coreseek(sphinx)全文检索

coreseek是基于sphinx全文检索引擎的，与sphinx相比就是支持中文分词。

在搜索这块有挺不错的性能，亲测50多万的数据，用coreseek的搜索方案比原始mysql的like查询性能提高10倍左右。

一、安装（前提安装好LAMP）

1、下载解压coreseek
tar -zxvf coreseek-3.2.14.tar.gz


2、先安装mmseg中文分词
cd coreseek-3.2.14/mmesg-3.2.14
./configure --prefix=/usr/local/mmseg
（如果报错误error: cannot find input file: src/Makefile.in
运行
aclocal
libtoolize --force 
automake --add-missing
autoconf
autoheader
make clean
再configure && make && make install ）


3、安装coreseek
cd coreseek-3.2.14/csft-3.2.14/

./configure --prefix=/usr/local/coreseek --with-mysql=/usr/local/mysql --with-mmseg=/usr/local/mmseg --with-mmseg-includes=/usr/local/mmseg/include/mmseg/ --with-mmseg-libs=/usr/local/mmseg/lib/

make && make install

二、配置测试coreseek

4、配置coreseek
cd /usr/local/coreseek/etc
cp sphinx.conf.dist csft.conf
vi csft.conf
csft.conf的结构为
source src1{ }    主数据源
source src1throttled : src1{ }  增量数据源
index test1{ }     主数据索引
index test1stemmed : test1{ }  增量数据索引
index dist1{ }     分布式索引
indexer              索引器
searchd             服务进程


4.1 在主数据源source src1{}里
更改数据库主机，用户名，密码，sql_sock


更改
# sql_query_pre = SET NAMES utf8
# sql_query_pre = SET SESSION query_cache_type=OFF
为
# sql_query_pre = SET NAMES utf8
sql_query_pre = SET SESSION query_cache_type=OFF


更改
sql_query               = \
SELECT id, group_id, UNIX_TIMESTAMP(date_added) AS date_added, title, content \
FROM documents
为
sql_query     =  select id,title,content from post;


更改
sql_attr_uint                  = group_id
为
#sql_attr_uint                  = group_id


更改
sql_attr_timestamp             = date_added
为
#sql_attr_timestamp             = date_added


更改
sql_query_info          = SELECT * FROM document WHERE id=$id 
为
sql_query_info          = SELECT * FROM post WHERE id=$id


4.2 在增量数据源source src1throttled : src1{ }里
注释整个source src1throttled : src1{ }，包括source src1throttled : src1{ }


4.3 在主数据索引index test1{ }里
更改
index test1{ }
为
index src1{ }


更改
source    =    test1
为
source    =    src1


更改
path   =  /usr/local/sphinx/var/data/test1
为
path   =  /usr/local/sphinx/var/data/src1


更改
stopwords                      = G:\data\stopwords.txt
为
#stopwords                      = G:\data\stopwords.txt


更改
wordforms                      = G:\data\wordforms.txt
为
#wordforms                      = G:\data\wordforms.txt


更改
exceptions             = /data/exceptions.txt
为
#exceptions             = /data/exceptions.txt


更改
charset_type = sbcs
为
charset_type  = zh_cn.utf-8
charset_dictpath = /usr/local/mmseg/etc/


4.4 注释全部  index test1stemmed : test1{ }  增量数据索引
4.5 注释全部  index dist1{ }     分布式索引


4.6 在索引器indexer里
更改
mem_limit   = 32M
为
mem_limit   = 128M


4.7 在服务进程searchd里不需更改

或者参照

</pre><pre name="code" class="html">#
# Sphinx configuration file sample
#
# WARNING! While this sample file mentions all available options,
# it contains (very) short helper descriptions only. Please refer to
# doc/sphinx.html for details.
#
#############################################################################
## data source definition
#############################################################################

source src1
{
	# data source type. mandatory, no default value
	# known types are mysql, pgsql, mssql, xmlpipe, xmlpipe2, odbc
	type					= mysql

	#####################################################################
	## SQL settings (for 'mysql' and 'pgsql' types)
	#####################################################################

	# some straightforward parameters for SQL source types
	sql_host				= localhost
	sql_user				= root
	sql_pass				= 111111
	sql_db					= mraz
	sql_port				= 3306	# optional, default is 3306

	# UNIX socket name
	# optional, default is empty (reuse client library defaults)
	# usually '/var/lib/mysql/mysql.sock' on Linux
	# usually '/tmp/mysql.sock' on FreeBSD
	#
	# sql_sock				= /tmp/mysql.sock


	# MySQL specific client connection flags
	# optional, default is 0
	#
	# mysql_connect_flags	= 32 # enable compression

	# MySQL specific SSL certificate settings
	# optional, defaults are empty
	#
	# mysql_ssl_cert		= /etc/ssl/client-cert.pem
	# mysql_ssl_key		= /etc/ssl/client-key.pem
	# mysql_ssl_ca		= /etc/ssl/cacert.pem

	# MS SQL specific Windows authentication mode flag
	# MUST be in sync with charset_type index-level setting
	# optional, default is 0
	#
	# mssql_winauth			= 1 # use currently logged on user credentials


	# MS SQL specific Unicode indexing flag
	# optional, default is 0 (request SBCS data)
	#
	# mssql_unicode			= 1 # request Unicode data from server


	# ODBC specific DSN (data source name)
	# mandatory for odbc source type, no default value
	#
	# odbc_dsn				= DBQ=C:\data;DefaultDir=C:\data;Driver={Microsoft Text Driver (*.txt; *.csv)};
	# sql_query				= SELECT id, data FROM documents.csv


	# pre-query, executed before the main fetch query
	# multi-value, optional, default is empty list of queries
	#
	# sql_query_pre			= SET NAMES utf8
	 sql_query_pre			= SET SESSION query_cache_type=OFF


	# main document fetch query
	# mandatory, integer document ID field MUST be the first selected column
	#sql_query				= \
	#	SELECT id, group_id, UNIX_TIMESTAMP(date_added) AS date_added, title, content \
	#	FROM documents

	sql_query = select id,title,content from tbContent;

	# range query setup, query that must return min and max ID values
	# optional, default is empty
	#
	# sql_query will need to reference $start and $end boundaries
	# if using ranged query:
	#
	# sql_query				= \
	#	SELECT doc.id, doc.id AS group, doc.title, doc.data \
	#	FROM documents doc \
	#	WHERE id>=$start AND id<=$end
	#
	# sql_query_range		= SELECT MIN(id),MAX(id) FROM documents


	# range query step
	# optional, default is 1024
	#
	# sql_range_step		= 1000


	# unsigned integer attribute declaration
	# multi-value (an arbitrary number of attributes is allowed), optional
	# optional bit size can be specified, default is 32
	#
	# sql_attr_uint			= author_id
	# sql_attr_uint			= forum_id:9 # 9 bits for forum_id
	#sql_attr_uint			= group_id

	# boolean attribute declaration
	# multi-value (an arbitrary number of attributes is allowed), optional
	# equivalent to sql_attr_uint with 1-bit size
	#
	# sql_attr_bool			= is_deleted


	# bigint attribute declaration
	# multi-value (an arbitrary number of attributes is allowed), optional
	# declares a signed (unlike uint!) 64-bit attribute
	#
	# sql_attr_bigint			= my_bigint_id


	# UNIX timestamp attribute declaration
	# multi-value (an arbitrary number of attributes is allowed), optional
	# similar to integer, but can also be used in date functions
	#
	# sql_attr_timestamp	= posted_ts
	# sql_attr_timestamp	= last_edited_ts
	#sql_attr_timestamp		= date_added

	# string ordinal attribute declaration
	# multi-value (an arbitrary number of attributes is allowed), optional
	# sorts strings (bytewise), and stores their indexes in the sorted list
	# sorting by this attr is equivalent to sorting by the original strings
	#
	# sql_attr_str2ordinal	= author_name


	# floating point attribute declaration
	# multi-value (an arbitrary number of attributes is allowed), optional
	# values are stored in single precision, 32-bit IEEE 754 format
	#
	# sql_attr_float = lat_radians
	# sql_attr_float = long_radians


	# multi-valued attribute (MVA) attribute declaration
	# multi-value (an arbitrary number of attributes is allowed), optional
	# MVA values are variable length lists of unsigned 32-bit integers
	#
	# syntax is ATTR-TYPE ATTR-NAME 'from' SOURCE-TYPE [;QUERY] [;RANGE-QUERY]
	# ATTR-TYPE is 'uint' or 'timestamp'
	# SOURCE-TYPE is 'field', 'query', or 'ranged-query'
	# QUERY is SQL query used to fetch all ( docid, attrvalue ) pairs
	# RANGE-QUERY is SQL query used to fetch min and max ID values, similar to 'sql_query_range'
	#
	# sql_attr_multi	= uint tag from query; SELECT id, tag FROM tags
	# sql_attr_multi	= uint tag from ranged-query; \
	#	SELECT id, tag FROM tags WHERE id>=$start AND id<=$end; \
	#	SELECT MIN(id), MAX(id) FROM tags


	# post-query, executed on sql_query completion
	# optional, default is empty
	#
	# sql_query_post		=

	
	# post-index-query, executed on successful indexing completion
	# optional, default is empty
	# $maxid expands to max document ID actually fetched from DB
	#
	# sql_query_post_index = REPLACE INTO counters ( id, val ) \
	#	VALUES ( 'max_indexed_id', $maxid )


	# ranged query throttling, in milliseconds
	# optional, default is 0 which means no delay
	# enforces given delay before each query step
	sql_ranged_throttle	= 0

	# document info query, ONLY for CLI search (ie. testing and debugging)
	# optional, default is empty
	# must contain $id macro and must fetch the document by that id
	#sql_query_info		= SELECT * FROM documents WHERE id=$id
	sql_query_info		= SELECT * FROM tbContent WHERE id=$id

	# kill-list query, fetches the document IDs for kill-list
	# k-list will suppress matches from preceding indexes in the same query
	# optional, default is empty
	#
	# sql_query_killlist	= SELECT id FROM documents WHERE edited>=@last_reindex


