001/* Utilities that manipulate strings using XSLT.
002
003 Copyright (c) 2002-2018 The Regents of the University of California.
004 All rights reserved.
005 Permission is hereby granted, without written agreement and without
006 license or royalty fees, to use, copy, modify, and distribute this
007 software and its documentation for any purpose, provided that the above
008 copyright notice and the following two paragraphs appear in all copies
009 of this software.
010
011 IN NO EVENT SHALL THE UNIVERSITY OF CALIFORNIA BE LIABLE TO ANY PARTY
012 FOR DIRECT, INDIRECT, SPECIAL, INCIDENTAL, OR CONSEQUENTIAL DAMAGES
013 ARISING OUT OF THE USE OF THIS SOFTWARE AND ITS DOCUMENTATION, EVEN IF
014 THE UNIVERSITY OF CALIFORNIA HAS BEEN ADVISED OF THE POSSIBILITY OF
015 SUCH DAMAGE.
016
017 THE UNIVERSITY OF CALIFORNIA SPECIFICALLY DISCLAIMS ANY WARRANTIES,
018 INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF
019 MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE. THE SOFTWARE
020 PROVIDED HEREUNDER IS ON AN "AS IS" BASIS, AND THE UNIVERSITY OF
021 CALIFORNIA HAS NO OBLIGATION TO PROVIDE MAINTENANCE, SUPPORT, UPDATES,
022 ENHANCEMENTS, OR MODIFICATIONS.
023
024 PT_COPYRIGHT_VERSION_2
025 COPYRIGHTENDKEY
026
027 */
028package ptolemy.util;
029
030import java.io.ByteArrayOutputStream;
031import java.io.FileWriter;
032import java.io.IOException;
033import java.io.OutputStream;
034import java.net.URL;
035import java.util.Iterator;
036import java.util.LinkedList;
037import java.util.List;
038
039import javax.xml.parsers.DocumentBuilder;
040import javax.xml.parsers.DocumentBuilderFactory;
041import javax.xml.parsers.ParserConfigurationException;
042import javax.xml.transform.ErrorListener;
043import javax.xml.transform.Transformer;
044import javax.xml.transform.TransformerException;
045import javax.xml.transform.TransformerFactory;
046import javax.xml.transform.dom.DOMResult;
047import javax.xml.transform.dom.DOMSource;
048import javax.xml.transform.stream.StreamResult;
049import javax.xml.transform.stream.StreamSource;
050
051import org.w3c.dom.Document;
052import org.xml.sax.InputSource;
053import org.xml.sax.SAXException;
054
055///////////////////////////////////////////////////////////////////
056//// XSLTUtilities
057
058/**
059 A collection of utilities for manipulating strings using XSLT.
060 These utilities do not depend on any other ptolemy.* packages.
061
062 <p>This file uses Saxon, the XSLT and XQuery Processor
063 <a href="http://saxon.sourceforge.net/#in_browser" target="_top">http://saxon.sourceforge.net</a>.
064
065 <p>Between Java 1.4.x and Java 1.5, Xalan was removed from the jar
066 files that are shipped.  Since Caltrop uses Saxon anyway, we now
067 use Saxon here as well.
068
069
070 @author Christopher Hylands, Haiyang Zheng
071 @version $Id$
072 @since Ptolemy II 2.1
073 @Pt.ProposedRating Green (eal)
074 Pending Java 1.5 changes
075 @Pt.AcceptedRating Yellow (cxh)
076 */
077public class XSLTUtilities {
078    /** Instances of this class cannot be created.
079     */
080    private XSLTUtilities() {
081    }
082
083    ///////////////////////////////////////////////////////////////////
084    ////                         public methods                    ////
085
086    /** Apply XSL transforms to an input file and generate an output file.
