001/*
002 * Licensed to the Apache Software Foundation (ASF) under one or more
003 * contributor license agreements.  See the NOTICE file distributed with
004 * this work for additional information regarding copyright ownership.
005 * The ASF licenses this file to You under the Apache License, Version 2.0
006 * (the "License"); you may not use this file except in compliance with
007 * the License.  You may obtain a copy of the License at
008 *
009 *      http://www.apache.org/licenses/LICENSE-2.0
010 *
011 * Unless required by applicable law or agreed to in writing, software
012 * distributed under the License is distributed on an "AS IS" BASIS,
013 * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
014 * See the License for the specific language governing permissions and
015 * limitations under the License.
016 */
017package org.apache.commons.fileupload.disk;
018
019import static java.lang.String.format;
020
021import java.io.ByteArrayInputStream;
022import java.io.File;
023import java.io.FileInputStream;
024import java.io.FileOutputStream;
025import java.io.IOException;
026import java.io.InputStream;
027import java.io.OutputStream;
028import java.io.UnsupportedEncodingException;
029import java.util.Map;
030import java.util.UUID;
031import java.util.concurrent.atomic.AtomicInteger;
032
033import org.apache.commons.fileupload.FileItem;
034import org.apache.commons.fileupload.FileItemHeaders;
035import org.apache.commons.fileupload.FileUploadException;
036import org.apache.commons.fileupload.ParameterParser;
037import org.apache.commons.fileupload.util.Streams;
038import org.apache.commons.io.FileUtils;
039import org.apache.commons.io.IOUtils;
040import org.apache.commons.io.output.DeferredFileOutputStream;
041
042/**
043 * <p> The default implementation of the
044 * {@link org.apache.commons.fileupload.FileItem FileItem} interface.
045 *
046 * <p> After retrieving an instance of this class from a {@link
047 * DiskFileItemFactory} instance (see
048 * {@link org.apache.commons.fileupload.servlet.ServletFileUpload
049 * #parseRequest(javax.servlet.http.HttpServletRequest)}), you may
050 * either request all contents of file at once using {@link #get()} or
051 * request an {@link java.io.InputStream InputStream} with
052 * {@link #getInputStream()} and process the file without attempting to load
053 * it into memory, which may come handy with large files.
054 *
055 * <p>Temporary files, which are created for file items, should be
056 * deleted later on. The best way to do this is using a
057 * {@link org.apache.commons.io.FileCleaningTracker}, which you can set on the
058 * {@link DiskFileItemFactory}. However, if you do use such a tracker,
059 * then you must consider the following: Temporary files are automatically
060 * deleted as soon as they are no longer needed. (More precisely, when the
061 * corresponding instance of {@link java.io.File} is garbage collected.)
062 * This is done by the so-called reaper thread, which is started and stopped
063 * automatically by the {@link org.apache.commons.io.FileCleaningTracker} when
064 * there are files to be tracked.
065 * It might make sense to terminate that thread, for example, if
066 * your web application ends. See the section on "Resource cleanup"
067 * in the users guide of commons-fileupload.</p>
068 *
069 * @since FileUpload 1.1
070 */
071public class DiskFileItem
072    implements FileItem {
073
074    // ----------------------------------------------------- Manifest constants
075
076    /**
077     * Default content charset to be used when no explicit charset
078     * parameter is provided by the sender. Media subtypes of the
079     * "text" type are defined to have a default charset value of
080     * "ISO-8859-1" when received via HTTP.
081     */
082    public static final String DEFAULT_CHARSET = "ISO-8859-1";
083
084    // ----------------------------------------------------------- Data members
085
086    /**
087     * UID used in unique file name generation.
088     */
089    private static final String UID =
090            UUID.randomUUID().toString().replace('-', '_');
091
092    /**
093     * Counter used in unique identifier generation.
094     */
095    private static final AtomicInteger COUNTER = new AtomicInteger(0);
096
097    /**
098     * The name of the form field as provided by the browser.
