private String cleanString(String string) {
string.trim();
String patt = "'";
Pattern r = Pattern.compile(patt);
StringBuilder sb = new StringBuilder(string);
Matcher m = r.matcher(sb);
int start = 0;
String replString = "\\'";
while (m.find(start)) {
sb.replace(m.start(), m.end(), "\\'");
start = m.start() + replString.length();
}
patt = "\n";
r = Pattern.compile(patt);
m = r.matcher(sb);
start = 0;
replString = "\' + char(13) + char(10) + \'";
while (m.find(start)) {
sb.replace(m.start(), m.end(), replString);
start = m.start() + replString.length();
}
return sb.toString();
}
/**
* Applies the size filter
*
* @param sb message to filter
* @param front true if apply front sizer, false for back sizer
* @param player whether or not it is for a player
*/
private void filterSizer(StringBuilder sb, boolean front, boolean player) {
Pattern regex = front ? EXPAND_FRONT : EXPAND_BACK;
Matcher match = regex.matcher(sb);
int size = sb.length();
while (match.find()) {
int playerSize = Integer.parseInt(match.group(2));
int consoleSize = Integer.parseInt(match.group(3));
String string = match.group(1);
if (player) {
sb.replace(
match.start() + sb.length() - size,
match.end(),
(TextSizer.measureString(string) > playerSize - 2
? string
: TextSizer.expand(string, playerSize, front)));
} else {
sb.replace(
match.start() + sb.length() - size,
match.end(),
(string.length() > consoleSize
? string
: TextSizer.expandConsole(string, consoleSize, front)));
}
}
}
/** Index initialization */
private void initialIndex() {
index = new HashMap<String, ArrayList<IndexWord>>();
int indexSpeech = 0;
for (Speech speech : subtitle) {
String text = speech.content;
Matcher matcherTag = Pattern.compile("<[^>]*>").matcher(text);
while (matcherTag.find()) {
String fill = repeatChar(' ', matcherTag.group().length());
text =
text.substring(0, matcherTag.start())
+ fill
+ text.substring(matcherTag.end(), text.length());
}
Pattern patternWord = Pattern.compile("[\\p{L}']{3,}");
Matcher matcherWord = patternWord.matcher(text);
while (matcherWord.find()) {
String word = matcherWord.group();
String stem = Stemmator.stemmingWord(word.toLowerCase(), language);
if (!index.containsKey(stem)) index.put(stem, new ArrayList<IndexWord>());
IndexWord indexWord =
new IndexWord(word, stem, indexSpeech, matcherWord.start(), matcherWord.end());
index.get(stem).add(indexWord);
}
indexSpeech++;
}
}
// priorityHelper parses the string and sets the Task's importance
private static boolean priorityHelper(Task task) {
String inputText = task.getValue(Task.TITLE);
String[] importanceStrings = {
"()((^|[^\\w!])!+|(^|[^\\w!])!\\d)($|[^\\w!])",
"()(?i)((\\s?bang){1,})$",
"(?i)(\\spriority\\s?(\\d)$)",
"(?i)(\\sbang\\s?(\\d)$)",
"(?i)()(\\shigh(est)?|\\slow(est)?|\\stop|\\sleast) ?priority$"
};
boolean result = false;
for (String importanceString : importanceStrings) {
Pattern importancePattern = Pattern.compile(importanceString);
while (true) {
Matcher m = importancePattern.matcher(inputText);
if (m.find()) {
result = true;
task.setValue(Task.IMPORTANCE, strToPriority(m.group(2).trim()));
int start = m.start() == 0 ? 0 : m.start() + 1;
inputText = inputText.substring(0, start) + inputText.substring(m.end());
} else {
break;
}
}
}
task.setValue(Task.TITLE, inputText.trim());
return result;
}
private Multiplier multiplier(String expression, int offset) {
Matcher matcher = MULTIPLIER_PREFIX_PATTERN.matcher(expression);
if (!matcher.find()) {
return null;
}
if (matcher.start() > 0) {
fail(offset + matcher.start(), "illegal multiplier position");
}
Matcher digitMatcher = DIGIT_PATTERN.matcher(expression);
if (!digitMatcher.find()) {
return null;
}
