View Javadoc

1   /**
2    * Licensed to the Apache Software Foundation (ASF) under one
3    * or more contributor license agreements.  See the NOTICE file
4    * distributed with this work for additional information
5    * regarding copyright ownership.  The ASF licenses this file
6    * to you under the Apache License, Version 2.0 (the
7    * "License"); you may not use this file except in compliance
8    * with the License.  You may obtain a copy of the License at
9    *
10   *     http://www.apache.org/licenses/LICENSE-2.0
11   *
12   * Unless required by applicable law or agreed to in writing, software
13   * distributed under the License is distributed on an "AS IS" BASIS,
14   * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
15   * See the License for the specific language governing permissions and
16   * limitations under the License.
17   */
18  package org.apache.hadoop.hbase.util;
19  
20  import static org.apache.hadoop.hbase.util.Order.ASCENDING;
21  import static org.apache.hadoop.hbase.util.Order.DESCENDING;
22  
23  import java.math.BigDecimal;
24  import java.math.BigInteger;
25  import java.math.MathContext;
26  import java.math.RoundingMode;
27  import java.nio.charset.Charset;
28  
29  import org.apache.hadoop.hbase.classification.InterfaceAudience;
30  import org.apache.hadoop.hbase.classification.InterfaceStability;
31  
32  /**
33   * Utility class that handles ordered byte arrays. That is, unlike
34   * {@link Bytes}, these methods produce byte arrays which maintain the sort
35   * order of the original values.
36   * <h3>Encoding Format summary</h3>
37   * <p>
38   * Each value is encoded as one or more bytes. The first byte of the encoding,
39   * its meaning, and a terse description of the bytes that follow is given by
40   * the following table:
41   * </p>
42   * <table summary="Encodings">
43   * <tr><th>Content Type</th><th>Encoding</th></tr>
44   * <tr><td>NULL</td><td>0x05</td></tr>
45   * <tr><td>negative infinity</td><td>0x07</td></tr>
46   * <tr><td>negative large</td><td>0x08, ~E, ~M</td></tr>
47   * <tr><td>negative medium</td><td>0x13-E, ~M</td></tr>
48   * <tr><td>negative small</td><td>0x14, -E, ~M</td></tr>
49   * <tr><td>zero</td><td>0x15</td></tr>
50   * <tr><td>positive small</td><td>0x16, ~-E, M</td></tr>
51   * <tr><td>positive medium</td><td>0x17+E, M</td></tr>
52   * <tr><td>positive large</td><td>0x22, E, M</td></tr>
53   * <tr><td>positive infinity</td><td>0x23</td></tr>
54   * <tr><td>NaN</td><td>0x25</td></tr>
55   * <tr><td>fixed-length 32-bit integer</td><td>0x27, I</td></tr>
56   * <tr><td>fixed-length 64-bit integer</td><td>0x28, I</td></tr>
57   * <tr><td>fixed-length 8-bit integer</td><td>0x29</td></tr>
58   * <tr><td>fixed-length 16-bit integer</td><td>0x2a</td></tr>
59   * <tr><td>fixed-length 32-bit float</td><td>0x30, F</td></tr>
60   * <tr><td>fixed-length 64-bit float</td><td>0x31, F</td></tr>
61   * <tr><td>TEXT</td><td>0x33, T</td></tr>
62   * <tr><td>variable length BLOB</td><td>0x35, B</td></tr>
63   * <tr><td>byte-for-byte BLOB</td><td>0x36, X</td></tr>
64   * </table>
65   *
66   * <h3>Null Encoding</h3>
67   * <p>
68   * Each value that is a NULL encodes as a single byte of 0x05. Since every
69   * other value encoding begins with a byte greater than 0x05, this forces NULL
70   * values to sort first.
71   * </p>
72   * <h3>Text Encoding</h3>
73   * <p>
74   * Each text value begins with a single byte of 0x33 and ends with a single
75   * byte of 0x00. There are zero or more intervening bytes that encode the text
76   * value. The intervening bytes are chosen so that the encoding will sort in
77   * the desired collating order. The intervening bytes may not contain a 0x00
78   * character; the only 0x00 byte allowed in a text encoding is the final byte.
79   * </p>
80   * <p>
81   * The text encoding ends in 0x00 in order to ensure that when there are two
82   * strings where one is a prefix of the other that the shorter string will
83   * sort first.
84   * </p>
85   * <h3>Binary Encoding</h3>
86   * <p>
87   * There are two encoding strategies for binary fields, referred to as
88   * "BlobVar" and "BlobCopy". BlobVar is less efficient in both space and
89   * encoding time. It has no limitations on the range of encoded values.
90   * BlobCopy is a byte-for-byte copy of the input data followed by a
91   * termination byte. It is extremely fast to encode and decode. It carries the
92   * restriction of not allowing a 0x00 value in the input byte[] as this value
93   * is used as the termination byte.
94   * </p>
95   * <h4>BlobVar</h4>
96   * <p>
97   * "BlobVar" encodes the input byte[] in a manner similar to a variable length
98   * integer encoding. As with the other {@code OrderedBytes} encodings,
99   * the first encoded byte is used to indicate what kind of value follows. This
100  * header byte is 0x37 for BlobVar encoded values. As with the traditional
101  * varint encoding, the most significant bit of each subsequent encoded
102  * {@code byte} is used as a continuation marker. The 7 remaining bits
103  * contain the 7 most significant bits of the first unencoded byte. The next
104  * encoded byte starts with a continuation marker in the MSB. The least
105  * significant bit from the first unencoded byte follows, and the remaining 6
106  * bits contain the 6 MSBs of the second unencoded byte. The encoding
107  * continues, encoding 7 bytes on to 8 encoded bytes. The MSB of the final
108  * encoded byte contains a termination marker rather than a continuation
109  * marker, and any remaining bits from the final input byte. Any trailing bits
110  * in the final encoded byte are zeros.
111  * </p>
112  * <h4>BlobCopy</h4>
113  * <p>
114  * "BlobCopy" is a simple byte-for-byte copy of the input data. It uses 0x38
115  * as the header byte, and is terminated by 0x00 in the DESCENDING case. This
116  * alternative encoding is faster and more space-efficient, but it cannot
117  * accept values containing a 0x00 byte in DESCENDING order.
118  * </p>
119  * <h3>Variable-length Numeric Encoding</h3>
120  * <p>
121  * Numeric values must be coded so as to sort in numeric order. We assume that
122  * numeric values can be both integer and floating point values. Clients must
123  * be careful to use inspection methods for encoded values (such as
124  * {@link #isNumericInfinite(PositionedByteRange)} and
125  * {@link #isNumericNaN(PositionedByteRange)} to protect against decoding
126  * values into object which do not support these numeric concepts (such as
127  * {@link Long} and {@link BigDecimal}).
128  * </p>
129  * <p>
130  * Simplest cases first: If the numeric value is a NaN, then the encoding is a
131  * single byte of 0x25. This causes NaN values to sort after every other
132  * numeric value.
133  * </p>
134  * <p>
135  * If the numeric value is a negative infinity then the encoding is a single
136  * byte of 0x07. Since every other numeric value except NaN has a larger
137  * initial byte, this encoding ensures that negative infinity will sort prior
138  * to every other numeric value other than NaN.
139  * </p>
140  * <p>
141  * If the numeric value is a positive infinity then the encoding is a single
142  * byte of 0x23. Every other numeric value encoding begins with a smaller
143  * byte, ensuring that positive infinity always sorts last among numeric
144  * values. 0x23 is also smaller than 0x33, the initial byte of a text value,
145  * ensuring that every numeric value sorts before every text value.
146  * </p>
147  * <p>
148  * If the numeric value is exactly zero then it is encoded as a single byte of
149  * 0x15. Finite negative values will have initial bytes of 0x08 through 0x14
150  * and finite positive values will have initial bytes of 0x16 through 0x22.
151  * </p>
152  * <p>
153  * For all numeric values, we compute a mantissa M and an exponent E. The
154  * mantissa is a base-100 representation of the value. The exponent E
155  * determines where to put the decimal point.
156  * </p>
157  * <p>
158  * Each centimal digit of the mantissa is stored in a byte. If the value of
159  * the centimal digit is X (hence X&ge;0 and X&le;99) then the byte value will
160  * be 2*X+1 for every byte of the mantissa, except for the last byte which
161  * will be 2*X+0. The mantissa must be the minimum number of bytes necessary
162  * to represent the value; trailing X==0 digits are omitted. This means that
163  * the mantissa will never contain a byte with the value 0x00.
164  * </p>
165  * <p>
166  * If we assume all digits of the mantissa occur to the right of the decimal
167  * point, then the exponent E is the power of one hundred by which one must
168  * multiply the mantissa to recover the original value.
169  * </p>
170  * <p>
171  * Values are classified as large, medium, or small according to the value of
172  * E. If E is 11 or more, the value is large. For E between 0 and 10, the
173  * value is medium. For E less than zero, the value is small.
174  * </p>
175  * <p>
176  * Large positive values are encoded as a single byte 0x22 followed by E as a
177  * varint and then M. Medium positive values are a single byte of 0x17+E
178  * followed by M. Small positive values are encoded as a single byte 0x16
179  * followed by the ones-complement of the varint for -E followed by M.
180  * </p>
181  * <p>
182  * Small negative values are encoded as a single byte 0x14 followed by -E as a
183  * varint and then the ones-complement of M. Medium negative values are
184  * encoded as a byte 0x13-E followed by the ones-complement of M. Large
185  * negative values consist of the single byte 0x08 followed by the
186  * ones-complement of the varint encoding of E followed by the ones-complement
187  * of M.
188  * </p>
189  * <h3>Fixed-length Integer Encoding</h3>
190  * <p>
191  * All 4-byte integers are serialized to a 5-byte, fixed-width, sortable byte
192  * format. All 8-byte integers are serialized to the equivelant 9-byte format.