099     */
100    private String fieldName;
101
102    /**
103     * The content type passed by the browser, or <code>null</code> if
104     * not defined.
105     */
106    private final String contentType;
107
108    /**
109     * Whether or not this item is a simple form field.
110     */
111    private boolean isFormField;
112
113    /**
114     * The original filename in the user's filesystem.
115     */
116    private final String fileName;
117
118    /**
119     * The size of the item, in bytes. This is used to cache the size when a
120     * file item is moved from its original location.
121     */
122    private long size = -1;
123
124
125    /**
126     * The threshold above which uploads will be stored on disk.
127     */
128    private final int sizeThreshold;
129
130    /**
131     * The directory in which uploaded files will be stored, if stored on disk.
132     */
133    private final File repository;
134
135    /**
136     * Cached contents of the file.
137     */
138    private byte[] cachedContent;
139
140    /**
141     * Output stream for this item.
142     */
143    private transient DeferredFileOutputStream dfos;
144
145    /**
146     * The temporary file to use.
147     */
148    private transient File tempFile;
149
150    /**
151     * The file items headers.
152     */
153    private FileItemHeaders headers;
154
155    /**
156     * Default content charset to be used when no explicit charset
157     * parameter is provided by the sender.
158     */
159    private String defaultCharset = DEFAULT_CHARSET;
160
161    // ----------------------------------------------------------- Constructors
162
163    /**
164     * Constructs a new <code>DiskFileItem</code> instance.
165     *
166     * @param fieldName     The name of the form field.
167     * @param contentType   The content type passed by the browser or
168     *                      <code>null</code> if not specified.
169     * @param isFormField   Whether or not this item is a plain form field, as
170     *                      opposed to a file upload.
171     * @param fileName      The original filename in the user's filesystem, or
172     *                      <code>null</code> if not specified.
173     * @param sizeThreshold The threshold, in bytes, below which items will be
174     *                      retained in memory and above which they will be
175     *                      stored as a file.
176     * @param repository    The data repository, which is the directory in
177     *                      which files will be created, should the item size
178     *                      exceed the threshold.
179     */
180    public DiskFileItem(String fieldName,
181            String contentType, boolean isFormField, String fileName,
182            int sizeThreshold, File repository) {
183        this.fieldName = fieldName;
184        this.contentType = contentType;
185        this.isFormField = isFormField;
186        this.fileName = fileName;
187        this.sizeThreshold = sizeThreshold;
188        this.repository = repository;
189    }
190
191    // ------------------------------- Methods from javax.activation.DataSource
192
193    /**
194     * Returns an {@link java.io.InputStream InputStream} that can be
195     * used to retrieve the contents of the file.
196     *
197     * @return An {@link java.io.InputStream InputStream} that can be
198     *         used to retrieve the contents of the file.
199     *
200     * @throws IOException if an error occurs.
201     */
202    @Override
203    public InputStream getInputStream()
204        throws IOException {
205        if (!isInMemory()) {
206            return new FileInputStream(dfos.getFile());
207        }
208
209        if (cachedContent == null) {
210            cachedContent = dfos.getData();
211        }
212        return new ByteArrayInputStream(cachedContent);
213    }
214
215    /**
216     * Returns the content type passed by the agent or <code>null</code> if
217     * not defined.
218     *
219     * @return The content type passed by the agent or <code>null</code> if
220     *         not defined.
221     */
222    @Override
223    public String getContentType() {
224        return contentType;
225    }
226
227    /**
228     * Returns the content charset passed by the agent or <code>null</code> if
229     * not defined.
230     *
231     * @return The content charset passed by the agent or <code>null</code> if
232     *         not defined.
