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 */
017
018package org.apache.commons.codec.net;
019
020import java.io.ByteArrayOutputStream;
021import java.io.UnsupportedEncodingException;
022import java.nio.charset.Charset;
023import java.nio.charset.IllegalCharsetNameException;
024import java.nio.charset.UnsupportedCharsetException;
025import java.util.BitSet;
026
027import org.apache.commons.codec.BinaryDecoder;
028import org.apache.commons.codec.BinaryEncoder;
029import org.apache.commons.codec.Charsets;
030import org.apache.commons.codec.DecoderException;
031import org.apache.commons.codec.EncoderException;
032import org.apache.commons.codec.StringDecoder;
033import org.apache.commons.codec.StringEncoder;
034import org.apache.commons.codec.binary.StringUtils;
035
036/**
037 * Codec for the Quoted-Printable section of <a href="http://www.ietf.org/rfc/rfc1521.txt">RFC 1521</a>.
038 * <p>
039 * The Quoted-Printable encoding is intended to represent data that largely consists of octets that correspond to
040 * printable characters in the ASCII character set. It encodes the data in such a way that the resulting octets are
041 * unlikely to be modified by mail transport. If the data being encoded are mostly ASCII text, the encoded form of the
042 * data remains largely recognizable by humans. A body which is entirely ASCII may also be encoded in Quoted-Printable
043 * to ensure the integrity of the data should the message pass through a character- translating, and/or line-wrapping
044 * gateway.
045 * <p>
046 * Note:
047 * <p>
048 * Depending on the selected {@code strict} parameter, this class will implement a different set of rules of the
049 * quoted-printable spec:
050 * <ul>
051 *   <li>{@code strict=false}: only rules #1 and #2 are implemented
052 *   <li>{@code strict=true}: all rules #1 through #5 are implemented
053 * </ul>
054 * Originally, this class only supported the non-strict mode, but the codec in this partial form could already be used
055 * for certain applications that do not require quoted-printable line formatting (rules #3, #4, #5), for instance
056 * Q codec. The strict mode has been added in 1.10.
057 * <p>
058 * This class is immutable and thread-safe.
059 *
060 * @see <a href="http://www.ietf.org/rfc/rfc1521.txt">RFC 1521 MIME (Multipurpose Internet Mail Extensions) Part One:
061 *          Mechanisms for Specifying and Describing the Format of Internet Message Bodies </a>
062 *
063 * @since 1.3
064 * @version $Id: QuotedPrintableCodec.java 1637069 2014-11-06 10:38:51Z tn $
065 */
066public class QuotedPrintableCodec implements BinaryEncoder, BinaryDecoder, StringEncoder, StringDecoder {
067    /**
068     * The default charset used for string decoding and encoding.
069     */
070    private final Charset charset;
071
072    /**
073     * Indicates whether soft line breaks shall be used during encoding (rule #3-5).
074     */
075    private final boolean strict;
076
077    /**
078     * BitSet of printable characters as defined in RFC 1521.
079     */
080    private static final BitSet PRINTABLE_CHARS = new BitSet(256);
081
082    private static final byte ESCAPE_CHAR = '=';
083
084    private static final byte TAB = 9;
085
086    private static final byte SPACE = 32;
087
088    private static final byte CR = 13;
089
090    private static final byte LF = 10;
091
092    /**
093     * Safe line length for quoted printable encoded text.
094     */
095    private static final int SAFE_LENGTH = 73;
096
097    // Static initializer for printable chars collection
098    static {
099        // alpha characters
100        for (int i = 33; i <= 60; i++) {
101            PRINTABLE_CHARS.set(i);
102        }
103        for (int i = 62; i <= 126; i++) {
104            PRINTABLE_CHARS.set(i);
105        }
106        PRINTABLE_CHARS.set(TAB);
107        PRINTABLE_CHARS.set(SPACE);
108    }
109
110    /**
111     * Default constructor, assumes default charset of {@link Charsets#UTF_8}
112     */
113    public QuotedPrintableCodec() {
114        this(Charsets.UTF_8, false);
115    }
116
117    /**
118     * Constructor which allows for the selection of the strict mode.