String digitStr = expression.substring(0, digitMatcher.end());
int number = 0;
try {
number = Integer.parseInt(digitStr);
} catch (NumberFormatException e) {
fail(offset, e);
}
if (number <= 0) {
fail(offset, "illegal 0 multiplier");
}
String subexpression = expression.substring(matcher.end(), expression.length() - 1);
return new Multiplier(number, subexpression, matcher.end());
}
public boolean addImages(Context context, Spannable spannable) {
Pattern refImg = Pattern.compile("\\Q[img src=\\E([a-zA-Z0-9_]+?)\\Q/]\\E");
boolean hasChanges = false;
Matcher matcher = refImg.matcher(spannable);
while (matcher.find()) {
boolean set = true;
for (ImageSpan span : spannable.getSpans(matcher.start(), matcher.end(), ImageSpan.class)) {
if (spannable.getSpanStart(span) >= matcher.start()
&& spannable.getSpanEnd(span) <= matcher.end()) {
spannable.removeSpan(span);
} else {
set = false;
break;
}
}
String resname = spannable.subSequence(matcher.start(1), matcher.end(1)).toString().trim();
int id = context.getResources().getIdentifier(resname, "drawable", context.getPackageName());
Drawable icon = context.getResources().getDrawable(id); // ,this.getTheme());
icon.setBounds(0, 0, tv_test1.getLineHeight(), tv_test1.getLineHeight());
if (set) {
hasChanges = true;
spannable.setSpan(
new ImageSpan(icon, ImageSpan.ALIGN_BASELINE),
matcher.start(),
matcher.end(),
Spannable.SPAN_EXCLUSIVE_EXCLUSIVE);
}
}
return hasChanges;
}
/** 对spanableString进行正则判断，如果符合要求，则以表情图片代替 */
public static void dealExpression(
Context context, SpannableString spannableString, Pattern patten, int start)
throws Exception {
Matcher matcher = patten.matcher(spannableString);
while (matcher.find()) {
String key = matcher.group();
Log.d("Key", key);
if (matcher.start() < start) {
continue;
}
Field field =
R.drawable.class.getDeclaredField(
"emoji_" + key.substring(key.indexOf("]") + 1, key.lastIndexOf("[")));
int resId = Integer.parseInt(field.get(null).toString());
if (resId != 0) {
Bitmap bitmap = BitmapFactory.decodeResource(context.getResources(), resId);
ImageSpan imageSpan = new ImageSpan(bitmap);
int end = matcher.start() + key.length();
spannableString.setSpan(
imageSpan, matcher.start(), end, Spannable.SPAN_INCLUSIVE_EXCLUSIVE);
if (end < spannableString.length()) {
dealExpression(context, spannableString, patten, end);
}
break;
}
}
}
/**
* replace existing spannable with smiles
*
* @param context
* @param spannable
* @return
*/
public static boolean addSmiles(Context context, Spannable spannable) {
boolean hasChanges = false;
for (Entry<Pattern, Integer> entry : emoticons.entrySet()) {
Matcher matcher = entry.getKey().matcher(spannable);
while (matcher.find()) {
boolean set = true;
for (ImageSpan span : spannable.getSpans(matcher.start(), matcher.end(), ImageSpan.class))
if (spannable.getSpanStart(span) >= matcher.start()
&& spannable.getSpanEnd(span) <= matcher.end()) spannable.removeSpan(span);
else {
set = false;
break;
}
if (set) {
hasChanges = true;
spannable.setSpan(
new ImageSpan(context, entry.getValue()),
matcher.start(),
matcher.end(),
Spannable.SPAN_EXCLUSIVE_EXCLUSIVE);
}
}
}
return hasChanges;
}
protected int continueSearch(Matcher matcher, int pos, boolean forward) {
boolean hasNext = false;
int start = 0, end = 0;
if (!forward) {
while (matcher.find()) {
hasNext = true;
start = matcher.start();
end = matcher.end();
}
} else {
hasNext = matcher.find();
if (!hasNext) {
return -1;
}
start = matcher.start();
end = matcher.end();
}
if (hasNext) {
Document doc = getDocument();
getCaret().setDot(pos + end);
getCaret().moveDot(pos + start);
getCaret().setSelectionVisible(true);
return pos + start;
}
return -1;
}
/**
* 关于微博的文本处理： @、##、http://t.cn/....