	# columns to unpack on indexer side when indexing
	# multi-value, optional, default is empty list
	#
	# unpack_zlib = zlib_column
	# unpack_mysqlcompress = compressed_column
	# unpack_mysqlcompress = compressed_column_2


	# maximum unpacked length allowed in MySQL COMPRESS() unpacker
	# optional, default is 16M
	#
	# unpack_mysqlcompress_maxsize = 16M


	#####################################################################
	## xmlpipe settings
	#####################################################################

	# type				= xmlpipe

	# shell command to invoke xmlpipe stream producer
	# mandatory
	#
	# xmlpipe_command	= cat /usr/local/coreseek/var/test.xml

	#####################################################################
	## xmlpipe2 settings
	#####################################################################

	# type				= xmlpipe2
	# xmlpipe_command	= cat /usr/local/coreseek/var/test2.xml


	# xmlpipe2 field declaration
	# multi-value, optional, default is empty
	#
	# xmlpipe_field				= subject
	# xmlpipe_field				= content


	# xmlpipe2 attribute declaration
	# multi-value, optional, default is empty
	# all xmlpipe_attr_XXX options are fully similar to sql_attr_XXX
	#
	# xmlpipe_attr_timestamp	= published
	# xmlpipe_attr_uint			= author_id


	# perform UTF-8 validation, and filter out incorrect codes
	# avoids XML parser choking on non-UTF-8 documents
	# optional, default is 0
	#
	# xmlpipe_fixup_utf8		= 1
}


# inherited source example
#
# all the parameters are copied from the parent source,
# and may then be overridden in this source definition
#source src1throttled : src1
#{
#	sql_ranged_throttle			= 100
#}

#############################################################################
## index definition
#############################################################################

# local index example
#
# this is an index which is stored locally in the filesystem
#
# all indexing-time options (such as morphology and charsets)
# are configured per local index
index src1
{
	# document source(s) to index
	# multi-value, mandatory
	# document IDs must be globally unique across all sources
	source			= src1

	# index files path and file name, without extension
	# mandatory, path must be writable, extensions will be auto-appended
	path			= /usr/local/coreseek/var/data/src1

	# document attribute values (docinfo) storage mode
	# optional, default is 'extern'
	# known values are 'none', 'extern' and 'inline'
	docinfo			= extern

	# memory locking for cached data (.spa and .spi), to prevent swapping
	# optional, default is 0 (do not mlock)
	# requires searchd to be run from root
	mlock			= 0

	# a list of morphology preprocessors to apply
	# optional, default is empty
	#
	# builtin preprocessors are 'none', 'stem_en', 'stem_ru', 'stem_enru',
	# 'soundex', and 'metaphone'; additional preprocessors available from
	# libstemmer are 'libstemmer_XXX', where XXX is algorithm code
	# (see libstemmer_c/libstemmer/modules.txt)
	#
	# morphology 	= stem_en, stem_ru, soundex
	# morphology	= libstemmer_german
	# morphology	= libstemmer_sv
	morphology		= none

	# minimum word length at which to enable stemming
	# optional, default is 1 (stem everything)
	#
	# min_stemming_len	= 1


	# stopword files list (space separated)
	# optional, default is empty
	# contents are plain text, charset_table and stemming are both applied
	#
	#stopwords			= G:\data\stopwords.txt


	# wordforms file, in "mapfrom > mapto" plain text format
	# optional, default is empty
	#
	#wordforms			= G:\data\wordforms.txt


	# tokenizing exceptions file
	# optional, default is empty
	#
	# plain text, case sensitive, space insensitive in map-from part
	# one "Map Several Words => ToASingleOne" entry per line
	#
	#exceptions		= /data/exceptions.txt


	# minimum indexed word length
	# default is 1 (index everything)
	min_word_len		= 1

	# charset encoding type
	# optional, default is 'sbcs'
	# known types are 'sbcs' (Single Byte CharSet) and 'utf-8'
	charset_type		= zh_cn.utf-8
	charset_dictpath = /usr/local/mmseg/etc/

	# charset definition and case folding rules "table"
	# optional, default value depends on charset_type
	#
	# defaults are configured to include English and Russian characters only
	# you need to change the table to include additional ones
	# this behavior MAY change in future versions
	#
	# 'sbcs' default value is
	# charset_table		= 0..9, A..Z->a..z, _, a..z, U+A8->U+B8, U+B8, U+C0..U+DF->U+E0..U+FF, U+E0..U+FF
	#
	# 'utf-8' default value is
	# charset_table		= 0..9, A..Z->a..z, _, a..z, U+410..U+42F->U+430..U+44F, U+430..U+44F


	# ignored characters list
	# optional, default value is empty
	#
	# ignore_chars		= U+00AD


	# minimum word prefix length to index
	# optional, default is 0 (do not index prefixes)
	#
	# min_prefix_len	= 0


	# minimum word infix length to index
	# optional, default is 0 (do not index infixes)
	#
	# min_infix_len		= 0


	# list of fields to limit prefix/infix indexing to
	# optional, default value is empty (index all fields in prefix/infix mode)
	#
	# prefix_fields		= filename
	# infix_fields		= url, domain


	# enable star-syntax (wildcards) when searching prefix/infix indexes
	# known values are 0 and 1
	# optional, default is 0 (do not use wildcard syntax)
	#
	# enable_star		= 1


	# n-gram length to index, for CJK indexing
	# only supports 0 and 1 for now, other lengths to be implemented
	# optional, default is 0 (disable n-grams)
	#
	# ngram_len				= 1


	# n-gram characters list, for CJK indexing
	# optional, default is empty
	#
	# ngram_chars			= U+3000..U+2FA1F


	# phrase boundary characters list
	# optional, default is empty
	#
	# phrase_boundary		= ., ?, !, U+2026 # horizontal ellipsis


	# phrase boundary word position increment
	# optional, default is 0
	#
	# phrase_boundary_step	= 100


	# whether to strip HTML tags from incoming documents
	# known values are 0 (do not strip) and 1 (do strip)
	# optional, default is 0
	html_strip				= 0

	# what HTML attributes to index if stripping HTML
	# optional, default is empty (do not index anything)
	#
	# html_index_attrs		= img=alt,title; a=title;


	# what HTML elements contents to strip
	# optional, default is empty (do not strip element contents)
	#
	# html_remove_elements	= style, script


	# whether to preopen index data files on startup
	# optional, default is 0 (do not preopen), searchd-only
	#
	# preopen					= 1


	# whether to keep dictionary (.spi) on disk, or cache it in RAM
	# optional, default is 0 (cache in RAM), searchd-only
	#
	# ondisk_dict				= 1


	# whether to enable in-place inversion (2x less disk, 90-95% speed)
	# optional, default is 0 (use separate temporary files), indexer-only
	#
	# inplace_enable			= 1


	# in-place fine-tuning options
	# optional, defaults are listed below
	#
	# inplace_hit_gap			= 0		# preallocated hitlist gap size
	# inplace_docinfo_gap		= 0		# preallocated docinfo gap size
	# inplace_reloc_factor	= 0.1	# relocation buffer size within arena
	# inplace_write_factor	= 0.1	# write buffer size within arena


	# whether to index original keywords along with stemmed versions
	# enables "=exactform" operator to work
	# optional, default is 0
	#
	# index_exact_words		= 1


	# position increment on overshort (less that min_word_len) words
	# optional, allowed values are 0 and 1, default is 1
	#
	# overshort_step			= 1


	# position increment on stopword
	# optional, allowed values are 0 and 1, default is 1
	#
	# stopword_step			= 1
}


# inherited index example
#
# all the parameters are copied from the parent index,
# and may then be overridden in this index definition
#index test1stemmed : test1
#{
#	path			= /usr/local/coreseek/var/data/test1stemmed
#	morphology		= stem_en
#}


# distributed index example
#
# this is a virtual index which can NOT be directly indexed,
# and only contains references to other local and/or remote indexes
#index dist1
#{
	# 'distributed' index type MUST be specified
#	type				= distributed

	# local index to be searched
	# there can be many local indexes configured
#	local				= test1
#	local				= test1stemmed

	# remote agent
	# multiple remote agents may be specified
	# syntax for TCP connections is 'hostname:port:index1,[index2[,...]]'
	# syntax for local UNIX connections is '/path/to/socket:index1,[index2[,...]]'
#	agent				= localhost:9313:remote1
#	agent				= localhost:9314:remote2,remote3
	# agent				= /var/run/searchd.sock:remote4

	# blackhole remote agent, for debugging/testing
	# network errors and search results will be ignored
	#
	# agent_blackhole		= testbox:9312:testindex1,testindex2


	# remote agent connection timeout, milliseconds
	# optional, default is 1000 ms, ie. 1 sec
#	agent_connect_timeout	= 1000

	# remote agent query timeout, milliseconds
	# optional, default is 3000 ms, ie. 3 sec
#	agent_query_timeout		= 3000
#}

#############################################################################
## indexer settings
#############################################################################

indexer
{
	# memory limit, in bytes, kiloytes (16384K) or megabytes (256M)
	# optional, default is 32M, max is 2047M, recommended is 256M to 1024M
	mem_limit			= 128M

	# maximum IO calls per second (for I/O throttling)
	# optional, default is 0 (unlimited)
	#
	# max_iops			= 40


	# maximum IO call size, bytes (for I/O throttling)
	# optional, default is 0 (unlimited)
	#
	# max_iosize		= 1048576


	# maximum xmlpipe2 field length, bytes
	# optional, default is 2M
	#
	# max_xmlpipe2_field	= 4M


	# write buffer size, bytes
	# several (currently up to 4) buffers will be allocated
	# write buffers are allocated in addition to mem_limit
	# optional, default is 1M
	#
	# write_buffer		= 1M
}

#############################################################################
## searchd settings
#############################################################################

searchd
{
	# hostname, port, or hostname:port, or /unix/socket/path to listen on
	# multi-value, multiple listen points are allowed
	# optional, default is 0.0.0.0:9312 (listen on all interfaces, port 9312)
	#
	# listen				= 127.0.0.1
	# listen				= 192.168.0.1:9312
	# listen				= 9312
	# listen				= /var/run/searchd.sock


