Examples with PatternMatcherInput org.apache.oro.text.regex.PatternMatcherInput used on opensource projects

Example 16 with PatternMatcherInput

use of org.apache.oro.text.regex.PatternMatcherInput in project Lucee by lucee.

the class Perl5Util method match.

public static Array match(String strPattern, String strInput, int offset, boolean caseSensitive) throws MalformedPatternException {
    Perl5Matcher matcher = new Perl5Matcher();
    PatternMatcherInput input = new PatternMatcherInput(strInput);
    int compileOptions = caseSensitive ? 0 : Perl5Compiler.CASE_INSENSITIVE_MASK;
    compileOptions += Perl5Compiler.MULTILINE_MASK;
    if (offset < 1)
        offset = 1;
    Pattern pattern = getPattern(strPattern, compileOptions);
    Array rtn = new ArrayImpl();
    MatchResult result;
    while (matcher.contains(input, pattern)) {
        result = matcher.getMatch();
        rtn.appendEL(result.toString());
    }
    return rtn;
}

Example 17 with PatternMatcherInput

use of org.apache.oro.text.regex.PatternMatcherInput in project nutch by apache.

the class OutlinkExtractor method getOutlinks.

/**
 * Extracts <code>Outlink</code> from given plain text and adds anchor to the
 * extracted <code>Outlink</code>s
 *
 * @param plainText
 *          the plain text from wich URLs should be extracted.
 * @param anchor
 *          the anchor of the url
 *
 * @return Array of <code>Outlink</code>s within found in plainText
 */
public static Outlink[] getOutlinks(final String plainText, String anchor, Configuration conf) {
    long start = System.currentTimeMillis();
    final List<Outlink> outlinks = new ArrayList<>();
    try {
        final PatternCompiler cp = new Perl5Compiler();
        final Pattern pattern = cp.compile(URL_PATTERN, Perl5Compiler.CASE_INSENSITIVE_MASK | Perl5Compiler.READ_ONLY_MASK | Perl5Compiler.MULTILINE_MASK);
        final PatternMatcher matcher = new Perl5Matcher();
        final PatternMatcherInput input = new PatternMatcherInput(plainText);
        MatchResult result;
        String url;
        // loop the matches
        while (matcher.contains(input, pattern)) {
            // do not unnecessarily hit this limit.)
            if (System.currentTimeMillis() - start >= 60000L) {
                if (LOG.isWarnEnabled()) {
                    LOG.warn("Time limit exceeded for getOutLinks");
                }
                break;
            }
            result = matcher.getMatch();
            url = result.group(0);
            try {
                outlinks.add(new Outlink(url, anchor));
            } catch (MalformedURLException mue) {
                LOG.warn("Invalid url: '" + url + "', skipping.");
            }
        }
    } catch (Exception ex) {
        // on
        if (LOG.isErrorEnabled()) {
            LOG.error("getOutlinks", ex);
        }
    }
    final Outlink[] retval;
    // create array of the Outlinks
    if (outlinks != null && outlinks.size() > 0) {
        retval = outlinks.toArray(new Outlink[0]);
    } else {
        retval = new Outlink[0];
    }
    return retval;
}

Example 18 with PatternMatcherInput

use of org.apache.oro.text.regex.PatternMatcherInput in project jmeter by apache.

the class TestHTTPSamplersAgainstHttpMirrorServer method getPositionOfBody.

private int getPositionOfBody(String stringToCheck) {
    Perl5Matcher localMatcher = JMeterUtils.getMatcher();
    // The headers and body are divided by a blank line
    String regularExpression = "^.$";
    Pattern pattern = JMeterUtils.getPattern(regularExpression, Perl5Compiler.READ_ONLY_MASK | Perl5Compiler.CASE_INSENSITIVE_MASK | Perl5Compiler.MULTILINE_MASK);
    PatternMatcherInput input = new PatternMatcherInput(stringToCheck);
    while (localMatcher.contains(input, pattern)) {
        MatchResult match = localMatcher.getMatch();
        return match.beginOffset(0);
    }
    // No divider was found
    return -1;
}

Example 19 with PatternMatcherInput

use of org.apache.oro.text.regex.PatternMatcherInput in project jmeter by apache.

the class RenderAsRegexp method process.

private String process(String textToParse) {
    Perl5Matcher matcher = new Perl5Matcher();
    PatternMatcherInput input = new PatternMatcherInput(textToParse);
    PatternCacheLRU pcLRU = new PatternCacheLRU();
    Pattern pattern;
    try {
        pattern = pcLRU.getPattern(regexpField.getText(), Perl5Compiler.READ_ONLY_MASK);
    } catch (MalformedCachePatternException e) {
        return e.toString();
    }
    List<MatchResult> matches = new LinkedList<>();
    while (matcher.contains(input, pattern)) {
        matches.add(matcher.getMatch());
    }
    // Construct a multi-line string with all matches
    StringBuilder sb = new StringBuilder();
    final int size = matches.size();
    sb.append("Match count: ").append(size).append("\n");
    for (int j = 0; j < size; j++) {
        MatchResult mr = matches.get(j);
        final int groups = mr.groups();
        for (int i = 0; i < groups; i++) {
            sb.append("Match[").append(j + 1).append("][").append(i).append("]=").append(mr.group(i)).append("\n");
        }
    }
    return sb.toString();
}

Example 20 with PatternMatcherInput

use of org.apache.oro.text.regex.PatternMatcherInput in project tdi-studio-se by Talend.

the class WebServiceExpressionParser method parseInTableEntryLocations.

public Map<String, String> parseInTableEntryLocations(String expression) {
    // resultSet.clear();
    Map<String, String> map = new HashMap<String, String>();
    if (expression != null) {
        matcher.setMultiline(true);
        if (patternMatcherInput == null) {
            patternMatcherInput = new PatternMatcherInput(expression);
        } else {
            patternMatcherInput.setInput(expression);
        }
        recompilePatternIfNecessary(locationPattern);
        while (matcher.contains(patternMatcherInput, pattern)) {
            MatchResult matchResult = matcher.getMatch();
            map.put(matchResult.group(2), matchResult.group(1));
        // resultSet.add(map);
        }
    }
    // .toArray(new TableEntryLocation[0]);
    return map;
}