119     *
120     * @param strict
121     *            if {@code true}, soft line breaks will be used
122     * @since 1.10
123     */
124    public QuotedPrintableCodec(final boolean strict) {
125        this(Charsets.UTF_8, strict);
126    }
127
128    /**
129     * Constructor which allows for the selection of a default charset.
130     *
131     * @param charset
132     *            the default string charset to use.
133     * @since 1.7
134     */
135    public QuotedPrintableCodec(final Charset charset) {
136        this(charset, false);
137    }
138
139    /**
140     * Constructor which allows for the selection of a default charset and strict mode.
141     *
142     * @param charset
143     *            the default string charset to use.
144     * @param strict
145     *            if {@code true}, soft line breaks will be used
146     * @since 1.10
147     */
148    public QuotedPrintableCodec(final Charset charset, final boolean strict) {
149        this.charset = charset;
150        this.strict = strict;
151    }
152
153    /**
154     * Constructor which allows for the selection of a default charset.
155     *
156     * @param charsetName
157     *            the default string charset to use.
158     * @throws UnsupportedCharsetException
159     *             If no support for the named charset is available
160     *             in this instance of the Java virtual machine
161     * @throws IllegalArgumentException
162     *             If the given charsetName is null
163     * @throws IllegalCharsetNameException
164     *             If the given charset name is illegal
165     *
166     * @since 1.7 throws UnsupportedCharsetException if the named charset is unavailable
167     */
168    public QuotedPrintableCodec(final String charsetName)
169            throws IllegalCharsetNameException, IllegalArgumentException, UnsupportedCharsetException {
170        this(Charset.forName(charsetName), false);
171    }
172
173    /**
174     * Encodes byte into its quoted-printable representation.
175     *
176     * @param b
177     *            byte to encode
178     * @param buffer
179     *            the buffer to write to
180     * @return The number of bytes written to the <code>buffer</code>
181     */
182    private static final int encodeQuotedPrintable(final int b, final ByteArrayOutputStream buffer) {
183        buffer.write(ESCAPE_CHAR);
184        final char hex1 = Character.toUpperCase(Character.forDigit((b >> 4) & 0xF, 16));
185        final char hex2 = Character.toUpperCase(Character.forDigit(b & 0xF, 16));
186        buffer.write(hex1);
187        buffer.write(hex2);
188        return 3;
189    }
190
191    /**
192     * Return the byte at position <code>index</code> of the byte array and
193     * make sure it is unsigned.
194     *
195     * @param index
196     *            position in the array
197     * @param bytes
198     *            the byte array
199     * @return the unsigned octet at position <code>index</code> from the array
200     */
201    private static int getUnsignedOctet(final int index, final byte[] bytes) {
202        int b = bytes[index];
203        if (b < 0) {
204            b = 256 + b;
205        }
206        return b;
207    }
208
209    /**
210     * Write a byte to the buffer.
211     *
212     * @param b
213     *            byte to write
214     * @param encode
215     *            indicates whether the octet shall be encoded
216     * @param buffer
217     *            the buffer to write to
218     * @return the number of bytes that have been written to the buffer
219     */
220    private static int encodeByte(final int b, final boolean encode,
221                                  final ByteArrayOutputStream buffer) {
222        if (encode) {
223            return encodeQuotedPrintable(b, buffer);
224        } else {
225            buffer.write(b);
226            return 1;
227        }
228    }
229
230    /**
231     * Checks whether the given byte is whitespace.
232     *
233     * @param b
234     *            byte to be checked
235     * @return <code>true</code> if the byte is either a space or tab character
236     */
237    private static boolean isWhitespace(final int b) {
238        return b == SPACE || b == TAB;
239    }
240
241    /**
242     * Encodes an array of bytes into an array of quoted-printable 7-bit characters. Unsafe characters are escaped.
243     * <p>
244     * This function implements a subset of quoted-printable encoding specification (rule #1 and rule #2) as defined in
245     * RFC 1521 and is suitable for encoding binary data and unformatted text.
246     *
247     * @param printable
248     *            bitset of characters deemed quoted-printable
249     * @param bytes
250     *            array of bytes to be encoded
251     * @return array of bytes containing quoted-printable data
252     */
253    public static final byte[] encodeQuotedPrintable(BitSet printable, final byte[] bytes) {
254        return encodeQuotedPrintable(printable, bytes, false);
255    }
256
257    /**
258     * Encodes an array of bytes into an array of quoted-printable 7-bit characters. Unsafe characters are escaped.