233     */
234    public String getCharSet() {
235        ParameterParser parser = new ParameterParser();
236        parser.setLowerCaseNames(true);
237        // Parameter parser can handle null input
238        Map<String, String> params = parser.parse(getContentType(), ';');
239        return params.get("charset");
240    }
241
242    /**
243     * Returns the original filename in the client's filesystem.
244     *
245     * @return The original filename in the client's filesystem.
246     * @throws org.apache.commons.fileupload.InvalidFileNameException The file name contains a NUL character,
247     *   which might be an indicator of a security attack. If you intend to
248     *   use the file name anyways, catch the exception and use
249     *   {@link org.apache.commons.fileupload.InvalidFileNameException#getName()}.
250     */
251    @Override
252    public String getName() {
253        return Streams.checkFileName(fileName);
254    }
255
256    // ------------------------------------------------------- FileItem methods
257
258    /**
259     * Provides a hint as to whether or not the file contents will be read
260     * from memory.
261     *
262     * @return <code>true</code> if the file contents will be read
263     *         from memory; <code>false</code> otherwise.
264     */
265    @Override
266    public boolean isInMemory() {
267        if (cachedContent != null) {
268            return true;
269        }
270        return dfos.isInMemory();
271    }
272
273    /**
274     * Returns the size of the file.
275     *
276     * @return The size of the file, in bytes.
277     */
278    @Override
279    public long getSize() {
280        if (size >= 0) {
281            return size;
282        } else if (cachedContent != null) {
283            return cachedContent.length;
284        } else if (dfos.isInMemory()) {
285            return dfos.getData().length;
286        } else {
287            return dfos.getFile().length();
288        }
289    }
290
291    /**
292     * Returns the contents of the file as an array of bytes.  If the
293     * contents of the file were not yet cached in memory, they will be
294     * loaded from the disk storage and cached.
295     *
296     * @return The contents of the file as an array of bytes
297     * or {@code null} if the data cannot be read
298     */
299    @Override
300    public byte[] get() {
301        if (isInMemory()) {
302            if (cachedContent == null && dfos != null) {
303                cachedContent = dfos.getData();
304            }
305            return cachedContent;
306        }
307
308        byte[] fileData = new byte[(int) getSize()];
309        InputStream fis = null;
310
311        try {
312            fis = new FileInputStream(dfos.getFile());
313            IOUtils.readFully(fis, fileData);
314        } catch (IOException e) {
315            fileData = null;
316        } finally {
317            IOUtils.closeQuietly(fis);
318        }
319
320        return fileData;
321    }
322
323    /**
324     * Returns the contents of the file as a String, using the specified
325     * encoding.  This method uses {@link #get()} to retrieve the
326     * contents of the file.
327     *
328     * @param charset The charset to use.
329     *
330     * @return The contents of the file, as a string.
331     *
332     * @throws UnsupportedEncodingException if the requested character
333     *                                      encoding is not available.
334     */
335    @Override
336    public String getString(final String charset)
337        throws UnsupportedEncodingException {
338        return new String(get(), charset);
339    }
340
341    /**
342     * Returns the contents of the file as a String, using the default
343     * character encoding.  This method uses {@link #get()} to retrieve the
344     * contents of the file.
345     *
346     * <b>TODO</b> Consider making this method throw UnsupportedEncodingException.
347     *
348     * @return The contents of the file, as a string.
349     */
350    @Override
351    public String getString() {
352        byte[] rawdata = get();
353        String charset = getCharSet();
354        if (charset == null) {
355            charset = defaultCharset;
356        }
357        try {
358            return new String(rawdata, charset);
359        } catch (UnsupportedEncodingException e) {
360            return new String(rawdata);
361        }
362    }
363
364    /**
365     * A convenience method to write an uploaded item to disk. The client code
366     * is not concerned with whether or not the item is stored in memory, or on
367     * disk in a temporary location. They just want to write the uploaded item
368     * to a file.
369     * <p>
370     * This implementation first attempts to rename the uploaded item to the
371     * specified destination file, if the item was originally written to disk.
