private String getMaintainers() {
final TreeMap<String, Language> list = new TreeMap<>();
for (final Language lang : Language.REAL_LANGUAGES) {
if (!lang.isVariant()) {
if (lang.getMaintainers() != null) {
list.put(messages.getString(lang.getShortName()), lang);
}
}
}
final StringBuilder maintainersInfo = new StringBuilder();
maintainersInfo.append("<table border=0 cellspacing=0 cellpadding=0>");
for (String lang : list.keySet()) {
maintainersInfo.append("<tr valign=\"top\"><td>");
maintainersInfo.append(lang);
maintainersInfo.append(":</td>");
maintainersInfo.append("<td> </td>");
maintainersInfo.append("<td>");
int i = 0;
for (Contributor contributor : list.get(lang).getMaintainers()) {
if (i > 0) {
maintainersInfo.append(", ");
if (i % 3 == 0) {
maintainersInfo.append("<br>");
}
}
maintainersInfo.append(contributor.getName());
i++;
}
maintainersInfo.append("</td></tr>");
}
maintainersInfo.append("</table>");
return maintainersInfo.toString();
}
private String dutchSpecialCase(
final String firstToken, final String secondToken, final AnalyzedTokenReadings[] tokens) {
if (!language.getShortName().equals("nl")) {
return null;
}
if (tokens.length >= 3 && firstToken.equals("'") && secondToken.matches("k|m|n|r|s|t")) {
return tokens[3].getToken();
}
return null;
}
private static Language getLanguageOrExit(final String lang) {
Language language = null;
boolean foundLanguage = false;
final List<String> supportedLanguages = new ArrayList<>();
for (final Language tmpLang : Languages.get()) {
supportedLanguages.add(tmpLang.getShortName());
if (lang.equals(tmpLang.getShortName())) {
language = tmpLang;
foundLanguage = true;
break;
}
}
if (!foundLanguage) {
System.out.println(
"Unknown language '" + lang + "'. Supported languages are: " + supportedLanguages);
exitWithUsageMessage();
}
return language;
}
private String dutchSpecialCase(
final String firstToken, final String secondToken, final AnalyzedTokenReadings[] tokens) {
if (!language.getShortName().equals("nl")) {
return null;
}
if (tokens.length >= 3
&& firstToken.equals("'")
&& DUTCH_SPECIAL_CASE.matcher(secondToken).matches()) {
return tokens[3].getToken();
}
return null;
}
private Date getLatestDate(String dateField, Language language) {
try {
String sql =
"SELECT "
+ dateField
+ " FROM feed_matches WHERE language_code = ? ORDER BY "
+ dateField
+ " DESC";
try (PreparedStatement prepSt = conn.prepareStatement(sql)) {
prepSt.setString(1, language.getShortName());
ResultSet resultSet = prepSt.executeQuery();
if (resultSet.next() && resultSet.getTimestamp(dateField) != null) {
return new Date(resultSet.getTimestamp(dateField).getTime());
}
}
} catch (Exception e) {
throw new RuntimeException("Could not get latest " + dateField + " from database", e);
}
return new Date(0);
}
@Nullable
private static MorfologikMultiSpeller getSpeller(Language language) {
if (!language.getShortName().equals(Locale.GERMAN.getLanguage())) {
throw new RuntimeException("Language is not a variant of German: " + language);
}
try {
String morfoFile = "/de/hunspell/de_" + language.getCountries()[0] + ".dict";
if (JLanguageTool.getDataBroker().resourceExists(morfoFile)) {
// spell data will not exist in LibreOffice/OpenOffice context
try (InputStream stream =
JLanguageTool.getDataBroker()
.getFromResourceDirAsStream("/de/hunspell/spelling.txt");
BufferedReader br = new BufferedReader(new InputStreamReader(stream, "utf-8"))) {
return new MorfologikMultiSpeller(morfoFile, new ExpandingReader(br), MAX_EDIT_DISTANCE);
}
} else {
return null;
}
} catch (IOException e) {
throw new RuntimeException("Could not set up morfologik spell checker", e);
}
}
private void run(Language lang) throws IOException {
File basePath = new File("/lt/git/languagetool/languagetool-language-modules");
if (!basePath.exists()) {
throw new RuntimeException("basePath does not exist: " + basePath);
}
String langCode = lang.getShortName();
File xml =
new File(
basePath,
"/"
+ langCode
+ "/src/main/resources/org/languagetool/rules/"
+ langCode
+ "/grammar.xml");
List<String> xmlLines = IOUtils.readLines(new FileReader(xml));