259     * <p>
260     * Depending on the selection of the {@code strict} parameter, this function either implements the full ruleset
261     * or only a subset of quoted-printable encoding specification (rule #1 and rule #2) as defined in
262     * RFC 1521 and is suitable for encoding binary data and unformatted text.
263     *
264     * @param printable
265     *            bitset of characters deemed quoted-printable
266     * @param bytes
267     *            array of bytes to be encoded
268     * @param strict
269     *            if {@code true} the full ruleset is used, otherwise only rule #1 and rule #2
270     * @return array of bytes containing quoted-printable data
271     * @since 1.10
272     */
273    public static final byte[] encodeQuotedPrintable(BitSet printable, final byte[] bytes, boolean strict) {
274        if (bytes == null) {
275            return null;
276        }
277        if (printable == null) {
278            printable = PRINTABLE_CHARS;
279        }
280        final ByteArrayOutputStream buffer = new ByteArrayOutputStream();
281
282        if (strict) {
283            int pos = 1;
284            // encode up to buffer.length - 3, the last three octets will be treated
285            // separately for simplification of note #3
286            for (int i = 0; i < bytes.length - 3; i++) {
287                int b = getUnsignedOctet(i, bytes);
288                if (pos < SAFE_LENGTH) {
289                    // up to this length it is safe to add any byte, encoded or not
290                    pos += encodeByte(b, !printable.get(b), buffer);
291                } else {
292                    // rule #3: whitespace at the end of a line *must* be encoded
293                    encodeByte(b, !printable.get(b) || isWhitespace(b), buffer);
294
295                    // rule #5: soft line break
296                    buffer.write(ESCAPE_CHAR);
297                    buffer.write(CR);
298                    buffer.write(LF);
299                    pos = 1;
300                }
301            }
302
303            // rule #3: whitespace at the end of a line *must* be encoded
304            // if we would do a soft break line after this octet, encode whitespace
305            int b = getUnsignedOctet(bytes.length - 3, bytes);
306            boolean encode = !printable.get(b) || (isWhitespace(b) && pos > SAFE_LENGTH - 5);
307            pos += encodeByte(b, encode, buffer);
308
309            // note #3: '=' *must not* be the ultimate or penultimate character
310            // simplification: if < 6 bytes left, do a soft line break as we may need
311            //                 exactly 6 bytes space for the last 2 bytes
312            if (pos > SAFE_LENGTH - 2) {
313                buffer.write(ESCAPE_CHAR);
314                buffer.write(CR);
315                buffer.write(LF);
316            }
317            for (int i = bytes.length - 2; i < bytes.length; i++) {
318                b = getUnsignedOctet(i, bytes);
319                // rule #3: trailing whitespace shall be encoded
320                encode = !printable.get(b) || (i > bytes.length - 2 && isWhitespace(b));
321                encodeByte(b, encode, buffer);
322            }
323        } else {
324            for (final byte c : bytes) {
325                int b = c;
326                if (b < 0) {
327                    b = 256 + b;
328                }
329                if (printable.get(b)) {
330                    buffer.write(b);
331                } else {
332                    encodeQuotedPrintable(b, buffer);
333                }
334            }
335        }
336        return buffer.toByteArray();
337    }
338
339    /**
340     * Decodes an array quoted-printable characters into an array of original bytes. Escaped characters are converted
341     * back to their original representation.
342     * <p>
343     * This function fully implements the quoted-printable encoding specification (rule #1 through rule #5) as
344     * defined in RFC 1521.
