博客园_nasdaqhe's blog

ubuntu+php5-fpm 下安装 memcached PHP扩展

2011-05-15T15:44:00Z

1.获取编译工具下面要用来编译源代码，如果已经安装可跳过这步

[bash] sudo apt-get install build-essential [/bash]

2.安装PEAR and memcached

[bash] sudo apt-get install php-pear memcached [/bash]

3.编译libmemcached

到 https://launchpad.net/libmemcached/+download 下载最新的libmemcached源代码，解压编译安装步骤如下：

Extract the files with your preferred method, or type:

[bash] tar -xzf libmemcached-0.37.tar.gz [/bash]

Move into the extracted directory:

[bash] cd libmemcached-0.37/ [/bash]

Configure the package.

[bash] ./configure –with-memcached [/bash]

Make.

[bash] make [/bash]

Make install. Note this probably will require sudo rights.

[bash] sudo make install [/bash]

4.PECL

[bash] sudo pecl install memcached [/bash]

When this process completed, you should see something like this:

[bash]

Build process completed successfully

Installing ‘/usr/lib/php5/20060613+lfs/memcached.so’

install ok: channel://pecl.php.net/memcached-1.0.0

Extension memcached enabled in php.ini

[/bash]

5.修改PHP.INI添加扩展

[bash] sudo vi /etc/php5/fpm/php.ini [/bash]

在php.ini 中添加“extension = memcached.so”

重启PHP-FPM服务

[bash] sudo service php5-fpm restart [/bash]

本文链接

cmd下使用telnet连接到memcached服务器操作

2011-05-15T15:25:00Z

//telnet IP Port 连接到memcached服务器
输入 telnet 192.168.12.14 11211
//查看服务器状态
输入 stats
返回一堆状态
//SetValue 格式 set keyName 0 过期时间数据长度，注意数据长度必须是Value的字节长度，如果两者不一致会报错
输入 set key1 0 10 5
输入 abcde
返回 STORED 表示设置成功
//GetValue
输入 get key1

本文链接

Lucence.Net 2.9.3 日期范围搜索

2011-05-10T09:38:00Z

关键步骤如下：

1.将 DateTime 转换成 long 型。

2.索引时要使用 NumericField 进行索引

3.搜索时使用 NumericRangeQuery 进行查询

事例代码如下：

[TestMethod()]

        public void RangTest()
        {
            RAMDirectory mDirInfo = new RAMDirectory();
            IndexWriter mIndexWriter = new IndexWriter(mDirInfo, CAnalyzerLoader.GetAnalyzer(), true, IndexWriter.MaxFieldLength.LIMITED);
            mIndexWriter.AddDocument(GetDocumentByModel(1, "姑姑，我是态儿", CDateTime.UNIX_TIMESTAMP(System.DateTime.Now.AddDays(-1))));
            mIndexWriter.AddDocument(GetDocumentByModel(2, "B儿，我是态儿", CDateTime.UNIX_TIMESTAMP(System.DateTime.Now.AddDays(-2))));
            mIndexWriter.AddDocument(GetDocumentByModel(3, "肝儿，我是态儿", CDateTime.UNIX_TIMESTAMP(System.DateTime.Now.AddDays(-3))));
            mIndexWriter.Close();
            long upper = CDateTime.UNIX_TIMESTAMP(DateTime.Now.AddDays(-1));
            long lower = CDateTime.UNIX_TIMESTAMP(DateTime.Now.AddDays(-3).AddHours(-1));
            NumericRangeQuery rangeQuery = NumericRangeQuery.NewLongRange("UpdateOn", lower, upper, true, true);

            IndexSearcher mSearcher = new IndexSearcher(mDirInfo, true);
            TopDocs topDocs = mSearcher.Search(rangeQuery, 100);
            Assert.AreEqual(2, topDocs.totalHits);
        }

        private Document GetDocumentByModel(int tid,string title,long time)
        {
            Document doc = new Document();
            doc.AddFiled("Id", tid.ToString(), Field.Store.YES, Field.Index.NOT_ANALYZED)
                .AddFiled("Title", title.Trim(), Field.Store.YES, Field.Index.ANALYZED)
                .AddLongFiled("UpdateOn", time, Field.Store.NO, true);
            return doc;
        }
   public static Document AddFiled(this Document self, string name, string value, Field.Store fStore, Field.Index fIndex)
        {
            Field filed = new Field(name, value, fStore, fIndex);
            self.Add(filed);
            return self;
        }
        public static Document AddLongFiled(this Document self, string name, long value, Field.Store fStore, bool bIndex)
        {
            NumericField filed = new NumericField(name, NumericUtils.PRECISION_STEP_DEFAULT, fStore, bIndex).SetLongValue(value);
            self.Add(filed);
            return self;
        }

