Merge pull request #1864 from webanno/feature/1828-pre-tokenized-text…

…-format-support

#1828  - A simple import of already tokenized text to which no further tokenization is applied

webanno-doc/src/main/resources/META-INF/asciidoc/user-guide/formats-text.adoc

@@ -17,7 +17,7 @@
 [[sect_formats_text]]
 = Plain Text
 
-Basic UTF-8 plain text.
+Basic UTF-8 plain text. Automatic sentence and token detection will be performed.
 
 [cols="2,1,1,1,3"]
 |====
@@ -29,3 +29,36 @@ Basic UTF-8 plain text.
 | no
 | No annotations
 |====
+
+[[sect_formats_text_sentence_per_line]]
+= Plain Text (one sentence per line)
+
+Basic UTF-8 plain text where each line is interpreted as one sentence.
+
+[cols="2,1,1,1,3"]
+|====
+| Format | Read | Write | Custom Layers | Description
+
+| Plain text
+| yes
+| no
+| no
+| No annotations
+|====
+
+[[sect_formats_text_pretokenized]]
+= Plain Text (pretokenized)
+
+Basic UTF-8 plain text. Tokens are taken to be separated by spaces. Each line is interpreted as a
+sentence.
+
+[cols="2,1,1,1,3"]
+|====
+| Format | Read | Write | Custom Layers | Description
+
+| Plain text
+| yes
+| no
+| no
+| No annotations
+|====

webanno-io-text/pom.xml

@@ -65,6 +65,11 @@
       <groupId>org.dkpro.core</groupId>
       <artifactId>dkpro-core-api-parameter-asl</artifactId>
     </dependency>
+    <dependency>
+      <groupId>org.assertj</groupId>
+      <artifactId>assertj-core</artifactId>
+      <scope>test</scope>
+    </dependency>
     <dependency>
       <groupId>junit</groupId>
       <artifactId>junit</artifactId>

webanno-io-text/src/main/java/de/tudarmstadt/ukp/clarin/webanno/text/PretokenizedLineOrientedTextReader.java

