Pattern[] matchRegexToDocument(String regex) {
regex = interpolateTextMarksIntoRegex(regex);
ArrayList<Pattern> patterns = new ArrayList<Pattern>();
String fullText = util.loadTextFromId(url.toExternalForm()).replaceAll("\\s+", " ");
tempRegex = fullText;
String negatedPrecedingString = getNegatedPrecedingCharacters(regex);
String negatedFollowingString = getNegatedFollowingChars(regex);
int precedingIndexAdjustment = negatedPrecedingString.length();
int followingIndexAdjustment = negatedFollowingString.length();
regex = util.substituteDiacritics(regex);
tempRegex += "<br><br>" + regex;
Pattern pattern =
Pattern.compile(
regex,
Pattern.CASE_INSENSITIVE | Pattern.UNICODE_CASE | Pattern.UNIX_LINES | Pattern.DOTALL);
Matcher matcher = pattern.matcher(fullText);
while (matcher.find()) {
int startIndex = matcher.start() - precedingIndexAdjustment;
int endIndex = matcher.end() + followingIndexAdjustment;
if (startIndex < 0) {
startIndex = 0;
}
if (endIndex > fullText.length()) {
endIndex = fullText.length();
}
String found = fullText.substring(startIndex, endIndex);
found = anchorAssertionAtStart(found, precedingIndexAdjustment);
found = anchorAssertionAtEnd(found, followingIndexAdjustment);
if ((precedingIndexAdjustment == 0
|| !negatedPrecedingString.equals(found.substring(0, precedingIndexAdjustment + 1)))
&& (followingIndexAdjustment == 0
|| !negatedFollowingString.equals(
found.substring(found.length() - followingIndexAdjustment)))) {
found = found.replaceAll("([()\\[\\]{}\\.])", ".");
found = found.replaceAll("\\s{2,}", "\\\\b");
String[] foundBits = found.split("[\\d]+\\.");
for (int i = 0; i < foundBits.length; i++) {
String fbit = foundBits[i].trim();
fbit = fbit.replaceAll("\\s", "\\\\s+");
highlightWords.add(fbit);
Pattern foundPattern =
Pattern.compile(
fbit,
Pattern.CASE_INSENSITIVE
| Pattern.UNICODE_CASE
| Pattern.UNIX_LINES
| Pattern.DOTALL);
patterns.add(foundPattern);
}
}
}
Pattern[] arrPatterns = new Pattern[patterns.size()];
return patterns.toArray(arrPatterns);
}
String getKWIC() {
StringBuilder html = new StringBuilder();
try {
List<String> kwix =
util.highlightMatches(util.loadTextFromId(url.toExternalForm()), highlightTerms);
html.append("<tr class=\"result-text\"><td class=\"kwic\" colspan=\"7\">");
for (String kwic : kwix) {
html.append(
kwic.replaceAll(
"\\s*ⓐ\\s*",
"")); // TODO: why is this character sneaking through when the user does a regex
// word-boundary (\b) search?
html.append("<br/>\n");
}
html.append("</td></tr>");
} catch (Exception e) {
// TODO: Need to do something sensible here with regard to highlighting
logger.error("Highlightling failure", e);
}
return html.toString();
}