danielnaber

package org.languagetool.dev;

import com.google.common.io.Files;
import org.languagetool.languagemodel.LuceneLanguageModel;

import java.io.File;
import java.io.IOException;
import java.nio.charset.StandardCharsets;
import java.nio.file.Paths;
import java.util.Arrays;

danielnaber

  public static void main(String[] args) throws IOException {
    Dictionary dictionary = Dictionary.read(JLanguageTool.getDataBroker().getFromResourceDirAsUrl("/de/german.dict"));
    DictionaryLookup dl = new DictionaryLookup(dictionary);
    Map<String, List<String>> res = new HashMap<>();
    for (WordData wd : dl) {
      String word = wd.getWord().toString();
      String base = wd.getStem().toString();
      String tag = wd.getTag().toString();
      if (tag.startsWith("ADJ:")) {
        if (tag.matches("ADJ:AKK:PLU:...:GRU:DEF")) {

danielnaber

fälschlich akzeptierte Wörter
lfdNr Wort Wort-ohne-er Wort-mit-erz

1 Abbaufelder Abbaufeld Abbaufelderz
2 Abbildungsfehler Abbildungsfehl Abbildungsfehlerz
3 Abblendlichter Abblendlicht Abblendlichterz
4 Abwehrspieler Abwehrspiel Abwehrspielerz
5 Abziehbilder Abziehbild Abziehbilderz
6 Achtender Achtend Achtenderz
7 Achter Acht Achterz

danielnaber

  public class SpecialCaseFinder {
    public static void main(String[] args) throws IOException {
      List<String> lines = Files.readAllLines(Paths.get("/home/dnaber/lt/git/german-pos-dict/x"));
      GermanSpellerRule speller = new GermanSpellerRule(JLanguageTool.getMessageBundle(), new GermanyGerman());
      int i = 1;
      for (String line : lines) {
        if (line.matches("[A-ZÖÄÜ].*") && line.endsWith("er") && !speller.isMisspelled(line)) {
          String shortened = line.replaceAll("er$", "");
          if (!speller.isMisspelled(shortened)) {
            String generated = shortened + "erz";

danielnaber

diff --git a/languagetool-language-modules/nl/src/main/java/org/languagetool/language/BelgianDutch.java b/languagetool-language-modules/nl/src/main/java/org/languagetool/language/BelgianDutch.java
new file mode 100644
index 0000000000..edabaaf58a
--- /dev/null
+++ b/languagetool-language-modules/nl/src/main/java/org/languagetool/language/BelgianDutch.java
@@ -0,0 +1,33 @@
+/* LanguageTool, a natural language style checker 
+ * Copyright (C) 2007 Daniel Naber (http://www.danielnaber.de)
+ * 
+ * This library is free software; you can redistribute it and/or

danielnaber

#!/bin/bash

result=tatoeba-sentences

function extract {
  echo "extract to >$result-$2-20191014.txt"
  grep "	$1	" sentences.csv | awk -F '\t' '{print $3}' >$result-$2-20191014.txt
  shuf $result-$2-20191014.txt | head -n 1000 >sentences-$2-20191014-top1000.txt
}

danielnaber

package com.optimaize.langdetect;

import com.google.common.base.Optional;
import com.optimaize.langdetect.i18n.LdLocale;
import com.optimaize.langdetect.ngram.NgramExtractors;
import com.optimaize.langdetect.profiles.LanguageProfile;
import com.optimaize.langdetect.profiles.LanguageProfileReader;

import java.io.IOException;
import java.util.Arrays;

danielnaber

#!/bin/sh

echo "Call from LT source top directory"

mvn package -DskipTests

LANG_CODE=fr
PREFIX=${LANG_CODE}_FR
# this is the list of words as exported from https://github.com/Fanaen/Hunspell2WordList:
WORD_LIST=/home/dnaber/lt/fr.out.uniq

danielnaber

#!/bin/bash

echo "Create morfologik spelling dictionary, based on Hunspell dictionary"
echo "This script assumes you have the full LanguageTool build environment"
echo "Please call this script from the LanguageTool top-level directory"
echo ""

if [ $# -ne 2 ]
then
  SCRIPT=`basename $0`

danielnaber

import java.util.Calendar;
import java.util.Locale;

public class GregorianCalender {

  public static void main(String[] args) {
    System.out.println("See Java deal with the date the Gregorian calender was introduced:");
    printDates(Locale.GERMANY);
    printDates(Locale.JAPAN);
  }