372     * Otherwise, the data will be copied to the specified file.
373     * <p>
374     * This method is only guaranteed to work <em>once</em>, the first time it
375     * is invoked for a particular item. This is because, in the event that the
376     * method renames a temporary file, that file will no longer be available
377     * to copy or rename again at a later time.
378     *
379     * @param file The <code>File</code> into which the uploaded item should
380     *             be stored.
381     *
382     * @throws Exception if an error occurs.
383     */
384    @Override
385    public void write(File file) throws Exception {
386        if (isInMemory()) {
387            FileOutputStream fout = null;
388            try {
389                fout = new FileOutputStream(file);
390                fout.write(get());
391                fout.close();
392            } finally {
393                IOUtils.closeQuietly(fout);
394            }
395        } else {
396            File outputFile = getStoreLocation();
397            if (outputFile != null) {
398                // Save the length of the file
399                size = outputFile.length();
400                /*
401                 * The uploaded file is being stored on disk
402                 * in a temporary location so move it to the
403                 * desired file.
404                 */
405                FileUtils.moveFile(outputFile, file);
406            } else {
407                /*
408                 * For whatever reason we cannot write the
409                 * file to disk.
410                 */
411                throw new FileUploadException(
412                    "Cannot write uploaded file to disk!");
413            }
414        }
415    }
416
417    /**
418     * Deletes the underlying storage for a file item, including deleting any
419     * associated temporary disk file. Although this storage will be deleted
420     * automatically when the <code>FileItem</code> instance is garbage
421     * collected, this method can be used to ensure that this is done at an
422     * earlier time, thus preserving system resources.
423     */
424    @Override
425    public void delete() {
426        cachedContent = null;
427        File outputFile = getStoreLocation();
428        if (outputFile != null && !isInMemory() && outputFile.exists()) {
429            outputFile.delete();
430        }
431    }
432
433    /**
434     * Returns the name of the field in the multipart form corresponding to
435     * this file item.
436     *
437     * @return The name of the form field.
438     *
439     * @see #setFieldName(java.lang.String)
440     *
441     */
442    @Override
443    public String getFieldName() {
444        return fieldName;
445    }
446
447    /**
448     * Sets the field name used to reference this file item.
449     *
450     * @param fieldName The name of the form field.
451     *
452     * @see #getFieldName()
453     *
454     */
455    @Override
456    public void setFieldName(String fieldName) {
457        this.fieldName = fieldName;
458    }
459
460    /**
461     * Determines whether or not a <code>FileItem</code> instance represents
462     * a simple form field.
463     *
464     * @return <code>true</code> if the instance represents a simple form
465     *         field; <code>false</code> if it represents an uploaded file.
466     *
467     * @see #setFormField(boolean)
468     *
469     */
470    @Override
471    public boolean isFormField() {
472        return isFormField;
473    }
474
475    /**
476     * Specifies whether or not a <code>FileItem</code> instance represents
477     * a simple form field.
478     *
479     * @param state <code>true</code> if the instance represents a simple form
480     *              field; <code>false</code> if it represents an uploaded file.
481     *
482     * @see #isFormField()
483     *
484     */
485    @Override
486    public void setFormField(boolean state) {
487        isFormField = state;
488    }
489
490    /**
491     * Returns an {@link java.io.OutputStream OutputStream} that can
492     * be used for storing the contents of the file.
493     *
494     * @return An {@link java.io.OutputStream OutputStream} that can be used
495     *         for storing the contents of the file.
496     *
497     * @throws IOException if an error occurs.