*
* @param tv
*/
public static void textFormat(TextView tv) {
Spannable sp = (Spannable) tv.getText();
String text = tv.getText().toString();
Matcher m1 = Pattern.compile("http://t.cn/[a-zA-Z0-9]+").matcher(text);
Matcher m2 = Pattern.compile("@[^:||^：||\\s]+").matcher(text);
Matcher m3 = Pattern.compile("#[^#]+#").matcher(text);
// 下面使用正则表达式
while (m1.find()) {
int start, end;
start = m1.start();
end = m1.end();
sp.setSpan(
new ForegroundColorSpan(0xff444444), start, end, Spannable.SPAN_EXCLUSIVE_EXCLUSIVE);
}
while (m2.find()) {
int start, end;
start = m2.start();
end = m2.end();
sp.setSpan(
new ForegroundColorSpan(0xff444444), start, end, Spannable.SPAN_EXCLUSIVE_EXCLUSIVE);
}
while (m3.find()) {
int start, end;
start = m3.start();
end = m3.end();
sp.setSpan(
new ForegroundColorSpan(0xff444444), start, end, Spannable.SPAN_EXCLUSIVE_EXCLUSIVE);
}
}
@Test
public void testCurrentVersionStructureAgainstGold() throws Throwable {
int latestVersion = CatalogVersions.getCurrentVersion().getSchemaVersion();
String[] lastKnown = UpgradeTestUtils.getGoldVersion(latestVersion);
String[] current = generateCatalog();
Pattern thanksHibernate =
Pattern.compile("(alter table )(\\w+)( add index )(\\w+)(.*add constraint )(\\w+)(.*)");
assertEquals("number of stmts in schema ddl", lastKnown.length, current.length - 1);
for (int i = 0; i < current.length - 1; i++) {
if (!lastKnown[i].equals(current[i])) {
// accept differences in index, constraint names
if (lastKnown[i].startsWith("alter table") && current[i].startsWith("alter table")) {
Matcher last = thanksHibernate.matcher(lastKnown[i]);
Matcher now = thanksHibernate.matcher(current[i]);
if (last.matches() && now.matches()) {
String knownConstraint = lastKnown[i].substring(last.start(6), last.end(6));
String knownIndex = lastKnown[i].substring(last.start(4), last.end(4));
StringBuilder buf = new StringBuilder();
buf.append(current[i].substring(0, now.start(4)));
buf.append(knownIndex);
buf.append(current[i].substring(now.end(4), now.start(6)));
buf.append(knownConstraint);
buf.append(current[i].substring(now.end(6)));
String munged = buf.toString();
if (lastKnown[i].equals(munged)) continue;
}
}
assertEquals("schema ddl stmt " + i, lastKnown[i], current[i]);
}
}
}
public String replaceTokens(String text, List<String> values) {
StringBuilder output = new StringBuilder();
Matcher tokenMatcher = tokenPattern.matcher(text);
int cursor = 0;
while (tokenMatcher.find()) {
// A token is defined as a sequence of the format "{...}".
// A key is defined as the content between the brackets.
int tokenStart = tokenMatcher.start();
int tokenEnd = tokenMatcher.end();
int keyStart = tokenMatcher.start(1);
int keyEnd = tokenMatcher.end(1);
output.append(text.substring(cursor, tokenStart));
String token = text.substring(tokenStart, tokenEnd);
String key = text.substring(keyStart, keyEnd);
Integer index = Integer.parseInt(key);
if (values.size() >= index) {
String value = values.get(index);
output.append(value);
} else {
output.append(token);
}
cursor = tokenEnd;
}
output.append(text.substring(cursor));
return output.toString();
}
public static Spannable getSmiledText(Context context, CharSequence text) {
Spannable spannable = spannableFactory.newSpannable(text);
for (Entry entry : ANDROID_EMOTICONS.entrySet()) {
Matcher matcher = ((Pattern) entry.getKey()).matcher(spannable);
while (matcher.find()) {
for (Object obj :
(ImageSpan[]) spannable.getSpans(matcher.start(), matcher.end(), ImageSpan.class)) {
if (spannable.getSpanStart(obj) < matcher.start()
|| spannable.getSpanEnd(obj) > matcher.end()) {
Object obj2 = null;
break;
}
spannable.removeSpan(obj);
}
int i = 1;
if (obj2 != null) {
spannable.setSpan(
new ImageSpan(context, ((Integer) entry.getValue()).intValue()),
matcher.start(),
matcher.end(),
33);
}
}
}
return spannable;
}
/**
* @param desc
* @return
*/
private String extractClassification(String desc) {
String result = desc;
Matcher mat = classificationPattern.matcher(desc);
boolean matched = false;
if (mat.find()) {
matched = true;
String category = mat.group(2);
result = desc.substring(0, mat.start()) + desc.substring(mat.end());
if (debug) System.out.println(" Category: " + category);
}
mat = Pattern.compile(",? ?(classified .*)").matcher(desc);
if (mat.find()) {
// strip out the excess text regardless
// XXX overwriting previous result.