193  * Serialization is performed by writing a header byte, inverting the integer
194  * sign bit and writing the resulting bytes to the byte array in big endian
195  * order.
196  * </p>
197  * <h3>Fixed-length Floating Point Encoding</h3>
198  * <p>
199  * 32-bit and 64-bit floating point numbers are encoded to a 5-byte and 9-byte
200  * encoding format, respectively. The format is identical, save for the
201  * precision respected in each step of the operation.
202  * <p>
203  * This format ensures the following total ordering of floating point values:
204  * Float.NEGATIVE_INFINITY &lt; -Float.MAX_VALUE &lt; ... &lt;
205  * -Float.MIN_VALUE &lt; -0.0 &lt; +0.0; &lt; Float.MIN_VALUE &lt; ... &lt;
206  * Float.MAX_VALUE &lt; Float.POSITIVE_INFINITY &lt; Float.NaN
207  * </p>
208  * <p>
209  * Floating point numbers are encoded as specified in IEEE 754. A 32-bit
210  * single precision float consists of a sign bit, 8-bit unsigned exponent
211  * encoded in offset-127 notation, and a 23-bit significand. The format is
212  * described further in the <a
213  * href="http://en.wikipedia.org/wiki/Single_precision"> Single Precision
214  * Floating Point Wikipedia page</a>
215  * </p>
216  * <p>
217  * The value of a normal float is -1 <sup>sign bit</sup> &times;
218  * 2<sup>exponent - 127</sup> &times; 1.significand
219  * </p>
220  * <p>
221  * The IEE754 floating point format already preserves sort ordering for
222  * positive floating point numbers when the raw bytes are compared in most
223  * significant byte order. This is discussed further at <a href=
224  * "http://www.cygnus-software.com/papers/comparingfloats/comparingfloats.htm">
225  * http://www.cygnus-software.com/papers/comparingfloats/comparingfloats.htm</a>
226  * </p>
227  * <p>
228  * Thus, we need only ensure that negative numbers sort in the the exact
229  * opposite order as positive numbers (so that say, negative infinity is less
230  * than negative 1), and that all negative numbers compare less than any
231  * positive number. To accomplish this, we invert the sign bit of all floating
232  * point numbers, and we also invert the exponent and significand bits if the
233  * floating point number was negative.
234  * </p>
235  * <p>
236  * More specifically, we first store the floating point bits into a 32-bit int
237  * {@code j} using {@link Float#floatToIntBits}. This method collapses
238  * all NaNs into a single, canonical NaN value but otherwise leaves the bits
239  * unchanged. We then compute
240  * </p>
241  *
242  * <pre>
243  * j &circ;= (j &gt;&gt; (Integer.SIZE - 1)) | Integer.MIN_SIZE
244  * </pre>
245  * <p>
246  * which inverts the sign bit and XOR's all other bits with the sign bit
247  * itself. Comparing the raw bytes of {@code j} in most significant byte
248  * order is equivalent to performing a single precision floating point
249  * comparison on the underlying bits (ignoring NaN comparisons, as NaNs don't
250  * compare equal to anything when performing floating point comparisons).
251  * </p>
252  * <p>
253  * The resulting integer is then converted into a byte array by serializing
254  * the integer one byte at a time in most significant byte order. The
255  * serialized integer is prefixed by a single header byte. All serialized
256  * values are 5 bytes in length.
257  * </p>
258  * <p>
259  * {@code OrderedBytes} encodings are heavily influenced by the
260  * <a href="http://sqlite.org/src4/doc/trunk/www/key_encoding.wiki">SQLite4 Key
261  * Encoding</a>. Slight deviations are make in the interest of order
262  * correctness and user extensibility. Fixed-width {@code Long} and
263  * {@link Double} encodings are based on implementations from the now defunct
264  * Orderly library.
265  * </p>
266  */
267 @InterfaceAudience.Public
268 @InterfaceStability.Evolving
269 public class OrderedBytes {
270 
271   /*
272    * These constants define header bytes used to identify encoded values. Note
273    * that the values here are not exhaustive as the Numeric format encodes
274    * portions of its value within the header byte. The values listed here are
275    * directly applied to persisted data -- DO NOT modify the values specified
276    * here. Instead, gaps are placed intentionally between values so that new
277    * implementations can be inserted into the total ordering enforced here.
278    */
279   private static final byte NULL = 0x05;
280   // room for 1 expansion type
281   private static final byte NEG_INF = 0x07;
282   private static final byte NEG_LARGE = 0x08;
283   private static final byte NEG_MED_MIN = 0x09;
284   private static final byte NEG_MED_MAX = 0x13;
285   private static final byte NEG_SMALL = 0x14;
286   private static final byte ZERO = 0x15;
287   private static final byte POS_SMALL = 0x16;
288   private static final byte POS_MED_MIN = 0x17;
289   private static final byte POS_MED_MAX = 0x21;
290   private static final byte POS_LARGE = 0x22;
291   private static final byte POS_INF = 0x23;
292   // room for 2 expansion type
293   private static final byte NAN = 0x26;
294   // room for 2 expansion types
295   private static final byte FIXED_INT8 = 0x29;
296   private static final byte FIXED_INT16 = 0x2a;
297   private static final byte FIXED_INT32 = 0x2b;
298   private static final byte FIXED_INT64 = 0x2c;
299   // room for 3 expansion types
300   private static final byte FIXED_FLOAT32 = 0x30;
301   private static final byte FIXED_FLOAT64 = 0x31;
302   // room for 2 expansion type
303   private static final byte TEXT = 0x34;
304   // room for 2 expansion type
305   private static final byte BLOB_VAR = 0x37;
306   private static final byte BLOB_COPY = 0x38;
307 
308   /*
309    * The following constant values are used by encoding implementations
310    */
311 
312   public static final Charset UTF8 = Charset.forName("UTF-8");
313   private static final byte TERM = 0x00;
314   private static final BigDecimal E8 = BigDecimal.valueOf(1e8);
315   private static final BigDecimal E32 = BigDecimal.valueOf(1e32);
316   private static final BigDecimal EN2 = BigDecimal.valueOf(1e-2);
317   private static final BigDecimal EN10 = BigDecimal.valueOf(1e-10);
318 
319   /**
320    * Max precision guaranteed to fit into a {@code long}.
321    */
322   public static final int MAX_PRECISION = 31;
323 
324   /**
325    * The context used to normalize {@link BigDecimal} values.
326    */
327   public static final MathContext DEFAULT_MATH_CONTEXT =
328       new MathContext(MAX_PRECISION, RoundingMode.HALF_UP);
329 
330   /**
331    * Creates the standard exception when the encoded header byte is unexpected for the decoding
332    * context.
333    * @param header value used in error message.
334    */
335   private static IllegalArgumentException unexpectedHeader(byte header) {
336     throw new IllegalArgumentException("unexpected value in first byte: 0x"
337         + Long.toHexString(header));
338   }
339 
340   /**
341    * Perform unsigned comparison between two long values. Conforms to the same interface as
342    * {@link Comparator#compare(Object, Object)}.
343    */
344   private static int unsignedCmp(long x1, long x2) {
345     int cmp;
346     if ((cmp = (x1 < x2 ? -1 : (x1 == x2 ? 0 : 1))) == 0) return 0;
347     // invert the result when either value is negative
348     if ((x1 < 0) != (x2 < 0)) return -cmp;
349     return cmp;
350   }
351 
352   /**
353    * Write a 32-bit unsigned integer to {@code dst} as 4 big-endian bytes.
354    * @return number of bytes written.
355    */
356   private static int putUint32(PositionedByteRange dst, int val) {
357     dst.put((byte) (val >>> 24))
358        .put((byte) (val >>> 16))
359        .put((byte) (val >>> 8))
360        .put((byte) val);
361     return 4;
362   }
363 
364   /**
365    * Encode an unsigned 64-bit unsigned integer {@code val} into {@code dst}.
366    * @param dst The destination to which encoded bytes are written.
367    * @param val The value to write.
368    * @param comp Compliment the encoded value when {@code comp} is true.
369    * @return number of bytes written.
370    */
371   @InterfaceAudience.Private
372   static int putVaruint64(PositionedByteRange dst, long val, boolean comp) {
373     int w, y, len = 0;
374     final int offset = dst.getOffset(), start = dst.getPosition();
375     byte[] a = dst.getBytes();
376     Order ord = comp ? DESCENDING : ASCENDING;
377     if (-1 == unsignedCmp(val, 241L)) {
378       dst.put((byte) val);
379       len = dst.getPosition() - start;
380       ord.apply(a, offset + start, len);
381       return len;
382     }
383     if (-1 == unsignedCmp(val, 2288L)) {
384       y = (int) (val - 240);
385       dst.put((byte) (y / 256 + 241))
386          .put((byte) (y % 256));
387       len = dst.getPosition() - start;
388       ord.apply(a, offset + start, len);
389       return len;
390     }
391     if (-1 == unsignedCmp(val, 67824L)) {
392       y = (int) (val - 2288);
393       dst.put((byte) 249)
394          .put((byte) (y / 256))
395          .put((byte) (y % 256));
396       len = dst.getPosition() - start;
397       ord.apply(a, offset + start, len);
398       return len;
399     }
400     y = (int) val;
401     w = (int) (val >>> 32);
402     if (w == 0) {
403       if (-1 == unsignedCmp(y, 16777216L)) {
404         dst.put((byte) 250)
405            .put((byte) (y >>> 16))
406            .put((byte) (y >>> 8))
407            .put((byte) y);
408         len = dst.getPosition() - start;
409         ord.apply(a, offset + start, len);
410         return len;
411       }
412       dst.put((byte) 251);
413       putUint32(dst, y);
414       len = dst.getPosition() - start;
415       ord.apply(a, offset + start, len);
416       return len;
417     }
418     if (-1 == unsignedCmp(w, 256L)) {
419       dst.put((byte) 252)
420          .put((byte) w);
421       putUint32(dst, y);
422       len = dst.getPosition() - start;
423       ord.apply(a, offset + start, len);
424       return len;
425     }
426     if (-1 == unsignedCmp(w, 65536L)) {
427       dst.put((byte) 253)
428          .put((byte) (w >>> 8))
429          .put((byte) w);
430       putUint32(dst, y);
431       len = dst.getPosition() - start;
432       ord.apply(a, offset + start, len);
433       return len;
434     }
435     if (-1 == unsignedCmp(w, 16777216L)) {
436       dst.put((byte) 254)
437          .put((byte) (w >>> 16))
438          .put((byte) (w >>> 8))
439          .put((byte) w);
440       putUint32(dst, y);
441       len = dst.getPosition() - start;
442       ord.apply(a, offset + start, len);
443       return len;
444     }
445     dst.put((byte) 255);
446     putUint32(dst, w);
447     putUint32(dst, y);
448     len = dst.getPosition() - start;
449     ord.apply(a, offset + start, len);
450     return len;
451   }
452 
453   /**
454    * Inspect {@code src} for an encoded varuint64 for its length in bytes.