498     */
499    @Override
500    public OutputStream getOutputStream()
501        throws IOException {
502        if (dfos == null) {
503            File outputFile = getTempFile();
504            dfos = new DeferredFileOutputStream(sizeThreshold, outputFile);
505        }
506        return dfos;
507    }
508
509    // --------------------------------------------------------- Public methods
510
511    /**
512     * Returns the {@link java.io.File} object for the <code>FileItem</code>'s
513     * data's temporary location on the disk. Note that for
514     * <code>FileItem</code>s that have their data stored in memory,
515     * this method will return <code>null</code>. When handling large
516     * files, you can use {@link java.io.File#renameTo(java.io.File)} to
517     * move the file to new location without copying the data, if the
518     * source and destination locations reside within the same logical
519     * volume.
520     *
521     * @return The data file, or <code>null</code> if the data is stored in
522     *         memory.
523     */
524    public File getStoreLocation() {
525        if (dfos == null) {
526            return null;
527        }
528        if (isInMemory()) {
529            return null;
530        }
531        return dfos.getFile();
532    }
533
534    // ------------------------------------------------------ Protected methods
535
536    /**
537     * Removes the file contents from the temporary storage.
538     */
539    @Override
540    protected void finalize() {
541        if (dfos == null || dfos.isInMemory()) {
542            return;
543        }
544        File outputFile = dfos.getFile();
545
546        if (outputFile != null && outputFile.exists()) {
547            outputFile.delete();
548        }
549    }
550
551    /**
552     * Creates and returns a {@link java.io.File File} representing a uniquely
553     * named temporary file in the configured repository path. The lifetime of
554     * the file is tied to the lifetime of the <code>FileItem</code> instance;
555     * the file will be deleted when the instance is garbage collected.
556     * <p>
557     * <b>Note: Subclasses that override this method must ensure that they return the
558     * same File each time.</b>
559     *
560     * @return The {@link java.io.File File} to be used for temporary storage.
561     */
562    protected File getTempFile() {
563        if (tempFile == null) {
564            File tempDir = repository;
565            if (tempDir == null) {
566                tempDir = new File(System.getProperty("java.io.tmpdir"));
567            }
568
569            String tempFileName = format("upload_%s_%s.tmp", UID, getUniqueId());
570
571            tempFile = new File(tempDir, tempFileName);
572        }
573        return tempFile;
574    }
575
576    // -------------------------------------------------------- Private methods
577
578    /**
579     * Returns an identifier that is unique within the class loader used to
580     * load this class, but does not have random-like appearance.
581     *
582     * @return A String with the non-random looking instance identifier.
583     */
584    private static String getUniqueId() {
585        final int limit = 100000000;
586        int current = COUNTER.getAndIncrement();
587        String id = Integer.toString(current);
588
589        // If you manage to get more than 100 million of ids, you'll
590        // start getting ids longer than 8 characters.
591        if (current < limit) {
592            id = ("00000000" + id).substring(id.length());
593        }
594        return id;
595    }
596
597    /**
598     * Returns a string representation of this object.
599     *
600     * @return a string representation of this object.
601     */
602    @Override
603    public String toString() {
604        return format("name=%s, StoreLocation=%s, size=%s bytes, isFormField=%s, FieldName=%s",
605                      getName(), getStoreLocation(), Long.valueOf(getSize()),
606                      Boolean.valueOf(isFormField()), getFieldName());
607    }
608
609    /**
610     * Returns the file item headers.
611     * @return The file items headers.
612     */
613    @Override
614    public FileItemHeaders getHeaders() {
615        return headers;
616    }
617
618    /**
619     * Sets the file item headers.
620     * @param pHeaders The file items headers.
621     */
622    @Override
623    public void setHeaders(FileItemHeaders pHeaders) {
624        headers = pHeaders;
625    }
626
627    /**
628     * Returns the default charset for use when no explicit charset
629     * parameter is provided by the sender.
630     * @return the default charset
631     */
632    public String getDefaultCharset() {
633        return defaultCharset;
634    }
635
636    /**
637     * Sets the default charset for use when no explicit charset
638     * parameter is provided by the sender.
639     * @param charset the default charset
640     */
641    public void setDefaultCharset(String charset) {
642        defaultCharset = charset;
643    }
644}