Example usage for edu.stanford.nlp.trees Tree toString

List of usage examples for edu.stanford.nlp.trees Tree toString

  1. HOME
  2. Java
  3. edu.stanford.nlp
  4. edu.stanford.nlp.trees.*
  5. Tree
  6. toString

Introduction

In this page you can find the example usage for edu.stanford.nlp.trees Tree toString.

Prototype

@Override
public String toString()

Source Link

Document

Converts parse tree to string in Penn Treebank format.

Usage

From source file:elkfed.mmax.importer.ImportOntonotes.java

License:Apache License

public MiniDiscourse importFile(String fname) {
    try {//ww  w  .j  ava  2s  .  co m
        boolean had_space = true;
        boolean need_bugfix = System.getProperty("elkfed.BuggyOntonotes", "no").matches("y|yes|true");
        List<Tag> names_stack = new ArrayList<Tag>();
        Alphabet<String> sets = new Alphabet<String>();
        sets.lookupIndex("*DUMMY*");
        int sent_id = 0;
        Tag sentence_tag = null;
        OntonotesReader reader = new OntonotesReader(new File(fname + ".coref"));
        OntonotesReader readerNE = new OntonotesReader(new File(fname + ".name"));
        TreeReader tr = new PennTreeReader(new FileReader(fname + ".parse"), new LabeledScoredTreeFactory(),
                new BobChrisTreeNormalizer());
        Tree tree = null;
        int eventType = reader.getNextEvent();
        boolean in_text = false;
        do {
            if (eventType == OntonotesReader.START_TAG && "COREF".equals(reader.getName())) {
                Tag t;
                if (need_bugfix) {
                    t = buggy_push_tag("coref", tag_stack);
                } else {
                    t = push_tag("coref");
                }
                if ("IDENT".equals(reader.getAttribute("TYPE"))) {
                    t.attrs.put("coref_set", "set_" + sets.lookupIndex(reader.getAttribute("ID")));
                }
                had_space = true;
            } else if (eventType == OntonotesReader.END_TAG && "COREF".equals(reader.getName())) {
                Tag t = pop_tag("coref");
                DetermineMinSpan.addMinSpan(sentence_tag.start, tree, t, tokens);
                had_space = true;
            } else if (in_text && eventType == OntonotesReader.TOKEN) {
                if (!reader.isTrace()) {
                    // process up to the next token in the names part
                    int names_event = readerNE.getNextEvent();
                    while (names_event != OntonotesReader.TOKEN) {
                        if (names_event == OntonotesReader.START_TAG && "ENAMEX".equals(readerNE.getName())) {
                            Tag t = push_tag("enamex", names_stack);
                            t.attrs.put("tag", readerNE.getAttribute("TYPE"));
                        } else if (names_event == OntonotesReader.END_TAG
                                && "ENAMEX".equals(readerNE.getName())) {
                            Tag t = pop_tag("enamex", names_stack);
                        } else {
                            throw new IllegalStateException("Unexpected event:" + names_event);
                        }
                        names_event = readerNE.getNextEvent();
                    }
                    assert (reader.getToken().equals(readerNE.getToken()));
                    String tok = reader.getToken();
                    if (tok.equals("-LRB-"))
                        tok = "(";
                    if (tok.equals("-RRB-"))
                        tok = ")";
                    if (tok.equals("-LSB-"))
                        tok = "[";
                    if (tok.equals("-RSB-"))
                        tok = "]";
                    if (tok.equals("-LCB-"))
                        tok = "{";
                    if (tok.equals("-RCB-"))
                        tok = "}";
                    add_token(tok);
                }
            } else if (in_text && eventType == OntonotesReader.NEWLINE) {
                //System.out.println("sentence break");
                if (sentence_tag != null) {
                    sentence_tag.end = tokens.size() - 1;
                    if (sentence_tag.end >= sentence_tag.start) {
                        tags.add(sentence_tag);
                        if (tree != null) {
                            Tag parse_tag = new Tag();
                            parse_tag.tag = "parse";
                            parse_tag.start = sentence_tag.start;
                            parse_tag.end = sentence_tag.end;
                            parse_tag.attrs.put("tag", tree.toString());
                            tags.add(parse_tag);
                            assert sentence_tag.end - sentence_tag.start + 1 == tree.yield().size() : String
                                    .format("%s / %s", tokens.subList(sentence_tag.start, sentence_tag.end + 1),
                                            tree.yield());
                            addParseInfo(sentence_tag.start, tree);
                        }
                    }
                }
                // process up to end of sentence in names annotation
                int names_event = readerNE.getNextEvent();
                while (names_event != OntonotesReader.NEWLINE) {
                    if (names_event == OntonotesReader.START_TAG && "ENAMEX".equals(readerNE.getName())) {
                        Tag t = push_tag("enamex", names_stack);
                        t.attrs.put("tag", readerNE.getAttribute("TYPE"));
                    } else if (names_event == OntonotesReader.END_TAG && "ENAMEX".equals(readerNE.getName())) {
                        Tag t = pop_tag("enamex", names_stack);
                    } else if (names_event == OntonotesReader.END_TAG && "DOC".equals(readerNE.getName())) {
                        // ignore
                    } else {
                        throw new IllegalStateException(
                                "Unexpected event:" + readerNE.describeEvent(names_event));
                    }
                    names_event = readerNE.getNextEvent();
                }
                // prepare new parse and sentence
                sentence_tag = new Tag();
                sentence_tag.start = tokens.size();
                sentence_tag.tag = "sentence";
                sentence_tag.attrs.put("orderid", "" + sent_id++);
                tree = tr.readTree();
            } else if (eventType == OntonotesReader.END_TAG && "DOCNO".equals(reader.getName())) {
                in_text = true;
                // go to the end of the DOCNO part in name doc
                int names_event = readerNE.getNextEvent();
                while (names_event != OntonotesReader.END_TAG || !"DOCNO".equals(reader.getName())) {
                    names_event = readerNE.getNextEvent();
                }
            } else if (eventType == OntonotesReader.START_TAG && "TURN".equals(reader.getName())) {
                int names_event = readerNE.getNextEvent();
                if (names_event != OntonotesReader.START_TAG || !"TURN".equals(readerNE.getName())) {
                    throw new UnsupportedOperationException("TURN in coref but not in names");
                }
                // parse level seems to be inconsistent... so don't check here :-|
                System.err.println("TURN parse:" + tree.toString());
                tree = tr.readTree();
                eventType = reader.getNextEvent();
                names_event = readerNE.getNextEvent();
                if (eventType != OntonotesReader.NEWLINE || names_event != OntonotesReader.NEWLINE) {
                    throw new UnsupportedOperationException("No Newline after TURN");
                }
            }
            eventType = reader.getNextEvent();
        } while (eventType != OntonotesReader.END_DOCUMENT);
        return create();
    } catch (IOException ex) {
        throw new RuntimeException("Cannot read file", ex);
    }