	# log file, searchd run info is logged here
	# optional, default is 'searchd.log'
	log					= /usr/local/coreseek/var/log/searchd.log

	# query log file, all search queries are logged here
	# optional, default is empty (do not log queries)
	query_log			= /usr/local/coreseek/var/log/query.log

	# client read timeout, seconds
	# optional, default is 5
	read_timeout		= 5

	# request timeout, seconds
	# optional, default is 5 minutes
	client_timeout		= 300

	# maximum amount of children to fork (concurrent searches to run)
	# optional, default is 0 (unlimited)
	max_children		= 30

	# PID file, searchd process ID file name
	# mandatory
	pid_file			= /usr/local/coreseek/var/log/searchd.pid

	# max amount of matches the daemon ever keeps in RAM, per-index
	# WARNING, THERE'S ALSO PER-QUERY LIMIT, SEE SetLimits() API CALL
	# default is 1000 (just like Google)
	max_matches			= 999999

	# seamless rotate, prevents rotate stalls if precaching huge datasets
	# optional, default is 1
	seamless_rotate		= 1

	# whether to forcibly preopen all indexes on startup
	# optional, default is 0 (do not preopen)
	preopen_indexes		= 0

	# whether to unlink .old index copies on succesful rotation.
	# optional, default is 1 (do unlink)
	unlink_old			= 1

	# attribute updates periodic flush timeout, seconds
	# updates will be automatically dumped to disk this frequently
	# optional, default is 0 (disable periodic flush)
	#
	# attr_flush_period	= 900


	# instance-wide ondisk_dict defaults (per-index value take precedence)
	# optional, default is 0 (precache all dictionaries in RAM)
	#
	# ondisk_dict_default	= 1


	# MVA updates pool size
	# shared between all instances of searchd, disables attr flushes!
	# optional, default size is 1M
	mva_updates_pool	= 1M

	# max allowed network packet size
	# limits both query packets from clients, and responses from agents
	# optional, default size is 8M
	max_packet_size		= 8M

	# crash log path
	# searchd will (try to) log crashed query to 'crash_log_path.PID' file
	# optional, default is empty (do not create crash logs)
	#
	# crash_log_path		= /usr/local/coreseek/var/log/crash


	# max allowed per-query filter count
	# optional, default is 256
	max_filters			= 256

	# max allowed per-filter values count
	# optional, default is 4096
	max_filter_values	= 4096


	# socket listen queue length
	# optional, default is 5
	#
	# listen_backlog		= 5


	# per-keyword read buffer size
	# optional, default is 256K
	#
	# read_buffer			= 256K


	# unhinted read size (currently used when reading hits)
	# optional, default is 32K
	#
	# read_unhinted		= 32K
}

# --eof--

4.8 测试，数据库加入测试数据
create table tbContent(id int(11) primary key auto_increment,title varchar(100),content varchar(255))default charset utf8;

insert into tbContent(title,content) values('php啊啊啊123','php123欢迎你');
insert into tbContent(title,content) values('php中国好123','php123欢迎你');
insert into tbContent(title,content) values('php中国好123','php123无限好');
insert into tbContent(title,content) values('hello123','php一起无限好');

insert into tbContent(title,content) values('hello123','php旅游无限好');

insert into tbContent(title,content) values('hello123','我们和你们一起去旅游吧');

insert into tbContent(title,content) select title,content from tbContent;##翻倍新增数据，增至50多万

4.9、为coreseek创建索引
/usr/local/coreseek/bin/indexer --all
或指定配置文件
/usr/local/coreseek/bin/indexer -c /usr/local/coreseek/etc/csft.conf --all
(如果启动了服务进程，则/usr/local/coreseek/bin/indexer -c /usr/local/coreseek/etc/csft.conf --all --rotate)


ps：如果报错，error while loading shared libraries: libmysqlclient.so.16: cannot open shared object file: No such file or directory
就执行
ln -s /usr/local/mysql/lib/mysql/libmysqlclient.so /usr/lib/libmysqlclient.so
还不行，则
vim /etc/ld.so.conf
添加/usr/local/mysql/lib
然后执行ldconfig


4.10、测试搜索
/usr/local/coreseek/bin/search 欢迎
或指定配置文件

/usr/local/coreseek/bin/search -c /usr/local/coreseek/etc/csft.conf 欢迎

能搜索处理结果表示配置成功

5、在/usr/local/apache/htdocs下，新建test_with_mysql.php和testCoreseek_with_mysql.php，以便测试用coreseek的搜索方案比原始mysql的like查询性能提高多少

test_with_mysql.php内容如下

<?php
$start_time = microtime();
header("content-type:text/html;charset=utf-8;");

$keyword = $_REQUEST['keyword'];

$mysql_server_name='localhost';
$mysql_username='root'; 
$mysql_password='111111'; 
$mysql_database='mraz'; 

$conn=mysql_connect($mysql_server_name,$mysql_username,$mysql_password) or die("error connecting") ;
mysql_query("set names utf8"); 
mysql_select_db($mysql_database); 
$sql ="select * from tbContent where title like '%".$keyword."%' or content like '%".$keyword."%' limit 1000"; 
//var_dump($sql);
$result = mysql_query($sql,$conn); 

//var_dump($result);

if($result){
	$end_time = microtime();
	var_dump( $end_time - $start_time );
}

//while($row = mysql_fetch_array($result)){
//  echo $row['title'] . "--" . $row['content'];
//  echo "<br />";
//}

mysql_close($conn);

testCoreseek_with_mysql.php内容如下

<?php
$start_time = microtime();
header("content-type:text/html;charset=utf-8;");

require_once("sphinxapi.php");

$sphinx = new SphinxClient();
$sphinx->SetServer("localhost",9312);
$sphinx->SetMatchMode(SPH_MATCH_ANY);
$keyword = $_REQUEST['keyword'];
$result = $sphinx->query($keyword,"*");

$id_arr = array();
$id_str = '';
foreach($result['matches'] as $id => $ret){
        $id_arr[] = $id;
		$id_str .= $id.",";
}

$id_str .= "0";

//var_dump($id_arr);//根据id找数据表对应内容

$mysql_server_name='localhost';
$mysql_username='root'; 
$mysql_password='111111'; 
$mysql_database='mraz'; 

$conn=mysql_connect($mysql_server_name,$mysql_username,$mysql_password) or die("error connecting") ;
mysql_query("set names utf8"); 
mysql_select_db($mysql_database); 
$sql ="select * from tbContent where id in (".$id_str.")"; 
//var_dump($sql);
$result = mysql_query($sql,$conn); 

//var_dump($result);

if($result){
	$end_time = microtime();
	var_dump( $end_time - $start_time );
}

//while($row = mysql_fetch_array($result)){
//  echo $row['title'] . "--" . $row['content'];
//  echo "<br />";
//}

mysql_close($conn);

testCoreseek_with_mysql.php需要引入的sphinxapi.php内容如下

<?php

//
// $Id: sphinxapi.php 2055 2009-11-06 23:09:58Z shodan $
//

//
// Copyright (c) 2001-2008, Andrew Aksyonoff. All rights reserved.
//
// This program is free software; you can redistribute it and/or modify
// it under the terms of the GNU General Public License. You should have
// received a copy of the GPL license along with this program; if you
// did not, you can find it at http://www.gnu.org/
//

/////////////////////////////////////////////////////////////////////////////
// PHP version of Sphinx searchd client (PHP API)
/////////////////////////////////////////////////////////////////////////////

/// known searchd commands
define ( "SEARCHD_COMMAND_SEARCH",	0 );
define ( "SEARCHD_COMMAND_EXCERPT",	1 );
define ( "SEARCHD_COMMAND_UPDATE",	2 );
define ( "SEARCHD_COMMAND_KEYWORDS",3 );
define ( "SEARCHD_COMMAND_PERSIST",	4 );
define ( "SEARCHD_COMMAND_STATUS",	5 );
define ( "SEARCHD_COMMAND_QUERY",	6 );

/// current client-side command implementation versions
define ( "VER_COMMAND_SEARCH",		0x116 );
define ( "VER_COMMAND_EXCERPT",		0x100 );
define ( "VER_COMMAND_UPDATE",		0x102 );
define ( "VER_COMMAND_KEYWORDS",	0x100 );
define ( "VER_COMMAND_STATUS",		0x100 );
define ( "VER_COMMAND_QUERY",		0x100 );

/// known searchd status codes
define ( "SEARCHD_OK",				0 );
define ( "SEARCHD_ERROR",			1 );
define ( "SEARCHD_RETRY",			2 );
define ( "SEARCHD_WARNING",			3 );

/// known match modes
define ( "SPH_MATCH_ALL",			0 );
define ( "SPH_MATCH_ANY",			1 );
define ( "SPH_MATCH_PHRASE",		2 );
define ( "SPH_MATCH_BOOLEAN",		3 );
define ( "SPH_MATCH_EXTENDED",		4 );
define ( "SPH_MATCH_FULLSCAN",		5 );
define ( "SPH_MATCH_EXTENDED2",		6 );	// extended engine V2 (TEMPORARY, WILL BE REMOVED)

/// known ranking modes (ext2 only)
define ( "SPH_RANK_PROXIMITY_BM25",	0 );	///< default mode, phrase proximity major factor and BM25 minor one
define ( "SPH_RANK_BM25",			1 );	///< statistical mode, BM25 ranking only (faster but worse quality)
define ( "SPH_RANK_NONE",			2 );	///< no ranking, all matches get a weight of 1
define ( "SPH_RANK_WORDCOUNT",		3 );	///< simple word-count weighting, rank is a weighted sum of per-field keyword occurence counts
define ( "SPH_RANK_PROXIMITY",		4 );
define ( "SPH_RANK_MATCHANY",		5 );
define ( "SPH_RANK_FIELDMASK",		6 );

/// known sort modes
define ( "SPH_SORT_RELEVANCE",		0 );
define ( "SPH_SORT_ATTR_DESC",		1 );
define ( "SPH_SORT_ATTR_ASC",		2 );
define ( "SPH_SORT_TIME_SEGMENTS", 	3 );
define ( "SPH_SORT_EXTENDED", 		4 );
define ( "SPH_SORT_EXPR", 			5 );