本文链接

SQL语句优化一例 row_number not in or

2011-04-07T01:29:00Z

select [id],[poster],[fid],[title],[editername],[modifytime] from(

select [id],[poster],[fid],[title],[editername],[modifytime], row_number() over

( order by id desc) as row from [news] WITH (NOLOCK)

where path like 'A_B%' and id not in

(select id from [news] where layer=7 or layer=5)) a

where row between 1 and 15

上面的查询很慢。如果只执行子查询就很快，但是做 where row between 1 and 15 分页就很慢很慢，原来是因为查询里面用了or，优化成下面的就不超时了

select [id],[poster],[fid],[title],[editername],[modifytime] from(

select [id],[poster],[fid],[title],[editername],[modifytime], row_number() over

( order by id desc) as row from [news] WITH (NOLOCK)

where path like 'A_B%' and id not in

(select id from [news] where layer=7 union select id from [news] where layer=5)) a

where row between 1 and 15

本文链接

flickr head中用到的标签

2010-12-23T13:39:00Z

通过rel="search"，把你的搜索功能放到浏览器的快捷搜索工具上

用firefox打开taobao时，看一下地址栏边上的快捷搜索时，可以填加一个“淘宝搜索”，完成后，以后就可以直接在那里搜索东西了！

实现方法：
在首页的head标签里加上 “<link rel="search" type="application/opensearchdescription+xml" href=provider.xml" title="关键词搜索" />”

provider.xml的写法为：
<?xml version="1.0" encoding="utf-8"?>
<OpenSearchDescription xmlns="http://a9.com/-/spec/opensearch/1.1/">
　<InputEncoding>UTF-8</InputEncoding>
　<ShortName>关键词搜索</ShortName>
　<Description>关键词搜索</Description>
　<Image height="16" width="16" type="image/vnd.microsoft.icon">images/favicon.ico</Image>
　<Url type="text/html" template="search.php?keyword={searchTerms}" />
</OpenSearchDescription>

-------------------------------------------------------------------------------------------------------------------------------------------------

通过name="viewport" 在iPhone的浏览器中页面将以原始大小显示

网页手机wap2.0网页的head里加入下面这条元标签，在iPhone的浏览器中页面将以原始大小显示，并不允许缩放。

<meta name="viewport" content="width=device-width, initial-scale=1.0, minimum-scale=1.0, maximum-scale=1.0, user-scalable=no">

其中：
width - viewport的宽度 height - viewport的高度
initial-scale - 初始的缩放比例
minimum-scale - 允许用户缩放到的最小比例
maximum-scale - 允许用户缩放到的最大比例
user-scalable - 用户是否可以手动缩放

本文链接

判断中文是否UTF8编码

2010-12-23T03:16:00Z

        #region 判断Url参数是否UTF8编码
        public static bool IsUTF8(string url)
        {
            byte[] buf = GetUrlCodingToBytes(url);
            return IsTextUTF8(buf);
        }
        private static bool IsTextUTF8(byte[] buf)
        {
            int i;
            byte cOctets = 0; // octets to go in this UTF-8 encoded character
            bool bAllAscii = true;
            long iLen = buf.Length;
            for (i = 0; i < iLen; i++)
            {
                if ((buf[i] & 0x80) != 0) bAllAscii = false;

                if (cOctets == 0)
                {
                    if (buf[i] >= 0x80)
                    {
                        do
                        {
                            buf[i] <<= 1;
                            cOctets++;
                        }
                        while ((buf[i] & 0x80) != 0);

                        cOctets--;
                        if (cOctets != 2)
                            return false;
                    }
                }
                else
                {
                    if ((buf[i] & 0xC0) != 0x80)
                        return false;
                    cOctets--;
                }
            }
            if (cOctets > 0)
                return false;
            if (bAllAscii)
                return false;
            return true;
        }
        private static byte[] GetUrlCodingToBytes(string url)
        {
            StringBuilder sb = new StringBuilder();

            int i = url.IndexOf('%');
            while (i >= 0)
            {
                if (url.Length < i + 3)
                {
                    break;
                }
                sb.Append(url.Substring(i, 3));
                url = url.Substring(i + 3);
                i = url.IndexOf('%');
            }

            string urlCoding = sb.ToString();
            if (string.IsNullOrEmpty(urlCoding))
                return new byte[0];

            urlCoding = urlCoding.Replace("%", string.Empty);
            int len = urlCoding.Length / 2;
            byte[] result = new byte[len];
            len *= 2;
            for (int index = 0; index < len; index++)
            {
                string s = urlCoding.Substring(index, 2);
                int b = int.Parse(s, System.Globalization.NumberStyles.HexNumber);
                result[index / 2] = (byte)b;
                index++;
            }
            return result;
        }
        #endregion 判断Url参数是否UTF8编码

UTF-8编码规则参考

http://blog.csdn.net/sandyen/archive/2006/08/23/1108168.aspx

上面代码是网络上找的，不过存在大部分不能识别的情况，后根据对于中文，UTF8 一定编码成 3 字节，这个原则

修改了一下，现在大部分情况下都能正确识别

本文链接

MSSQL备忘

2010-12-15T14:41:00Z

只有mdf 文件，没有日志文件附加数据库的方法

EXEC sp_attach_single_file_db @dbname= 'test'
, @physname= N'E:\Program Files\Microsoft SQL Server\MSSQL.1\MSSQL\Data\test.mdf'

本文链接

新浪微博产品图 (Vincent.H手笔) MindManage

2010-12-01T00:42:00Z

2010-12-01T00:42:00Z2010-12-01T00:42:00Znasdaqhehttp://www.cnblogs.com/nasdaqhe/

本文链接

HTML5学习资料整理

2010-06-17T13:59:00Z

用JavaScript玩转游戏物理(一)运动学模拟与粒子系统
http://www.cnblogs.com/miloyip/archive/2010/06/14/Kinematics_ParticleSystem.html

本文链接

ubuntu备忘

2010-06-13T13:01:00Z

开启SSH

sudo apt-get install openssh-server

列出分区信息

sudo fdisk -l

挂载分区

mount /dev/sda8 /mnt

Ubuntu下VirtualBox使用笔记

http://lhflinux.blog.51cto.com/1961662/515811

关闭防火墙

sudo ufw disable

本文链接