Restructuration of the code and adding the JSON annotation files as p…

…arameters

pom.xml

@@ -4,7 +4,7 @@
 
 	<groupId>ca.crim.nlp</groupId>
 	<artifactId>crim-heideltime</artifactId>
-	<version>1.0.0-SNAPSHOT</version>
+	<version>3.0.0-SNAPSHOT</version>
 
 	<name>HeidelTime</name>
 	<description> This version of HeidelTime extends the well-known multilingual cross-domain temporal tagger (com.github.hiedltime) that extracts temporal expressions from documents and normalizes them according to the TIMEX3 annotation standard.</description>
@@ -54,7 +54,7 @@
 	</scm>
 
 	<build>
-		<sourceDirectory>src</sourceDirectory>
+		<sourceDirectory>src/main/java</sourceDirectory>
 		<outputDirectory>${basedir}/class</outputDirectory>
 		<resources>
 			<resource>
@@ -64,7 +64,7 @@
 				</includes>
 			</resource>
 			<resource>
-				<directory>resources/</directory>
+				<directory>${basedir}/main/resources/</directory>
 				<includes>
 					<include>**/*.txt</include>
 				</includes>
@@ -182,7 +182,7 @@
     		<groupId>junit</groupId>
     		<artifactId>junit</artifactId>
    			<version>4.11</version>
-    		<scope>test</scope>
+			<scope>test</scope>
 		</dependency>
 		<dependency>
  			<groupId>com.github.stefanbirkner</groupId>

src/de/unihd/dbs/heideltime/standalone/CLISwitch.java → src/main/java/de/unihd/dbs/heideltime/standalone/CLISwitch.java

@@ -24,6 +24,8 @@ public enum CLISwitch {
 	LOCALE		("Locale", "-locale", null),
 	POSTAGGER	("Part of Speech tagger", "-pos", POSTagger.TREETAGGER),
 	INTERVALS	("Interval Tagger", "-it"),
+	POSFILE		("Path to the JSON-file describing the POS", "-pf"),
+	SENTENCEFILE("Path to the JSON-file describing the sentences", "-sf"),
 	HELP		("This screen", "-h"),
 	;
 

src/de/unihd/dbs/heideltime/standalone/HeidelTimeStandalone.java → src/main/java/de/unihd/dbs/heideltime/standalone/HeidelTimeStandalone.java

@@ -14,6 +14,7 @@
 
 package de.unihd.dbs.heideltime.standalone;
 
+import java.io.File;
 import java.io.FileInputStream;
 import java.io.FileNotFoundException;
 import java.io.IOException;
@@ -332,9 +333,9 @@ private void provideDocumentCreationTime(JCas jcas,
 	 * @param jcas
 	 * @throws Exception 
 	 */
-	private void establishHeidelTimePreconditions(JCas jcas)throws Exception {
+	private void establishHeidelTimePreconditions(JCas jcas, String pos_file, String sentence_file)throws Exception {
 		// Token information & sentence structure
-		establishPartOfSpeechInformation(jcas);
+		establishPartOfSpeechInformation(jcas, pos_file, sentence_file);
 	}
 
 	/**
@@ -343,7 +344,7 @@ private void establishHeidelTimePreconditions(JCas jcas)throws Exception {
 	 * @param jcas
 	 * @throws Exception 
 	 */
-	private void establishPartOfSpeechInformation(JCas jcas) throws Exception {
+	private void establishPartOfSpeechInformation(JCas jcas, String pos_file, String sentence_file) throws Exception {
 		logger.log(Level.FINEST, "Establishing part of speech information...");
 
 		PartOfSpeechTagger partOfSpeechTagger = null;
@@ -437,7 +438,10 @@ private void establishPartOfSpeechInformation(JCas jcas) throws Exception {
 				}
 		}
 		partOfSpeechTagger.initialize(settings);
-		partOfSpeechTagger.process(jcas);
+		if (!POSTagger.JSON.equals(posTagger))
+			partOfSpeechTagger.process(jcas);
+		else 
+			((JSONTaggerWrapper)partOfSpeechTagger).process(jcas, pos_file, sentence_file);
 		partOfSpeechTagger.reset();
 
 		logger.log(Level.FINEST, "Part of speech information established");
@@ -466,7 +470,7 @@ private ResultFormatter getFormatter() {
 	 */
 	public String process(String document)
 			throws DocumentCreationTimeMissingException {
-		return process(document, null, getFormatter());
+		return process(document, null, getFormatter(), null, null);
 	}
 
