizeLocale(m_header.getSourceLang()); m_sourceLocales.add(sourceLocale); }
From source file:com.globalsight.everest.tm.importer.TmxReaderThread.java
License:Apache License
public void run() { try {/*w ww.j av a2 s .c om*/ SAXReader reader = new SAXReader(); reader.setXMLReaderClassName("org.apache.xerces.parsers.SAXParser"); // Read the DTD and validate. reader.setEntityResolver(DtdResolver.getInstance()); reader.setValidation(true); reader.addHandler("/tmx", new ElementHandler() { public void onStart(ElementPath path) { Element element = path.getCurrent(); m_tmxVersion = element.attributeValue(Tmx.VERSION); } public void onEnd(ElementPath path) { } }); reader.addHandler("/tmx/header", new ElementHandler() { public void onStart(ElementPath path) { } public void onEnd(ElementPath path) { Element element = path.getCurrent(); element.detach(); m_tmx = new Tmx(element); m_tmx.setTmxVersion(m_tmxVersion); m_defaultSrcLang = ImportUtil.normalizeLocale(m_tmx.getSourceLang()); } }); // enable pruning to call me back as each Element is complete reader.addHandler("/tmx/body/tu", new ElementHandler() { public void onStart(ElementPath path) { m_count++; } public void onEnd(ElementPath path) { Element element = path.getCurrent(); element.detach(); m_result = m_results.hireResult(); try { // Normalize spelling of locales. normalizeTu(element); // Filter out targets not to be imported. filterTu(element); // Validate we have source and target. validateTu(element); // Create TU objects SegmentTmTu tu = createTu(element); if (CATEGORY.isDebugEnabled()) { CATEGORY.debug(tu.toDebugString(true)); } m_result.setResultObject(tu); } catch (Throwable ex) { String msg = "Entry " + m_count + ": " + ex.getMessage(); m_result.setError(msg); if (CATEGORY.isDebugEnabled()) { CATEGORY.debug(msg, ex); } else { CATEGORY.warn(msg); } } boolean done = m_results.put(m_result); m_result = null; // Stop reading the TMX file. if (done) { throw new ThreadDeath(); } } }); String url = m_options.getFileName(); Document document = reader.read(url); } catch (ThreadDeath ignore) { CATEGORY.info("ReaderThread: interrupted."); } catch (Throwable ignore) { // Should never happen, and I don't know how to handle // this case other than passing the exception in // m_results, which I won't do for now. CATEGORY.error("unexpected error", ignore); } finally { if (m_result != null) { m_results.fireResult(m_result); } m_results.producerDone(); m_results = null; CATEGORY.debug("ReaderThread: done."); } }
From source file:com.globalsight.everest.tm.util.TmxAnalyzer.java
License:Apache License
public void analyze(String p_url) throws Exception { m_tuCount = 0;/*from w w w .java 2s . c o m*/ m_tuvCount = 0; m_localeCount = 0; m_locales = new HashSet(); m_tmxVersion = ""; m_tmx = null; SAXReader reader = new SAXReader(); reader.setXMLReaderClassName("org.apache.xerces.parsers.SAXParser"); reader.setEntityResolver(DtdResolver.getInstance()); reader.setValidation(true); log("Analyzing document: " + p_url); reader.addHandler("/tmx", new ElementHandler() { public void onStart(ElementPath path) { Element element = path.getCurrent(); m_tmxVersion = element.attributeValue("version"); } public void onEnd(ElementPath path) { } }); reader.addHandler("/tmx/header", new ElementHandler() { public void onStart(ElementPath path) { } public void onEnd(ElementPath path) { Element element = path.getCurrent(); element.detach(); m_tmx = new Tmx(element); m_tmx.setTmxVersion(m_tmxVersion); } }); // enable element complete notifications to conserve memory reader.addHandler("/tmx/body/tu", new ElementHandler() { public void onStart(ElementPath path) { ++m_tuCount; if (m_tuCount % 1000 == 0) { log("TU " + m_tuCount); } } public void onEnd(ElementPath path) { Element element = path.getCurrent(); List tuvs = element.selectNodes("//tuv"); m_tuvCount += tuvs.size(); for (int i = 0, max = tuvs.size(); i < max; i++) { Element tuv = (Element) tuvs.get(i); String locale = tuv.attributeValue("lang"); m_locales.add(locale); } // prune the current element to reduce memory element.detach(); element = null; } }); Document document = reader.read(p_url); m_localeCount = m_locales.size(); log("File: " + p_url); log("TMX version: " + m_tmxVersion); log("Total TUs: " + m_tuCount); log("Total TUVs: " + m_tuvCount); log("Total Locales: " + m_localeCount); for (Iterator it = m_locales.iterator(); it.hasNext();) { String locale = (String) it.next(); log(locale); } // all done }
From source file:com.globalsight.everest.tm.util.TmxLevelSplitter.java
License:Apache License