result = desc.substring(0, mat.start()) + desc.substring(mat.end());
if (!matched) {
// if we didn't previously match a sub-set of the clause, record the category
String category = mat.group(1);
if (debug) System.out.println(" Category: " + category);
}
}
mat = Pattern.compile("text too brief to classify,? ?").matcher(result);
if (mat.find()) {
result = result.substring(0, mat.start()) + result.substring(mat.end());
if (debug) System.out.println(" Category: text too brief to classify");
}
return result;
}
/**
* @param desc
* @return
*/
private String extractContents(String desc) {
Pattern p = Pattern.compile("((?:an?|the) (.*?)(?: of (.*))?)(?:,|\\s*located)");
Pattern p2 = Pattern.compile("((incomplete .*?)(?: of (.*))?)(?:,|\\s*located)");
String result = desc;
Matcher mat = p.matcher(desc);
Matcher mat2 = p2.matcher(desc);
mat = mat.find() ? mat : mat2.find() ? mat2 : null;
if (mat != null) {
String matchedText = mat.group(1);
int start = mat.start();
int end =
(matchedText.indexOf("located") >= 0)
? end = mat.start(1) + matchedText.indexOf("located")
: mat.end(1);
String contents = desc.substring(mat.start(1), end);
result = desc.substring(0, start) + desc.substring(end);
contents = StringUtils.trimToEmpty(contents);
if (contents.lastIndexOf(",") == contents.length() - 1) {
contents = contents.substring(0, contents.length() - 1);
}
ms.setContents(contents);
LOGGER.debug(" Contents: " + contents);
}
return result;
}
/** recursive method that finds the matches. */
private void findMatch() {
int startLine;
int startColumn;
int endLine;
int endColumn;
boolean foundMatch;
boolean ignore = false;
foundMatch = mMatcher.find();
if (!foundMatch && !mIllegalPattern && (mMatchCount == 0)) {
logMessage(0);
} else if (foundMatch) {
startLine = (mCharacters.get(mMatcher.start()))[0].intValue();
startColumn = (mCharacters.get(mMatcher.start()))[1].intValue();
endLine = (mCharacters.get(mMatcher.end() - 1))[0].intValue();
endColumn = (mCharacters.get(mMatcher.end() - 1))[1].intValue();
if (mIgnoreComments) {
final FileContents theFileContents = getFileContents();
ignore =
theFileContents.hasIntersectionWithComment(startLine, startColumn, endLine, endColumn);
}
if (!ignore) {
mMatchCount++;
if (mIllegalPattern || (mCheckForDuplicates && ((mMatchCount - 1) > mDuplicateLimit))) {
mErrorCount++;
logMessage(startLine);
}
}
if ((mErrorCount < mErrorLimit) && (ignore || mIllegalPattern || mCheckForDuplicates)) {
findMatch();
}
}
}
private static String filter(String str) {
if (str.length() > 63) {
throw new IllegalArgumentException(
str
+ " is too long to be a "
+ "valid PostgreSQL name. By default names must be shorter "
+ "than 64, but it has "
+ str.length()
+ " characters");
}
Pattern quotesPattern = Pattern.compile("(\"+)");
Matcher matcher = quotesPattern.matcher(str);
while (matcher.find()) {
if (((matcher.end() - matcher.start()) & 1) == 1) { // lenght is uneven
throw new IllegalArgumentException(
"The name '"
+ str
+ "' is"
+ "illegal because contains an open quote at "
+ matcher.start());
}
}
return str;
}
private Pattern toPattern(final String pattern) {
final StringBuilder sb = new StringBuilder();
sb.append('^');
final Matcher m = SPLIT_PATTERN.matcher(pattern);
int lastEnd = 0;
while (m.find()) {
if (lastEnd < m.start()) {
sb.append(Pattern.quote(pattern.substring(lastEnd, m.start())));
}
final String matched = pattern.substring(m.start(), m.end());
if ("*".equals(matched)) {
sb.append(".*");
} else if ("?".equals(matched)) {
sb.append(".");
} else {
throw new IllegalStateException("Wildcard character does not match * nor ?");
}
lastEnd = m.end();
}
if (lastEnd < pattern.length()) {
sb.append(Pattern.compile(pattern.substring(lastEnd)));
}
sb.append('$');
return Pattern.compile(sb.toString());
}