345     *
346     * @param bytes
347     *            array of quoted-printable characters
348     * @return array of original bytes
349     * @throws DecoderException
350     *             Thrown if quoted-printable decoding is unsuccessful
351     */
352    public static final byte[] decodeQuotedPrintable(final byte[] bytes) throws DecoderException {
353        if (bytes == null) {
354            return null;
355        }
356        final ByteArrayOutputStream buffer = new ByteArrayOutputStream();
357        for (int i = 0; i < bytes.length; i++) {
358            final int b = bytes[i];
359            if (b == ESCAPE_CHAR) {
360                try {
361                    // if the next octet is a CR we have found a soft line break
362                    if (bytes[++i] == CR) {
363                        continue;
364                    }
365                    final int u = Utils.digit16(bytes[i]);
366                    final int l = Utils.digit16(bytes[++i]);
367                    buffer.write((char) ((u << 4) + l));
368                } catch (final ArrayIndexOutOfBoundsException e) {
369                    throw new DecoderException("Invalid quoted-printable encoding", e);
370                }
371            } else if (b != CR && b != LF) {
372                // every other octet is appended except for CR & LF
373                buffer.write(b);
374            }
375        }
376        return buffer.toByteArray();
377    }
378
379    /**
380     * Encodes an array of bytes into an array of quoted-printable 7-bit characters. Unsafe characters are escaped.
381     * <p>
382     * Depending on the selection of the {@code strict} parameter, this function either implements the full ruleset
383     * or only a subset of quoted-printable encoding specification (rule #1 and rule #2) as defined in
384     * RFC 1521 and is suitable for encoding binary data and unformatted text.
385     *
386     * @param bytes
387     *            array of bytes to be encoded
388     * @return array of bytes containing quoted-printable data
389     */
390    @Override
391    public byte[] encode(final byte[] bytes) {
392        return encodeQuotedPrintable(PRINTABLE_CHARS, bytes, strict);
393    }
394
395    /**
396     * Decodes an array of quoted-printable characters into an array of original bytes. Escaped characters are converted
397     * back to their original representation.
398     * <p>
399     * This function fully implements the quoted-printable encoding specification (rule #1 through rule #5) as
400     * defined in RFC 1521.
401     *
402     * @param bytes
403     *            array of quoted-printable characters
404     * @return array of original bytes
405     * @throws DecoderException
406     *             Thrown if quoted-printable decoding is unsuccessful
407     */
408    @Override
409    public byte[] decode(final byte[] bytes) throws DecoderException {
410        return decodeQuotedPrintable(bytes);
411    }
412
413    /**
414     * Encodes a string into its quoted-printable form using the default string charset. Unsafe characters are escaped.
415     * <p>
416     * Depending on the selection of the {@code strict} parameter, this function either implements the full ruleset
417     * or only a subset of quoted-printable encoding specification (rule #1 and rule #2) as defined in
418     * RFC 1521 and is suitable for encoding binary data and unformatted text.
419     *
420     * @param str
421     *            string to convert to quoted-printable form
422     * @return quoted-printable string
423     * @throws EncoderException
424     *             Thrown if quoted-printable encoding is unsuccessful
425     *
426     * @see #getCharset()
427     */
428    @Override
429    public String encode(final String str) throws EncoderException {
430        return this.encode(str, getCharset());
431    }
432
433    /**
434     * Decodes a quoted-printable string into its original form using the specified string charset. Escaped characters
435     * are converted back to their original representation.
436     *
437     * @param str
438     *            quoted-printable string to convert into its original form
439     * @param charset
440     *            the original string charset
441     * @return original string
442     * @throws DecoderException
443     *             Thrown if quoted-printable decoding is unsuccessful
444     * @since 1.7
445     */
446    public String decode(final String str, final Charset charset) throws DecoderException {
447        if (str == null) {
448            return null;
449        }
450        return new String(this.decode(StringUtils.getBytesUsAscii(str)), charset);
451    }
452
453    /**
454     * Decodes a quoted-printable string into its original form using the specified string charset. Escaped characters
455     * are converted back to their original representation.
456     *
457     * @param str
458     *            quoted-printable string to convert into its original form
459     * @param charset
460     *            the original string charset
461     * @return original string
462     * @throws DecoderException
463     *             Thrown if quoted-printable decoding is unsuccessful
464     * @throws UnsupportedEncodingException
465     *             Thrown if charset is not supported
466     */
467    public String decode(final String str, final String charset) throws DecoderException, UnsupportedEncodingException {
468        if (str == null) {
469            return null;
470        }
471        return new String(decode(StringUtils.getBytesUsAscii(str)), charset);
472    }
473
474    /**
475     * Decodes a quoted-printable string into its original form using the default string charset. Escaped characters are
476     * converted back to their original representation.