JLanguageTool tool = new JLanguageTool(lang);
int totalRules = 0;
for (Rule rule : tool.getAllActiveRules()) {
if (!(rule instanceof PatternRule)) {
continue;
}
PatternRule patternRule = (PatternRule) rule;
String id = patternRule.getFullId();
if (isSimple((PatternRule) rule)) {
System.err.println("Simplifying: " + id);
simplify(patternRule, xmlLines);
} else {
System.err.println("Can't simplify: " + id);
}
totalRules++;
}
System.err.println("touchedRulesCount: " + touchedRulesCount + " out of " + totalRules);
for (String xmlLine : xmlLines) {
System.out.println(xmlLine);
}
}
@Override
public final RuleMatch[] match(final AnalyzedSentence text) {
final List<RuleMatch> ruleMatches = new ArrayList<RuleMatch>();
final AnalyzedTokenReadings[] tokens = text.getTokensWithoutWhitespace();
if (tokens.length < 2) {
return toRuleMatchArray(ruleMatches);
}
int matchTokenPos = 1; // 0 = SENT_START
final String firstToken = tokens[matchTokenPos].getToken();
String secondToken = null;
String thirdToken = null;
// ignore quote characters:
if (tokens.length >= 3
&& ("'".equals(firstToken) || "\"".equals(firstToken) || "„".equals(firstToken))) {
matchTokenPos = 2;
secondToken = tokens[matchTokenPos].getToken();
}
final String firstDutchToken = dutchSpecialCase(firstToken, secondToken, tokens);
if (firstDutchToken != null) {
thirdToken = firstDutchToken;
matchTokenPos = 3;
}
String checkToken = firstToken;
if (thirdToken != null) {
checkToken = thirdToken;
} else if (secondToken != null) {
checkToken = secondToken;
}
String lastToken = tokens[tokens.length - 1].getToken();
if (lastToken.matches("[ \"'„»«“]") && tokens.length >= 2) {
// ignore trailing whitespace or quote
lastToken = tokens[tokens.length - 2].getToken();
}
boolean preventError = false;
// TODO: why do only *these* languages have that special case?
final String langCode = language.getShortName();
final boolean languageHasSpecialCases =
langCode.equals("ru")
|| langCode.equals("pl")
|| langCode.equals("uk")
|| langCode.equals("be")
|| langCode.equals(Locale.ENGLISH.getLanguage())
|| langCode.equals(Locale.ITALIAN.getLanguage())
|| langCode.equals(Locale.GERMAN.getLanguage());
if (languageHasSpecialCases) {
// fix for lists; note - this will not always work for the last point in OOo,
// as OOo might serve paragraphs in any order.
if (";".equals(lastParagraphString)
|| ";".equals(lastToken)
|| ",".equals(lastParagraphString)
|| ",".equals(lastToken)) {
preventError = true;
}
// fix for words in table (not sentences); note - this will not always work for the last point
// in OOo,
// as OOo might serve paragraphs in any order.
if (!lastToken.matches("[.?!…]")) {
preventError = true;
}
}
lastParagraphString = lastToken;
if (checkToken.length() > 0) {
final char firstChar = checkToken.charAt(0);
if (!preventError && Character.isLowerCase(firstChar)) {
final RuleMatch ruleMatch =
new RuleMatch(
this,
tokens[matchTokenPos].getStartPos(),
tokens[matchTokenPos].getStartPos() + tokens[matchTokenPos].getToken().length(),
messages.getString("incorrect_case"));
ruleMatch.setSuggestedReplacement(StringTools.uppercaseFirstChar(checkToken));
ruleMatches.add(ruleMatch);
}
}
return toRuleMatchArray(ruleMatches);
}
/**
* Get the name of the prohibit file, which lists words not to be accepted, even when the spell
* checker would accept them.
*
* @since 2.8
*/
protected String getProhibitFileName() {
return language.getShortName() + SPELLING_PROHIBIT_FILE;
}
/**
* Get the name of the spelling file, which lists words to be accepted and used for suggestions,
* even when the spell checker would not accept them.
*
* @since 2.9
*/
protected String getSpellingFileName() {
return language.getShortName() + SPELLING_FILE;
}
/**
* Get the name of the ignore file, which lists words to be accepted, even when the spell checker
* would not accept them. Unlike with {@link #getSpellingFileName()} the words in this file will
* not be used for creating suggestions for misspelled words.
*
* @since 2.7
*/
protected String getIgnoreFileName() {
return language.getShortName() + SPELLING_IGNORE_FILE;
}