private static String processHtml(final String source, StringBuilder errorMessages) {
if (M_evilTags == null) init();
// normalize all variants of the "<br>" HTML tag to be "<br />\n"
// TODO call a method to do this in each process routine
String Html = M_patternTagBr.matcher(source).replaceAll("<br />");
// process text and tags
StringBuilder buf = new StringBuilder();
if (Html != null) {
try {
int start = 0;
Matcher m = M_patternTag.matcher(Html);
// if there are no tags, return as is
if (!m.find()) return Html;
m.reset(Html);
// if there are tags, make sure they are safe
while (m.find()) {
// append text that isn't part of a tag
if (m.start() > start) buf.append(Html.substring(start, m.start()));
start = m.end();
buf.append(checkTag(m.group(), errorMessages));
}
// tail
if (Html.length() > start) buf.append(Html.substring((start)));
} catch (Exception e) {
M_log.warn("FormattedText.processEscapedHtml M_patternTag.matcher(Html):", e);
}
}
return new String(buf.toString());
}
public static List getNcMLElements(String path, Document doc) {
// XPath doesn't support default namespaces, so we add nc as a prefix for the tags within the
// namespace!!!
if (!path.startsWith(NS_PREFIX_ON_TAG) && !path.startsWith("/")) path = NS_PREFIX_ON_TAG + path;
Pattern pattern = Pattern.compile("/\\w");
Matcher matcher = pattern.matcher(path);
StringBuilder sb = new StringBuilder();
int currentChar = 0;
while (matcher.find()) {
sb.append(path.substring(currentChar, matcher.start() - currentChar + 1));
if (!sb.toString().endsWith("/")) sb.append("/");
sb.append(NS_PREFIX_ON_TAG);
currentChar = matcher.start() + 1;
}
sb.append(path.substring(currentChar, path.length()));
XPath xpath;
try {
xpath = XPath.newInstance(sb.toString());
xpath.addNamespace(NS_PREFIX, doc.getRootElement().getNamespaceURI());
return xpath.selectNodes(doc);
} catch (JDOMException e) {
e.printStackTrace();
}
return null;
}
public String documentedJson(String json) {
JSONReaderImpl jsonReader = new JSONReaderImpl();
JsonKeyFinder keyFinder = new JsonKeyFinder(jsonReader);
jsonReader.setContentHandler(keyFinder);
try {
jsonReader.parse(json);
} catch (SAJException e) {
// TODO Auto-generated catch block
e.printStackTrace();
}
StringBuffer documentedJson = new StringBuffer();
Pattern pattern = Pattern.compile("\\\"([a-zA-Z_]+)\\\"\\:");
Matcher matcher = pattern.matcher(json);
int afterLastMatch = 0;
while (matcher.find()) {
String dotNotatedKey =
keyFinder.getKeys().get(matcher.start() + matcher.group(0).length() - 1);
String keyDoc = docReference.getDocumentationForKey(dotNotatedKey);
// after last match to beginning of this one:
documentedJson.append(json.substring(afterLastMatch, matcher.start()));
// documented key
if (keyDoc != null) {
documentedJson.append(surroundWith.getDocumentedKey(matcher.group(0), keyDoc));
} else documentedJson.append(matcher.group(0));
afterLastMatch = matcher.start() + matcher.group(0).length();
}
documentedJson.append(json.substring(afterLastMatch));
return documentedJson.toString();
}
private void replace(Reader in, Writer out, boolean refs) throws IOException {
final String template = IoUtils.read(in);
final Matcher matcher = refStart.matcher(template);
int matchPos = 0;
int appendPos = 0;
while (matcher.find(matchPos)) {
final String name = matcher.group(1);
if (!snippets.containsKey(name)) {
throw new IllegalArgumentException("Snippet '" + name + "' not defined.");
}
if (refs) {
out.write(template.substring(appendPos, matcher.start()));
matchPos = appendPos = matcher.end();
} else {
out.write(template.substring(appendPos, matcher.end()));
appendPos = template.indexOf(refEnd, matcher.end());
if (appendPos < 0) {
throw new IllegalArgumentException(
"No refEnd marker found for refStart '"