455    * Preserves the state of {@code src}.
456    * @param src source buffer
457    * @param comp if true, parse the compliment of the value.
458    * @return the number of bytes consumed by this value.
459    */
460   @InterfaceAudience.Private
461   static int lengthVaruint64(PositionedByteRange src, boolean comp) {
462     int a0 = (comp ? DESCENDING : ASCENDING).apply(src.peek()) & 0xff;
463     if (a0 <= 240) return 1;
464     if (a0 <= 248) return 2;
465     if (a0 == 249) return 3;
466     if (a0 == 250) return 4;
467     if (a0 == 251) return 5;
468     if (a0 == 252) return 6;
469     if (a0 == 253) return 7;
470     if (a0 == 254) return 8;
471     if (a0 == 255) return 9;
472     throw unexpectedHeader(src.peek());
473   }
474 
475   /**
476    * Skip {@code src} over the encoded varuint64.
477    * @param src source buffer
478    * @param cmp if true, parse the compliment of the value.
479    * @return the number of bytes skipped.
480    */
481   @InterfaceAudience.Private
482   static int skipVaruint64(PositionedByteRange src, boolean cmp) {
483     final int len = lengthVaruint64(src, cmp);
484     src.setPosition(src.getPosition() + len);
485     return len;
486   }
487 
488   /**
489    * Decode a sequence of bytes in {@code src} as a varuint64. Compliment the
490    * encoded value when {@code comp} is true.
491    * @return the decoded value.
492    */
493   @InterfaceAudience.Private
494   static long getVaruint64(PositionedByteRange src, boolean comp) {
495     assert src.getRemaining() >= lengthVaruint64(src, comp);
496     final long ret;
497     Order ord = comp ? DESCENDING : ASCENDING;
498     byte x = src.get();
499     final int a0 = ord.apply(x) & 0xff, a1, a2, a3, a4, a5, a6, a7, a8;
500     if (-1 == unsignedCmp(a0, 241)) {
501       return a0;
502     }
503     x = src.get();
504     a1 = ord.apply(x) & 0xff;
505     if (-1 == unsignedCmp(a0, 249)) {
506       return (a0 - 241L) * 256 + a1 + 240;
507     }
508     x = src.get();
509     a2 = ord.apply(x) & 0xff;
510     if (a0 == 249) {
511       return 2288L + 256 * a1 + a2;
512     }
513     x = src.get();
514     a3 = ord.apply(x) & 0xff;
515     if (a0 == 250) {
516       return ((long) a1 << 16L) | (a2 << 8) | a3;
517     }
518     x = src.get();
519     a4 = ord.apply(x) & 0xff;
520     ret = (((long) a1) << 24) | (a2 << 16) | (a3 << 8) | a4;
521     if (a0 == 251) {
522       return ret;
523     }
524     x = src.get();
525     a5 = ord.apply(x) & 0xff;
526     if (a0 == 252) {
527       return (ret << 8) | a5;
528     }
529     x = src.get();
530     a6 = ord.apply(x) & 0xff;
531     if (a0 == 253) {
532       return (ret << 16) | (a5 << 8) | a6;
533     }
534     x = src.get();
535     a7 = ord.apply(x) & 0xff;
536     if (a0 == 254) {
537       return (ret << 24) | (a5 << 16) | (a6 << 8) | a7;
538     }
539     x = src.get();
540     a8 = ord.apply(x) & 0xff;
541     return (ret << 32) | (((long) a5) << 24) | (a6 << 16) | (a7 << 8) | a8;
542   }
543 
544   /**
545    * Strip all trailing zeros to ensure that no digit will be zero and round
546    * using our default context to ensure precision doesn't exceed max allowed.
547    * From Phoenix's {@code NumberUtil}.
548    * @return new {@link BigDecimal} instance
549    */
550   @InterfaceAudience.Private
551   static BigDecimal normalize(BigDecimal val) {
552     return null == val ? null : val.stripTrailingZeros().round(DEFAULT_MATH_CONTEXT);
553   }
554 
555   /**
556    * Read significand digits from {@code src} according to the magnitude
557    * of {@code e}.
558    * @param src The source from which to read encoded digits.
559    * @param e The magnitude of the first digit read.
560    * @param comp Treat encoded bytes as compliments when {@code comp} is true.
561    * @return The decoded value.
562    * @throws IllegalArgumentException when read exceeds the remaining length
563    *     of {@code src}.
564    */
565   private static BigDecimal decodeSignificand(PositionedByteRange src, int e, boolean comp) {
566     // TODO: can this be made faster?
567     byte[] a = src.getBytes();
568     final int start = src.getPosition(), offset = src.getOffset(), remaining = src.getRemaining();
569     Order ord = comp ? DESCENDING : ASCENDING;
570     BigDecimal m = BigDecimal.ZERO;
571     e--;
572     for (int i = 0;; i++) {
573       if (i > remaining) {
574         // we've exceeded this range's window
575         src.setPosition(start);
576         throw new IllegalArgumentException(
577             "Read exceeds range before termination byte found. offset: " + offset + " position: "
578                 + (start + i));
579       }
580       // base-100 digits are encoded as val * 2 + 1 except for the termination digit.
581       m = m.add( // m +=
582         new BigDecimal(BigInteger.ONE, e * -2).multiply( // 100 ^ p * [decoded digit]
583           BigDecimal.valueOf((ord.apply(a[offset + start + i]) & 0xff) / 2)));
584       e--;
585       // detect termination digit
586       if ((ord.apply(a[offset + start + i]) & 1) == 0) {
587         src.setPosition(start + i + 1);
588         break;
589       }
590     }
591     return normalize(m);
592   }
593 
594   /**
595    * Skip {@code src} over the significand bytes.
596    * @param src The source from which to read encoded digits.
597    * @param comp Treat encoded bytes as compliments when {@code comp} is true.
598    * @return the number of bytes skipped.
599    */
600   private static int skipSignificand(PositionedByteRange src, boolean comp) {
601     byte[] a = src.getBytes();
602     final int offset = src.getOffset(), start = src.getPosition();
603     int i = src.getPosition();
604     while (((comp ? DESCENDING : ASCENDING).apply(a[offset + i++]) & 1) != 0)
605       ;
606     src.setPosition(i);
607     return i - start;
608   }
609 
610   /**
611    * <p>
612    * Encode the small magnitude floating point number {@code val} using the
613    * key encoding. The caller guarantees that 1.0 > abs(val) > 0.0.
614    * </p>
615    * <p>
616    * A floating point value is encoded as an integer exponent {@code E} and a
617    * mantissa {@code M}. The original value is equal to {@code (M * 100^E)}.
618    * {@code E} is set to the smallest value possible without making {@code M}
619    * greater than or equal to 1.0.
620    * </p>
621    * <p>
622    * For this routine, {@code E} will always be zero or negative, since the
623    * original value is less than one. The encoding written by this routine is
624    * the ones-complement of the varint of the negative of {@code E} followed
625    * by the mantissa:
626    * <pre>
627    *   Encoding:   ~-E  M
628    * </pre>
629    * </p>
630    * @param dst The destination to which encoded digits are written.
631    * @param val The value to encode.
632    * @return the number of bytes written.
633    */
634   private static int encodeNumericSmall(PositionedByteRange dst, BigDecimal val) {
635     // TODO: this can be done faster?
636     // assert 1.0 > abs(val) > 0.0
637     BigDecimal abs = val.abs();
638     assert BigDecimal.ZERO.compareTo(abs) < 0 && BigDecimal.ONE.compareTo(abs) > 0;
639     byte[] a = dst.getBytes();
640     boolean isNeg = val.signum() == -1;
641     final int offset = dst.getOffset(), start = dst.getPosition();
642     int e = 0, d, startM;
643 
644     if (isNeg) { /* Small negative number: 0x14, -E, ~M */
645       dst.put(NEG_SMALL);
646     } else { /* Small positive number: 0x16, ~-E, M */
647       dst.put(POS_SMALL);
648     }
649 
650     // normalize abs(val) to determine E
651     while (abs.compareTo(EN10) < 0) { abs = abs.movePointRight(8); e += 4; }
652     while (abs.compareTo(EN2) < 0) { abs = abs.movePointRight(2); e++; }
653 
654     putVaruint64(dst, e, !isNeg); // encode appropriate E value.
655 
656     // encode M by peeling off centimal digits, encoding x as 2x+1
657     startM = dst.getPosition();
658     // TODO: 18 is an arbitrary encoding limit. Reevaluate once we have a better handling of
659     // numeric scale.