@@ -0,0 +1,165 @@
+/*
+ * Licensed to the Technische Universität Darmstadt under one
+ * or more contributor license agreements.  See the NOTICE file
+ * distributed with this work for additional information
+ * regarding copyright ownership.  The Technische Universität Darmstadt 
+ * licenses this file to you under the Apache License, Version 2.0 (the
+ * "License"); you may not use this file except in compliance
+ * with the License.
+ *  
+ * http://www.apache.org/licenses/LICENSE-2.0
+ * 
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS IS" BASIS,
+ * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+package de.tudarmstadt.ukp.clarin.webanno.text;
+
+import java.io.BufferedInputStream;
+import java.io.IOException;
+import java.io.InputStream;
+import java.util.regex.Matcher;
+import java.util.regex.Pattern;
+
+import org.apache.commons.io.IOUtils;
+import org.apache.uima.collection.CollectionException;
+import org.apache.uima.fit.descriptor.TypeCapability;
+import org.apache.uima.jcas.JCas;
+import org.dkpro.core.api.io.JCasResourceCollectionReader_ImplBase;
+
+import de.tudarmstadt.ukp.dkpro.core.api.segmentation.type.Sentence;
+import de.tudarmstadt.ukp.dkpro.core.api.segmentation.type.Token;
+
+/**
+ * UIMA collection reader for plain text files, whitespace-separated tokens, one sentence per line.
+ */
+@TypeCapability(outputs = { "de.tudarmstadt.ukp.dkpro.core.api.metadata.type.DocumentMetaData" })
+public class PretokenizedLineOrientedTextReader
+    extends JCasResourceCollectionReader_ImplBase
+{
+    private static final Pattern whitespace = Pattern.compile("\\s+");
+
+    @Override
+    public void getNext(JCas aJCas) throws IOException, CollectionException
+    {
+        Resource res = nextFile();
+        initCas(aJCas, res);
+
+        try (InputStream is = new BufferedInputStream(res.getInputStream())) {
+            aJCas.setDocumentText(IOUtils.toString(is, "UTF-8"));
+        }
+
+        String t = aJCas.getDocumentText();
+        int start = 0;
+        int end = t.indexOf('\n');
+        while (end >= 0) {
+            createSentence(aJCas, start, end);
+            start = end + 1;
+            if (start < t.length()) {
+                end = t.indexOf('\n', start);
+            }
+            else {
+                end = -1;
+            }
+        }
+
+        if (start < t.length()) {
+            createSentence(aJCas, start, t.length());
+        }
+    }
+
+    protected Sentence createSentence(final JCas aJCas, final int aBegin, final int aEnd)
+    {
+        int[] span = new int[] { aBegin, aEnd };
+        trim(aJCas.getDocumentText(), span);
+        if (!isEmpty(span[0], span[1])) {
+            Sentence seg = new Sentence(aJCas, span[0], span[1]);
+            seg.addToIndexes(aJCas);
+
+            Matcher whitespaceMatcher = whitespace.matcher(seg.getCoveredText());
+            int prevBegin = 0;
+            while (whitespaceMatcher.find()) {
+                int end = whitespaceMatcher.start();
+                createToken(aJCas, seg.getBegin() + prevBegin, seg.getBegin() + end);
+                prevBegin = whitespaceMatcher.end();
+            }
+
+            if (prevBegin < aEnd) {
+                createToken(aJCas, seg.getBegin() + prevBegin, seg.getEnd());
+            }
+
+            return seg;
+        }
+        else {
+            return null;
+        }
+    }
+
+    protected Token createToken(final JCas aJCas, final int aBegin, final int aEnd)
+    {
+        int[] span = new int[] { aBegin, aEnd };
+        trim(aJCas.getDocumentText(), span);
+        if (!isEmpty(span[0], span[1])) {
+            Token seg = new Token(aJCas, span[0], span[1]);
+            seg.addToIndexes(aJCas);
+            return seg;
+        }
+        else {
+            return null;
+        }
+    }
+
+    /**
+     * Remove trailing or leading whitespace from the annotation.
+     * 
+     * @param aText
+     *            the text.
+     * @param aSpan
+     *            the offsets.
+     */
+    public void trim(String aText, int[] aSpan)
+    {
+        int begin = aSpan[0];
+        int end = aSpan[1] - 1;
+
+        while ((begin < (aText.length() - 1)) && trimChar(aText.charAt(begin))) {
+            begin++;
+        }
+        while ((end > 0) && trimChar(aText.charAt(end))) {
+            end--;
+        }
+        end++;
+
+        aSpan[0] = begin;
+        aSpan[1] = end;
+    }
+
+    public boolean isEmpty(int aBegin, int aEnd)
+    {
+        return aBegin >= aEnd;
+    }
+
+    public boolean trimChar(final char aChar)
+    {
+        switch (aChar) {
+        case '\n':
+            return true; // Line break
+        case '\r':
+            return true; // Carriage return
+        case '\t':
+            return true; // Tab
+        case '\u200E':
+            return true; // LEFT-TO-RIGHT MARK
+        case '\u200F':
+            return true; // RIGHT-TO-LEFT MARK
+        case '\u2028':
+            return true; // LINE SEPARATOR
+        case '\u2029':
+            return true; // PARAGRAPH SEPARATOR
+        default:
+            return Character.isWhitespace(aChar);
+        }
+    }
+}

webanno-io-text/src/main/java/de/tudarmstadt/ukp/clarin/webanno/text/PretokenizedTextFormatSupport.java

@@ -0,0 +1,60 @@
+/*
+ * Licensed to the Technische Universität Darmstadt under one
+ * or more contributor license agreements.  See the NOTICE file
+ * distributed with this work for additional information
+ * regarding copyright ownership.  The Technische Universität Darmstadt 
+ * licenses this file to you under the Apache License, Version 2.0 (the
+ * "License"); you may not use this file except in compliance
+ * with the License.
+ *  
+ * http://www.apache.org/licenses/LICENSE-2.0
+ * 
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS IS" BASIS,
+ * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+package de.tudarmstadt.ukp.clarin.webanno.text;
+
+import static org.apache.uima.fit.factory.CollectionReaderFactory.createReaderDescription;
+
+import org.apache.uima.collection.CollectionReaderDescription;
+import org.apache.uima.resource.ResourceInitializationException;
+import org.apache.uima.resource.metadata.TypeSystemDescription;
+import org.springframework.stereotype.Component;
+
+import de.tudarmstadt.ukp.clarin.webanno.api.format.FormatSupport;
+
+@Component
+public class PretokenizedTextFormatSupport
+    implements FormatSupport
+{
+    public static final String ID = "pretokenized-textlines";
+    public static final String NAME = "Plain text (space-separated tokens, one sentence per line)";
+
+    @Override
+    public String getId()
+    {
+        return ID;
+    }
+
+    @Override
+    public String getName()
+    {
+        return NAME;
+    }
+
+    @Override
+    public boolean isReadable()
+    {
+        return true;
+    }
+
+    @Override
+    public CollectionReaderDescription getReaderDescription(TypeSystemDescription aTSD)
+        throws ResourceInitializationException
+    {
+        return createReaderDescription(PretokenizedLineOrientedTextReader.class, aTSD);
+    }
+}

webanno-io-text/src/test/java/de/tudarmstadt/ukp/clarin/webanno/text/PretokenizedLineOrientedTextReaderTest.java