/// known filter types
define ( "SPH_FILTER_VALUES",		0 );
define ( "SPH_FILTER_RANGE",		1 );
define ( "SPH_FILTER_FLOATRANGE",	2 );

/// known attribute types
define ( "SPH_ATTR_INTEGER",		1 );
define ( "SPH_ATTR_TIMESTAMP",		2 );
define ( "SPH_ATTR_ORDINAL",		3 );
define ( "SPH_ATTR_BOOL",			4 );
define ( "SPH_ATTR_FLOAT",			5 );
define ( "SPH_ATTR_BIGINT",			6 );
define ( "SPH_ATTR_MULTI",			0x40000000 );

/// known grouping functions
define ( "SPH_GROUPBY_DAY",			0 );
define ( "SPH_GROUPBY_WEEK",		1 );
define ( "SPH_GROUPBY_MONTH",		2 );
define ( "SPH_GROUPBY_YEAR",		3 );
define ( "SPH_GROUPBY_ATTR",		4 );
define ( "SPH_GROUPBY_ATTRPAIR",	5 );

// important properties of PHP's integers:
//  - always signed (one bit short of PHP_INT_SIZE)
//  - conversion from string to int is saturated
//  - float is double
//  - div converts arguments to floats
//  - mod converts arguments to ints

// the packing code below works as follows:
//  - when we got an int, just pack it
//    if performance is a problem, this is the branch users should aim for
//
//  - otherwise, we got a number in string form
//    this might be due to different reasons, but we assume that this is
//    because it didn't fit into PHP int
//
//  - factor the string into high and low ints for packing
//    - if we have bcmath, then it is used
//    - if we don't, we have to do it manually (this is the fun part)
//
//    - x64 branch does factoring using ints
//    - x32 (ab)uses floats, since we can't fit unsigned 32-bit number into an int
//
// unpacking routines are pretty much the same.
//  - return ints if we can
//  - otherwise format number into a string

/// pack 64-bit signed
function sphPackI64 ( $v )
{
	assert ( is_numeric($v) );
	
	// x64
	if ( PHP_INT_SIZE>=8 )
	{
		$v = (int)$v;
		return pack ( "NN", $v>>32, $v&0xFFFFFFFF );
	}

	// x32, int
	if ( is_int($v) )
		return pack ( "NN", $v < 0 ? -1 : 0, $v );

	// x32, bcmath	
	if ( function_exists("bcmul") )
	{
		if ( bccomp ( $v, 0 ) == -1 )
			$v = bcadd ( "18446744073709551616", $v );
		$h = bcdiv ( $v, "4294967296", 0 );
		$l = bcmod ( $v, "4294967296" );
		return pack ( "NN", (float)$h, (float)$l ); // conversion to float is intentional; int would lose 31st bit
	}

	// x32, no-bcmath
	$p = max(0, strlen($v) - 13);
	$lo = abs((float)substr($v, $p));
	$hi = abs((float)substr($v, 0, $p));

	$m = $lo + $hi*1316134912.0; // (10 ^ 13) % (1 << 32) = 1316134912
	$q = floor($m/4294967296.0);
	$l = $m - ($q*4294967296.0);
	$h = $hi*2328.0 + $q; // (10 ^ 13) / (1 << 32) = 2328

	if ( $v<0 )
	{
		if ( $l==0 )
			$h = 4294967296.0 - $h;
		else
		{
			$h = 4294967295.0 - $h;
			$l = 4294967296.0 - $l;
		}
	}
	return pack ( "NN", $h, $l );
}

/// pack 64-bit unsigned
function sphPackU64 ( $v )
{
	assert ( is_numeric($v) );
	
	// x64
	if ( PHP_INT_SIZE>=8 )
	{
		assert ( $v>=0 );
		
		// x64, int
		if ( is_int($v) )
			return pack ( "NN", $v>>32, $v&0xFFFFFFFF );
						  
		// x64, bcmath
		if ( function_exists("bcmul") )
		{
			$h = bcdiv ( $v, 4294967296, 0 );
			$l = bcmod ( $v, 4294967296 );
			return pack ( "NN", $h, $l );
		}
		
		// x64, no-bcmath
		$p = max ( 0, strlen($v) - 13 );
		$lo = (int)substr ( $v, $p );
		$hi = (int)substr ( $v, 0, $p );
	
		$m = $lo + $hi*1316134912;
		$l = $m % 4294967296;
		$h = $hi*2328 + (int)($m/4294967296);

		return pack ( "NN", $h, $l );
	}

	// x32, int
	if ( is_int($v) )
		return pack ( "NN", 0, $v );
	
	// x32, bcmath
	if ( function_exists("bcmul") )
	{
		$h = bcdiv ( $v, "4294967296", 0 );
		$l = bcmod ( $v, "4294967296" );
		return pack ( "NN", (float)$h, (float)$l ); // conversion to float is intentional; int would lose 31st bit
	}

	// x32, no-bcmath
	$p = max(0, strlen($v) - 13);
	$lo = (float)substr($v, $p);
	$hi = (float)substr($v, 0, $p);
	
	$m = $lo + $hi*1316134912.0;
	$q = floor($m / 4294967296.0);
	$l = $m - ($q * 4294967296.0);
	$h = $hi*2328.0 + $q;

	return pack ( "NN", $h, $l );
}

// unpack 64-bit unsigned
function sphUnpackU64 ( $v )
{
	list ( $hi, $lo ) = array_values ( unpack ( "N*N*", $v ) );

	if ( PHP_INT_SIZE>=8 )
	{
		if ( $hi<0 ) $hi += (1<<32); // because php 5.2.2 to 5.2.5 is totally fucked up again
		if ( $lo<0 ) $lo += (1<<32);

		// x64, int
		if ( $hi<=2147483647 )
			return ($hi<<32) + $lo;

		// x64, bcmath
		if ( function_exists("bcmul") )
			return bcadd ( $lo, bcmul ( $hi, "4294967296" ) );

		// x64, no-bcmath
		$C = 100000;
		$h = ((int)($hi / $C) << 32) + (int)($lo / $C);
		$l = (($hi % $C) << 32) + ($lo % $C);
		if ( $l>$C )
		{
			$h += (int)($l / $C);
			$l  = $l % $C;
		}

		if ( $h==0 )
			return $l;
		return sprintf ( "%d%05d", $h, $l );
	}

	// x32, int
	if ( $hi==0 )
	{
		if ( $lo>0 )
			return $lo;
		return sprintf ( "%u", $lo );
	}

	$hi = sprintf ( "%u", $hi );
	$lo = sprintf ( "%u", $lo );

	// x32, bcmath
	if ( function_exists("bcmul") )
		return bcadd ( $lo, bcmul ( $hi, "4294967296" ) );
	
	// x32, no-bcmath
	$hi = (float)$hi;
	$lo = (float)$lo;
	
	$q = floor($hi/10000000.0);
	$r = $hi - $q*10000000.0;
	$m = $lo + $r*4967296.0;
	$mq = floor($m/10000000.0);
	$l = $m - $mq*10000000.0;
	$h = $q*4294967296.0 + $r*429.0 + $mq;

	$h = sprintf ( "%.0f", $h );
	$l = sprintf ( "%07.0f", $l );
	if ( $h=="0" )
		return sprintf( "%.0f", (float)$l );
	return $h . $l;
}

// unpack 64-bit signed
function sphUnpackI64 ( $v )
{
	list ( $hi, $lo ) = array_values ( unpack ( "N*N*", $v ) );

	// x64
	if ( PHP_INT_SIZE>=8 )
	{
		if ( $hi<0 ) $hi += (1<<32); // because php 5.2.2 to 5.2.5 is totally fucked up again
		if ( $lo<0 ) $lo += (1<<32);

		return ($hi<<32) + $lo;
	}

	// x32, int
	if ( $hi==0 )
	{
		if ( $lo>0 )
			return $lo;
		return sprintf ( "%u", $lo );
	}
	// x32, int
	elseif ( $hi==-1 )
	{
		if ( $lo<0 )
			return $lo;
		return sprintf ( "%.0f", $lo - 4294967296.0 );
	}
	
	$neg = "";
	$c = 0;
	if ( $hi<0 )
	{
		$hi = ~$hi;
		$lo = ~$lo;
		$c = 1;
		$neg = "-";
	}	

	$hi = sprintf ( "%u", $hi );
	$lo = sprintf ( "%u", $lo );

	// x32, bcmath
	if ( function_exists("bcmul") )
		return $neg . bcadd ( bcadd ( $lo, bcmul ( $hi, "4294967296" ) ), $c );

	// x32, no-bcmath
	$hi = (float)$hi;
	$lo = (float)$lo;
	
	$q = floor($hi/10000000.0);
	$r = $hi - $q*10000000.0;
	$m = $lo + $r*4967296.0;
	$mq = floor($m/10000000.0);
	$l = $m - $mq*10000000.0 + $c;
	$h = $q*4294967296.0 + $r*429.0 + $mq;
	if ( $l==10000000 )
	{
		$l = 0;
		$h += 1;
	}

	$h = sprintf ( "%.0f", $h );
	$l = sprintf ( "%07.0f", $l );
	if ( $h=="0" )
		return $neg . sprintf( "%.0f", (float)$l );
	return $neg . $h . $l;
}


function sphFixUint ( $value )
{
	if ( PHP_INT_SIZE>=8 )
	{
		// x64 route, workaround broken unpack() in 5.2.2+
		if ( $value<0 ) $value += (1<<32);
		return $value;
	}
	else
	{
		// x32 route, workaround php signed/unsigned braindamage
		return sprintf ( "%u", $value );
	}
}