660     for (int i = 0; i < 18 && abs.compareTo(BigDecimal.ZERO) != 0; i++) {
661       abs = abs.movePointRight(2);
662       d = abs.intValue();
663       dst.put((byte) ((2 * d + 1) & 0xff));
664       abs = abs.subtract(BigDecimal.valueOf(d));
665     }
666     // terminal digit should be 2x
667     a[offset + dst.getPosition() - 1] = (byte) (a[offset + dst.getPosition() - 1] & 0xfe);
668     if (isNeg) {
669       // negative values encoded as ~M
670       DESCENDING.apply(a, offset + startM, dst.getPosition() - startM);
671     }
672     return dst.getPosition() - start;
673   }
674 
675   /**
676    * Encode the large magnitude floating point number {@code val} using
677    * the key encoding. The caller guarantees that {@code val} will be
678    * finite and abs(val) >= 1.0.
679    * <p>
680    * A floating point value is encoded as an integer exponent {@code E}
681    * and a mantissa {@code M}. The original value is equal to
682    * {@code (M * 100^E)}. {@code E} is set to the smallest value
683    * possible without making {@code M} greater than or equal to 1.0.
684    * </p>
685    * <p>
686    * Each centimal digit of the mantissa is stored in a byte. If the value of
687    * the centimal digit is {@code X} (hence {@code X>=0} and
688    * {@code X<=99}) then the byte value will be {@code 2*X+1} for
689    * every byte of the mantissa, except for the last byte which will be
690    * {@code 2*X+0}. The mantissa must be the minimum number of bytes
691    * necessary to represent the value; trailing {@code X==0} digits are
692    * omitted. This means that the mantissa will never contain a byte with the
693    * value {@code 0x00}.
694    * </p>
695    * <p>
696    * If {@code E > 10}, then this routine writes of {@code E} as a
697    * varint followed by the mantissa as described above. Otherwise, if
698    * {@code E <= 10}, this routine only writes the mantissa and leaves
699    * the {@code E} value to be encoded as part of the opening byte of the
700    * field by the calling function.
701    *
702    * <pre>
703    *   Encoding:  M       (if E<=10)
704    *              E M     (if E>10)
705    * </pre>
706    * </p>
707    * @param dst The destination to which encoded digits are written.
708    * @param val The value to encode.
709    * @return the number of bytes written.
710    */
711   private static int encodeNumericLarge(PositionedByteRange dst, BigDecimal val) {
712     // TODO: this can be done faster
713     BigDecimal abs = val.abs();
714     byte[] a = dst.getBytes();
715     boolean isNeg = val.signum() == -1;
716     final int start = dst.getPosition(), offset = dst.getOffset();
717     int e = 0, d, startM;
718 
719     if (isNeg) { /* Large negative number: 0x08, ~E, ~M */
720       dst.put(NEG_LARGE);
721     } else { /* Large positive number: 0x22, E, M */
722       dst.put(POS_LARGE);
723     }
724 
725     // normalize abs(val) to determine E
726     while (abs.compareTo(E32) >= 0 && e <= 350) { abs = abs.movePointLeft(32); e +=16; }
727     while (abs.compareTo(E8) >= 0 && e <= 350) { abs = abs.movePointLeft(8); e+= 4; }
728     while (abs.compareTo(BigDecimal.ONE) >= 0 && e <= 350) { abs = abs.movePointLeft(2); e++; }
729 
730     // encode appropriate header byte and/or E value.
731     if (e > 10) { /* large number, write out {~,}E */
732       putVaruint64(dst, e, isNeg);
733     } else {
734       if (isNeg) { /* Medium negative number: 0x13-E, ~M */
735         dst.put(start, (byte) (NEG_MED_MAX - e));
736       } else { /* Medium positive number: 0x17+E, M */
737         dst.put(start, (byte) (POS_MED_MIN + e));
738       }
739     }
740 
741     // encode M by peeling off centimal digits, encoding x as 2x+1
742     startM = dst.getPosition();
743     // TODO: 18 is an arbitrary encoding limit. Reevaluate once we have a better handling of
744     // numeric scale.
745     for (int i = 0; i < 18 && abs.compareTo(BigDecimal.ZERO) != 0; i++) {
746       abs = abs.movePointRight(2);
747       d = abs.intValue();
748       dst.put((byte) (2 * d + 1));
749       abs = abs.subtract(BigDecimal.valueOf(d));
750     }
751     // terminal digit should be 2x
752     a[offset + dst.getPosition() - 1] = (byte) (a[offset + dst.getPosition() - 1] & 0xfe);
753     if (isNeg) {
754       // negative values encoded as ~M
755       DESCENDING.apply(a, offset + startM, dst.getPosition() - startM);
756     }
757     return dst.getPosition() - start;
758   }
759 
760   /**
761    * Encode a numerical value using the variable-length encoding.
762    * @param dst The destination to which encoded digits are written.
763    * @param val The value to encode.
764    * @param ord The {@link Order} to respect while encoding {@code val}.
765    * @return the number of bytes written.
766    */
767   public static int encodeNumeric(PositionedByteRange dst, long val, Order ord) {
768     return encodeNumeric(dst, BigDecimal.valueOf(val), ord);
769   }
770 
771   /**
772    * Encode a numerical value using the variable-length encoding.
773    * @param dst The destination to which encoded digits are written.
774    * @param val The value to encode.
775    * @param ord The {@link Order} to respect while encoding {@code val}.
776    * @return the number of bytes written.
777    */
778   public static int encodeNumeric(PositionedByteRange dst, double val, Order ord) {
779     if (val == 0.0) {
780       dst.put(ord.apply(ZERO));
781       return 1;
782     }
783     if (Double.isNaN(val)) {
784       dst.put(ord.apply(NAN));
785       return 1;
786     }
787     if (val == Double.NEGATIVE_INFINITY) {
788       dst.put(ord.apply(NEG_INF));
789       return 1;
790     }
791     if (val == Double.POSITIVE_INFINITY) {
792       dst.put(ord.apply(POS_INF));
793       return 1;
794     }
795     return encodeNumeric(dst, BigDecimal.valueOf(val), ord);
796   }
797 
798   /**
799    * Encode a numerical value using the variable-length encoding.
800    * @param dst The destination to which encoded digits are written.
801    * @param val The value to encode.
802    * @param ord The {@link Order} to respect while encoding {@code val}.
803    * @return the number of bytes written.
804    */
805   public static int encodeNumeric(PositionedByteRange dst, BigDecimal val, Order ord) {
806     final int len, offset = dst.getOffset(), start = dst.getPosition();
807     if (null == val) {
808       return encodeNull(dst, ord);
809     } else if (BigDecimal.ZERO.compareTo(val) == 0) {
810       dst.put(ord.apply(ZERO));
811       return 1;
812     }
813     BigDecimal abs = val.abs();
814     if (BigDecimal.ONE.compareTo(abs) <= 0) { // abs(v) >= 1.0
815       len = encodeNumericLarge(dst, normalize(val));
816     } else { // 1.0 > abs(v) >= 0.0
817       len = encodeNumericSmall(dst, normalize(val));
818     }
819     ord.apply(dst.getBytes(), offset + start, len);
820     return len;
821   }
822 
823   /**
824    * Decode a {@link BigDecimal} from {@code src}. Assumes {@code src} encodes
825    * a value in Numeric encoding and is within the valid range of
826    * {@link BigDecimal} values. {@link BigDecimal} does not support {@code NaN}
827    * or {@code Infinte} values.
828    * @see #decodeNumericAsDouble(PositionedByteRange)
829    */
830   private static BigDecimal decodeNumericValue(PositionedByteRange src) {
831     final int e;
832     byte header = src.get();
833     boolean dsc = -1 == Integer.signum(header);
834     header = dsc ? DESCENDING.apply(header) : header;
835 
836     if (header == NULL) return null;
837     if (header == NEG_LARGE) { /* Large negative number: 0x08, ~E, ~M */
838       e = (int) getVaruint64(src, !dsc);
839       return decodeSignificand(src, e, !dsc).negate();
840     }
841     if (header >= NEG_MED_MIN && header <= NEG_MED_MAX) {
842       /* Medium negative number: 0x13-E, ~M */
843       e = NEG_MED_MAX - header;
844       return decodeSignificand(src, e, !dsc).negate();
845     }
846     if (header == NEG_SMALL) { /* Small negative number: 0x14, -E, ~M */
847       e = (int) -getVaruint64(src, dsc);
848       return decodeSignificand(src, e, !dsc).negate();
849     }
850     if (header == ZERO) {
851       return BigDecimal.ZERO;
852     }
853     if (header == POS_SMALL) { /* Small positive number: 0x16, ~-E, M */
854       e = (int) -getVaruint64(src, !dsc);
855       return decodeSignificand(src, e, dsc);
856     }
857     if (header >= POS_MED_MIN && header <= POS_MED_MAX) {
858       /* Medium positive number: 0x17+E, M */
859       e = header - POS_MED_MIN;
860       return decodeSignificand(src, e, dsc);
861     }
862     if (header == POS_LARGE) { /* Large positive number: 0x22, E, M */
863       e = (int) getVaruint64(src, dsc);
864       return decodeSignificand(src, e, dsc);
865     }
866     throw unexpectedHeader(header);
867   }
868 
869   /**
870    * Decode a primitive {@code double} value from the Numeric encoding. Numeric
871    * encoding is based on {@link BigDecimal}; in the event the encoded value is
872    * larger than can be represented in a {@code double}, this method performs
873    * an implicit narrowing conversion as described in
874    * {@link BigDecimal#doubleValue()}.
875    * @throws NullPointerException when the encoded value is {@code NULL}.
876    * @throws IllegalArgumentException when the encoded value is not a Numeric.
877    * @see #encodeNumeric(PositionedByteRange, double, Order)
878    * @see BigDecimal#doubleValue()
879    */
880   public static double decodeNumericAsDouble(PositionedByteRange src) {
881     // TODO: should an encoded NULL value throw unexpectedHeader() instead?