@@ -0,0 +1,88 @@
+/*
+ * Licensed to the Technische Universität Darmstadt under one
+ * or more contributor license agreements.  See the NOTICE file
+ * distributed with this work for additional information
+ * regarding copyright ownership.  The Technische Universität Darmstadt 
+ * licenses this file to you under the Apache License, Version 2.0 (the
+ * "License"); you may not use this file except in compliance
+ * with the License.
+ *  
+ * http://www.apache.org/licenses/LICENSE-2.0
+ * 
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS IS" BASIS,
+ * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+package de.tudarmstadt.ukp.clarin.webanno.text;
+
+import static org.apache.uima.fit.factory.CollectionReaderFactory.createReader;
+import static org.apache.uima.fit.util.JCasUtil.select;
+import static org.assertj.core.api.Assertions.assertThat;
+import static org.junit.Assert.assertEquals;
+
+import org.apache.uima.collection.CollectionReader;
+import org.apache.uima.fit.factory.JCasFactory;
+import org.apache.uima.jcas.JCas;
+import org.junit.Test;
+
+import de.tudarmstadt.ukp.dkpro.core.api.segmentation.type.Sentence;
+import de.tudarmstadt.ukp.dkpro.core.api.segmentation.type.Token;
+
+public class PretokenizedLineOrientedTextReaderTest
+{
+    @Test
+    public void test() throws Exception
+    {
+        JCas doc = JCasFactory.createJCas();
+
+        CollectionReader reader = createReader(PretokenizedLineOrientedTextReader.class,
+                PretokenizedLineOrientedTextReader.PARAM_SOURCE_LOCATION, "LICENSE.txt");
+
+        reader.getNext(doc.getCas());
+
+        assertEquals(169, select(doc, Sentence.class).size());
+        assertEquals(1581, select(doc, Token.class).size());
+    }
+
+    @Test
+    public void testSpaceSeparatedNoFinalLineBreak() throws Exception
+    {
+        JCas doc = JCasFactory.createJCas();
+
+        CollectionReader reader = createReader(PretokenizedLineOrientedTextReader.class,
+                PretokenizedLineOrientedTextReader.PARAM_SOURCE_LOCATION,
+                "src/test/resources/text/space-separated-no-final-line-break.txt");
+
+        reader.getNext(doc.getCas());
+
+        assertThat(select(doc, Sentence.class)) //
+                .extracting(s -> s.getCoveredText()) //
+                .containsExactly("1 2 3", "4 5 6", "7 8 9");
+
+        assertThat(select(doc, Token.class)) //
+                .extracting(t -> t.getCoveredText()) //
+                .containsExactly("1", "2", "3", "4", "5", "6", "7", "8", "9");
+    }
+
+    @Test
+    public void testTextWithPunctuation() throws Exception
+    {
+        JCas doc = JCasFactory.createJCas();
+
+        CollectionReader reader = createReader(PretokenizedLineOrientedTextReader.class,
+                PretokenizedLineOrientedTextReader.PARAM_SOURCE_LOCATION,
+                "src/test/resources/text/text-with-punctuation.txt");
+
+        reader.getNext(doc.getCas());
+
+        assertThat(select(doc, Sentence.class)) //
+                .extracting(s -> s.getCoveredText()) //
+                .containsExactly("Yesterday, I left the house.");
+
+        assertThat(select(doc, Token.class)) //
+                .extracting(t -> t.getCoveredText()) //
+                .containsExactly("Yesterday,", "I", "left", "the", "house.");
+    }
+}

webanno-io-text/src/test/resources/text/space-separated-no-final-line-break.txt

@@ -0,0 +1,3 @@
+1 2 3
+4 5 6
+7 8 9

webanno-io-text/src/test/resources/text/text-with-punctuation.txt

@@ -0,0 +1 @@
+Yesterday, I left the house.