/// sphinx searchd client class
class SphinxClient
{
	var $_host;			///< searchd host (default is "localhost")
	var $_port;			///< searchd port (default is 9312)
	var $_offset;		///< how many records to seek from result-set start (default is 0)
	var $_limit;		///< how many records to return from result-set starting at offset (default is 20)
	var $_mode;			///< query matching mode (default is SPH_MATCH_ALL)
	var $_weights;		///< per-field weights (default is 1 for all fields)
	var $_sort;			///< match sorting mode (default is SPH_SORT_RELEVANCE)
	var $_sortby;		///< attribute to sort by (defualt is "")
	var $_min_id;		///< min ID to match (default is 0, which means no limit)
	var $_max_id;		///< max ID to match (default is 0, which means no limit)
	var $_filters;		///< search filters
	var $_groupby;		///< group-by attribute name
	var $_groupfunc;	///< group-by function (to pre-process group-by attribute value with)
	var $_groupsort;	///< group-by sorting clause (to sort groups in result set with)
	var $_groupdistinct;///< group-by count-distinct attribute
	var $_maxmatches;	///< max matches to retrieve
	var $_cutoff;		///< cutoff to stop searching at (default is 0)
	var $_retrycount;	///< distributed retries count
	var $_retrydelay;	///< distributed retries delay
	var $_anchor;		///< geographical anchor point
	var $_indexweights;	///< per-index weights
	var $_ranker;		///< ranking mode (default is SPH_RANK_PROXIMITY_BM25)
	var $_maxquerytime;	///< max query time, milliseconds (default is 0, do not limit)
	var $_fieldweights;	///< per-field-name weights
	var $_overrides;	///< per-query attribute values overrides
	var $_select;		///< select-list (attributes or expressions, with optional aliases)

	var $_error;		///< last error message
	var $_warning;		///< last warning message
	var $_connerror;		///< connection error vs remote error flag

	var $_reqs;			///< requests array for multi-query
	var $_mbenc;		///< stored mbstring encoding
	var $_arrayresult;	///< whether $result["matches"] should be a hash or an array
	var $_timeout;		///< connect timeout

	/////////////////////////////////////////////////////////////////////////////
	// common stuff
	/////////////////////////////////////////////////////////////////////////////

	/// create a new client object and fill defaults
	function SphinxClient ()
	{
		// per-client-object settings
		$this->_host		= "localhost";
		$this->_port		= 9312;
		$this->_path		= false;
		$this->_socket		= false;

		// per-query settings
		$this->_offset		= 0;
		$this->_limit		= 999999;
		$this->_mode		= SPH_MATCH_ALL;
		$this->_weights		= array ();
		$this->_sort		= SPH_SORT_RELEVANCE;
		$this->_sortby		= "";
		$this->_min_id		= 0;
		$this->_max_id		= 0;
		$this->_filters		= array ();
		$this->_groupby		= "";
		$this->_groupfunc	= SPH_GROUPBY_DAY;
		$this->_groupsort	= "@group desc";
		$this->_groupdistinct= "";
		$this->_maxmatches	= 1000;
		$this->_cutoff		= 0;
		$this->_retrycount	= 0;
		$this->_retrydelay	= 0;
		$this->_anchor		= array ();
		$this->_indexweights= array ();
		$this->_ranker		= SPH_RANK_PROXIMITY_BM25;
		$this->_maxquerytime= 0;
		$this->_fieldweights= array();
		$this->_overrides 	= array();
		$this->_select		= "*";

		$this->_error		= ""; // per-reply fields (for single-query case)
		$this->_warning		= "";
		$this->_connerror	= false;

		$this->_reqs		= array ();	// requests storage (for multi-query case)
		$this->_mbenc		= "";
		$this->_arrayresult	= false;
		$this->_timeout		= 0;
	}

	function __destruct()
	{
		if ( $this->_socket !== false )
			fclose ( $this->_socket );
	}

	/// get last error message (string)
	function GetLastError ()
	{
		return $this->_error;
	}

	/// get last warning message (string)
	function GetLastWarning ()
	{
		return $this->_warning;
	}

	/// get last error flag (to tell network connection errors from searchd errors or broken responses)
	function IsConnectError()
	{
		return $this->_connerror;
	}

	/// set searchd host name (string) and port (integer)
	function SetServer ( $host, $port = 0 )
	{
		assert ( is_string($host) );
		if ( $host[0] == '/')
		{
			$this->_path = 'unix://' . $host;
			return;
		}
		if ( substr ( $host, 0, 7 )=="unix://" )
		{
			$this->_path = $host;
			return;
		}
				
		assert ( is_int($port) );
		$this->_host = $host;
		$this->_port = $port;
		$this->_path = '';

	}

	/// set server connection timeout (0 to remove)
	function SetConnectTimeout ( $timeout )
	{
		assert ( is_numeric($timeout) );
		$this->_timeout = $timeout;
	}


	function _Send ( $handle, $data, $length )
	{
		if ( feof($handle) || fwrite ( $handle, $data, $length ) !== $length )
		{
			$this->_error = 'connection unexpectedly closed (timed out?)';
			$this->_connerror = true;
			return false;
		}
		return true;
	}

	/////////////////////////////////////////////////////////////////////////////

	/// enter mbstring workaround mode
	function _MBPush ()
	{
		$this->_mbenc = "";
		if ( ini_get ( "mbstring.func_overload" ) & 2 )
		{
			$this->_mbenc = mb_internal_encoding();
			mb_internal_encoding ( "latin1" );
		}
    }

	/// leave mbstring workaround mode
	function _MBPop ()
	{
		if ( $this->_mbenc )
			mb_internal_encoding ( $this->_mbenc );
	}

	/// connect to searchd server
	function _Connect ()
	{
		if ( $this->_socket!==false )
		{
			// we are in persistent connection mode, so we have a socket
			// however, need to check whether it's still alive
			if ( !@feof ( $this->_socket ) )
				return $this->_socket;

			// force reopen
			$this->_socket = false;
		}

		$errno = 0;
		$errstr = "";
		$this->_connerror = false;

		if ( $this->_path )
		{
			$host = $this->_path;
			$port = 0;
		}
		else
		{
			$host = $this->_host;
			$port = $this->_port;
		}

		if ( $this->_timeout<=0 )
			$fp = @fsockopen ( $host, $port, $errno, $errstr );
		else
			$fp = @fsockopen ( $host, $port, $errno, $errstr, $this->_timeout );
		
		if ( !$fp )
		{
			if ( $this->_path )
				$location = $this->_path;
			else
				$location = "{$this->_host}:{$this->_port}";
			
			$errstr = trim ( $errstr );
			$this->_error = "connection to $location failed (errno=$errno, msg=$errstr)";
			$this->_connerror = true;
			return false;
		}

		// send my version
		// this is a subtle part. we must do it before (!) reading back from searchd.
		// because otherwise under some conditions (reported on FreeBSD for instance)
		// TCP stack could throttle write-write-read pattern because of Nagle.
		if ( !$this->_Send ( $fp, pack ( "N", 1 ), 4 ) )
		{
			fclose ( $fp );
			$this->_error = "failed to send client protocol version";
			return false;
		}

		// check version
		list(,$v) = unpack ( "N*", fread ( $fp, 4 ) );
		$v = (int)$v;
		if ( $v<1 )
		{
			fclose ( $fp );
			$this->_error = "expected searchd protocol version 1+, got version '$v'";
			return false;
		}

		return $fp;
	}

	/// get and check response packet from searchd server
	function _GetResponse ( $fp, $client_ver )
	{
		$response = "";
		$len = 0;

		$header = fread ( $fp, 8 );
		if ( strlen($header)==8 )
		{
			list ( $status, $ver, $len ) = array_values ( unpack ( "n2a/Nb", $header ) );
			$left = $len;
			while ( $left>0 && !feof($fp) )
			{
				$chunk = fread ( $fp, $left );
				if ( $chunk )
				{
					$response .= $chunk;
					$left -= strlen($chunk);
				}
			}
		}
		if ( $this->_socket === false )
			fclose ( $fp );

		// check response
		$read = strlen ( $response );
		if ( !$response || $read!=$len )
		{
			$this->_error = $len
				? "failed to read searchd response (status=$status, ver=$ver, len=$len, read=$read)"
				: "received zero-sized searchd response";
			return false;
		}

		// check status
		if ( $status==SEARCHD_WARNING )
		{
			list(,$wlen) = unpack ( "N*", substr ( $response, 0, 4 ) );
			$this->_warning = substr ( $response, 4, $wlen );
			return substr ( $response, 4+$wlen );
		}
		if ( $status==SEARCHD_ERROR )
		{
			$this->_error = "searchd error: " . substr ( $response, 4 );
			return false;
		}
		if ( $status==SEARCHD_RETRY )
		{
			$this->_error = "temporary searchd error: " . substr ( $response, 4 );
			return false;
		}
		if ( $status!=SEARCHD_OK )
		{
			$this->_error = "unknown status code '$status'";
			return false;
		}

		// check version
		if ( $ver<$client_ver )
		{
			$this->_warning = sprintf ( "searchd command v.%d.%d older than client's v.%d.%d, some options might not work",
				$ver>>8, $ver&0xff, $client_ver>>8, $client_ver&0xff );
		}

		return $response;
	}

	/////////////////////////////////////////////////////////////////////////////
	// searching
	/////////////////////////////////////////////////////////////////////////////

	/// set offset and count into result set,
	/// and optionally set max-matches and cutoff limits
	function SetLimits ( $offset, $limit, $max=0, $cutoff=0 )
	{
		assert ( is_int($offset) );
		assert ( is_int($limit) );
		assert ( $offset>=0 );
		assert ( $limit>0 );
		assert ( $max>=0 );
		$this->_offset = $offset;
		$this->_limit = $limit;
		if ( $max>0 )
			$this->_maxmatches = $max;
		if ( $cutoff>0 )
			$this->_cutoff = $cutoff;
	}

	/// set maximum query time, in milliseconds, per-index
	/// integer, 0 means "do not limit"
	function SetMaxQueryTime ( $max )
	{
		assert ( is_int($max) );
		assert ( $max>=0 );
		$this->_maxquerytime = $max;
	}

	/// set matching mode
	function SetMatchMode ( $mode )
	{
		assert ( $mode==SPH_MATCH_ALL
			|| $mode==SPH_MATCH_ANY
			|| $mode==SPH_MATCH_PHRASE
			|| $mode==SPH_MATCH_BOOLEAN
			|| $mode==SPH_MATCH_EXTENDED
			|| $mode==SPH_MATCH_FULLSCAN
			|| $mode==SPH_MATCH_EXTENDED2 );
		$this->_mode = $mode;
	}