 	/**
@@ -480,9 +484,9 @@ public String process(String document)
 	 *             {@link #process(String, Date)} instead to provide document
 	 *             creation time!
 	 */
-	public String process(String document, Date documentCreationTime)
+	public String process(String document, Date documentCreationTime, String pos_file, String sentence_file)
 			throws DocumentCreationTimeMissingException {
-		return process(document, documentCreationTime, getFormatter());
+		return process(document, documentCreationTime, getFormatter(), pos_file, sentence_file);
 	}
 
 	/**
@@ -498,7 +502,7 @@ public String process(String document, Date documentCreationTime)
 	 */
 	public String process(String document, ResultFormatter resultFormatter)
 			throws DocumentCreationTimeMissingException {
-		return process(document, null, resultFormatter);
+		return process(document, null, resultFormatter, null, null);
 	}
 
 	/**
@@ -513,7 +517,7 @@ public String process(String document, ResultFormatter resultFormatter)
 	 *             If document creation time is missing when processing a
 	 *             document of type {@link DocumentType#NEWS}
 	 */
-	public String process(String document, Date documentCreationTime, ResultFormatter resultFormatter)
+	public String process(String document, Date documentCreationTime, ResultFormatter resultFormatter, String pos_file, String sentence_file)
 			throws DocumentCreationTimeMissingException {
 		logger.log(Level.INFO, "Processing started");
 
@@ -533,7 +537,7 @@ public String process(String document, Date documentCreationTime, ResultFormatte
 		try {
 			logger.log(Level.FINER, "Establishing preconditions...");
 			provideDocumentCreationTime(jcas, documentCreationTime);
-			establishHeidelTimePreconditions(jcas);
+			establishHeidelTimePreconditions(jcas, pos_file, sentence_file);
 			logger.log(Level.FINER, "Preconditions established");
 
 			heidelTime.process(jcas);
@@ -756,8 +760,52 @@ public static void main(String[] args) {
 		} else {
 			// Type not found
 			posTagger = (POSTagger) CLISwitch.POSTAGGER.getValue();
-			logger.log(Level.INFO, "POS Tagger '-pos': NOT FOUND OR RECOGNIZED; set to "+posTagger.toString().toUpperCase());
+			logger.log(Level.INFO, "POS Tagger '-pos': NOT FOUND OR RECOGNIZED; set to "+ posTagger.toString().toUpperCase());
 		}
+
+		// If POS tagger is JSONTagger, make sure you have a path for the sentence annotations and the POS annotations
+		String pos_file = "";
+		String sentence_file = "";
+		if (posTagger == POSTagger.JSON) {
+			if(CLISwitch.POSFILE.getIsActive()) {
+				try {
+					pos_file = CLISwitch.POSFILE.getValue().toString();
+					File f = new File(pos_file);
+					if( !f.exists() || f.isDirectory()) { 
+					    throw new IllegalArgumentException();
+					}
+
+				} catch(IllegalArgumentException e) {
+					logger.log(Level.WARNING, "Given part-of-speech JSON file doesn't exist.");
+					System.exit(-1);
+				}
+				logger.log(Level.INFO, "Part-of-speech JSON file '-pf': "+ pos_file);
+			} else {
+				// Path to part-of-speech JSON file needed
+				logger.log(Level.INFO, "Path to the part-of-speech JSON file needed");
+				System.exit(-1);
+			}
+
+			if(CLISwitch.SENTENCEFILE.getIsActive()) {
+				try {
+					sentence_file = CLISwitch.SENTENCEFILE.getValue().toString();
+					File f = new File(sentence_file);
+					if( !f.exists() || f.isDirectory()) { 
+					    throw new IllegalArgumentException();
+					}
+
+				} catch(IllegalArgumentException e) {
+					logger.log(Level.WARNING, "Given sentence JSON file doesn't exist.");
+					System.exit(-1);
+				}
+				logger.log(Level.INFO, "Sentence JSON file '-sf': "+ sentence_file);
+			} else {
+				// Path to part-of-speech JSON file needed
+				logger.log(Level.INFO, "Path to the sentence JSON file needed");
+				System.exit(-1);
+			}
+		}
+
 
 		// Set whether or not to use the Interval Tagger
 		Boolean doIntervalTagging = false;
@@ -799,7 +847,7 @@ public static void main(String[] args) {
 			String input = new String(new String(inArr, encodingType).getBytes("UTF-8"), "UTF-8");
 
 			HeidelTimeStandalone standalone = new HeidelTimeStandalone(language, type, outputType, null, posTagger, doIntervalTagging);
-			String out = standalone.process(input, dct);
+			String out = standalone.process(input, dct, pos_file, sentence_file);
 
