centos6.5安装coreseek替换wordpress的搜索功能
- centos6.5安装coreseek替换wordpress的搜索功能
- 引言
- 安装coreseek
- 配置coreseek支持mysql数据源
- coreseek+php使用方式
- coreseek日常维护
- 后记
引言
实在受不了wordpress乌龟般慢的搜索速度了,感觉网站挂了似得。
之前已经苦逼过怎么优化wordpress了,参见另一篇文章数据过万后wordpress优化过程记录2, 也换成了全世界最好的编程语言php7,过程见于安装php7为wordpress提速。这次是要把wordpress自带的mysql like的搜索替换成全文搜索引擎。
安装coreseek
安装依赖包
yum install make gcc g++ gcc-c++ libtool autoconf automake imake mysql-devel libxml2-devel expat-devel* mysql-server mysql
coreseek下载
官网打不开了,网上找了个包,放到了csdn上。
tar xzvf coreseek-3.2.14.tar.gz
cd coreseek-3.2.14
安装mmseg
$ cd mmseg-3.2.14
$ ./bootstrap #输出的warning信息可以忽略,如果出现error则需要解决
$ ./configure --prefix=/usr/local/mmseg3
$ make && make install
$ cd ..
##如果提示libtool: unrecognized option `--tag=CC' ,请查看libtool问题解决方案
##安装完成后,mmseg使用的词典和配置文件,将自动安装到/usr/local/mmseg3/etc中
##中文分词测试,如果显示不正常,请检查当前环境下的locale和UTF-8中文字符显示设置
$ /usr/local/mmseg3/bin/mmseg -d /usr/local/mmseg3/etc mmseg-3.2.14/src/t1.txt
中文/x 分/x 词/x 测试/x
中国人/x 上海市/x
Word Splite took: 1 ms.
安装coreseek
$ cd csft-4.1
#执行configure,进行编译配置:
$ sh buildconf.sh
$ ./configure --prefix=/usr/local/coreseek --without-unixodbc --with-mmseg --with-mmseg-includes=/usr/local/mmseg3/include/mmseg/ --with-mmseg-libs=/usr/local/mmseg3/lib/ --with-mysql
$ make && make install
测试coreseek
cd ../testpack
$ /usr/local/coreseek/bin/indexer -c etc/csft.conf
以下为正常情况下的提示信息:
Coreseek Fulltext 4.1 [ Sphinx 2.0.2-dev (r2922)]
Copyright (c) 2007-2011,
Beijing Choice Software Technologies Inc (http://www.coreseek.com)
ERROR: nothing to do.
$ /usr/local/coreseek/bin/indexer -c etc/csft.conf --all
以下为正常索引全部数据时的提示信息:
Coreseek Fulltext 4.1 [ Sphinx 2.0.2-dev (r2922)]
Copyright (c) 2007-2011,
Beijing Choice Software Technologies Inc (http://www.coreseek.com)
using config file ‘etc/csft.conf’…
indexing index ‘xml’…
collected 3 docs, 0.0 MB
sorted 0.0 Mhits, 100.0% done
total 3 docs, 7585 bytes
total 0.007 sec, 979088 bytes/sec, 387.24 docs/sec
total 3 reads, 0.000 sec, 2.8 kb/call avg, 0.0 msec/call avg
total 9 writes, 0.000 sec, 2.2 kb/call avg, 0.0 msec/call avg
$ /usr/local/coreseek/bin/indexer -c etc/csft.conf xml
以下为正常索引指定数据时的提示信息:
Coreseek Fulltext 4.1 [ Sphinx 2.0.2-dev (r2922)]
Copyright (c) 2007-2011,
Beijing Choice Software Technologies Inc (http://www.coreseek.com)
using config file ‘etc/csft.conf’…
indexing index ‘xml’…
collected 3 docs, 0.0 MB
sorted 0.0 Mhits, 100.0% done
total 3 docs, 7585 bytes
total 0.006 sec, 1260594 bytes/sec, 498.58 docs/sec
total 3 reads, 0.000 sec, 2.8 kb/call avg, 0.0 msec/call avg
total 9 writes, 0.000 sec, 2.2 kb/call avg, 0.0 msec/call avg
$ /usr/local/coreseek/bin/search -c etc/csft.conf
以下为正常测试搜索时的提示信息:
Coreseek Fulltext 4.1 [ Sphinx 2.0.2-dev (r2922)]
Copyright (c) 2007-2011,
Beijing Choice Software Technologies Inc (http://www.coreseek.com)
using config file ‘etc/csft.conf’…
index ‘xml’: query ”: returned 3 matches of 3 total in 0.000 sec
displaying matches:
1. document=1, weight=1, published=Thu Apr 1 22:20:07 2010, author_id=1
2. document=2, weight=1, published=Thu Apr 1 23:25:48 2010, author_id=1
3. document=3, weight=1, published=Thu Apr 1 12:01:00 2010, author_id=2
words:
$ /usr/local/coreseek/bin/search -c etc/csft.conf -a Twittter和Opera都提供了搜索服务
以下为正常测试搜索关键词时的提示信息:
Coreseek Fulltext 4.1 [ Sphinx 2.0.2-dev (r2922)]
Copyright (c) 2007-2011,
Beijing Choice Software Technologies Inc (http://www.coreseek.com)
using config file ‘etc/csft.conf’…
index ‘xml’: query ‘Twittter和Opera都提供了搜索服务 ‘: returned 0 matches of 0 total in 0.015 sec
words:
1. ‘twittter’: 1 documents, 3 hits
2. ‘和’: 3 documents, 15 hits
3. ‘opera’: 1 documents, 25 hits
4. ‘都’: 2 documents, 4 hits
5. ‘提供’: 0 documents, 0 hits
6. ‘了’: 3 documents, 18 hits
7. ‘搜索’: 2 documents, 5 hits
8. ‘服务’: 1 documents, 1 hits
$ /usr/local/coreseek/bin/searchd -c etc/csft.conf
以下为正常开启搜索服务时的提示信息:(csft-4.0版类似)
Coreseek Fulltext 4.1 [ Sphinx 2.0.2-dev (r2922)]
Copyright (c) 2007-2011,
Beijing Choice Software Technologies Inc (http://www.coreseek.com)
using config file ‘etc/csft.conf’…
WARNING: compat_sphinxql_magics=1 is deprecated; please update your application and config
listening on all interfaces, port=9312
precaching index ‘xml’
precached 1 indexes in 0.001 se
配置coreseek支持mysql数据源
配置csft_mysql.conf文件
复制mysql配置文件到coreseek安装目录etc/下(比如/usr/local/coreseek/etc/)
cp /usr/local/src/coreseek-3.2.14/testpack/etc/csft_mysql.conf /usr/local/coreseek/etc/
cd /usr/local/coreseek/etc/
vi csft_mysql.conf
源定义
source phperz
{
type = mysql
sql_host = localhost
sql_user = root
sql_pass = xxxx
sql_db = phperz
sql_port = 3306
sql_query_pre = SET NAMES utf8
sql_query = SELECT ID, post_author, UNIX_TIMESTAMP(post_date) AS post_date, post_title, post_content FROM wp_posts where post_status='publish' and post_type in ('page','post')
#sql_query第一列id需为整数
#title、content作为字符串/文本字段,被全文索引
sql_attr_uint = post_author#从SQL读取到的值必须为整数
sql_attr_timestamp = post_date#从SQL读取到的值必须为整数,作为时间属性
sql_query_info_pre = SET NAMES utf8 #命令行查询时,设置正确的字符集
sql_query_info = SELECT * FROM wp_posts WHERE ID=$id #命令行查询时,从数据库读取原始数据信息
}
index定义
index phperz
{
source = phperz #对应的source名称
path = /usr/local/coreseek/var/data/phperz #请修改为实际使用的绝对路径,例如:/usr/local/coreseek/var/…
docinfo = extern
mlock = 0
morphology = none
min_word_len = 1
html_strip = 0
#中文分词配置,详情请查看:http://www.coreseek.cn/products-install/coreseek_mmseg/
charset_dictpath = /usr/local/mmseg3/etc/ #BSD、Linux环境下设置,/符号结尾
#charset_dictpath = etc/ #Windows环境下设置,/符号结尾,最好给出绝对路径,例如:C:/usr/local/coreseek/etc/...
charset_type = zh_cn.utf-8
}
全局index定义
indexer
{
mem_limit = 128M
}
searchd服务定义
searchd
{
listen = 9312
read_timeout = 5
max_children = 30
max_matches = 1000
seamless_rotate = 0
preopen_indexes = 0
unlink_old = 1
pid_file = /usr/local/coreseek/var/log/searchd_mysql.pid #请修改为实际使用的绝对路径,例如:/usr/local/coreseek/var/…
log = /usr/local/coreseek/var/log/searchd_mysql.log #请修改为实际使用的绝对路径,例如:/usr/local/coreseek/var/…
query_log = /usr/local/coreseek/var/log/query_mysql.log #请修改为实际使用的绝对路径,例如:/usr/local/coreseek/var/…
}
建立索引
路经部分需要改成你自己的地址
/usr/local/coreseek/bin/indexer -c /usr/local/coreseek/etc/csft_mysql.conf –all
可能出现的错误
ERROR: index ‘phperz’: sql_connect: Can’t connect to local MySQL server through socket ‘/var/lib/mysql/mysql.sock’ (2) (DSN=mysql://root:***@localhost:3306/phperz).
这是因为mysql的sock文件路经不正确导致的.