	/// set ranking mode
	function SetRankingMode ( $ranker )
	{
		assert ( $ranker==SPH_RANK_PROXIMITY_BM25
			|| $ranker==SPH_RANK_BM25
			|| $ranker==SPH_RANK_NONE
			|| $ranker==SPH_RANK_WORDCOUNT
			|| $ranker==SPH_RANK_PROXIMITY );
		$this->_ranker = $ranker;
	}

	/// set matches sorting mode
	function SetSortMode ( $mode, $sortby="" )
	{
		assert (
			$mode==SPH_SORT_RELEVANCE ||
			$mode==SPH_SORT_ATTR_DESC ||
			$mode==SPH_SORT_ATTR_ASC ||
			$mode==SPH_SORT_TIME_SEGMENTS ||
			$mode==SPH_SORT_EXTENDED ||
			$mode==SPH_SORT_EXPR );
		assert ( is_string($sortby) );
		assert ( $mode==SPH_SORT_RELEVANCE || strlen($sortby)>0 );

		$this->_sort = $mode;
		$this->_sortby = $sortby;
	}

	/// bind per-field weights by order
	/// DEPRECATED; use SetFieldWeights() instead
	function SetWeights ( $weights )
	{
		assert ( is_array($weights) );
		foreach ( $weights as $weight )
			assert ( is_int($weight) );

		$this->_weights = $weights;
	}

	/// bind per-field weights by name
	function SetFieldWeights ( $weights )
	{
		assert ( is_array($weights) );
		foreach ( $weights as $name=>$weight )
		{
			assert ( is_string($name) );
			assert ( is_int($weight) );
		}
		$this->_fieldweights = $weights;
	}

	/// bind per-index weights by name
	function SetIndexWeights ( $weights )
	{
		assert ( is_array($weights) );
		foreach ( $weights as $index=>$weight )
		{
			assert ( is_string($index) );
			assert ( is_int($weight) );
		}
		$this->_indexweights = $weights;
	}

	/// set IDs range to match
	/// only match records if document ID is beetwen $min and $max (inclusive)
	function SetIDRange ( $min, $max )
	{
		assert ( is_numeric($min) );
		assert ( is_numeric($max) );
		assert ( $min<=$max );
		$this->_min_id = $min;
		$this->_max_id = $max;
	}

	/// set values set filter
	/// only match records where $attribute value is in given set
	function SetFilter ( $attribute, $values, $exclude=false )
	{
		assert ( is_string($attribute) );
		assert ( is_array($values) );
		assert ( count($values) );

		if ( is_array($values) && count($values) )
		{
			foreach ( $values as $value )
				assert ( is_numeric($value) );

			$this->_filters[] = array ( "type"=>SPH_FILTER_VALUES, "attr"=>$attribute, "exclude"=>$exclude, "values"=>$values );
		}
	}

	/// set range filter
	/// only match records if $attribute value is beetwen $min and $max (inclusive)
	function SetFilterRange ( $attribute, $min, $max, $exclude=false )
	{
		assert ( is_string($attribute) );
		assert ( is_numeric($min) );
		assert ( is_numeric($max) );
		assert ( $min<=$max );

		$this->_filters[] = array ( "type"=>SPH_FILTER_RANGE, "attr"=>$attribute, "exclude"=>$exclude, "min"=>$min, "max"=>$max );
	}

	/// set float range filter
	/// only match records if $attribute value is beetwen $min and $max (inclusive)
	function SetFilterFloatRange ( $attribute, $min, $max, $exclude=false )
	{
		assert ( is_string($attribute) );
		assert ( is_float($min) );
		assert ( is_float($max) );
		assert ( $min<=$max );

		$this->_filters[] = array ( "type"=>SPH_FILTER_FLOATRANGE, "attr"=>$attribute, "exclude"=>$exclude, "min"=>$min, "max"=>$max );
	}

	/// setup anchor point for geosphere distance calculations
	/// required to use @geodist in filters and sorting
	/// latitude and longitude must be in radians
	function SetGeoAnchor ( $attrlat, $attrlong, $lat, $long )
	{
		assert ( is_string($attrlat) );
		assert ( is_string($attrlong) );
		assert ( is_float($lat) );
		assert ( is_float($long) );

		$this->_anchor = array ( "attrlat"=>$attrlat, "attrlong"=>$attrlong, "lat"=>$lat, "long"=>$long );
	}

	/// set grouping attribute and function
	function SetGroupBy ( $attribute, $func, $groupsort="@group desc" )
	{
		assert ( is_string($attribute) );
		assert ( is_string($groupsort) );
		assert ( $func==SPH_GROUPBY_DAY
			|| $func==SPH_GROUPBY_WEEK
			|| $func==SPH_GROUPBY_MONTH
			|| $func==SPH_GROUPBY_YEAR
			|| $func==SPH_GROUPBY_ATTR
			|| $func==SPH_GROUPBY_ATTRPAIR );

		$this->_groupby = $attribute;
		$this->_groupfunc = $func;
		$this->_groupsort = $groupsort;
	}

	/// set count-distinct attribute for group-by queries
	function SetGroupDistinct ( $attribute )
	{
		assert ( is_string($attribute) );
		$this->_groupdistinct = $attribute;
	}

	/// set distributed retries count and delay
	function SetRetries ( $count, $delay=0 )
	{
		assert ( is_int($count) && $count>=0 );
		assert ( is_int($delay) && $delay>=0 );
		$this->_retrycount = $count;
		$this->_retrydelay = $delay;
	}

	/// set result set format (hash or array; hash by default)
	/// PHP specific; needed for group-by-MVA result sets that may contain duplicate IDs
	function SetArrayResult ( $arrayresult )
	{
		assert ( is_bool($arrayresult) );
		$this->_arrayresult = $arrayresult;
	}

	/// set attribute values override
	/// there can be only one override per attribute
	/// $values must be a hash that maps document IDs to attribute values
	function SetOverride ( $attrname, $attrtype, $values )
	{
		assert ( is_string ( $attrname ) );
		assert ( in_array ( $attrtype, array ( SPH_ATTR_INTEGER, SPH_ATTR_TIMESTAMP, SPH_ATTR_BOOL, SPH_ATTR_FLOAT, SPH_ATTR_BIGINT ) ) );
		assert ( is_array ( $values ) );

		$this->_overrides[$attrname] = array ( "attr"=>$attrname, "type"=>$attrtype, "values"=>$values );
	}

	/// set select-list (attributes or expressions), SQL-like syntax
	function SetSelect ( $select )
	{
		assert ( is_string ( $select ) );
		$this->_select = $select;
	}

	//////////////////////////////////////////////////////////////////////////////

	/// clear all filters (for multi-queries)
	function ResetFilters ()
	{
		$this->_filters = array();
		$this->_anchor = array();
	}

	/// clear groupby settings (for multi-queries)
	function ResetGroupBy ()
	{
		$this->_groupby		= "";
		$this->_groupfunc	= SPH_GROUPBY_DAY;
		$this->_groupsort	= "@group desc";
		$this->_groupdistinct= "";
	}

	/// clear all attribute value overrides (for multi-queries)
	function ResetOverrides ()
    {
    	$this->_overrides = array ();
    }

	//////////////////////////////////////////////////////////////////////////////

	/// connect to searchd server, run given search query through given indexes,
	/// and return the search results
	function Query ( $query, $index="*", $comment="" )
	{
		assert ( empty($this->_reqs) );

		$this->AddQuery ( $query, $index, $comment );
		$results = $this->RunQueries ();
		$this->_reqs = array (); // just in case it failed too early

		if ( !is_array($results) )
			return false; // probably network error; error message should be already filled

		$this->_error = $results[0]["error"];
		$this->_warning = $results[0]["warning"];
		if ( $results[0]["status"]==SEARCHD_ERROR )
			return false;
		else
			return $results[0];
	}

	/// helper to pack floats in network byte order
	function _PackFloat ( $f )
	{
		$t1 = pack ( "f", $f ); // machine order
		list(,$t2) = unpack ( "L*", $t1 ); // int in machine order
		return pack ( "N", $t2 );
	}

	/// add query to multi-query batch
	/// returns index into results array from RunQueries() call
	function AddQuery ( $query, $index="*", $comment="" )
	{
		// mbstring workaround
		$this->_MBPush ();

		// build request
		$req = pack ( "NNNNN", $this->_offset, $this->_limit, $this->_mode, $this->_ranker, $this->_sort ); // mode and limits
		$req .= pack ( "N", strlen($this->_sortby) ) . $this->_sortby;
		$req .= pack ( "N", strlen($query) ) . $query; // query itself
		$req .= pack ( "N", count($this->_weights) ); // weights
		foreach ( $this->_weights as $weight )
			$req .= pack ( "N", (int)$weight );
		$req .= pack ( "N", strlen($index) ) . $index; // indexes
		$req .= pack ( "N", 1 ); // id64 range marker
		$req .= sphPackU64 ( $this->_min_id ) . sphPackU64 ( $this->_max_id ); // id64 range

		// filters
		$req .= pack ( "N", count($this->_filters) );
		foreach ( $this->_filters as $filter )
		{
			$req .= pack ( "N", strlen($filter["attr"]) ) . $filter["attr"];
			$req .= pack ( "N", $filter["type"] );
			switch ( $filter["type"] )
			{
				case SPH_FILTER_VALUES:
					$req .= pack ( "N", count($filter["values"]) );
					foreach ( $filter["values"] as $value )
						$req .= sphPackI64 ( $value );
					break;

				case SPH_FILTER_RANGE:
					$req .= sphPackI64 ( $filter["min"] ) . sphPackI64 ( $filter["max"] );
					break;

				case SPH_FILTER_FLOATRANGE:
					$req .= $this->_PackFloat ( $filter["min"] ) . $this->_PackFloat ( $filter["max"] );
					break;

				default:
					assert ( 0 && "internal error: unhandled filter type" );
			}
			$req .= pack ( "N", $filter["exclude"] );
		}