882     if (isNull(src)) {
883       throw new NullPointerException("A null value cannot be decoded to a double.");
884     }
885     if (isNumericNaN(src)) {
886       src.get();
887       return Double.NaN;
888     }
889     if (isNumericZero(src)) {
890       src.get();
891       return Double.valueOf(0.0);
892     }
893 
894     byte header = -1 == Integer.signum(src.peek()) ? DESCENDING.apply(src.peek()) : src.peek();
895 
896     if (header == NEG_INF) {
897       src.get();
898       return Double.NEGATIVE_INFINITY;
899     } else if (header == POS_INF) {
900       src.get();
901       return Double.POSITIVE_INFINITY;
902     } else {
903       return decodeNumericValue(src).doubleValue();
904     }
905   }
906 
907   /**
908    * Decode a primitive {@code long} value from the Numeric encoding. Numeric
909    * encoding is based on {@link BigDecimal}; in the event the encoded value is
910    * larger than can be represented in a {@code long}, this method performs an
911    * implicit narrowing conversion as described in
912    * {@link BigDecimal#doubleValue()}.
913    * @throws NullPointerException when the encoded value is {@code NULL}.
914    * @throws IllegalArgumentException when the encoded value is not a Numeric.
915    * @see #encodeNumeric(PositionedByteRange, long, Order)
916    * @see BigDecimal#longValue()
917    */
918   public static long decodeNumericAsLong(PositionedByteRange src) {
919     // TODO: should an encoded NULL value throw unexpectedHeader() instead?
920     if (isNull(src)) throw new NullPointerException();
921     if (!isNumeric(src)) throw unexpectedHeader(src.peek());
922     if (isNumericNaN(src)) throw unexpectedHeader(src.peek());
923     if (isNumericInfinite(src)) throw unexpectedHeader(src.peek());
924 
925     if (isNumericZero(src)) {
926       src.get();
927       return Long.valueOf(0);
928     }
929     return decodeNumericValue(src).longValue();
930   }
931 
932   /**
933    * Decode a {@link BigDecimal} value from the variable-length encoding.
934    * @throws IllegalArgumentException when the encoded value is not a Numeric.
935    * @see #encodeNumeric(PositionedByteRange, BigDecimal, Order)
936    */
937   public static BigDecimal decodeNumericAsBigDecimal(PositionedByteRange src) {
938     if (isNull(src)) {
939       src.get();
940       return null;
941     }
942     if (!isNumeric(src)) throw unexpectedHeader(src.peek());
943     if (isNumericNaN(src)) throw unexpectedHeader(src.peek());
944     if (isNumericInfinite(src)) throw unexpectedHeader(src.peek());
945     return decodeNumericValue(src);
946   }
947 
948   /**
949    * Encode a String value. String encoding is 0x00-terminated and so it does
950    * not support {@code \u0000} codepoints in the value.
951    * @param dst The destination to which the encoded value is written.
952    * @param val The value to encode.
953    * @param ord The {@link Order} to respect while encoding {@code val}.
954    * @return the number of bytes written.
955    * @throws IllegalArgumentException when {@code val} contains a {@code \u0000}.
956    */
957   public static int encodeString(PositionedByteRange dst, String val, Order ord) {
958     if (null == val) {
959       return encodeNull(dst, ord);
960     }
961     if (val.contains("\u0000"))
962       throw new IllegalArgumentException("Cannot encode String values containing '\\u0000'");
963     final int offset = dst.getOffset(), start = dst.getPosition();
964     dst.put(TEXT);
965     // TODO: is there no way to decode into dst directly?
966     dst.put(val.getBytes(UTF8));
967     dst.put(TERM);
968     ord.apply(dst.getBytes(), offset + start, dst.getPosition() - start);
969     return dst.getPosition() - start;
970   }
971 
972   /**
973    * Decode a String value.
974    */
975   public static String decodeString(PositionedByteRange src) {
976     final byte header = src.get();
977     if (header == NULL || header == DESCENDING.apply(NULL))
978       return null;
979     assert header == TEXT || header == DESCENDING.apply(TEXT);
980     Order ord = header == TEXT ? ASCENDING : DESCENDING;
981     byte[] a = src.getBytes();
982     final int offset = src.getOffset(), start = src.getPosition();
983     final byte terminator = ord.apply(TERM);
984     int rawStartPos = offset + start, rawTermPos = rawStartPos;
985     for (; a[rawTermPos] != terminator; rawTermPos++)
986       ;
987     src.setPosition(rawTermPos - offset + 1); // advance position to TERM + 1
988     if (DESCENDING == ord) {
989       // make a copy so that we don't disturb encoded value with ord.
990       byte[] copy = new byte[rawTermPos - rawStartPos];
991       System.arraycopy(a, rawStartPos, copy, 0, copy.length);
992       ord.apply(copy);
993       return new String(copy, UTF8);
994     } else {
995       return new String(a, rawStartPos, rawTermPos - rawStartPos, UTF8);
996     }
997   }
998 
999   /**
1000    * Calculate the expected BlobVar encoded length based on unencoded length.
1001    */
1002   public static int blobVarEncodedLength(int len) {
1003     if (0 == len)
1004       return 2; // 1-byte header + 1-byte terminator
1005     else
1006       return (int)
1007           Math.ceil(
1008             (len * 8) // 8-bits per input byte
1009             / 7.0)    // 7-bits of input data per encoded byte, rounded up
1010           + 1;        // + 1-byte header
1011   }
1012 
1013   /**
1014    * Calculate the expected BlobVar decoded length based on encoded length.
1015    */
1016   @InterfaceAudience.Private
1017   static int blobVarDecodedLength(int len) {
1018     return
1019         ((len
1020           - 1) // 1-byte header
1021           * 7) // 7-bits of payload per encoded byte
1022           / 8; // 8-bits per byte
1023   }
1024 
1025   /**
1026    * Encode a Blob value using a modified varint encoding scheme.
1027    * <p>
1028    * This format encodes a byte[] value such that no limitations on the input
1029    * value are imposed. The first byte encodes the encoding scheme that
1030    * follows, {@link #BLOB_VAR}. Each encoded byte thereafter consists of a
1031    * header bit followed by 7 bits of payload. A header bit of '1' indicates
1032    * continuation of the encoding. A header bit of '0' indicates this byte
1033    * contains the last of the payload. An empty input value is encoded as the
1034    * header byte immediately followed by a termination byte {@code 0x00}. This
1035    * is not ambiguous with the encoded value of {@code []}, which results in
1036    * {@code [0x80, 0x00]}.
1037    * </p>
1038    * @return the number of bytes written.
1039    */
1040   public static int encodeBlobVar(PositionedByteRange dst, byte[] val, int voff, int vlen,
1041       Order ord) {
1042     if (null == val) {
1043       return encodeNull(dst, ord);
1044     }
1045     // Empty value is null-terminated. All other values are encoded as 7-bits per byte.
1046     assert dst.getRemaining() >= blobVarEncodedLength(vlen) : "buffer overflow expected.";
1047     final int offset = dst.getOffset(), start = dst.getPosition();
1048     dst.put(BLOB_VAR);
1049     if (0 == vlen) {
1050       dst.put(TERM);
1051     } else {
1052       byte s = 1, t = 0;
1053       for (int i = voff; i < vlen; i++) {
1054         dst.put((byte) (0x80 | t | ((val[i] & 0xff) >>> s)));
1055         if (s < 7) {
1056           t = (byte) (val[i] << (7 - s));
1057           s++;
1058         } else {
1059           dst.put((byte) (0x80 | val[i]));
1060           s = 1;
1061           t = 0;
1062         }
1063       }
1064       if (s > 1) {
1065         dst.put((byte) (0x7f & t));
1066       } else {
1067         dst.getBytes()[offset + dst.getPosition() - 1] =
1068           (byte) (dst.getBytes()[offset + dst.getPosition() - 1] & 0x7f);
1069       }
1070     }
1071     ord.apply(dst.getBytes(), offset + start, dst.getPosition() - start);
1072     return dst.getPosition() - start;
1073   }
1074 
1075   /**
1076    * Encode a blob value using a modified varint encoding scheme.
1077    * @return the number of bytes written.
1078    * @see #encodeBlobVar(PositionedByteRange, byte[], int, int, Order)
1079    */
1080   public static int encodeBlobVar(PositionedByteRange dst, byte[] val, Order ord) {
1081     return encodeBlobVar(dst, val, 0, null != val ? val.length : 0, ord);
1082   }
1083 
1084   /**
1085    * Decode a blob value that was encoded using BlobVar encoding.
1086    */
1087   public static byte[] decodeBlobVar(PositionedByteRange src) {
1088     final byte header = src.get();
1089     if (header == NULL || header == DESCENDING.apply(NULL)) {
1090       return null;
1091     }
1092     assert header == BLOB_VAR || header == DESCENDING.apply(BLOB_VAR);
1093     Order ord = BLOB_VAR == header ? ASCENDING : DESCENDING;
1094     if (src.peek() == ord.apply(TERM)) {
1095       // skip empty input buffer.