087     *  <p>Example use:
088     * <pre>
089     *  java -classpath $PTII ptolemy.util.XSLTUtilities
090     *       $PTII/ptolemy/hsif/demo/SwimmingPool/SwimmingPool.xml \
091     *       $PTII/ptolemy/hsif/xsl/GlobalVariablePreprocessor.xsl \
092     *       exportMoMLDTD \
093     *       /tmp/SwimmingPool_1.xml
094     * </pre>
095
096     * @param args At least three arguments:
097     * <ul>
098     * <li> The first argument is the input file name.</li>
099     * <li> The second through n-1 arguments are the named xsl files.</li>
100     * <li> The final argument is the output file name.</li>
101     * </ul>
102     * @exception Exception If there are problems with the transform.
103     */
104    public static void main(String[] args) throws Exception {
105
106        String lastArg = args[args.length - 1];
107        String outputFileName = lastArg;
108        int numberOfTransformers = args.length - 2;
109        if (lastArg.compareTo("exportMoMLDTD") == 0) {
110            _exportDTD = true;
111            outputFileName = args[args.length - 2];
112            numberOfTransformers = args.length - 3;
113        } else {
114            _exportDTD = false;
115        }
116
117        if (_exportDTD && args.length < 4 || !_exportDTD && args.length < 3) {
118            System.err.println("Usage: java -classpath $PTII "
119                    + "ptolemy.util.XSLTUtilities inputFile "
120                    + "xslFile1 [xslFile2 . . .] outputFile "
121                    + "[exportMoMLDTD]");
122            StringUtilities.exit(2);
123        }
124
125        // Make sure we can write the output first
126        FileWriter fileWriter = null;
127
128        try {
129            fileWriter = new FileWriter(outputFileName);
130
131            Document inputDocument = parse(args[0]);
132
133            List transforms = new LinkedList();
134
135            for (int i = 1; i < numberOfTransformers + 1; i++) {
136                transforms.add(args[i]);
137            }
138
139            Document outputDocument = XSLTUtilities.transform(inputDocument,
140                    transforms);
141            _writeOutput(XSLTUtilities.toString(outputDocument), fileWriter);
142        } finally {
143            if (fileWriter != null) {
144                fileWriter.close();
145            }
146        }
147    }
148
149    /** Parse an XML document using Saxon.
150     *  @param filename The file name of the xml file to be read in
151     *  The filename is passed to org.xml.sax.InputSource(String),
152     *  so it may be a file name or a URL.
153     *  @return the parsed document.
154     *  @exception ParserConfigurationException If there is a problem
155     *  creating the DocumentBuilder.
156     *  @exception IOException If the filename could not be parsed.
157     */
158    public static Document parse(String filename)
159            throws ParserConfigurationException, IOException {
160        // FIXME: Throw something other than Exception
161        //        System.setProperty("javax.xml.parsers.DocumentBuilderFactory",
162        //                "net.sf.saxon.om.DocumentBuilderFactoryImpl");
163
164        DocumentBuilderFactory factory = DocumentBuilderFactory.newInstance();
165        DocumentBuilder builder = factory.newDocumentBuilder();
166
167        // We use InputSource here so that we can specify the filename
168        // argument as a jar url so that HSIFToMoML works under Web Start.
169        try {
170            return builder.parse(new InputSource(filename));
171        } catch (SAXException ex) {
172            // Rethrow this with the filename included.
173            IOException exception = new IOException(
174                    "Failed to parse '" + filename + "'");
175            exception.initCause(ex);
176            throw exception;
177        }
178    }
179
180    /** Set the flag indicating whether to export DTD specification when
181     *  transforming XML files. By default, the transformer does not export
182     *  DTD.
183     *  @param exportDTD True for export DTD, false for not.
184     */
185    public static void setExportDTD(boolean exportDTD) {
186        _exportDTD = exportDTD;
187    }
188
189    /** Given a Document, generate a String.
190     *  @param document The document to be converted to a string.
191     *  @return A string representation of the Document.
192     *  @exception TransformerException If there is a
193     *  a problem creating a new Transformer or parser.