}

From source file:knowledgeextraction.EntityAttributeGraph.java

public static void main(String[] args) throws IOException {
    // TODO code application logic here

    BufferedReader reader = new BufferedReader(new FileReader(filePath));
    String text = reader.readLine();

    Annotation document = new Annotation(text);
    Properties props = new Properties();
    props.put("annotators", "tokenize, ssplit");//, pos, lemma, ner, parse, dcoref");
    StanfordCoreNLP pipeline = new StanfordCoreNLP(props);
    pipeline.annotate(document);// ww w . j av  a2  s  .  c o m
    List<CoreMap> sentences = document.get(CoreAnnotations.SentencesAnnotation.class);

    for (CoreMap sentence : sentences) {
        String input = sentence.toString();
        System.out.println(input);
        Tree tree = new EntityAttributeGraph().parse(input);
        System.out.println("tree: " + tree.toString());

        // Get dependency tree
        TreebankLanguagePack tlp = new PennTreebankLanguagePack();
        GrammaticalStructureFactory gsf = tlp.grammaticalStructureFactory();
        GrammaticalStructure gs = gsf.newGrammaticalStructure(tree);
        Collection<TypedDependency> td = gs.typedDependenciesCollapsed();
        //System.out.println(td);

        Object[] list = td.toArray();
        //System.out.println(list.length);
        PrintBestPath(list);
        //System.out.println();
    }
}

From source file:knu.univ.lingvo.coref.MentionExtractor.java

License:Open Source License

private String treeToKey(Tree t) {
    int idx = getHeadIndex(t);
    String key = Integer.toString(idx) + ':' + t.toString();
    return key;// w w w. j  a v  a  2s. c o  m
}