 			// Print output always as UTF-8
 			pwOut = new PrintWriter(new OutputStreamWriter(System.out, "UTF-8"));

src/de/unihd/dbs/heideltime/standalone/components/impl/JSONTaggerWrapper.java → src/main/java/de/unihd/dbs/heideltime/standalone/components/impl/JSONTaggerWrapper.java

@@ -19,6 +19,7 @@
 import de.unihd.dbs.uima.types.heideltime.Sentence;
 import de.unihd.dbs.uima.types.heideltime.Token;
 
+
 /**
  * The JSON tagger wrapper reads JSON files containing the sentence annotations and the POS annotations
  * The path to these two files should be indicated in the environment 
@@ -29,12 +30,16 @@ public class JSONTaggerWrapper implements PartOfSpeechTagger {
 
 	private String configFile;
 	private Hashtable<String, List<String>> defs = new Hashtable<String, List<String>>();
-
+
+
+	public String SENTENCES = "sentences";
 	public String SENTENCE_BEGIN = "sentence_begin";
 	public String SENTENCE_END = "sentence_end";
+	public String TOKENS = "tokens";
 	public String TOKEN_BEGIN = "token_begin";
 	public String TOKEN_END = "token_end";
 	public String TOKEN_POS = "token_pos";
+
 
 	@Override
 	public void initialize(Properties settings) throws IllegalArgumentException, IOException {
@@ -64,31 +69,52 @@ public void initialize(Properties settings) throws IllegalArgumentException, IOE
 	    			|| defs.get(SENTENCE_END)== null 
 	    			|| defs.get(TOKEN_BEGIN)== null 
 	    			|| defs.get(TOKEN_END)== null 
-	    			||defs.get(TOKEN_POS)== null)
+	    			||defs.get(TOKEN_POS)== null
+	    			||defs.get(TOKENS)== null
+	    			||defs.get(SENTENCES)== null)
 	    		throw new IOException(String.format("The JSON configfile '%s' is not valid.", configFile));
 	    }
 	}
 
 	@Override
 	public void process(JCas jcas) throws Exception {
-		String sentence_annotation_filepath = System.getenv("SENTENCE_ANNOTATION_FILE_PATH");
-		String token_annotation_filepath = System.getenv("POS_ANNOTATION_FILE_PATH");
+		throw new Exception("Missing arguments : sentence annotation file path and part-of-speech annotation file path.");   
+	}
+
+	public void process(JCas jcas, String token_annotation_filepath, String sentence_annotation_filepath) throws Exception {
 		JSONParser parser = new JSONParser();
 
-		JSONArray sentences = (JSONArray) parser.parse(new FileReader(sentence_annotation_filepath));
+		// getting sentences
+		List<String> path_to_sentences = defs.get(SENTENCES);
+		JSONArray sentences;
+		if (!path_to_sentences.get(0).equalsIgnoreCase("None"))
+			sentences = (JSONArray) getInfo(parser.parse(new FileReader(sentence_annotation_filepath)), path_to_sentences);
+		else 
+			sentences= (JSONArray) parser.parse(new FileReader(sentence_annotation_filepath));
 		for (Object o : sentences) {
 			Sentence s = new Sentence(jcas, (int)(long)getInfo(o, defs.get(SENTENCE_BEGIN)), (int)(long)getInfo(o, defs.get(SENTENCE_END)));
 			s.addToIndexes();		
 		}
-
-		JSONArray tokens = (JSONArray) parser.parse(new FileReader(token_annotation_filepath));
+
+		// getting tokens
+		List<String> path_to_tokens = defs.get(TOKENS);
+		JSONArray tokens ;
+		if (!path_to_tokens.get(0).equalsIgnoreCase("None"))
+			tokens = (JSONArray) getInfo(parser.parse(new FileReader(token_annotation_filepath)), path_to_tokens);
+		else 
+			tokens= (JSONArray) parser.parse(new FileReader(token_annotation_filepath));
 		for (Object o : tokens) {
-			Token t = new Token(jcas, (int)(long)getInfo(o, defs.get(TOKEN_BEGIN)), (int)(long)getInfo(o, defs.get(TOKEN_END)));
-			t.setPos((String) getInfo(o, defs.get(TOKEN_POS)));
-			t.addToIndexes();		
+			String pos = (String) getInfo(o, defs.get(TOKEN_POS));
+			if ( pos != null  
+				&& pos.length() > 0
+				){
+				Token t = new Token(jcas, (int)(long)getInfo(o, defs.get(TOKEN_BEGIN)), (int)(long)getInfo(o, defs.get(TOKEN_END)));
+				t.setPos(pos);
+				t.addToIndexes();
+			}
 		}    
 	}
-
+	
 	@Override
 	public void reset() {
 		// TODO Auto-generated method stub