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

Example 21 with Matcher

use of java.util.regex.Matcher in project lucida by claritylab.

the class CorefResolver method isAnswerTypeThing.

/*
	 * isAnswerTypeThing returns true if AnswerType is thing, else false @param
	 * stemmed question string @return boolean
	 * 
	 */
private static boolean isAnswerTypeThing(String question) {
    ArrayList<Pattern> patterns = new ArrayList<Pattern>();
    boolean f = false;
    String qn = QuestionNormalizer.normalize(question);
    String stemmed = QuestionNormalizer.stemVerbsAndNouns(qn);
    String[] tokens = new String[1];
    tokens[0] = "(what|which)";
    for (int i = 0; i < tokens.length; i++) {
        patterns.add(Pattern.compile("\\b" + tokens[i] + "\\b", Pattern.CASE_INSENSITIVE));
    }
    for (int i = 0; i < tokens.length; i++) {
        Matcher m = patterns.get(i).matcher(stemmed);
        if (m.find()) {
            f = true;
        }
    }
    return f;
}

Example 22 with Matcher

use of java.util.regex.Matcher in project lucida by claritylab.

the class EphyraTREC13To16 method main.

/**
	 * Runs Ephyra on the TREC questions.
	 * 
	 * @param args argument 1: questionfile<br>
	 * 			   [argument 2: tag=runtag (uniquely identifies the run, also
	 * 							used as output file name, if not set an
	 * 							unambiguous tag is generated automatically)]<br>
	 * 			   [argument 3: log=logfile (if not set an unambiguous file name
	 * 							is generated automatically)]<br>
	 * 			   [argument 4: lin=input_logfile (specifies a separate logfile
	 * 							that is used as a source for answers to some of
	 * 							the	questions, if not set the standard log file
	 * 							is used)]<br>
	 * 			   [argument 5: lflags=[f][l][o] (answers to these types of
	 * 							questions are loaded from the log file instead
	 * 							of querying Ephyra,	e.g. "flo" for factoid, list
	 * 							and other questions)]<br>
	 * 			   [argument 6: fp=factoid_patternfile]<br>
	 * 			   [argument 7: lp=list_patternfile]
	 */
public static void main(String[] args) {
    // enable output of status and error messages
    MsgPrinter.enableStatusMsgs(true);
    MsgPrinter.enableErrorMsgs(true);
    if (args.length < 1) {
        MsgPrinter.printUsage("java EphyraTREC13To16 questionfile " + "[tag=runtag] [log=logfile] " + "[lin=input_logfile] " + "[lflags=[f][l][o]] " + "[fp=factoid_patternfile] " + "[lp=list_patternfile]");
        System.exit(1);
    }
    // load targets
    targets = TREC13To16Parser.loadTargets(args[0]);
    for (int i = 1; i < args.length; i++) if (args[i].matches("tag=.*")) {
        // set run tag
        runTag = args[i].substring(4);
    } else if (args[i].matches("log=.*")) {
        // set log file
        logFile = args[i].substring(4);
    } else if (args[i].matches("lin=.*")) {
        // set separate input log file
        inputLogFile = args[i].substring(4);
    } else if (args[i].matches("lflags=.*")) {
        // answers for some question types are loaded from log file
        String flags = args[i].substring(7).toLowerCase();
        if (flags.contains("f"))
            factoidLog = true;
        if (flags.contains("l"))
            listLog = true;
        if (flags.contains("o"))
            otherLog = true;
    } else if (args[i].matches("fp=.*")) {
        // load factoid patterns
        factoidPatterns = TREC13To16Parser.loadPatterns(args[i].substring(3));
    } else if (args[i].matches("lp=.*")) {
        // load list patterns
        listPatterns = TREC13To16Parser.loadPatterns(args[i].substring(3));
    }
    // if run tag or log file not set, generate unambiguous names
    if (runTag == null || logFile == null) {
        String n = "";
        Matcher m = Pattern.compile("\\d++").matcher(args[0]);
        if (m.find())
            n = m.group(0);
        String date = "";
        Calendar c = new GregorianCalendar();
        date += c.get(Calendar.DAY_OF_MONTH);
        if (date.length() == 1)
            date = "0" + date;
        date = (c.get(Calendar.MONTH) + 1) + date;
        if (date.length() == 3)
            date = "0" + date;
        date = c.get(Calendar.YEAR) + date;
        if (runTag == null)
            runTag = "TREC" + n + "_" + date + "_out";
        if (logFile == null)
            logFile = "log/TREC" + n + "_" + date;
    }
    // if input log file not set, use standard log file
    if (inputLogFile == null)
        inputLogFile = logFile;
    // enable logging
    Logger.setLogfile(logFile);
    Logger.enableLogging(true);
    // run Ephyra on questions, evaluate answers if patterns available
    runAndEval();
}