From source file:MedArkRef.AnalysisUtilities.java

License:Open Source License

public arkref.parsestuff.AnalysisUtilities.ParseResult parseSentence(String sentence) {
    String result = "";
    //System.err.println(sentence);
    //see if a parser socket server is available
    int port = new Integer(GlobalProperties.getProperties().getProperty("parserServerPort", "5556"));
    String host = "127.0.0.1";
    Socket client;/*  w  w  w  .  j av a 2 s. co m*/
    PrintWriter pw;
    BufferedReader br;
    String line;
    Tree parse = null;
    double parseScore = Double.MIN_VALUE;

    try {
        client = new Socket(host, port);

        pw = new PrintWriter(client.getOutputStream());
        br = new BufferedReader(new InputStreamReader(client.getInputStream()));
        pw.println(sentence);
        pw.flush(); //flush to complete the transmission

        while ((line = br.readLine()) != null) {
            //if(!line.matches(".*\\S.*")){
            //        System.out.println();
            //}
            if (br.ready()) {
                line = line.replaceAll("\n", "");
                line = line.replaceAll("\\s+", " ");
                result += line + " ";
            } else {
                parseScore = new Double(line);
            }
        }

        br.close();
        pw.close();
        client.close();

        if (parse == null) {
            parse = readTreeFromString("(ROOT (. .))");
            parseScore = -99999.0;
        }

        if (GlobalProperties.getDebug())
            System.err.println("result (parse):" + result);
        parse = readTreeFromString(result);
        return new arkref.parsestuff.AnalysisUtilities.ParseResult(true, parse, parseScore);

    } catch (Exception ex) {
        if (GlobalProperties.getDebug())
            System.err.println("Could not connect to parser server.");
        //ex.printStackTrace();
    }

    System.err.println("parsing:" + sentence);

    //if socket server not available, then use a local parser object
    if (parser == null) {
        try {
            Options op = new Options();
            String serializedInputFileOrUrl = GlobalProperties.getProperties().getProperty("parserGrammarFile",
                    "config" + File.separator + "englishFactored.ser.gz");
            parser = new LexicalizedParser(serializedInputFileOrUrl, op);
            int maxLength = new Integer(GlobalProperties.getProperties().getProperty("parserMaxLength", "40"))
                    .intValue();
            parser.setMaxLength(maxLength);
            parser.setOptionFlags("-outputFormat", "oneline");
        } catch (Exception e) {
            e.printStackTrace();
        }
    }

    try {
        if (parser.parse(sentence)) {
            parse = parser.getBestParse();

            //remove all the parent annotations (this is a hacky way to do it)
            String ps = parse.toString().replaceAll("\\[[^\\]]+/[^\\]]+\\]", "");
            parse = AnalysisUtilities.getInstance().readTreeFromString(ps);

            parseScore = parser.getPCFGScore();
            return new arkref.parsestuff.AnalysisUtilities.ParseResult(true, parse, parseScore);
        }
    } catch (Exception e) {
    }

    parse = readTreeFromString("(ROOT (. .))");
    parseScore = -99999.0;
    return new arkref.parsestuff.AnalysisUtilities.ParseResult(false, parse, parseScore);
}

From source file:opennlp.tools.parse_thicket.kernel_interface.MultiSentenceExtendedForestSearchResultsProcessorSetFormer.java

License:Apache License

protected List<String[]> formTreeKernelStructure(String searchResultText, int count, List<HitBase> hits,
        boolean isPositive) {
    List<String[]> treeBankBuffer = new ArrayList<String[]>();
    try {//ww w .j  a  va  2  s .c o m
        // get the parses from original documents, and form the training
        // dataset
        ParseThicket pt = matcher.buildParseThicketFromTextWithRST(searchResultText);
        List<Tree> forest = pt.getSentences();
        // if from the first half or ranked docs, then positive, otherwise
        // negative
        String posOrNeg = null;
        if (isPositive)
            posOrNeg = " 1 ";
        else
            posOrNeg = " -1 ";
        // form the list of training samples
        for (Tree t : forest) {
            treeBankBuffer.add(new String[] { posOrNeg + " |BT| " + t.toString() + " |ET|" });
        }
    } catch (Exception e) {
        e.printStackTrace();
    }
    return treeBankBuffer;
}

From source file:opennlp.tools.parse_thicket.kernel_interface.MultiSentenceKernelBasedSearchResultsProcessor.java