+ template.substring(matcher.start(), matcher.end())
+ "'");
}
matchPos = appendPos + refEnd.length();
}
out.write(prefix);
out.write(snippets.get(name));
out.write(postfix);
}
out.write(template.substring(appendPos));
}
private String addDataRights(String content, String classification, Artifact artifact) {
String toReturn = content;
PageOrientation orientation = WordRendererUtil.getPageOrientation(artifact);
DataRightInput request = new DataRightInput();
request.addData(artifact.getGuid(), classification, orientation, 0);
DataRightProvider provider = new DataRightProviderImpl();
DataRightResult dataRights = provider.getDataRights(request);
String footer = dataRights.getContent(artifact.getGuid(), orientation);
Matcher startFtr = START_PATTERN.matcher(footer);
Matcher endFtr = END_PATTERN.matcher(footer);
if (startFtr.find() && endFtr.find()) {
ChangeSet ftrCs = new ChangeSet(footer);
ftrCs.delete(0, startFtr.end());
ftrCs.delete(endFtr.start(), footer.length());
footer = ftrCs.applyChangesToSelf().toString();
}
startFtr.reset(content);
endFtr.reset(content);
ChangeSet cs = new ChangeSet(content);
while (startFtr.find()) {
if (endFtr.find()) {
cs.replace(startFtr.end(), endFtr.start(), footer);
}
}
toReturn = cs.applyChangesToSelf().toString();
return toReturn;
}
public Token getWordAt(int offs, Pattern p) {
Token word = null;
try {
Element line = getParagraphElement(offs);
if (line == null) {
return word;
}
int lineStart = line.getStartOffset();
int lineEnd = Math.min(line.getEndOffset(), getLength());
Segment seg = new Segment();
getText(lineStart, lineEnd - lineStart, seg);
if (seg.count > 0) {
// we need to get the word using the words pattern p
Matcher m = p.matcher(seg);
int o = offs - lineStart;
while (m.find()) {
if (m.start() <= o && o <= m.end()) {
word = new Token(TokenType.DEFAULT, m.start() + lineStart, m.end() - m.start());
break;
}
}
}
} catch (BadLocationException ex) {
Logger.getLogger(SyntaxDocument.class.getName()).log(Level.SEVERE, null, ex);
} finally {
return word;
}
}
public static void main(String[] args) throws PatternSyntaxException {
Scanner in = new Scanner(System.in);
System.out.println("Enter pattern: ");
String patternString = in.nextLine();
Pattern pattern = Pattern.compile(patternString);
while (true) {
System.out.println("Enter string to match: ");
String input = in.nextLine();
if (input == null || input.equals("")) return;
Matcher matcher = pattern.matcher(input);
if (matcher.matches()) {
System.out.println("Match");
int g = matcher.groupCount();
if (g > 0) {
for (int i = 0; i < input.length(); i++) {
// Print any empty groups
for (int j = 1; j <= g; j++)
if (i == matcher.start(j) && i == matcher.end(j)) System.out.print("()");
// Print ( for non-empty groups starting here
for (int j = 1; j <= g; j++)
if (i == matcher.start(j) && i != matcher.end(j)) System.out.print('(');
System.out.print(input.charAt(i));
// Print ) for non-empty groups ending here
for (int j = 1; j <= g; j++)
if (i + 1 != matcher.start(j) && i + 1 == matcher.end(j)) System.out.print(')');
}
System.out.println();
}
} else System.out.println("No match");
}
}
private void splitHead(String intro) {
Pattern pattern = Pattern.compile("^[0-9]{0,2}(\\. ){0,1}" + intro + "$", Pattern.MULTILINE);
Matcher matcher = pattern.matcher(fullText);
if (matcher.find()) {
if (debug) logger.info("Found " + intro + " at " + matcher.end());
head = fullText.substring(0, matcher.start());
body = fullText.substring(matcher.start());
} else {
// try "...." after Abstract
if (debug) logger.info("Trying to find abstract");
Pattern abstractPattern = Pattern.compile("\\s[0-9]*Abstract\\s");
Matcher abstractMatcher = abstractPattern.matcher(fullText);
int abstractOffSet = 0;
if (abstractMatcher.find()) {
if (debug) logger.info("Found Abstract");