194     *  @exception IOException If there is a problem closing the output
195     *  stream.
196     */
197    public static String toString(Document document)
198            throws TransformerException, IOException {
199        // FIXME: Joern's sample code had this in it, but if we
200        // include it, then we get Provider   not found errors
201        //String defaultDBFI =
202        //    System.getProperty("javax.xml.parsers.DocumentBuilderFactory");
203        //System.setProperty("javax.xml.parsers.DocumentBuilderFactory",
204        //                   defaultDBFI == null ? "" : defaultDBFI);
205        OutputStream outputStream = null;
206
207        try {
208            outputStream = new ByteArrayOutputStream();
209
210            StreamResult result = new StreamResult(outputStream);
211            TransformerFactory transformerFactory = TransformerFactory
212                    .newInstance();
213            Transformer serializer = transformerFactory.newTransformer();
214            serializer.transform(new DOMSource(document), result);
215            return outputStream.toString();
216        } finally {
217            if (outputStream != null) {
218                outputStream.close();
219            }
220        }
221    }
222
223    /** Transform a document.
224     *  @param inputDocument The Document to be transformed
225     *  @param xslFileName The file name of the xsl file to be used.
226     *  If the file cannot be found, then we look up the file in the classpath.
227     *  @return a transformed document
228     *  @exception TransformerException If there is a problem with the
229     *  transform.
230     *  @exception IOException If there is a problem finding the
231     *  transform file.
232     */
233    public static Document transform(Document inputDocument, String xslFileName)
234            throws TransformerException, IOException {
235        TransformerFactory transformerFactory = TransformerFactory
236                .newInstance();
237        // Use an ErrorListener so as to avoid error output on stderr
238        // which causes problems with the test harness
239        transformerFactory.setErrorListener(new ErrorListener() {
240            /** Receive notification of a recoverable error. */
241            @Override
242            public void error(TransformerException exception)
243                    throws TransformerException {
244                throw exception;
245            }
246
247            /** Receive notification of a non-recoverable error. */
248            @Override
249            public void fatalError(TransformerException exception)
250                    throws TransformerException {
251                throw exception;
252            }
253
254            /** Receive notification of a warning. */
255            @Override
256            public void warning(TransformerException exception) {
257                System.err.println("ptolemy.util.XSLTUtilities.transform()"
258                        + ": Warning: " + exception);
259            }
260        });
261        Transformer transformer = null;
262
263        // Set a valid transformer.
264        try {
265            transformer = transformerFactory
266                    .newTransformer(new StreamSource(xslFileName));
267        } catch (javax.xml.transform.TransformerConfigurationException ex) {
268            try {
269                // We might be in the Swing Event thread, so
270                // Thread.currentThread().getContextClassLoader()
271                // .getResource(entry) probably will not work.
272                Class refClass = Class.forName("ptolemy.util.XSLTUtilities");
273                URL entryURL = refClass.getClassLoader()
274                        .getResource(xslFileName);
275
276                if (entryURL != null) {
277                    transformer = transformerFactory.newTransformer(
278                            new StreamSource(entryURL.toString()));
279                } else {
280                    IOException exception = new IOException(
281                            "Failed to open '" + xslFileName + "'");
282                    exception.initCause(ex);
283                    throw exception;
284                }
285            } catch (Exception ex2) {
286                IOException exception = new IOException(
287                        "Failed to open \"" + xslFileName + "\".\n"
288                                + "Searching the classpath threw:\n" + ex2);
289                exception.initCause(ex);
290                throw exception;
291            }
292        }
293
294        DOMResult result = new DOMResult();
295        transformer.transform(new DOMSource(inputDocument), result);
296        return (Document) result.getNode();
297    }
298
299    /** Transform a document by applying a list of transforms.
300     *  @param inputDocument The Document to be transformed
301     *  @param xslFileNames A list of Strings naming the
302     *  xsl files to be applied sequentially.