public void split(String p_url) throws Exception { final String baseName = getBaseName(p_url); final String extension = getExtension(p_url); m_entryCount = 0;/* ww w. j av a 2 s. c om*/ SAXReader reader = new SAXReader(); reader.setXMLReaderClassName("org.apache.xerces.parsers.SAXParser"); reader.setEntityResolver(DtdResolver.getInstance()); reader.setValidation(true); log("Splitting document `" + p_url + "'"); // enable element complete notifications to conserve memory reader.addHandler("/tmx", new ElementHandler() { public void onStart(ElementPath path) { Element element = path.getCurrent(); m_version = element.attributeValue("version"); } public void onEnd(ElementPath path) { } }); // enable element complete notifications to conserve memory reader.addHandler("/tmx/header", new ElementHandler() { public void onStart(ElementPath path) { } public void onEnd(ElementPath path) { Element element = path.getCurrent(); m_header = element; try { startFiles(baseName, extension); } catch (Exception ex) { log(ex.toString()); System.exit(1); } // prune the current element to reduce memory element.detach(); element = null; } }); // enable element complete notifications to conserve memory reader.addHandler("/tmx/body/tu", new ElementHandler() { public void onStart(ElementPath path) { ++m_entryCount; } public void onEnd(ElementPath path) { Element element = path.getCurrent(); if (containsTags(element)) { writeTagsEntry(element.asXML()); m_tagsCount++; } else { writeTextEntry(element.asXML()); m_textCount++; } // prune the current element to reduce memory element.detach(); element = null; } }); Document document = reader.read(p_url); closeFiles(); log("Processed " + m_entryCount + " TUs, " + m_textCount + " level 1 (text), " + m_tagsCount + " level 2 (tags)"); // all done }
From source file:com.globalsight.everest.tm.util.TmxSplitter.java
License:Apache License
public void split(String p_url, String p_numEntries) throws Exception { final int maxEntries = Integer.parseInt(p_numEntries); final String baseName = getBaseName(p_url); final String extension = getExtension(p_url); m_entryCount = 0;//w ww. j av a 2 s . c o m SAXReader reader = new SAXReader(); reader.setXMLReaderClassName("org.apache.xerces.parsers.SAXParser"); reader.setEntityResolver(DtdResolver.getInstance()); reader.setValidation(true); log("Splitting document `" + p_url + "'"); // enable element complete notifications to conserve memory reader.addHandler("/tmx", new ElementHandler() { public void onStart(ElementPath path) { Element element = path.getCurrent(); m_version = element.attributeValue("version"); } public void onEnd(ElementPath path) { } }); // enable element complete notifications to conserve memory reader.addHandler("/tmx/header", new ElementHandler() { public void onStart(ElementPath path) { } public void onEnd(ElementPath path) { Element element = path.getCurrent(); m_header = element; try { startFile(baseName, extension); } catch (Exception ex) { log(ex.toString()); System.exit(1); } // prune the current element to reduce memory element.detach(); element = null; } }); // enable element complete notifications to conserve memory reader.addHandler("/tmx/body/tu", new ElementHandler() { public void onStart(ElementPath path) { ++m_entryCount; if (m_entryCount % maxEntries == 0) { try { closeFile(); startFile(baseName, extension); } catch (Exception ex) { log(ex.toString()); System.exit(1); } } } public void onEnd(ElementPath path) { Element element = path.getCurrent(); writeEntry(element.asXML()); // prune the current element to reduce memory element.detach(); element = null; } }); Document document = reader.read(p_url); closeFile(); // all done }
From source file:com.globalsight.everest.tm.util.trados.TradosFmSgmlTmxToGxml.java
License:Apache License
/** * Main method to call, returns the new filename of the result. *//*from w w w . jav a 2 s . c om*/ public String convertToGxml(String p_url) throws Exception { final String baseName = getBaseName(p_url); final String extension = getExtension(p_url); info("Converting TMX file to GXML: `" + p_url + "'"); startOutputFile(baseName); m_entryCount = 0; // Reading from a file, need to use Xerces. SAXReader reader = new SAXReader(); reader.setXMLReaderClassName("org.apache.xerces.parsers.SAXParser"); reader.setEntityResolver(DtdResolver.getInstance()); reader.setValidation(true); // enable element complete notifications to conserve memory reader.addHandler("/tmx", new ElementHandler() { final public void onStart(ElementPath path) { Element element = path.getCurrent(); m_version = element.attributeValue("version"); } final public void onEnd(ElementPath