Examples with Pattern java.util.regex.Pattern used on opensource projects

Example 31 with Pattern

use of java.util.regex.Pattern in project CoreNLP by stanfordnlp.

the class CorefScorer method getFinalConllScore.

public static double getFinalConllScore(String summary) {
    Pattern f1 = Pattern.compile("Coreference:.*F1: (.*)%");
    Matcher f1Matcher = f1.matcher(summary);
    double[] F1s = new double[5];
    int i = 0;
    while (f1Matcher.find()) {
        F1s[i++] = Double.parseDouble(f1Matcher.group(1));
    }
    double finalScore = (F1s[0] + F1s[1] + F1s[3]) / 3;
    return finalScore;
}

Example 32 with Pattern

use of java.util.regex.Pattern in project CoreNLP by stanfordnlp.

the class NERGUI method extract.

private void extract() {
    log.info("content type: " + editorPane.getContentType());
    if (!editorPane.getContentType().equals("text/html")) {
        DefaultStyledDocument doc = (DefaultStyledDocument) editorPane.getDocument();
        String text = null;
        try {
            text = doc.getText(0, doc.getLength());
        } catch (Exception e) {
            e.printStackTrace();
        }
        String labeledText = classifier.classifyWithInlineXML(text);
        taggedContents = labeledText;
        Set<String> tags = classifier.labels();
        String background = classifier.backgroundSymbol();
        StringBuilder tagPattern = new StringBuilder();
        for (String tag : tags) {
            if (background.equals(tag)) {
                continue;
            }
            if (tagPattern.length() > 0) {
                tagPattern.append('|');
            }
            tagPattern.append(tag);
        }
        Pattern startPattern = Pattern.compile("<(" + tagPattern + ")>");
        Pattern endPattern = Pattern.compile("</(" + tagPattern + ")>");
        String finalText = labeledText;
        Matcher m = startPattern.matcher(finalText);
        while (m.find()) {
            int start = m.start();
            finalText = m.replaceFirst("");
            m = endPattern.matcher(finalText);
            if (m.find()) {
                int end = m.start();
                String tag = m.group(1);
                finalText = m.replaceFirst("");
                AttributeSet attSet = getAttributeSet(tag);
                try {
                    String entity = finalText.substring(start, end);
                    doc.setCharacterAttributes(start, entity.length(), attSet, false);
                } catch (Exception ex) {
                    ex.printStackTrace();
                    System.exit(-1);
                }
                log.info(tag + ": " + finalText.substring(start, end));
            } else {
                log.info("Couldn't find end pattern!");
            }
            m = startPattern.matcher(finalText);
        }
        editorPane.revalidate();
        editorPane.repaint();
    } else {
        String untaggedContents = editorPane.getText();
        if (untaggedContents == null) {
            untaggedContents = "";
        }
        taggedContents = classifier.classifyWithInlineXML(untaggedContents);
        Set<String> tags = classifier.labels();
        String background = classifier.backgroundSymbol();
        StringBuilder tagPattern = new StringBuilder();
        for (String tag : tags) {
            if (background.equals(tag)) {
                continue;
            }
            if (tagPattern.length() > 0) {
                tagPattern.append('|');
            }
            tagPattern.append(tag);
        }
        Pattern startPattern = Pattern.compile("<(" + tagPattern + ")>");
        Pattern endPattern = Pattern.compile("</(" + tagPattern + ")>");
        String finalText = taggedContents;
        Matcher m = startPattern.matcher(finalText);
        while (m.find()) {
            String tag = m.group(1);
            String color = colorToHTML(tagToColorMap.get(tag));
            String newTag = "<span style=\"background-color: " + color + "; color: white\">";
            finalText = m.replaceFirst(newTag);
            int start = m.start() + newTag.length();
            Matcher m1 = endPattern.matcher(finalText);
            m1.find(m.end());
            String entity = finalText.substring(start, m1.start());
            log.info(tag + ": " + entity);
            finalText = m1.replaceFirst("</span>");
            m = startPattern.matcher(finalText);
        }
        // System.out.println(finalText);
        editorPane.setText(finalText);
        editorPane.revalidate();
        editorPane.repaint();
    // log.info(finalText);
    }
    saveTaggedAs.setEnabled(true);
}

Example 33 with Pattern

use of java.util.regex.Pattern in project CoreNLP by stanfordnlp.

the class ChineseQuantifiableEntityNormalizer method normalizeDateString.

