| 1 | /***************************************************************************** | 
| 2 |  | 
| 3 | Copyright (c) 2007, 2011, Oracle and/or its affiliates. All Rights Reserved. | 
| 4 |  | 
| 5 | This program is free software; you can redistribute it and/or modify it under | 
| 6 | the terms of the GNU General Public License as published by the Free Software | 
| 7 | Foundation; version 2 of the License. | 
| 8 |  | 
| 9 | This program is distributed in the hope that it will be useful, but WITHOUT | 
| 10 | ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS | 
| 11 | FOR A PARTICULAR PURPOSE. See the GNU General Public License for more details. | 
| 12 |  | 
| 13 | You should have received a copy of the GNU General Public License along with | 
| 14 | this program; if not, write to the Free Software Foundation, Inc., | 
| 15 | 51 Franklin Street, Suite 500, Boston, MA 02110-1335 USA | 
| 16 |  | 
| 17 | *****************************************************************************/ | 
| 18 |  | 
| 19 | /******************************************************************//** | 
| 20 | @file include/fts0vlc.ic | 
| 21 | Full text variable length integer encoding/decoding. | 
| 22 |  | 
| 23 | Created 2007-03-27 Sunny Bains | 
| 24 | *******************************************************/ | 
| 25 |  | 
| 26 | #ifndef INNOBASE_FTS0VLC_IC | 
| 27 | #define INNOBASE_FTS0VLC_IC | 
| 28 |  | 
| 29 | #include "fts0types.h" | 
| 30 |  | 
| 31 | /******************************************************************//** | 
| 32 | Return length of val if it were encoded using our VLC scheme. | 
| 33 | FIXME: We will need to be able encode 8 bytes value | 
| 34 | @return length of value encoded, in bytes */ | 
| 35 | UNIV_INLINE | 
| 36 | ulint | 
| 37 | fts_get_encoded_len( | 
| 38 | /*================*/ | 
| 39 | 	ulint	val)	/* in: value to encode */ | 
| 40 | { | 
| 41 | 	if (val <= 127) { | 
| 42 | 		return(1); | 
| 43 | 	} else if (val <= 16383) { | 
| 44 | 		return(2); | 
| 45 | 	} else if (val <= 2097151) { | 
| 46 | 		return(3); | 
| 47 | 	} else if (val <= 268435455) { | 
| 48 | 		return(4); | 
| 49 | 	} else { | 
| 50 | 		/* Possibly we should care that on 64-bit machines ulint can | 
| 51 | 		contain values that we can't encode in 5 bytes, but | 
| 52 | 		fts_encode_int doesn't handle them either so it doesn't much | 
| 53 | 		matter. */ | 
| 54 |  | 
| 55 | 		return(5); | 
| 56 | 	} | 
| 57 | } | 
| 58 |  | 
| 59 | /******************************************************************//** | 
| 60 | Encode an integer using our VLC scheme and return the length in bytes. | 
| 61 | @return length of value encoded, in bytes */ | 
| 62 | UNIV_INLINE | 
| 63 | ulint | 
| 64 | fts_encode_int( | 
| 65 | /*===========*/ | 
| 66 | 	ulint	val,	/* in: value to encode */ | 
| 67 | 	byte*	buf)	/* in: buffer, must have enough space */ | 
| 68 | { | 
| 69 | 	ulint	len; | 
| 70 |  | 
| 71 | 	if (val <= 127) { | 
| 72 | 		*buf = (byte) val; | 
| 73 |  | 
| 74 | 		len = 1; | 
| 75 | 	} else if (val <= 16383) { | 
| 76 | 		*buf++ = (byte)(val >> 7); | 
| 77 | 		*buf = (byte)(val & 0x7F); | 
| 78 |  | 
| 79 | 		len = 2; | 
| 80 | 	} else if (val <= 2097151) { | 
| 81 | 		*buf++ = (byte)(val >> 14); | 
| 82 | 		*buf++ = (byte)((val >> 7) & 0x7F); | 
| 83 | 		*buf = (byte)(val & 0x7F); | 
| 84 |  | 
| 85 | 		len = 3; | 
| 86 | 	} else if (val <= 268435455) { | 
| 87 | 		*buf++ = (byte)(val >> 21); | 
| 88 | 		*buf++ = (byte)((val >> 14) & 0x7F); | 
| 89 | 		*buf++ = (byte)((val >> 7) & 0x7F); | 
| 90 | 		*buf = (byte)(val & 0x7F); | 
| 91 |  | 
| 92 | 		len = 4; | 
| 93 | 	} else { | 
| 94 | 		/* Best to keep the limitations of the 32/64 bit versions | 
| 95 | 		identical, at least for the time being. */ | 
| 96 | 		ut_ad(val <= 4294967295u); | 
| 97 |  | 
| 98 | 		*buf++ = (byte)(val >> 28); | 
| 99 | 		*buf++ = (byte)((val >> 21) & 0x7F); | 
| 100 | 		*buf++ = (byte)((val >> 14) & 0x7F); | 
| 101 | 		*buf++ = (byte)((val >> 7) & 0x7F); | 
| 102 | 		*buf = (byte)(val & 0x7F); | 
| 103 |  | 
| 104 | 		len = 5; | 
| 105 | 	} | 
| 106 |  | 
| 107 | 	/* High-bit on means "last byte in the encoded integer". */ | 
| 108 | 	*buf |= 0x80; | 
| 109 |  | 
| 110 | 	return(len); | 
| 111 | } | 
| 112 |  | 
| 113 | /******************************************************************//** | 
| 114 | Decode and return the integer that was encoded using our VLC scheme. | 
| 115 | @return value decoded */ | 
| 116 | UNIV_INLINE | 
| 117 | ulint | 
| 118 | fts_decode_vlc( | 
| 119 | /*===========*/ | 
| 120 | 	byte**	ptr)	/* in: ptr to decode from, this ptr is | 
| 121 | 			incremented by the number of bytes decoded */ | 
| 122 | { | 
| 123 | 	ulint	val = 0; | 
| 124 |  | 
| 125 | 	for (;;) { | 
| 126 | 		byte	b = **ptr; | 
| 127 |  | 
| 128 | 		++*ptr; | 
| 129 | 		val |= (b & 0x7F); | 
| 130 |  | 
| 131 | 		/* High-bit on means "last byte in the encoded integer". */ | 
| 132 | 		if (b & 0x80) { | 
| 133 | 			break; | 
| 134 | 		} else { | 
| 135 | 			val <<= 7; | 
| 136 | 		} | 
| 137 | 	} | 
| 138 |  | 
| 139 | 	return(val); | 
| 140 | } | 
| 141 |  | 
| 142 | #endif | 
| 143 |  |