License:Apache License

protected List<String[]> formTreeKernelStructure(String searchResultText, int count, List<HitBase> hits) {
    List<String[]> treeBankBuffer = new ArrayList<String[]>();
    try {//from   www. j  a  va  2 s .c o  m
        // get the parses from original documents, and form the training
        // dataset
        ParseThicket pt = matcher.buildParseThicketFromTextWithRST(searchResultText);
        List<Tree> forest = pt.getSentences();
        // if from the first half or ranked docs, then positive, otherwise
        // negative
        String posOrNeg = null;
        if (count < hits.size() / 2)
            posOrNeg = " 1 ";
        else
            posOrNeg = " -1 ";
        // form the list of training samples
        for (Tree t : forest) {
            treeBankBuffer.add(new String[] { posOrNeg + " |BT| " + t.toString() + " |ET|" });
        }
    } catch (Exception e) {
        // TODO Auto-generated catch block
        e.printStackTrace();
    }
    return treeBankBuffer;
}

From source file:opennlp.tools.parse_thicket.kernel_interface.PT2ExtendedTreeForestBuilder.java

License:Apache License

private List<String[]> formTrainingSetFromText(String para, boolean positive) {
    String prefix = null;/*  w  w  w.  ja v  a  2s .  c o  m*/
    if (positive)
        prefix = " 1 ";
    else
        prefix = " -1 ";

    ParseThicket pt = matcher.buildParseThicketFromTextWithRST(para);
    List<Tree> forest = pt.getSentences();
    List<String[]> treeBankBuffer = new ArrayList<String[]>();
    for (Tree t : forest) {
        treeBankBuffer.add(new String[] { prefix + "|BT| " + t.toString() + " |ET|" });
    }
    return treeBankBuffer;
}

From source file:opennlp.tools.parse_thicket.kernel_interface.PT2ExtendedTreeForestBuilder.java

License:Apache License

private String formTrainingSetFromTextOneLine(String para, boolean positive) {
    String prefix = null;/* ww  w  .  j  av  a2  s .c om*/
    if (positive)
        prefix = " 1 ";
    else
        prefix = " -1 ";

    ParseThicket pt = matcher.buildParseThicketFromTextWithRST(para);
    List<Tree> forest = pt.getSentences();
    String line = prefix;
    for (Tree t : forest) {
        line += "|BT| " + t.toString() + " |ET| ";
    }
    return line;
}

From source file:opennlp.tools.parse_thicket.kernel_interface.PT2ExtendedTreeForestBuilder.java

License:Apache License

public void classifySentences(String sentences, String path) {
    ParseThicket pt = matcher.buildParseThicketFromTextWithRST(sentences);
    List<Tree> forest = pt.getSentences();
    List<String[]> treeBankBuffer = new ArrayList<String[]>();
    for (Tree t : forest) {
        treeBankBuffer.add(new String[] { " 0 |BT| " + t.toString() + " |ET|" });
    }/*w  ww. j a v  a  2 s.c  o  m*/

    ProfileReaderWriter.writeReport(treeBankBuffer, path + "unknown.txt", ' ');
    tkRunner.runClassifier(path, "unknown.txt", modelFileName, "classifier_output.txt");
}

From source file:opennlp.tools.parse_thicket.kernel_interface.TreeExtenderByAnotherLinkedTree.java

License:Apache License

public List<String> buildForestForCorefArcs(ParseThicket pt) {
    List<String> results = new ArrayList<String>();
    for (WordWordInterSentenceRelationArc arc : pt.getArcs()) {
        // if (!arc.getArcType().getType().startsWith("coref"))
        // continue;
        int fromSent = arc.getCodeFrom().getFirst();
        int toSent = arc.getCodeTo().getFirst();
        if (fromSent < 1 || toSent < 1) // TODO problem in sentence
            // enumeration => skip building
            // extended trees
            return results;

        String wordFrom = arc.getLemmaFrom();
        String wordTo = arc.getLemmaTo();

        List<Tree> trees = getASubtreeWithRootAsNodeForWord1(pt.getSentences().get(fromSent - 1),
                pt.getSentences().get(fromSent - 1), new String[] { wordFrom });
        if (trees == null || trees.size() < 1)
            continue;
        System.out.println(trees);
        StringBuilder sb = new StringBuilder(10000);
        toStringBuilderExtenderByAnotherLinkedTree1(sb, pt.getSentences().get(toSent - 1), trees.get(0),
                new String[] { wordTo });
        System.out.println(sb.toString());
        results.add(sb.toString());/*from w w w.  j av a2  s  .co m*/
    }
    // if no arcs then orig sentences
    if (results.isEmpty()) {
        for (Tree t : pt.getSentences()) {
            results.add(t.toString());
        }
    }
    return results;
}