abstractOffSet = abstractMatcher.end();
Pattern pointPattern = Pattern.compile("\\.{4,}");
Matcher pointMatcher = pointPattern.matcher(fullText);
while (pointMatcher.find()) {
if (pointMatcher.end() > abstractOffSet) {
head = fullText.substring(0, pointMatcher.end());
body = fullText.substring(pointMatcher.end());
return;
}
}
head = fullText.substring(0, abstractMatcher.start());
body = fullText.substring(abstractMatcher.start());
}
// Apparently abstract wasn't divided by points
splitByHeading();
}
}
public void prepareRequireSearch(final String file) {
// if an extension is specified, try more targetted searches
if (file.lastIndexOf('.') > file.lastIndexOf('/')) {
Matcher matcher = null;
if ((matcher = sourcePattern.matcher(file)).find()) {
// source extensions
suffixType = SuffixType.Source;
// trim extension to try other options
searchFile = file.substring(0, matcher.start());
} else if ((matcher = extensionPattern.matcher(file)).find()) {
// extension extensions
suffixType = SuffixType.Extension;
// trim extension to try other options
searchFile = file.substring(0, matcher.start());
} else {
// unknown extension, fall back to search with extensions
suffixType = SuffixType.Both;
searchFile = file;
}
} else {
// try all extensions
suffixType = SuffixType.Both;
searchFile = file;
}
}
/**
* Method to split a text by headings. As of now we assume a Heading has a leading number followed
* by a whitespace character and some text beginning with upper case letters, such as "3 Related
* Work"
*/
private void splitByHeading() {
Pattern pattern = Pattern.compile("^[0-9]*\\s[A-Z].*", Pattern.MULTILINE);
Matcher matcher;
int add = 0;
// try to find headings after abstract
Pattern abstractPattern = Pattern.compile("^(Abstract).{0,5}$", Pattern.MULTILINE);
Matcher abstractMatcher = abstractPattern.matcher(body);
if (abstractMatcher.find()) {
add = abstractMatcher.end();
matcher = pattern.matcher(body.substring(abstractMatcher.end()));
} else {
matcher = pattern.matcher(body);
}
if (matcher.find()) {
// found at least once
if (debug)
logger.info(
"Splitting by heading at "
+ add
+ matcher.start()
+ " which is the heading: "
+ matcher.group());
head = body.substring(0, add + matcher.start());
body = body.substring(add + matcher.start());
}
}
/** Returns the places of possible breaks between sentences. */
private static List<BreakPosition> getBreaks(String paragraph, Rule rule) {
List<BreakPosition> res = new ArrayList<BreakPosition>();
Matcher bbm = null;
if (rule.getBeforebreak() != null) bbm = rule.getCompiledBeforebreak().matcher(paragraph);
Matcher abm = null;
if (rule.getAfterbreak() != null) abm = rule.getCompiledAfterbreak().matcher(paragraph);
if (bbm == null && abm == null) return res;
if (abm != null) if (!abm.find()) return res;
if (bbm == null) bbm = DEFAULT_BEFOREBREAK_PATTERN.matcher(paragraph);
while (bbm.find()) {
int bbe = bbm.end();
if (abm == null) res.add(new BreakPosition(bbe, rule));
else {
int abs = abm.start();
while (abs < bbe) {
boolean found = abm.find();
if (!found) return res;
abs = abm.start();
}
if (abs == bbe) res.add(new BreakPosition(bbe, rule));
}
}
return res;
}
public static String interpret(String source) {
StringBuilder result = new StringBuilder();
Matcher matcher = PATTERN_VARIABLE.matcher(source);
int current = 0;
while (matcher.find()) {
if (matcher.group(1) == null) {
String var = matcher.group(3);
String value = System.getProperty(var);
if (value != null) {
result.append(source, current, matcher.start());
result.append(value);
current = matcher.end();
}
} else {
result.append(source, current, matcher.start());
current = matcher.start(2);
}
}
if (current < source.length()) {
result.append(source, current, source.length());
}
return result.toString();
}