path) { } }); // enable element complete notifications to conserve memory reader.addHandler("/tmx/header", new ElementHandler() { final public void onStart(ElementPath path) { } final public void onEnd(ElementPath path) { Element element = path.getCurrent(); setOldHeader(element); createNewHeader(); // prune the current element to reduce memory element.detach(); element = null; } }); // enable element complete notifications to conserve memory reader.addHandler("/tmx/body/tu", new ElementHandler() { final public void onStart(ElementPath path) { ++m_entryCount; m_tuError = false; } final public void onEnd(ElementPath path) { Element element = path.getCurrent(); if (m_tuError) { m_errorCount++; } else { writeEntry(element.asXML()); } // prune the current element to reduce memory element.detach(); element = null; if (m_entryCount % 1000 == 0) { debug("Entry " + m_entryCount); } } }); // enable element complete notifications to conserve memory reader.addHandler("/tmx/body/tu/tuv/seg", new ElementHandler() { final public void onStart(ElementPath path) { } final public void onEnd(ElementPath path) { Element element = path.getCurrent(); try { String gxml = handleTuv(element); Document doc = parse("<root>" + gxml + "</root>"); // Remove old content of seg List content = element.content(); for (int i = content.size() - 1; i >= 0; --i) { ((Node) content.get(i)).detach(); } // Add new GXML content (backwards) content = doc.getRootElement().content(); Collections.reverse(content); for (int i = content.size() - 1; i >= 0; --i) { Node node = (Node) content.get(i); element.add(node.detach()); } } catch (Throwable ex) { m_tuError = true; } } }); Document document = reader.read(p_url); closeOutputFile(); info("Processed " + m_entryCount + " TUs " + "into file `" + m_filename + "', " + m_errorCount + " errors."); return m_filename; }
From source file:com.globalsight.everest.tm.util.trados.TradosFmTmxToGxml.java
License:Apache License
/** * Main method to call, returns the new filename of the result. */// w w w . j a va2 s.co m public String convertToGxml(String p_url) throws Exception { final String baseName = getBaseName(p_url); final String extension = getExtension(p_url); info("Converting TMX file to GXML: `" + p_url + "'"); startOutputFile(baseName); m_entryCount = 0; // Reading from a file, need to use Xerces. SAXReader reader = new SAXReader(); reader.setXMLReaderClassName("org.apache.xerces.parsers.SAXParser"); reader.setEntityResolver(DtdResolver.getInstance()); reader.setValidation(true); // enable element complete notifications to conserve memory reader.addHandler("/tmx", new ElementHandler() { public void onStart(ElementPath path) { Element element = path.getCurrent(); m_version = element.attributeValue("version"); } public void onEnd(ElementPath path) { } }); // enable element complete notifications to conserve memory reader.addHandler("/tmx/header", new ElementHandler() { public void onStart(ElementPath path) { } public void onEnd(ElementPath path) { Element element = path.getCurrent(); setOldHeader(element); createNewHeader(); // prune the current element to reduce memory element.detach(); element = null; } }); // enable element complete notifications to conserve memory reader.addHandler("/tmx/body/tu", new ElementHandler() { public void onStart(ElementPath path) { ++m_entryCount; m_tuError = false; } public void onEnd(ElementPath path) { Element element = path.getCurrent(); if (m_tuError) { m_errorCount++; } else { writeEntry(element.asXML()); } // prune the current element to reduce memory element.detach(); element = null; if (m_entryCount % 1000 == 0) { debug("Entry " + m_entryCount); } } }); // enable element complete notifications to conserve memory reader.addHandler("/tmx/body/tu/tuv/seg", new ElementHandler() { public void onStart(ElementPath path) { } public void onEnd(ElementPath path) { Element element = path.getCurrent(); try { String gxml = handleTuv(element); Document doc = parse("<root>" + gxml + "</root>"); // Remove old content of seg List content = element.content(); for (int i = content.size() - 1; i >= 0; --i) { ((Node) content.get(i)).detach(); } // Add new GXML content (backwards) content = doc.getRootElement().content(); Collections.reverse(content); for (int i = content.size() - 1; i >= 0; --i) { Node node = (Node) content.get(i); element.add(node.detach()); } } catch (Throwable ex) { m_tuError = true; } } }); Document document = reader.read(p_url); closeOutputFile(); info("Processed " + m_entryCount + " TUs " + "into file `" + m_filename + "', " + m_errorCount + " errors."); return m_filename; }