1096       src.get();
1097       return new byte[0];
1098     }
1099     final int offset = src.getOffset(), start = src.getPosition();
1100     int end;
1101     byte[] a = src.getBytes();
1102     for (end = start; (byte) (ord.apply(a[offset + end]) & 0x80) != TERM; end++)
1103       ;
1104     end++; // increment end to 1-past last byte
1105     // create ret buffer using length of encoded data + 1 (header byte)
1106     PositionedByteRange ret = new SimplePositionedMutableByteRange(blobVarDecodedLength(end - start
1107         + 1));
1108     int s = 6;
1109     byte t = (byte) ((ord.apply(a[offset + start]) << 1) & 0xff);
1110     for (int i = start + 1; i < end; i++) {
1111       if (s == 7) {
1112         ret.put((byte) (t | (ord.apply(a[offset + i]) & 0x7f)));
1113         i++;
1114                // explicitly reset t -- clean up overflow buffer after decoding
1115                // a full cycle and retain assertion condition below. This happens
1116         t = 0; // when the LSB in the last encoded byte is 1. (HBASE-9893)
1117       } else {
1118         ret.put((byte) (t | ((ord.apply(a[offset + i]) & 0x7f) >>> s)));
1119       }
1120       if (i == end) break;
1121       t = (byte) ((ord.apply(a[offset + i]) << (8 - s)) & 0xff);
1122       s = s == 1 ? 7 : s - 1;
1123     }
1124     src.setPosition(end);
1125     assert t == 0 : "Unexpected bits remaining after decoding blob.";
1126     assert ret.getPosition() == ret.getLength() : "Allocated unnecessarily large return buffer.";
1127     return ret.getBytes();
1128   }
1129 
1130   /**
1131    * Encode a Blob value as a byte-for-byte copy. BlobCopy encoding in
1132    * DESCENDING order is NULL terminated so as to preserve proper sorting of
1133    * {@code []} and so it does not support {@code 0x00} in the value.
1134    * @return the number of bytes written.
1135    * @throws IllegalArgumentException when {@code ord} is DESCENDING and
1136    *    {@code val} contains a {@code 0x00} byte.
1137    */
1138   public static int encodeBlobCopy(PositionedByteRange dst, byte[] val, int voff, int vlen,
1139       Order ord) {
1140     if (null == val) {
1141       encodeNull(dst, ord);
1142       if (ASCENDING == ord) return 1;
1143       else {
1144         // DESCENDING ordered BlobCopy requires a termination bit to preserve
1145         // sort-order semantics of null values.
1146         dst.put(ord.apply(TERM));
1147         return 2;
1148       }
1149     }
1150     // Blobs as final entry in a compound key are written unencoded.
1151     assert dst.getRemaining() >= vlen + (ASCENDING == ord ? 1 : 2);
1152     if (DESCENDING == ord) {
1153       for (int i = 0; i < vlen; i++) {
1154         if (TERM == val[voff + i]) {
1155           throw new IllegalArgumentException("0x00 bytes not permitted in value.");
1156         }
1157       }
1158     }
1159     final int offset = dst.getOffset(), start = dst.getPosition();
1160     dst.put(BLOB_COPY);
1161     dst.put(val, voff, vlen);
1162     // DESCENDING ordered BlobCopy requires a termination bit to preserve
1163     // sort-order semantics of null values.
1164     if (DESCENDING == ord) dst.put(TERM);
1165     ord.apply(dst.getBytes(), offset + start, dst.getPosition() - start);
1166     return dst.getPosition() - start;
1167   }
1168 
1169   /**
1170    * Encode a Blob value as a byte-for-byte copy. BlobCopy encoding in
1171    * DESCENDING order is NULL terminated so as to preserve proper sorting of
1172    * {@code []} and so it does not support {@code 0x00} in the value.
1173    * @return the number of bytes written.
1174    * @throws IllegalArgumentException when {@code ord} is DESCENDING and
1175    *    {@code val} contains a {@code 0x00} byte.
1176    * @see #encodeBlobCopy(PositionedByteRange, byte[], int, int, Order)
1177    */
1178   public static int encodeBlobCopy(PositionedByteRange dst, byte[] val, Order ord) {
1179     return encodeBlobCopy(dst, val, 0, null != val ? val.length : 0, ord);
1180   }
1181 
1182   /**
1183    * Decode a Blob value, byte-for-byte copy.
1184    * @see #encodeBlobCopy(PositionedByteRange, byte[], int, int, Order)
1185    */
1186   public static byte[] decodeBlobCopy(PositionedByteRange src) {
1187     byte header = src.get();
1188     if (header == NULL || header == DESCENDING.apply(NULL)) {
1189       return null;
1190     }
1191     assert header == BLOB_COPY || header == DESCENDING.apply(BLOB_COPY);
1192     Order ord = header == BLOB_COPY ? ASCENDING : DESCENDING;
1193     final int length = src.getRemaining() - (ASCENDING == ord ? 0 : 1);
1194     byte[] ret = new byte[length];
1195     src.get(ret);
1196     ord.apply(ret, 0, ret.length);
1197     // DESCENDING ordered BlobCopy requires a termination bit to preserve
1198     // sort-order semantics of null values.
1199     if (DESCENDING == ord) src.get();
1200     return ret;
1201   }
1202 
1203   /**
1204    * Encode a null value.
1205    * @param dst The destination to which encoded digits are written.
1206    * @param ord The {@link Order} to respect while encoding {@code val}.
1207    * @return the number of bytes written.
1208    */
1209   public static int encodeNull(PositionedByteRange dst, Order ord) {
1210     dst.put(ord.apply(NULL));
1211     return 1;
1212   }
1213 
1214   /**
1215    * Encode an {@code int8} value using the fixed-length encoding.
1216    * @return the number of bytes written.
1217    * @see #encodeInt64(PositionedByteRange, long, Order)
1218    * @see #decodeInt8(PositionedByteRange)
1219    */
1220   public static int encodeInt8(PositionedByteRange dst, byte val, Order ord) {
1221     final int offset = dst.getOffset(), start = dst.getPosition();
1222     dst.put(FIXED_INT8)
1223        .put((byte) (val ^ 0x80));
1224     ord.apply(dst.getBytes(), offset + start, 2);
1225     return 2;
1226   }
1227 
1228   /**
1229    * Decode an {@code int8} value.
1230    * @see #encodeInt8(PositionedByteRange, byte, Order)
1231    */
1232   public static byte decodeInt8(PositionedByteRange src) {
1233     final byte header = src.get();
1234     assert header == FIXED_INT8 || header == DESCENDING.apply(FIXED_INT8);
1235     Order ord = header == FIXED_INT8 ? ASCENDING : DESCENDING;
1236     return (byte)((ord.apply(src.get()) ^ 0x80) & 0xff);
1237   }
1238 
1239   /**
1240    * Encode an {@code int16} value using the fixed-length encoding.
1241    * @return the number of bytes written.
1242    * @see #encodeInt64(PositionedByteRange, long, Order)
1243    * @see #decodeInt16(PositionedByteRange)
1244    */
1245   public static int encodeInt16(PositionedByteRange dst, short val, Order ord) {
1246     final int offset = dst.getOffset(), start = dst.getPosition();
1247     dst.put(FIXED_INT16)
1248        .put((byte) ((val >> 8) ^ 0x80))
1249        .put((byte) val);
1250     ord.apply(dst.getBytes(), offset + start, 3);
1251     return 3;
1252   }
1253 
1254   /**
1255    * Decode an {@code int16} value.
1256    * @see #encodeInt16(PositionedByteRange, short, Order)
1257    */
1258   public static short decodeInt16(PositionedByteRange src) {
1259     final byte header = src.get();
1260     assert header == FIXED_INT16 || header == DESCENDING.apply(FIXED_INT16);
1261     Order ord = header == FIXED_INT16 ? ASCENDING : DESCENDING;
1262     short val = (short) ((ord.apply(src.get()) ^ 0x80) & 0xff);
1263     val = (short) ((val << 8) + (ord.apply(src.get()) & 0xff));
1264     return val;
1265   }
1266 
1267   /**
1268    * Encode an {@code int32} value using the fixed-length encoding.
1269    * @return the number of bytes written.
1270    * @see #encodeInt64(PositionedByteRange, long, Order)
1271    * @see #decodeInt32(PositionedByteRange)
1272    */
1273   public static int encodeInt32(PositionedByteRange dst, int val, Order ord) {
1274     final int offset = dst.getOffset(), start = dst.getPosition();
1275     dst.put(FIXED_INT32)
1276         .put((byte) ((val >> 24) ^ 0x80))
1277         .put((byte) (val >> 16))
1278         .put((byte) (val >> 8))
1279         .put((byte) val);
1280     ord.apply(dst.getBytes(), offset + start, 5);
1281     return 5;
1282   }
1283 
1284   /**
1285    * Decode an {@code int32} value.
1286    * @see #encodeInt32(PositionedByteRange, int, Order)
1287    */
1288   public static int decodeInt32(PositionedByteRange src) {
1289     final byte header = src.get();
1290     assert header == FIXED_INT32 || header == DESCENDING.apply(FIXED_INT32);
1291     Order ord = header == FIXED_INT32 ? ASCENDING : DESCENDING;
1292     int val = (ord.apply(src.get()) ^ 0x80) & 0xff;
1293     for (int i = 1; i < 4; i++) {
1294       val = (val << 8) + (ord.apply(src.get()) & 0xff);
1295     }
1296     return val;
1297   }
1298 
1299   /**
1300    * Encode an {@code int64} value using the fixed-length encoding.
1301    * <p>
1302    * This format ensures that all longs sort in their natural order, as they
1303    * would sort when using signed long comparison.
1304    * </p>
1305    * <p>
1306    * All Longs are serialized to an 8-byte, fixed-width sortable byte format.
1307    * Serialization is performed by inverting the integer sign bit and writing
1308    * the resulting bytes to the byte array in big endian order. The encoded
1309    * value is prefixed by the {@link #FIXED_INT64} header byte. This encoding
1310    * is designed to handle java language primitives and so Null values are NOT
1311    * supported by this implementation.
1312    * </p>
1313    * <p>
1314    * For example:
1315    * </p>
1316    * <pre>
1317    * Input:   0x0000000000000005 (5)
1318    * Result:  0x288000000000000005
1319    *
1320    * Input:   0xfffffffffffffffb (-4)
1321    * Result:  0x280000000000000004
1322    *
1323    * Input:   0x7fffffffffffffff (Long.MAX_VALUE)
1324    * Result:  0x28ffffffffffffffff
1325    *
1326    * Input:   0x8000000000000000 (Long.MIN_VALUE)
1327    * Result:  0x287fffffffffffffff
1328    * </pre>
1329    * <p>
1330    * This encoding format, and much of this documentation string, is based on
1331    * Orderly's {@code FixedIntWritableRowKey}.