确认一下你的mysql.sock路经,建立一个软连接,比如
ln -s /tmp/mysql.sock /var/lib/mysql/mysql.sock
coreseek+php使用方式
复制安装目录下的/usr/local/src/coreseek-4.1-beta/testpack/api/sphinxapi.php文件到你的项目里
你的程序里include sphinxapi.php
php使用方法见/usr/local/src/coreseek-4.1-beta/testpack/api/test.php
wordpress修改, 把api目录拷贝到主题目录中后,修改search.php,我的修改如下(和主题相关,不能完全照抄,仅供参考)
SetServer ( '127.0.0.1', 9312);
$cl->SetConnectTimeout(3);
$cl->SetMatchMode(1);
//以下设置用于返回数组形式的结果
$cl->SetArrayResult ( true );
$paged = (get_query_var('paged')) ? get_query_var('paged') : 1;
$start = ($paged - 1) * 10;
$cl->SetLimits($start,10);
$keyword = $s = isset($_GET['s']) ? htmlspecialchars(trim($_GET['s'])) : ''; //获取搜索词
$res = $cl->Query ( "$keyword", "*" );
#print_r($res); //查看全文索引结果
$total = $res['total']; //所有返回文章数,用于分页
if ( $res===false )
{
print "Query failed: " . $cl->GetLastError() . ".\n";
} else
{
if ( $cl->GetLastWarning() )
print "WARNING: " . $cl->GetLastWarning() . "\n\n";
}
$have_posts = true;
if(isset($res['matches']) && !empty($res['matches'])) {
foreach($res['matches'] as $value) {
$id_arr[] = $value['id'];
}
}
else
{
$have_posts = false;
}
$id_str = @implode(",", $id_arr);
$args = array();
$args = array( 'include' => $id_arr);
wp_reset_query();
$sql = "select * from wp_posts where ID in($id_str) and post_type in ('post','page') AND post_status = 'publish'"; //根据ID读取文章数据
$data = $wpdb->get_results($sql); //输出数据,之后你可以使用foreach数据$post object
?>
'.hui_get_adcode('ads_search_01').'' : '' ?>
coreseek日常维护
启动
/usr/local/coreseek/bin/searchd -c /usr/local/coreseek/etc/csft_mysql.conf
停止
/usr/local/coreseek/bin/searchd -c /usr/local/coreseek/etc/csft_mysql.conf –stop
建立索引
/usr/local/coreseek/bin/indexer -c /usr/local/coreseek/etc/csft_mysql.conf –all
重建索引
/usr/local/coreseek/bin/indexer -c /usr/local/coreseek/etc/csft_mysql.conf –all –rotate
你需要把启动命令加到开机自启动里
把重建索引命令加到计划任务里每天执行
后记
折腾了3小时才搞定,希望对大家有用。可以访问我的网站试试搜索效果。写这篇博客的时候,我的博客已经快3万了哦。
经查证,直接修改search.php没有用。因为sql查询在之前就已经查了,search.php相当于一个模板页面了。
阅读完原代码,修改了wp-include/query.php, 终于优化完成了。
优化前:
Page generated in 2.18491s, 38.61% PHP, 61.39% MySQL
优化后:
Page generated in 0.54773s, 95.51% PHP, 4.49% MySQL
wordpress4.9.8 只需要修改wp-includes/class-wp-query.php即可。
如下: 主要是修改
//增加函数
private function get_ids_by_coreseek($q)
{
include_once( dirname(__FILE__) . "/sphinxapi.php" );
$cl = new SphinxClient ();
$cl->SetServer ( '127.0.0.1', 9312);
$cl->SetConnectTimeout(3);
$cl->SetMatchMode(1);
//以下设置用于返回数组形式的结果
$cl->SetArrayResult ( true );
$paged = ($q['paged']) ? $q['paged'] : 1;
$posts_per_page = isset($q['posts_per_page'])? $q['posts_per_page']:10;
$start = ($paged - 1) * $posts_per_page;
$cl->SetLimits($start,$posts_per_page);
$keyword = $s = isset($q['s']) ? htmlspecialchars(trim($q['s'])) : ''; //获取搜索词
$res = $cl->Query ( "$keyword", "*" );
$total = $res['total']; //所有返回文章数,用于分页
$have_posts = true;
if(isset($res['matches']) && !empty($res['matches'])) {
foreach($res['matches'] as $value) {
$id_arr[] = $value['id'];
}
}
else
{
$have_posts = false;
}
if($have_posts)
{
return array('total'=>$total, 'ids'=>$id_arr);
}
else
{
return array('total'=>0,'ids'=>array());
}
}
//修改调用
public function get_posts() {
$split_the_query = apply_filters( 'split_the_query', $split_the_query, $this );
if(isset($q['s']) && !empty($q['s']) && !isset($q['debug']))
{
$coreseekResult = $this->get_ids_by_coreseek($q);
$this->posts = $coreseekResult['ids'];
$this->found_posts = $coreseekResult['total'];
$this->max_num_pages = ceil( $this->found_posts / $q['posts_per_page'] );
}else if ( $split_the_query ) {
// 保持原样
} else {
// 保持原样
}
}