Example 23 with Matcher

use of java.util.regex.Matcher in project cogtool by cogtool.

the class StandardDrawingEditor method setZoomFromUserInput.

protected void setZoomFromUserInput(String userInput) {
    // TODO: this still really isn't the right UI. Instead of validating
    //       the value entered and then fussing at the user if it's not
    //       right, we should really be looking after every key press
    //       or paste operation and only allowing legal values to be
    //       entered at all.
    // TODO: deal with localization here. Number formats are also locale
    //       specific. When we deal with that in general, it needs to be
    //       dealt with here. The major issue is the symbol used as a
    //       radix point. As of right now ZOOM_VALUE_PATTERN can be
    //       localized, but the call to parseDouble below is not locale-
    //       sensitive, and needs eventually to be replaced by something
    //       that is.
    Matcher zoomMatcher = ZOOM_VALUE_PATTERN.matcher(userInput);
    if (zoomMatcher.matches()) {
        try {
            double newZoom = Double.parseDouble(zoomMatcher.group(1)) / 100.0;
            if (newZoom < MINIMUM_ZOOM) {
                // TODO: interact with the user, indicating that the
                //       value is too small (see comment at top of this
                //       file regarding likely need for an Interaction).
                newZoom = MINIMUM_ZOOM;
            } else if (newZoom > MAXIMUM_ZOOM) {
                // TODO: interact with the user, indicating that the
                //       value is too large (see comment at top of this
                //       file regarding likely need for an Interaction).
                newZoom = MAXIMUM_ZOOM;
            }
            zoomable.performZoom(newZoom);
        } catch (NumberFormatException e) {
            throw new RcvrUIException("Number format mismatch in zoom value.");
        }
    } else {
    // TODO: interact with the user, indicating that the value is
    //       not of the expected format; throw exception when
    //       exception architecture is "completed"
    }
}

Example 24 with Matcher

use of java.util.regex.Matcher in project crate by crate.

the class ExpressionAnalyzer method getQuotedSubscriptLiteral.

@Nullable
protected static String getQuotedSubscriptLiteral(String nodeName) {
    Matcher matcher = SUBSCRIPT_SPLIT_PATTERN.matcher(nodeName);
    if (matcher.matches()) {
        StringBuilder quoted = new StringBuilder();
        String group1 = matcher.group(1);
        if (!group1.isEmpty()) {
            quoted.append("\"").append(group1).append("\"");
        } else {
            quoted.append(group1);
        }
        String group2 = matcher.group(2);
        String group3 = matcher.group(3);
        if (!group2.isEmpty() && !group3.isEmpty()) {
            quoted.append(matcher.group(2));
            quoted.append("\"").append(group3).append("\"");
        } else if (!group2.isEmpty() && group3.isEmpty()) {
            return null;
        }
        quoted.append(matcher.group(4));
        return quoted.toString();
    } else {
        return null;
    }
}

Example 25 with Matcher

use of java.util.regex.Matcher in project elasticsearch by elastic.

the class MetaDataStateFormat method findMaxStateId.

long findMaxStateId(final String prefix, Path... locations) throws IOException {
    long maxId = -1;
    for (Path dataLocation : locations) {
        final Path resolve = dataLocation.resolve(STATE_DIR_NAME);
        if (Files.exists(resolve)) {
            try (DirectoryStream<Path> stream = Files.newDirectoryStream(resolve, prefix + "*")) {
                for (Path stateFile : stream) {
                    final Matcher matcher = stateFilePattern.matcher(stateFile.getFileName().toString());
                    if (matcher.matches()) {
                        final long id = Long.parseLong(matcher.group(1));
                        maxId = Math.max(maxId, id);
                    }
                }
            }
        }
    }
    return maxId;
}