477     *
478     * @param str
479     *            quoted-printable string to convert into its original form
480     * @return original string
481     * @throws DecoderException
482     *             Thrown if quoted-printable decoding is unsuccessful. Thrown if charset is not supported.
483     * @see #getCharset()
484     */
485    @Override
486    public String decode(final String str) throws DecoderException {
487        return this.decode(str, this.getCharset());
488    }
489
490    /**
491     * Encodes an object into its quoted-printable safe form. Unsafe characters are escaped.
492     *
493     * @param obj
494     *            string to convert to a quoted-printable form
495     * @return quoted-printable object
496     * @throws EncoderException
497     *             Thrown if quoted-printable encoding is not applicable to objects of this type or if encoding is
498     *             unsuccessful
499     */
500    @Override
501    public Object encode(final Object obj) throws EncoderException {
502        if (obj == null) {
503            return null;
504        } else if (obj instanceof byte[]) {
505            return encode((byte[]) obj);
506        } else if (obj instanceof String) {
507            return encode((String) obj);
508        } else {
509            throw new EncoderException("Objects of type " +
510                  obj.getClass().getName() +
511                  " cannot be quoted-printable encoded");
512        }
513    }
514
515    /**
516     * Decodes a quoted-printable object into its original form. Escaped characters are converted back to their original
517     * representation.
518     *
519     * @param obj
520     *            quoted-printable object to convert into its original form
521     * @return original object
522     * @throws DecoderException
523     *             Thrown if the argument is not a <code>String</code> or <code>byte[]</code>. Thrown if a failure
524     *             condition is encountered during the decode process.
525     */
526    @Override
527    public Object decode(final Object obj) throws DecoderException {
528        if (obj == null) {
529            return null;
530        } else if (obj instanceof byte[]) {
531            return decode((byte[]) obj);
532        } else if (obj instanceof String) {
533            return decode((String) obj);
534        } else {
535            throw new DecoderException("Objects of type " +
536                  obj.getClass().getName() +
537                  " cannot be quoted-printable decoded");
538        }
539    }
540
541    /**
542     * Gets the default charset name used for string decoding and encoding.
543     *
544     * @return the default charset name
545     * @since 1.7
546     */
547    public Charset getCharset() {
548        return this.charset;
549    }
550
551    /**
552     * Gets the default charset name used for string decoding and encoding.
553     *
554     * @return the default charset name
555     */
556    public String getDefaultCharset() {
557        return this.charset.name();
558    }
559
560    /**
561     * Encodes a string into its quoted-printable form using the specified charset. Unsafe characters are escaped.
562     * <p>
563     * Depending on the selection of the {@code strict} parameter, this function either implements the full ruleset
564     * or only a subset of quoted-printable encoding specification (rule #1 and rule #2) as defined in
565     * RFC 1521 and is suitable for encoding binary data and unformatted text.
566     *
567     * @param str
568     *            string to convert to quoted-printable form
569     * @param charset
570     *            the charset for str
571     * @return quoted-printable string
572     * @since 1.7
573     */
574    public String encode(final String str, final Charset charset) {
575        if (str == null) {
576            return null;
577        }
578        return StringUtils.newStringUsAscii(this.encode(str.getBytes(charset)));
579    }
580
581    /**
582     * Encodes a string into its quoted-printable form using the specified charset. Unsafe characters are escaped.
583     * <p>
584     * Depending on the selection of the {@code strict} parameter, this function either implements the full ruleset
585     * or only a subset of quoted-printable encoding specification (rule #1 and rule #2) as defined in
586     * RFC 1521 and is suitable for encoding binary data and unformatted text.
587     *
588     * @param str
589     *            string to convert to quoted-printable form
590     * @param charset
591     *            the charset for str
592     * @return quoted-printable string
593     * @throws UnsupportedEncodingException
594     *             Thrown if the charset is not supported
595     */
596    public String encode(final String str, final String charset) throws UnsupportedEncodingException {
597        if (str == null) {
598            return null;
599        }
600        return StringUtils.newStringUsAscii(encode(str.getBytes(charset)));
601    }
602}