/**
   * Normalizes date strings.
   * @param s Input date string
   * @param ctxdate Context date (usually doc_date)
   * @return Normalized Timex expression of the input date string
     */
public static String normalizeDateString(String s, String ctxdate) {
    // TODO [pengqi]: need to handle basic localization ("在七月二日到[八日]间")
    // TODO [pengqi]: need to handle literal numeral dates (usually used in events, e.g. "三一五" for 03-15)
    // TODO [pengqi]: might need to add a pattern for centuries ("上世纪90年代")?
    Pattern p;
    Matcher m;
    String ctxyear = "XXXX", ctxmonth = "XX", ctxday = "XX";
    // set up context date
    if (ctxdate != null) {
        p = Pattern.compile("^" + BASIC_YYYYMMDD_PATTERN + "$");
        m = p.matcher(ctxdate);
        if (m.find() && m.groupCount() == 3) {
            ctxyear = m.group(1);
            ctxmonth = m.group(2);
            ctxday = m.group(3);
        }
    }
    p = Pattern.compile("^" + BIRTH_DECADE_PATTERN + "$");
    m = p.matcher(s);
    if (m.find() && m.groupCount() == 1) {
        StringBuilder res = new StringBuilder();
        res.append(normalizeYear(m.group(1), ctxyear, true).substring(0, 3) + "X");
        res.append("-XX-XX");
        return res.toString();
    }
    p = Pattern.compile("^" + RELATIVE_TIME_PATTERN + "$");
    m = p.matcher(s);
    if (m.find() && m.groupCount() == 1) {
        StringBuilder res = new StringBuilder();
        res.append(ctxyear);
        res.append("-");
        res.append(ctxmonth);
        res.append("-");
        res.append(normalizeMonthOrDay(m.group(1), ctxday));
        return res.toString();
    }
    p = Pattern.compile("^" + BASIC_YYYYMMDD_PATTERN + "$");
    m = p.matcher(s);
    if (m.find() && m.groupCount() == 3) {
        StringBuilder res = new StringBuilder();
        res.append(normalizeYear(m.group(1), ctxyear));
        res.append("-");
        res.append(normalizeMonthOrDay(m.group(2), ctxmonth));
        res.append("-");
        res.append(normalizeMonthOrDay(m.group(3), ctxday));
        return res.toString();
    }
    p = Pattern.compile("^" + BASIC_MMDD_PATTERN + "$");
    m = p.matcher(s);
    if (m.find() && m.groupCount() == 2) {
        StringBuilder res = new StringBuilder();
        res.append(ctxyear);
        res.append("-");
        res.append(normalizeMonthOrDay(m.group(1), ctxmonth));
        res.append("-");
        res.append(normalizeMonthOrDay(m.group(2), ctxday));
        return res.toString();
    }
    p = Pattern.compile("^" + BASIC_DD_PATTERN + "$");
    m = p.matcher(s);
    if (m.find() && m.groupCount() == 1) {
        StringBuilder res = new StringBuilder();
        res.append(ctxyear);
        res.append("-");
        res.append(ctxmonth);
        res.append("-");
        res.append(normalizeMonthOrDay(m.group(1), ctxday));
        return res.toString();
    }
    p = Pattern.compile("^" + ENGLISH_MMDDYYYY_PATTERN + "$");
    m = p.matcher(s);
    if (m.find() && m.groupCount() == 3) {
        StringBuilder res = new StringBuilder();
        if (m.group(3) == null)
            res.append(ctxyear);
        else
            res.append(normalizeYear(m.group(3), ctxyear));
        res.append("-");
        res.append(normalizeMonthOrDay(m.group(1), ctxmonth));
        res.append("-");
        res.append(normalizeMonthOrDay(m.group(2), ctxday));
        return res.toString();
    }
    return s;
}

Example 34 with Pattern

use of java.util.regex.Pattern in project CoreNLP by stanfordnlp.

the class ISODateInstance method tokenizeDate.

/* -------------------------- Tokenization and Field Extraction -------------------------- */
//These methods are taken directly from or modified slightly from {@link DateInstance}
private void tokenizeDate(String inputDate) {
    tokens = new ArrayList<>();
    Pattern pat = Pattern.compile("[-]");
    if (inputDate == null) {
        System.out.println("Null input date");
    }
    Matcher m = pat.matcher(inputDate);
    String str = m.replaceAll(" - ");
    str = str.replaceAll(",", " ");
    PTBTokenizer<Word> tokenizer = PTBTokenizer.newPTBTokenizer(new BufferedReader(new StringReader(str)));
    while (tokenizer.hasNext()) {
        Word nextToken = tokenizer.next();
        tokens.add(nextToken.toString());
    }
    if (DEBUG) {
        System.out.println("tokens:" + tokens);
    }
}

Example 35 with Pattern

use of java.util.regex.Pattern in project CoreNLP by stanfordnlp.

the class UnitPrefix method loadPrefixes.

public static List<UnitPrefix> loadPrefixes(String filename) throws IOException {
    Pattern commaPattern = Pattern.compile("\\s*,\\s*");
    BufferedReader br = IOUtils.getBufferedFileReader(filename);
    String headerString = br.readLine();
    String[] header = commaPattern.split(headerString);
    Map<String, Integer> headerIndex = new HashMap<>();
    for (int i = 0; i < header.length; i++) {
        headerIndex.put(header[i], i);
    }
    int iName = headerIndex.get("name");
    int iPrefix = headerIndex.get("prefix");
    int iBase = headerIndex.get("base");
    int iExp = headerIndex.get("exp");
    int iSystem = headerIndex.get("system");
    String line;
    List<UnitPrefix> list = new ArrayList<>();
    while ((line = br.readLine()) != null) {
        String[] fields = commaPattern.split(line);
        double base = Double.parseDouble(fields[iBase]);
        double exp = Double.parseDouble(fields[iExp]);
        double scale = Math.pow(base, exp);
        UnitPrefix unitPrefix = new UnitPrefix(fields[iName], fields[iPrefix], scale, fields[iSystem]);
        list.add(unitPrefix);
    }
    br.close();
    return list;
}