303     *  @return a transformed document
304     *  @exception TransformerException If there is a
305     *  a problem creating a new Transformer or parser.
306     *  @exception IOException If there is a problem closing the output
307     *  stream.
308     */
309    public static Document transform(Document inputDocument, List xslFileNames)
310            throws TransformerException, IOException {
311        Iterator fileNames = xslFileNames.iterator();
312
313        while (fileNames.hasNext()) {
314            String fileName = (String) fileNames.next();
315            inputDocument = transform(inputDocument, fileName);
316        }
317
318        return inputDocument;
319    }
320
321    /** Transform a file by applying a list of XSL transforms.
322     *  @param input The XML to be transformed
323     *  @param fileWriter A FileWriter that will write to the MoML
324     *  file.  The caller of this method is responsible for closing
325     *  the the FileWriter.
326     *  @param xslFileNames A list of Strings naming the
327     *  xsl files to be applied sequentially.
328     *  @exception ParserConfigurationException If there is a problem
329     *  creating the DocumentBuilder.
330     *  @exception TransformerException If there is a
331     *  a problem with the transform.
332     *  @exception IOException If there is a problem
333     *  finding a transform file or applying a transform.
334     */
335    public static void transform(String input, FileWriter fileWriter,
336            List xslFileNames) throws ParserConfigurationException,
337            TransformerException, IOException {
338        // This method takes a FileWriter so that the user can
339        // ensure that the FileWriter exists and is writable before going
340        // through the trouble of doing the conversion.
341        Document inputDocument = null;
342
343        try {
344            inputDocument = XSLTUtilities.parse(input);
345        } catch (IOException ex) {
346            // net.sf.saxon.om.DocumentBuilderImpl.parse()
347            // can throw a javax.xml.transform.TransformerException
348            // which extends Exception, but has  IOException as a cause,
349            // so we must catch Exception here, not IOException.
350            // Try it as a jar url
351            try {
352                URL jarURL = ClassUtilities.jarURLEntryResource(input);
353
354                if (jarURL == null) {
355                    throw new IOException("'" + input + "' was not a jar "
356                            + "URL, or was not found");
357                }
358
359                inputDocument = XSLTUtilities.parse(jarURL.toString());
360            } catch (IOException ex2) {
361                // Rethrow the original exception
362                throw ex;
363            }
364        }
365
366        Document outputDocument = XSLTUtilities.transform(inputDocument,
367                xslFileNames);
368
369        _writeOutput(XSLTUtilities.toString(outputDocument), fileWriter);
370
371        // Let the caller close the fileWriter.
372        //fileWriter.close();
373    }
374
375    ///////////////////////////////////////////////////////////////////
376    ////                         private methods                   ////
377
378    private static void _writeOutput(String outputString, FileWriter fileWriter)
379            throws IOException {
380        if (_exportDTD) {
381            // XSLT discards the DTD decalration.
382            // The following code inserts the specified DTD decalrations.
383            // Write the following into the filewriter first:
384            // <?xml version="1.0" encoding="UTF-8"?>,
385            // which appears just before the comment, <!--
386            int positionToInsertDTD = outputString.indexOf("<!--");
387            fileWriter.write(outputString, 0, positionToInsertDTD);
388            // FIXME: So far, only MoML DTD can be exported. If the support
389            // of more DTDs is necessray, modify the main() method and
390            // setExportDTD() method to allow configuration of DTD.
391            fileWriter.write(
392                    "\r\n<!DOCTYPE entity PUBLIC \"-//UC Berkeley//DTD MoML 1//EN\" "
393                            + "\r\n\"http://ptolemy.eecs.berkeley.edu/xml/dtd/MoML_1.dtd\">");
394            fileWriter.write(outputString.substring(positionToInsertDTD));
395        } else {
396            fileWriter.write(outputString);
397        }
398    }
399
400    ///////////////////////////////////////////////////////////////////
401    ////                         private variables                 ////
402
403    private static boolean _exportDTD = false;
404}