		// group-by clause, max-matches count, group-sort clause, cutoff count
		$req .= pack ( "NN", $this->_groupfunc, strlen($this->_groupby) ) . $this->_groupby;
		$req .= pack ( "N", $this->_maxmatches );
		$req .= pack ( "N", strlen($this->_groupsort) ) . $this->_groupsort;
		$req .= pack ( "NNN", $this->_cutoff, $this->_retrycount, $this->_retrydelay );
		$req .= pack ( "N", strlen($this->_groupdistinct) ) . $this->_groupdistinct;

		// anchor point
		if ( empty($this->_anchor) )
		{
			$req .= pack ( "N", 0 );
		} else
		{
			$a =& $this->_anchor;
			$req .= pack ( "N", 1 );
			$req .= pack ( "N", strlen($a["attrlat"]) ) . $a["attrlat"];
			$req .= pack ( "N", strlen($a["attrlong"]) ) . $a["attrlong"];
			$req .= $this->_PackFloat ( $a["lat"] ) . $this->_PackFloat ( $a["long"] );
		}

		// per-index weights
		$req .= pack ( "N", count($this->_indexweights) );
		foreach ( $this->_indexweights as $idx=>$weight )
			$req .= pack ( "N", strlen($idx) ) . $idx . pack ( "N", $weight );

		// max query time
		$req .= pack ( "N", $this->_maxquerytime );

		// per-field weights
		$req .= pack ( "N", count($this->_fieldweights) );
		foreach ( $this->_fieldweights as $field=>$weight )
			$req .= pack ( "N", strlen($field) ) . $field . pack ( "N", $weight );

		// comment
		$req .= pack ( "N", strlen($comment) ) . $comment;

		// attribute overrides
		$req .= pack ( "N", count($this->_overrides) );
		foreach ( $this->_overrides as $key => $entry )
		{
			$req .= pack ( "N", strlen($entry["attr"]) ) . $entry["attr"];
			$req .= pack ( "NN", $entry["type"], count($entry["values"]) );
			foreach ( $entry["values"] as $id=>$val )
			{
				assert ( is_numeric($id) );
				assert ( is_numeric($val) );

				$req .= sphPackU64 ( $id );
				switch ( $entry["type"] )
				{
					case SPH_ATTR_FLOAT:	$req .= $this->_PackFloat ( $val ); break;
					case SPH_ATTR_BIGINT:	$req .= sphPackI64 ( $val ); break;
					default:				$req .= pack ( "N", $val ); break;
				}
			}
		}

		// select-list
		$req .= pack ( "N", strlen($this->_select) ) . $this->_select;

		// mbstring workaround
		$this->_MBPop ();

		// store request to requests array
		$this->_reqs[] = $req;
		return count($this->_reqs)-1;
	}

	/// connect to searchd, run queries batch, and return an array of result sets
	function RunQueries ()
	{
		if ( empty($this->_reqs) )
		{
			$this->_error = "no queries defined, issue AddQuery() first";
			return false;
		}

		// mbstring workaround
		$this->_MBPush ();

		if (!( $fp = $this->_Connect() ))
		{
			$this->_MBPop ();
			return false;
		}

		// send query, get response
		$nreqs = count($this->_reqs);
		$req = join ( "", $this->_reqs );
		$len = 4+strlen($req);
		$req = pack ( "nnNN", SEARCHD_COMMAND_SEARCH, VER_COMMAND_SEARCH, $len, $nreqs ) . $req; // add header

		if ( !( $this->_Send ( $fp, $req, $len+8 ) ) ||
			 !( $response = $this->_GetResponse ( $fp, VER_COMMAND_SEARCH ) ) )
		{
			$this->_MBPop ();
			return false;
		}

		// query sent ok; we can reset reqs now
		$this->_reqs = array ();

		// parse and return response
		return $this->_ParseSearchResponse ( $response, $nreqs );
	}

	/// parse and return search query (or queries) response
	function _ParseSearchResponse ( $response, $nreqs )
	{
		$p = 0; // current position
		$max = strlen($response); // max position for checks, to protect against broken responses

		$results = array ();
		for ( $ires=0; $ires<$nreqs && $p<$max; $ires++ )
		{
			$results[] = array();
			$result =& $results[$ires];

			$result["error"] = "";
			$result["warning"] = "";

			// extract status
			list(,$status) = unpack ( "N*", substr ( $response, $p, 4 ) ); $p += 4;
			$result["status"] = $status;
			if ( $status!=SEARCHD_OK )
			{
				list(,$len) = unpack ( "N*", substr ( $response, $p, 4 ) ); $p += 4;
				$message = substr ( $response, $p, $len ); $p += $len;

				if ( $status==SEARCHD_WARNING )
				{
					$result["warning"] = $message;
				} else
				{
					$result["error"] = $message;
					continue;
				}
			}

			// read schema
			$fields = array ();
			$attrs = array ();

			list(,$nfields) = unpack ( "N*", substr ( $response, $p, 4 ) ); $p += 4;
			while ( $nfields-->0 && $p<$max )
			{
				list(,$len) = unpack ( "N*", substr ( $response, $p, 4 ) ); $p += 4;
				$fields[] = substr ( $response, $p, $len ); $p += $len;
			}
			$result["fields"] = $fields;

			list(,$nattrs) = unpack ( "N*", substr ( $response, $p, 4 ) ); $p += 4;
			while ( $nattrs-->0 && $p<$max  )
			{
				list(,$len) = unpack ( "N*", substr ( $response, $p, 4 ) ); $p += 4;
				$attr = substr ( $response, $p, $len ); $p += $len;
				list(,$type) = unpack ( "N*", substr ( $response, $p, 4 ) ); $p += 4;
				$attrs[$attr] = $type;
			}
			$result["attrs"] = $attrs;

			// read match count
			list(,$count) = unpack ( "N*", substr ( $response, $p, 4 ) ); $p += 4;
			list(,$id64) = unpack ( "N*", substr ( $response, $p, 4 ) ); $p += 4;

			// read matches
			$idx = -1;
			while ( $count-->0 && $p<$max )
			{
				// index into result array
				$idx++;

				// parse document id and weight
				if ( $id64 )
				{
					$doc = sphUnpackU64 ( substr ( $response, $p, 8 ) ); $p += 8;
					list(,$weight) = unpack ( "N*", substr ( $response, $p, 4 ) ); $p += 4;
				}
				else
				{
					list ( $doc, $weight ) = array_values ( unpack ( "N*N*",
						substr ( $response, $p, 8 ) ) );
					$p += 8;
					$doc = sphFixUint($doc);
				}
				$weight = sprintf ( "%u", $weight );

				// create match entry
				if ( $this->_arrayresult )
					$result["matches"][$idx] = array ( "id"=>$doc, "weight"=>$weight );
				else
					$result["matches"][$doc]["weight"] = $weight;

				// parse and create attributes
				$attrvals = array ();
				foreach ( $attrs as $attr=>$type )
				{
					// handle 64bit ints
					if ( $type==SPH_ATTR_BIGINT )
					{
						$attrvals[$attr] = sphUnpackI64 ( substr ( $response, $p, 8 ) ); $p += 8;
						continue;
					}

					// handle floats
					if ( $type==SPH_ATTR_FLOAT )
					{
						list(,$uval) = unpack ( "N*", substr ( $response, $p, 4 ) ); $p += 4;
						list(,$fval) = unpack ( "f*", pack ( "L", $uval ) ); 
						$attrvals[$attr] = $fval;
						continue;
					}

					// handle everything else as unsigned ints
					list(,$val) = unpack ( "N*", substr ( $response, $p, 4 ) ); $p += 4;
					if ( $type & SPH_ATTR_MULTI )
					{
						$attrvals[$attr] = array ();
						$nvalues = $val;
						while ( $nvalues-->0 && $p<$max )
						{
							list(,$val) = unpack ( "N*", substr ( $response, $p, 4 ) ); $p += 4;
							$attrvals[$attr][] = sphFixUint($val);
						}
					} else
					{
						$attrvals[$attr] = sphFixUint($val);
					}
				}

				if ( $this->_arrayresult )
					$result["matches"][$idx]["attrs"] = $attrvals;
				else
					$result["matches"][$doc]["attrs"] = $attrvals;
			}

			list ( $total, $total_found, $msecs, $words ) =
				array_values ( unpack ( "N*N*N*N*", substr ( $response, $p, 16 ) ) );
			$result["total"] = sprintf ( "%u", $total );
			$result["total_found"] = sprintf ( "%u", $total_found );
			$result["time"] = sprintf ( "%.3f", $msecs/1000 );
			$p += 16;

			while ( $words-->0 && $p<$max )
			{
				list(,$len) = unpack ( "N*", substr ( $response, $p, 4 ) ); $p += 4;
				$word = substr ( $response, $p, $len ); $p += $len;
				list ( $docs, $hits ) = array_values ( unpack ( "N*N*", substr ( $response, $p, 8 ) ) ); $p += 8;
				$result["words"][$word] = array (
					"docs"=>sprintf ( "%u", $docs ),
					"hits"=>sprintf ( "%u", $hits ) );
			}
		}

		$this->_MBPop ();
		return $results;
	}

	/////////////////////////////////////////////////////////////////////////////
	// excerpts generation
	/////////////////////////////////////////////////////////////////////////////

	/// connect to searchd server, and generate exceprts (snippets)
	/// of given documents for given query. returns false on failure,
	/// an array of snippets on success
	function BuildExcerpts ( $docs, $index, $words, $opts=array() )
	{
		assert ( is_array($docs) );
		assert ( is_string($index) );
		assert ( is_string($words) );
		assert ( is_array($opts) );

		$this->_MBPush ();

		if (!( $fp = $this->_Connect() ))
		{
			$this->_MBPop();
			return false;
		}

		/////////////////
		// fixup options
		/////////////////

		if ( !isset($opts["before_match"]) )		$opts["before_match"] = "<b>";
		if ( !isset($opts["after_match"]) )			$opts["after_match"] = "</b>";
		if ( !isset($opts["chunk_separator"]) )		$opts["chunk_separator"] = " ... ";
		if ( !isset($opts["limit"]) )				$opts["limit"] = 256;
		if ( !isset($opts["around"]) )				$opts["around"] = 5;
		if ( !isset($opts["exact_phrase"]) )		$opts["exact_phrase"] = false;
		if ( !isset($opts["single_passage"]) )		$opts["single_passage"] = false;
		if ( !isset($opts["use_boundaries"]) )		$opts["use_boundaries"] = false;
		if ( !isset($opts["weight_order"]) )		$opts["weight_order"] = false;

