Examples with English org.languagetool.language.English used on opensource projects

Example 11 with English

use of org.languagetool.language.English in project languagetool by languagetool-org.

the class WikipediaSentenceSourceTest method testWikipediaSource.

@Test
public void testWikipediaSource() throws XMLStreamException, IOException {
    InputStream stream = WikipediaSentenceSourceTest.class.getResourceAsStream("/org/languagetool/dev/wikipedia/wikipedia-en.xml");
    WikipediaSentenceSource source = new WikipediaSentenceSource(stream, new English());
    assertTrue(source.hasNext());
    assertThat(source.next().getText(), is("This is the first document."));
    assertThat(source.next().getText(), is("It has three sentences."));
    assertThat(source.next().getText(), is("Here's the last sentence."));
    assertThat(source.next().getText(), is("This is the second document."));
    assertThat(source.next().getText(), is("It has two sentences."));
    assertFalse(source.hasNext());
}

Example 12 with English

use of org.languagetool.language.English in project languagetool by languagetool-org.

the class IndexerSearcherTest method testAllRules.

@Ignore("ignored as long as it doesn't work 100%")
public void testAllRules() throws Exception {
    long startTime = System.currentTimeMillis();
    // comment in to test with external index:
    //directory = new SimpleFSDirectory(new File("/media/external-disk/corpus/languagetool/fast-rule-evaluation-de/"));
    //errorSearcher = new Searcher(directory);
    // TODO: make this work for all languages
    Language language = new English();
    //Language language = new French();
    //Language language = new Spanish();
    //Language language = new Polish();
    //Language language = new German();
    JLanguageTool lt = new JLanguageTool(language);
    System.out.println("Creating index for " + language + "...");
    int ruleCount = createIndex(lt);
    System.out.println("Index created with " + ruleCount + " rules");
    int ruleCounter = 0;
    int ruleProblems = 0;
    int exceptionCount = 0;
    List<Rule> rules = lt.getAllActiveRules();
    for (Rule rule : rules) {
        if (rule instanceof PatternRule && !rule.isDefaultOff()) {
            PatternRule patternRule = (PatternRule) rule;
            try {
                ruleCounter++;
                SearcherResult searcherResult = errorSearcher.findRuleMatchesOnIndex(patternRule, language);
                List<MatchingSentence> matchingSentences = searcherResult.getMatchingSentences();
                boolean foundExpectedMatch = false;
                for (MatchingSentence matchingSentence : matchingSentences) {
                    List<RuleMatch> ruleMatches = matchingSentence.getRuleMatches();
                    List<String> ruleMatchIds = getRuleMatchIds(ruleMatches);
                    if (ruleMatchIds.contains(patternRule.getFullId())) {
                        // TODO: there can be more than one expected match, can't it?
                        foundExpectedMatch = true;
                        break;
                    }
                }
                if (!foundExpectedMatch) {
                    System.out.println("Error: No match found for " + patternRule);
                    System.out.println("Query      : " + searcherResult.getRelaxedQuery().toString(FIELD_NAME_LOWERCASE));
                    System.out.println("Default field: " + FIELD_NAME_LOWERCASE);
                    System.out.println("Lucene Hits: " + searcherResult.getLuceneMatchCount());
                    System.out.println("Matches    : " + matchingSentences);
                    System.out.println("Examples   : " + rule.getIncorrectExamples());
                    System.out.println();
                    ruleProblems++;
                } else {
                //long time = System.currentTimeMillis() - startTime;
                //System.out.println("Tested " + matchingSentences.size() + " sentences in " + time + "ms for rule " + patternRule);
                }
            } catch (UnsupportedPatternRuleException e) {
                System.out.println("UnsupportedPatternRuleException searching for rule " + patternRule.getFullId() + ": " + e.getMessage());
                ruleProblems++;
            } catch (Exception e) {
                System.out.println("Exception searching for rule " + patternRule.getFullId() + ": " + e.getMessage());
                e.printStackTrace(System.out);
                exceptionCount++;
            }
        }
    }
    System.out.println(language + ": problems: " + ruleProblems + ", total rules: " + ruleCounter);
    System.out.println(language + ": exceptions: " + exceptionCount + " (including timeouts)");
    System.out.println("Total time: " + (System.currentTimeMillis() - startTime) + "ms");
}

Example 13 with English

use of org.languagetool.language.English in project languagetool by languagetool-org.

the class IndexerSearcherTest method createIndex.

private void createIndex(String content) throws IOException {
    directory = new RAMDirectory();
    //directory = FSDirectory.open(new File("/tmp/lucenetest"));  // for debugging
    Indexer.run(content, directory, new English());
    errorSearcher = new Searcher(directory);
}

Example 14 with English

use of org.languagetool.language.English in project languagetool by languagetool-org.

the class IndexerSearcherTest method testIndexerSearcherWithEnglish.

public void testIndexerSearcherWithEnglish() throws Exception {
    // Note that the second sentence ends with "lid" instead of "lids" (the inflated one)
    createIndex("How to move back and fourth from linux to xmb? Calcium deposits on eye lid.");
    English language = new English();
    SearcherResult searcherResult = errorSearcher.findRuleMatchesOnIndex(getFirstRule("BACK_AND_FOURTH", language), language);
    assertEquals(2, searcherResult.getCheckedSentences());
    assertEquals(false, searcherResult.isResultIsTimeLimited());
    assertEquals(1, searcherResult.getMatchingSentences().size());
    searcherResult = errorSearcher.findRuleMatchesOnIndex(getFirstRule("EYE_BROW", language), language);
    assertEquals(2, searcherResult.getCheckedSentences());
    assertEquals(false, searcherResult.isResultIsTimeLimited());
    assertEquals(1, searcherResult.getMatchingSentences().size());
    searcherResult = errorSearcher.findRuleMatchesOnIndex(getFirstRule("ALL_OVER_THE_WORD", language), language);
    assertEquals(2, searcherResult.getCheckedSentences());
    assertEquals(false, searcherResult.isResultIsTimeLimited());
    assertEquals(0, searcherResult.getMatchingSentences().size());
    try {
        errorSearcher.findRuleMatchesOnIndex(getFirstRule("Invalid Rule Id", language), language);
        fail("Exception should be thrown for invalid rule id.");
    } catch (PatternRuleNotFoundException ignored) {
    }
}

Example 15 with English

use of org.languagetool.language.English in project languagetool by languagetool-org.

the class IndexerSearcherTest method testWithNewRule.

public void testWithNewRule() throws Exception {
    createIndex("How to move back and fourth from linux to xmb?");
    List<PatternToken> patternTokens = Arrays.asList(new PatternToken("move", false, false, false), new PatternToken("back", false, false, false));
    PatternRule rule1 = new PatternRule("RULE1", new English(), patternTokens, "desc", "msg", "shortMsg");
    Searcher errorSearcher = new Searcher(directory);
    SearcherResult searcherResult = errorSearcher.findRuleMatchesOnIndex(rule1, new English());
    assertEquals(1, searcherResult.getCheckedSentences());
    assertEquals(1, searcherResult.getMatchingSentences().size());
    List<RuleMatch> ruleMatches = searcherResult.getMatchingSentences().get(0).getRuleMatches();
    assertEquals(1, ruleMatches.size());
    Rule rule = ruleMatches.get(0).getRule();
    assertEquals("RULE1", rule.getId());
}