private static JCas makeJCasOneSentence() throws UIMAException { JCas jcas = makeJCas(); TokenBuilder<Token, Sentence> tb = new TokenBuilder<>(Token.class, Sentence.class); tb.buildTokens(jcas, "This is a test ."); return jcas; }
private static JCas makeJCasTwoSentences() throws UIMAException { JCas jcas = makeJCas(); TokenBuilder<Token, Sentence> tb = new TokenBuilder<>(Token.class, Sentence.class); tb.buildTokens(jcas, "He loves her .\nShe loves him not ."); assertEquals(2, select(jcas, Sentence.class).size()); return jcas; }
private static JCas makeJCasOneSentence(String aText) throws UIMAException { JCas jcas = makeJCas(); TokenBuilder<Token, Sentence> tb = new TokenBuilder<>(Token.class, Sentence.class); tb.buildTokens(jcas, aText); // Remove the sentences generated by the token builder which treats the line break as a // sentence break for (Sentence s : select(jcas, Sentence.class)) { s.removeFromIndexes(); } // Add a new sentence covering the whole text new Sentence(jcas, 0, jcas.getDocumentText().length()).addToIndexes(); return jcas; }
@Test public void testZeroLengthSpanBetweenAdjacentTokens() throws Exception { JCas jcas = makeJCas(); jcas.setDocumentText("word."); new Token(jcas, 0,4).addToIndexes(); new Token(jcas, 4,5).addToIndexes(); new Sentence(jcas, 0,5).addToIndexes(); CAS cas = jcas.getCas(); Type simpleSpanType = cas.getTypeSystem().getType("webanno.custom.SimpleSpan"); // Insert zero-width annotation between the adjacent tokens (at end of first token). AnnotationFS fs1a = cas.createAnnotation(simpleSpanType, 4, 4); cas.addFsToIndexes(fs1a); writeAndAssertEquals(jcas, WebannoTsv3Writer.PARAM_SPAN_LAYERS, asList("webanno.custom.SimpleSpan")); }