		/////////////////
		// build request
		/////////////////

		// v.1.0 req
		$flags = 1; // remove spaces
		if ( $opts["exact_phrase"] )	$flags |= 2;
		if ( $opts["single_passage"] )	$flags |= 4;
		if ( $opts["use_boundaries"] )	$flags |= 8;
		if ( $opts["weight_order"] )	$flags |= 16;
		$req = pack ( "NN", 0, $flags ); // mode=0, flags=$flags
		$req .= pack ( "N", strlen($index) ) . $index; // req index
		$req .= pack ( "N", strlen($words) ) . $words; // req words

		// options
		$req .= pack ( "N", strlen($opts["before_match"]) ) . $opts["before_match"];
		$req .= pack ( "N", strlen($opts["after_match"]) ) . $opts["after_match"];
		$req .= pack ( "N", strlen($opts["chunk_separator"]) ) . $opts["chunk_separator"];
		$req .= pack ( "N", (int)$opts["limit"] );
		$req .= pack ( "N", (int)$opts["around"] );

		// documents
		$req .= pack ( "N", count($docs) );
		foreach ( $docs as $doc )
		{
			assert ( is_string($doc) );
			$req .= pack ( "N", strlen($doc) ) . $doc;
		}

		////////////////////////////
		// send query, get response
		////////////////////////////

		$len = strlen($req);
		$req = pack ( "nnN", SEARCHD_COMMAND_EXCERPT, VER_COMMAND_EXCERPT, $len ) . $req; // add header
		if ( !( $this->_Send ( $fp, $req, $len+8 ) ) ||
			 !( $response = $this->_GetResponse ( $fp, VER_COMMAND_EXCERPT ) ) )
		{
			$this->_MBPop ();
			return false;
		}

		//////////////////
		// parse response
		//////////////////

		$pos = 0;
		$res = array ();
		$rlen = strlen($response);
		for ( $i=0; $i<count($docs); $i++ )
		{
			list(,$len) = unpack ( "N*", substr ( $response, $pos, 4 ) );
			$pos += 4;

			if ( $pos+$len > $rlen )
			{
				$this->_error = "incomplete reply";
				$this->_MBPop ();
				return false;
			}
			$res[] = $len ? substr ( $response, $pos, $len ) : "";
			$pos += $len;
		}

		$this->_MBPop ();
		return $res;
	}


	/////////////////////////////////////////////////////////////////////////////
	// keyword generation
	/////////////////////////////////////////////////////////////////////////////

	/// connect to searchd server, and generate keyword list for a given query
	/// returns false on failure,
	/// an array of words on success
	function BuildKeywords ( $query, $index, $hits )
	{
		assert ( is_string($query) );
		assert ( is_string($index) );
		assert ( is_bool($hits) );

		$this->_MBPush ();

		if (!( $fp = $this->_Connect() ))
		{
			$this->_MBPop();
			return false;
		}

		/////////////////
		// build request
		/////////////////

		// v.1.0 req
		$req  = pack ( "N", strlen($query) ) . $query; // req query
		$req .= pack ( "N", strlen($index) ) . $index; // req index
		$req .= pack ( "N", (int)$hits );

		////////////////////////////
		// send query, get response
		////////////////////////////

		$len = strlen($req);
		$req = pack ( "nnN", SEARCHD_COMMAND_KEYWORDS, VER_COMMAND_KEYWORDS, $len ) . $req; // add header
		if ( !( $this->_Send ( $fp, $req, $len+8 ) ) ||
			 !( $response = $this->_GetResponse ( $fp, VER_COMMAND_KEYWORDS ) ) )
		{
			$this->_MBPop ();
			return false;
		}

		//////////////////
		// parse response
		//////////////////

		$pos = 0;
		$res = array ();
		$rlen = strlen($response);
		list(,$nwords) = unpack ( "N*", substr ( $response, $pos, 4 ) );
		$pos += 4;
		for ( $i=0; $i<$nwords; $i++ )
		{
			list(,$len) = unpack ( "N*", substr ( $response, $pos, 4 ) );	$pos += 4;
			$tokenized = $len ? substr ( $response, $pos, $len ) : "";
			$pos += $len;

			list(,$len) = unpack ( "N*", substr ( $response, $pos, 4 ) );	$pos += 4;
			$normalized = $len ? substr ( $response, $pos, $len ) : "";
			$pos += $len;

			$res[] = array ( "tokenized"=>$tokenized, "normalized"=>$normalized );

			if ( $hits )
			{
				list($ndocs,$nhits) = array_values ( unpack ( "N*N*", substr ( $response, $pos, 8 ) ) );
				$pos += 8;
				$res [$i]["docs"] = $ndocs;
				$res [$i]["hits"] = $nhits;
			}

			if ( $pos > $rlen )
			{
				$this->_error = "incomplete reply";
				$this->_MBPop ();
				return false;
			}
		}

		$this->_MBPop ();
		return $res;
	}

	function EscapeString ( $string )
	{
		$from = array ( '\\', '(',')','|','-','!','@','~','"','&', '/', '^', '$', '=' );
		$to   = array ( '\\\\', '\(','\)','\|','\-','\!','\@','\~','\"', '\&', '\/', '\^', '\$', '\=' );

		return str_replace ( $from, $to, $string );
	}

	/////////////////////////////////////////////////////////////////////////////
	// attribute updates
	/////////////////////////////////////////////////////////////////////////////

	/// batch update given attributes in given rows in given indexes
	/// returns amount of updated documents (0 or more) on success, or -1 on failure
	function UpdateAttributes ( $index, $attrs, $values, $mva=false )
	{
		// verify everything
		assert ( is_string($index) );
		assert ( is_bool($mva) );

		assert ( is_array($attrs) );
		foreach ( $attrs as $attr )
			assert ( is_string($attr) );

		assert ( is_array($values) );
		foreach ( $values as $id=>$entry )
		{
			assert ( is_numeric($id) );
			assert ( is_array($entry) );
			assert ( count($entry)==count($attrs) );
			foreach ( $entry as $v )
			{
				if ( $mva )
				{
					assert ( is_array($v) );
					foreach ( $v as $vv )
						assert ( is_int($vv) );
				} else
					assert ( is_int($v) );
			}
		}

		// build request
		$req = pack ( "N", strlen($index) ) . $index;

		$req .= pack ( "N", count($attrs) );
		foreach ( $attrs as $attr )
		{
			$req .= pack ( "N", strlen($attr) ) . $attr;
			$req .= pack ( "N", $mva ? 1 : 0 );
		}

		$req .= pack ( "N", count($values) );
		foreach ( $values as $id=>$entry )
		{
			$req .= sphPackU64 ( $id );
			foreach ( $entry as $v )
			{
				$req .= pack ( "N", $mva ? count($v) : $v );
				if ( $mva )
					foreach ( $v as $vv )
						$req .= pack ( "N", $vv );
			}
		}

		// connect, send query, get response
		if (!( $fp = $this->_Connect() ))
			return -1;

		$len = strlen($req);
		$req = pack ( "nnN", SEARCHD_COMMAND_UPDATE, VER_COMMAND_UPDATE, $len ) . $req; // add header
		if ( !$this->_Send ( $fp, $req, $len+8 ) )
			return -1;

		if (!( $response = $this->_GetResponse ( $fp, VER_COMMAND_UPDATE ) ))
			return -1;

		// parse response
		list(,$updated) = unpack ( "N*", substr ( $response, 0, 4 ) );
		return $updated;
	}

	/////////////////////////////////////////////////////////////////////////////
	// persistent connections
	/////////////////////////////////////////////////////////////////////////////

	function Open()
	{
		if ( $this->_socket !== false )
		{
			$this->_error = 'already connected';
			return false;
		}
		if ( !$fp = $this->_Connect() )
			return false;

		// command, command version = 0, body length = 4, body = 1
		$req = pack ( "nnNN", SEARCHD_COMMAND_PERSIST, 0, 4, 1 );
		if ( !$this->_Send ( $fp, $req, 12 ) )
			return false;

		$this->_socket = $fp;
		return true;
	}

	function Close()
	{
		if ( $this->_socket === false )
		{
			$this->_error = 'not connected';
			return false;
		}

		fclose ( $this->_socket );
		$this->_socket = false;
		
		return true;
	}

	//////////////////////////////////////////////////////////////////////////
	// status
	//////////////////////////////////////////////////////////////////////////

	function Status ()
	{
		$this->_MBPush ();
		if (!( $fp = $this->_Connect() ))
		{
			$this->_MBPop();
			return false;
		}

		$req = pack ( "nnNN", SEARCHD_COMMAND_STATUS, VER_COMMAND_STATUS, 4, 1 ); // len=4, body=1
		if ( !( $this->_Send ( $fp, $req, 12 ) ) ||
			 !( $response = $this->_GetResponse ( $fp, VER_COMMAND_STATUS ) ) )
		{
			$this->_MBPop ();
			return false;
		}

		$res = substr ( $response, 4 ); // just ignore length, error handling, etc
		$p = 0;
		list ( $rows, $cols ) = array_values ( unpack ( "N*N*", substr ( $response, $p, 8 ) ) ); $p += 8;

		$res = array();
		for ( $i=0; $i<$rows; $i++ )
			for ( $j=0; $j<$cols; $j++ )
		{
			list(,$len) = unpack ( "N*", substr ( $response, $p, 4 ) ); $p += 4;
			$res[$i][] = substr ( $response, $p, $len ); $p += $len;
		}

		$this->_MBPop ();
		return $res;
	}
}

//
// $Id: sphinxapi.php 2055 2009-11-06 23:09:58Z shodan $
//

亲测性能提高10倍左右~效果还是挺显著的~