1332    * </p>
1333    * @return the number of bytes written.
1334    * @see #decodeInt64(PositionedByteRange)
1335    */
1336   public static int encodeInt64(PositionedByteRange dst, long val, Order ord) {
1337     final int offset = dst.getOffset(), start = dst.getPosition();
1338     dst.put(FIXED_INT64)
1339        .put((byte) ((val >> 56) ^ 0x80))
1340        .put((byte) (val >> 48))
1341        .put((byte) (val >> 40))
1342        .put((byte) (val >> 32))
1343        .put((byte) (val >> 24))
1344        .put((byte) (val >> 16))
1345        .put((byte) (val >> 8))
1346        .put((byte) val);
1347     ord.apply(dst.getBytes(), offset + start, 9);
1348     return 9;
1349   }
1350 
1351   /**
1352    * Decode an {@code int64} value.
1353    * @see #encodeInt64(PositionedByteRange, long, Order)
1354    */
1355   public static long decodeInt64(PositionedByteRange src) {
1356     final byte header = src.get();
1357     assert header == FIXED_INT64 || header == DESCENDING.apply(FIXED_INT64);
1358     Order ord = header == FIXED_INT64 ? ASCENDING : DESCENDING;
1359     long val = (ord.apply(src.get()) ^ 0x80) & 0xff;
1360     for (int i = 1; i < 8; i++) {
1361       val = (val << 8) + (ord.apply(src.get()) & 0xff);
1362     }
1363     return val;
1364   }
1365 
1366   /**
1367    * Encode a 32-bit floating point value using the fixed-length encoding.
1368    * Encoding format is described at length in
1369    * {@link #encodeFloat64(PositionedByteRange, double, Order)}.
1370    * @return the number of bytes written.
1371    * @see #decodeFloat32(PositionedByteRange)
1372    * @see #encodeFloat64(PositionedByteRange, double, Order)
1373    */
1374   public static int encodeFloat32(PositionedByteRange dst, float val, Order ord) {
1375     final int offset = dst.getOffset(), start = dst.getPosition();
1376     int i = Float.floatToIntBits(val);
1377     i ^= ((i >> (Integer.SIZE - 1)) | Integer.MIN_VALUE);
1378     dst.put(FIXED_FLOAT32)
1379         .put((byte) (i >> 24))
1380         .put((byte) (i >> 16))
1381         .put((byte) (i >> 8))
1382         .put((byte) i);
1383     ord.apply(dst.getBytes(), offset + start, 5);
1384     return 5;
1385   }
1386 
1387   /**
1388    * Decode a 32-bit floating point value using the fixed-length encoding.
1389    * @see #encodeFloat32(PositionedByteRange, float, Order)
1390    */
1391   public static float decodeFloat32(PositionedByteRange src) {
1392     final byte header = src.get();
1393     assert header == FIXED_FLOAT32 || header == DESCENDING.apply(FIXED_FLOAT32);
1394     Order ord = header == FIXED_FLOAT32 ? ASCENDING : DESCENDING;
1395     int val = ord.apply(src.get()) & 0xff;
1396     for (int i = 1; i < 4; i++) {
1397       val = (val << 8) + (ord.apply(src.get()) & 0xff);
1398     }
1399     val ^= (~val >> (Integer.SIZE - 1)) | Integer.MIN_VALUE;
1400     return Float.intBitsToFloat(val);
1401   }
1402 
1403   /**
1404    * Encode a 64-bit floating point value using the fixed-length encoding.
1405    * <p>
1406    * This format ensures the following total ordering of floating point
1407    * values: Double.NEGATIVE_INFINITY &lt; -Double.MAX_VALUE &lt; ... &lt;
1408    * -Double.MIN_VALUE &lt; -0.0 &lt; +0.0; &lt; Double.MIN_VALUE &lt; ...
1409    * &lt; Double.MAX_VALUE &lt; Double.POSITIVE_INFINITY &lt; Double.NaN
1410    * </p>
1411    * <p>
1412    * Floating point numbers are encoded as specified in IEEE 754. A 64-bit
1413    * double precision float consists of a sign bit, 11-bit unsigned exponent
1414    * encoded in offset-1023 notation, and a 52-bit significand. The format is
1415    * described further in the <a
1416    * href="http://en.wikipedia.org/wiki/Double_precision"> Double Precision
1417    * Floating Point Wikipedia page</a> </p>
1418    * <p>
1419    * The value of a normal float is -1 <sup>sign bit</sup> &times;
1420    * 2<sup>exponent - 1023</sup> &times; 1.significand
1421    * </p>
1422    * <p>
1423    * The IEE754 floating point format already preserves sort ordering for
1424    * positive floating point numbers when the raw bytes are compared in most
1425    * significant byte order. This is discussed further at <a href=
1426    * "http://www.cygnus-software.com/papers/comparingfloats/comparingfloats.htm"
1427    * > http://www.cygnus-software.com/papers/comparingfloats/comparingfloats.
1428    * htm</a>
1429    * </p>
1430    * <p>
1431    * Thus, we need only ensure that negative numbers sort in the the exact
1432    * opposite order as positive numbers (so that say, negative infinity is
1433    * less than negative 1), and that all negative numbers compare less than
1434    * any positive number. To accomplish this, we invert the sign bit of all
1435    * floating point numbers, and we also invert the exponent and significand
1436    * bits if the floating point number was negative.
1437    * </p>
1438    * <p>
1439    * More specifically, we first store the floating point bits into a 64-bit
1440    * long {@code l} using {@link Double#doubleToLongBits}. This method
1441    * collapses all NaNs into a single, canonical NaN value but otherwise
1442    * leaves the bits unchanged. We then compute
1443    * </p>
1444    * <pre>
1445    * l &circ;= (l &gt;&gt; (Long.SIZE - 1)) | Long.MIN_SIZE
1446    * </pre>
1447    * <p>
1448    * which inverts the sign bit and XOR's all other bits with the sign bit
1449    * itself. Comparing the raw bytes of {@code l} in most significant
1450    * byte order is equivalent to performing a double precision floating point
1451    * comparison on the underlying bits (ignoring NaN comparisons, as NaNs
1452    * don't compare equal to anything when performing floating point
1453    * comparisons).
1454    * </p>
1455    * <p>
1456    * The resulting long integer is then converted into a byte array by
1457    * serializing the long one byte at a time in most significant byte order.
1458    * The serialized integer is prefixed by a single header byte. All
1459    * serialized values are 9 bytes in length.
1460    * </p>
1461    * <p>
1462    * This encoding format, and much of this highly detailed documentation
1463    * string, is based on Orderly's {@code DoubleWritableRowKey}.
1464    * </p>
1465    * @return the number of bytes written.
1466    * @see #decodeFloat64(PositionedByteRange)
1467    */
1468   public static int encodeFloat64(PositionedByteRange dst, double val, Order ord) {
1469     final int offset = dst.getOffset(), start = dst.getPosition();
1470     long lng = Double.doubleToLongBits(val);
1471     lng ^= ((lng >> (Long.SIZE - 1)) | Long.MIN_VALUE);
1472     dst.put(FIXED_FLOAT64)
1473         .put((byte) (lng >> 56))
1474         .put((byte) (lng >> 48))
1475         .put((byte) (lng >> 40))
1476         .put((byte) (lng >> 32))
1477         .put((byte) (lng >> 24))
1478         .put((byte) (lng >> 16))
1479         .put((byte) (lng >> 8))
1480         .put((byte) lng);
1481     ord.apply(dst.getBytes(), offset + start, 9);
1482     return 9;
1483   }
1484 
1485   /**
1486    * Decode a 64-bit floating point value using the fixed-length encoding.
1487    * @see #encodeFloat64(PositionedByteRange, double, Order)
1488    */
1489   public static double decodeFloat64(PositionedByteRange src) {
1490     final byte header = src.get();
1491     assert header == FIXED_FLOAT64 || header == DESCENDING.apply(FIXED_FLOAT64);
1492     Order ord = header == FIXED_FLOAT64 ? ASCENDING : DESCENDING;
1493     long val = ord.apply(src.get()) & 0xff;
1494     for (int i = 1; i < 8; i++) {
1495       val = (val << 8) + (ord.apply(src.get()) & 0xff);
1496     }
1497     val ^= (~val >> (Long.SIZE - 1)) | Long.MIN_VALUE;
1498     return Double.longBitsToDouble(val);
1499   }
1500 
1501   /**
1502    * Returns true when {@code src} appears to be positioned an encoded value,
1503    * false otherwise.
1504    */
1505   public static boolean isEncodedValue(PositionedByteRange src) {
1506     return isNull(src) || isNumeric(src) || isFixedInt8(src) || isFixedInt16(src)
1507         || isFixedInt32(src) || isFixedInt64(src)
1508         || isFixedFloat32(src) || isFixedFloat64(src) || isText(src) || isBlobCopy(src)
1509         || isBlobVar(src);
1510   }
1511 
1512   /**
1513    * Return true when the next encoded value in {@code src} is null, false
1514    * otherwise.
1515    */
1516   public static boolean isNull(PositionedByteRange src) {
1517     return NULL ==
1518         (-1 == Integer.signum(src.peek()) ? DESCENDING : ASCENDING).apply(src.peek());
1519   }
1520 
1521   /**
1522    * Return true when the next encoded value in {@code src} uses Numeric
1523    * encoding, false otherwise. {@code NaN}, {@code +/-Inf} are valid Numeric
1524    * values.
1525    */
1526   public static boolean isNumeric(PositionedByteRange src) {
1527     byte x = (-1 == Integer.signum(src.peek()) ? DESCENDING : ASCENDING).apply(src.peek());
1528     return x >= NEG_INF && x <= NAN;
1529   }
1530 
1531   /**
1532    * Return true when the next encoded value in {@code src} uses Numeric
1533    * encoding and is {@code Infinite}, false otherwise.
1534    */
1535   public static boolean isNumericInfinite(PositionedByteRange src) {
1536     byte x = (-1 == Integer.signum(src.peek()) ? DESCENDING : ASCENDING).apply(src.peek());
1537     return NEG_INF == x || POS_INF == x;
1538   }
1539 
1540   /**
1541    * Return true when the next encoded value in {@code src} uses Numeric
1542    * encoding and is {@code NaN}, false otherwise.
1543    */
1544   public static boolean isNumericNaN(PositionedByteRange src) {
1545     return NAN == (-1 == Integer.signum(src.peek()) ? DESCENDING : ASCENDING).apply(src.peek());
1546   }
1547 
1548   /**
1549    * Return true when the next encoded value in {@code src} uses Numeric
1550    * encoding and is {@code 0}, false otherwise.
1551    */
1552   public static boolean isNumericZero(PositionedByteRange src) {
1553     return ZERO ==
1554         (-1 == Integer.signum(src.peek()) ? DESCENDING : ASCENDING).apply(src.peek());
1555   }
1556 
1557   /**
1558    * Return true when the next encoded value in {@code src} uses fixed-width
1559    * Int8 encoding, false otherwise.
1560    */
1561   public static boolean isFixedInt8(PositionedByteRange src) {
1562     return FIXED_INT8 ==
1563         (-1 == Integer.signum(src.peek()) ? DESCENDING : ASCENDING).apply(src.peek());
1564   }
1565 
1566   /**
1567    * Return true when the next encoded value in {@code src} uses fixed-width
1568    * Int16 encoding, false otherwise.
1569    */
1570   public static boolean isFixedInt16(PositionedByteRange src) {
1571     return FIXED_INT16 ==
1572         (-1 == Integer.signum(src.peek()) ? DESCENDING : ASCENDING).apply(src.peek());
1573   }
1574 
1575   /**
1576    * Return true when the next encoded value in {@code src} uses fixed-width
1577    * Int32 encoding, false otherwise.
1578    */
1579   public static boolean isFixedInt32(PositionedByteRange src) {
1580     return FIXED_INT32 ==
1581         (-1 == Integer.signum(src.peek()) ? DESCENDING : ASCENDING).apply(src.peek());
1582   }
1583 
1584   /**
1585    * Return true when the next encoded value in {@code src} uses fixed-width
1586    * Int64 encoding, false otherwise.
1587    */
1588   public static boolean isFixedInt64(PositionedByteRange src) {
1589     return FIXED_INT64 ==
1590         (-1 == Integer.signum(src.peek()) ? DESCENDING : ASCENDING).apply(src.peek());
1591   }
1592 
1593   /**
1594    * Return true when the next encoded value in {@code src} uses fixed-width
1595    * Float32 encoding, false otherwise.
1596    */
1597   public static boolean isFixedFloat32(PositionedByteRange src) {
1598     return FIXED_FLOAT32 ==
1599         (-1 == Integer.signum(src.peek()) ? DESCENDING : ASCENDING).apply(src.peek());
1600   }
1601 
1602   /**
1603    * Return true when the next encoded value in {@code src} uses fixed-width
1604    * Float64 encoding, false otherwise.
1605    */
1606   public static boolean isFixedFloat64(PositionedByteRange src) {
1607     return FIXED_FLOAT64 ==
1608         (-1 == Integer.signum(src.peek()) ? DESCENDING : ASCENDING).apply(src.peek());
1609   }
1610 
1611   /**
1612    * Return true when the next encoded value in {@code src} uses Text encoding,
1613    * false otherwise.
1614    */
1615   public static boolean isText(PositionedByteRange src) {
1616     return TEXT ==
1617         (-1 == Integer.signum(src.peek()) ? DESCENDING : ASCENDING).apply(src.peek());
1618   }
1619 
1620   /**
1621    * Return true when the next encoded value in {@code src} uses BlobVar
1622    * encoding, false otherwise.
1623    */
1624   public static boolean isBlobVar(PositionedByteRange src) {
1625     return BLOB_VAR ==
1626         (-1 == Integer.signum(src.peek()) ? DESCENDING : ASCENDING).apply(src.peek());
1627   }
1628 
1629   /**
1630    * Return true when the next encoded value in {@code src} uses BlobCopy
1631    * encoding, false otherwise.
1632    */
1633   public static boolean isBlobCopy(PositionedByteRange src) {
1634     return BLOB_COPY ==
1635         (-1 == Integer.signum(src.peek()) ? DESCENDING : ASCENDING).apply(src.peek());
1636   }
1637 
1638   /**
1639    * Skip {@code buff}'s position forward over one encoded value.
1640    * @return number of bytes skipped.
1641    */
1642   public static int skip(PositionedByteRange src) {
1643     final int start = src.getPosition();
1644     byte header = src.get();
1645     Order ord = (-1 == Integer.signum(header)) ? DESCENDING : ASCENDING;
1646     header = ord.apply(header);
1647 
1648     switch (header) {
1649       case NULL:
1650       case NEG_INF:
1651         return 1;
1652       case NEG_LARGE: /* Large negative number: 0x08, ~E, ~M */
1653         skipVaruint64(src, DESCENDING != ord);
1654         skipSignificand(src, DESCENDING != ord);
1655         return src.getPosition() - start;
1656       case NEG_MED_MIN: /* Medium negative number: 0x13-E, ~M */
1657       case NEG_MED_MIN + 0x01:
1658       case NEG_MED_MIN + 0x02:
1659       case NEG_MED_MIN + 0x03:
1660       case NEG_MED_MIN + 0x04:
1661       case NEG_MED_MIN + 0x05:
1662       case NEG_MED_MIN + 0x06:
1663       case NEG_MED_MIN + 0x07:
1664       case NEG_MED_MIN + 0x08:
1665       case NEG_MED_MIN + 0x09:
1666       case NEG_MED_MAX:
1667         skipSignificand(src, DESCENDING != ord);
1668         return src.getPosition() - start;
1669       case NEG_SMALL: /* Small negative number: 0x14, -E, ~M */
1670         skipVaruint64(src, DESCENDING == ord);
1671         skipSignificand(src, DESCENDING != ord);
1672         return src.getPosition() - start;
1673       case ZERO:
1674         return 1;
1675       case POS_SMALL: /* Small positive number: 0x16, ~-E, M */
1676         skipVaruint64(src, DESCENDING != ord);
1677         skipSignificand(src, DESCENDING == ord);
1678         return src.getPosition() - start;
1679       case POS_MED_MIN: /* Medium positive number: 0x17+E, M */
1680       case POS_MED_MIN + 0x01:
1681       case POS_MED_MIN + 0x02:
1682       case POS_MED_MIN + 0x03:
1683       case POS_MED_MIN + 0x04:
1684       case POS_MED_MIN + 0x05:
1685       case POS_MED_MIN + 0x06:
1686       case POS_MED_MIN + 0x07:
1687       case POS_MED_MIN + 0x08:
1688       case POS_MED_MIN + 0x09:
1689       case POS_MED_MAX:
1690         skipSignificand(src, DESCENDING == ord);
1691         return src.getPosition() - start;
1692       case POS_LARGE: /* Large positive number: 0x22, E, M */
1693         skipVaruint64(src, DESCENDING == ord);
1694         skipSignificand(src, DESCENDING == ord);
1695         return src.getPosition() - start;
1696       case POS_INF:
1697         return 1;
1698       case NAN:
1699         return 1;
1700       case FIXED_INT8:
1701         src.setPosition(src.getPosition() + 1);
1702         return src.getPosition() - start;
1703       case FIXED_INT16:
1704         src.setPosition(src.getPosition() + 2);
1705         return src.getPosition() - start;
1706       case FIXED_INT32:
1707         src.setPosition(src.getPosition() + 4);
1708         return src.getPosition() - start;
1709       case FIXED_INT64:
1710         src.setPosition(src.getPosition() + 8);
1711         return src.getPosition() - start;
1712       case FIXED_FLOAT32:
1713         src.setPosition(src.getPosition() + 4);
1714         return src.getPosition() - start;
1715       case FIXED_FLOAT64:
1716         src.setPosition(src.getPosition() + 8);
1717         return src.getPosition() - start;
1718       case TEXT:
1719         // for null-terminated values, skip to the end.
1720         do {
1721           header = ord.apply(src.get());
1722         } while (header != TERM);
1723         return src.getPosition() - start;
1724       case BLOB_VAR:
1725         // read until we find a 0 in the MSB
1726         do {
1727           header = ord.apply(src.get());
1728         } while ((byte) (header & 0x80) != TERM);
1729         return src.getPosition() - start;
1730       case BLOB_COPY:
1731         if (Order.DESCENDING == ord) {
1732           // if descending, read to termination byte.
1733           do {
1734             header = ord.apply(src.get());
1735           } while (header != TERM);
1736           return src.getPosition() - start;
1737         } else {
1738           // otherwise, just skip to the end.
1739           src.setPosition(src.getLength());
1740           return src.getPosition() - start;
1741         }
1742       default:
1743         throw unexpectedHeader(header);
1744     }
1745   }
1746 
1747   /**
1748    * Return the number of encoded entries remaining in {@code buff}. The
1749    * state of {@code buff} is not modified through use of this method.
1750    */
1751   public static int length(PositionedByteRange buff) {
1752     PositionedByteRange b =
1753         new SimplePositionedMutableByteRange(buff.getBytes(), buff.getOffset(), buff.getLength());
1754     b.setPosition(buff.getPosition());
1755     int cnt = 0;
1756     for (; isEncodedValue(b); skip(b), cnt++)
1757       ;
1758     return cnt;
1759   }
1760 }