其实原理很简单，贴上我的fonts.conf
不用安装什么美化包
也不用改什么字体的路径，以为内系统字体是有用的
 
 
首先，确保你的debian上有windows的字体
其次，修改fonts.conf,把我下面fonts.conf和你的比较一下
 
 
我的debian sarge4就是这么用的，这个方法我用了很久，没想到竟然最适合Debian
 
<?xml version="1.0"?>
<!DOCTYPE fontconfig SYSTEM "fonts.dtd">
<!-- /etc/fonts/fonts.conf file to configure system font access -->
<fontconfig>
<!--
 DO NOT EDIT THIS FILE.
 IT WILL BE REPLACED WHEN FONTCONFIG IS UPDATED.
 LOCAL CHANGES BELONG IN 'local.conf'.
 The intent of this standard configuration file is to be adequate for
 most environments.  If you have a reasonably normal environment and
 have found problems with this configuration, they are probably
 things that others will also want fixed.  Please submit any
 problems to the fontconfig bugzilla system located at fontconfig.org
 Note that the normal 'make install' procedure for fontconfig is to
 replace any existing fonts.conf file with the new version.  Place
 any local customizations in local.conf which this file references.
 Keith Packard
-->
<!-- Font directory list -->
 <dir>/usr/share/fonts</dir>
 <dir>/usr/X11R6/lib/X11/fonts</dir> <dir>/usr/local/share/fonts</dir>
 <dir>~/.fonts</dir>
<!--
  Accept deprecated 'mono' alias, replacing it with 'monospace'
-->
 <match target="pattern">
  <test qual="any" name="family">
   <string>mono</string>
  </test>
  <edit name="family" mode="assign">
   <string>monospace</string>
  </edit>
 </match>
<!--
  Accept alternate 'sans serif' spelling, replacing it with 'sans-serif'
-->
 <match target="pattern">
  <test qual="any" name="family">
   <string>sans serif</string>
  </test>
  <edit name="family" mode="assign">
   <string>sans-serif</string>
  </edit>
 </match>
<!--
  Accept deprecated 'sans' alias, replacing it with 'sans-serif'
-->
 <match target="pattern">
  <test qual="any" name="family">
   <string>sans</string>
  </test>
  <edit name="family" mode="assign">
   <string>sans-serif</string>
  </edit>
 </match>
<!--
  Mark common families with their generics so we'll get
  something reasonable
-->
<!--
  Serif faces
 -->
 <alias>
  <family>Tahoma</family>
  <family>SimSun</family>
  <family>NSimSun</family>
  <family>Arial</family>
  <family>Bitstream Vera Serif</family>
  <family>Times New Roman</family>
  <family>Thorndale AMT</family>
  <family>Times</family>
  <family>Nimbus Roman No9 L</family>
  <family>Luxi Serif</family>
  <family>Kochi Mincho</family>
  <family>AR PL SungtiL GB</family>
  <family>AR PL Mingti2L Big5</family>
  <family>ＭＳ 明朝</family>
  <family>Baekmuk Batang</family>
  <family>FreeSerif</family>
  <default><family>SimSun</family></default>
 </alias>
<!--
  Sans-serif faces
 -->
 <alias>
  <family>Tahoma</family>
  <family>SimSun</family>
  <family>NSimSun</family>
  <family>Arial</family>
  <family>Bitstream Vera Sans</family>
  <family>Helvetica</family>
  <family>Arial</family>
  <family>Verdana</family>
  <family>Albany AMT</family>
  <family>Nimbus Sans L</family>
  <family>Luxi Sans</family>
  <family>Kochi Gothic</family>
  <family>AR PL KaitiM GB</family>
  <family>AR PL KaitiM Big5</family>
  <family>ＭＳ ゴシック</family>
  <family>Baekmuk Dotum</family>
  <family>SimSun</family>
  <family>FreeSans</family>
  <default><family>Tahoma</family></default>
 </alias>
<!--
  Monospace faces
 -->
  <alias>
  <family>Tahoma</family>
  <family>SimSun</family>
  <family>NSimSun</family>
  <family>Arial</family>
  <family>Bitstream Vera Sans Mono</family>
  <family>Courier</family>
  <family>Courier New</family>
  <family>Andale Mono</family>
  <family>Luxi Mono</family>
  <family>Cumberland AMT</family>
  <family>Nimbus Mono L</family>
  <family>NSimSun</family>
  <family>FreeMono</family>
  <default><family>Tahoma</family></default>
 </alias>
<!--
  If the font still has no generic name, add sans-serif
 -->
 <match target="pattern">
  <test qual="all" name="family" compare="not_eq">
   <string>sans-serif</string>
  </test>
  <test qual="all" name="family" compare="not_eq">
   <string>serif</string>
  </test>
  <test qual="all" name="family" compare="not_eq">
   <string>monospace</string>
  </test>
  <edit name="family" mode="append_last">
   <string>sans-serif</string>
  </edit>
 </match>
<!--
  URW provides metric and shape compatible fonts for these 3 Adobe families
  Mark these as effective replacements by binding the replacement
  family names strongly
  -->
        <match target="pattern">
   <test name="family"><string>Times</string></test>
   <edit name="family" mode="append" binding="same">
     <string>Nimbus Roman No9 L</string>
   </edit>
 </match>
        <match target="pattern">
   <test name="family"><string>Helvetica</string></test>
   <edit name="family" mode="append" binding="same">
   <string>Nimbus Sans L</string>
   </edit>
 </match>
        <match target="pattern">
   <test name="family"><string>Courier</string></test>
   <edit name="family" mode="append" binding="same">
     <string>Nimbus Mono L</string>
   </edit>
 </match>
<!--
 AMT provides metric and shape compatible fonts for these three web font
 families.  Bind them weakly as matching here is not as important as
 with web fonts.
 -->
  <alias>
  <family>Times New Roman</family>
  <family>Tahoma</family>
  <family>SimSun</family>
  <family>NSimSun</family>
  <family>Arial</family>
  <accept><family>Thorndale AMT</family></accept>
 </alias>
  <alias>
  <family>Arial</family>
  <family>Tahoma</family>
  <family>SimSun</family>
  <family>NSimSun</family>
  <family>Arial</family>
  <accept><family>Albany AMT</family></accept>
 </alias>
  <alias>
  <family>Courier New</family>
  <accept><family>Cumberland AMT</family></accept>
 </alias>
 
<!--
  Some Asian fonts misadvertise themselves as monospaced when
  in fact they are dual-spaced (half and full).  This makes
  FreeType very confused as it forces all widths to match.
  Undo this magic by disabling the width forcing code -->
 <match target="font">
  <test name="family"><string>GulimChe</string></test>
  <edit name="globaladvance"><bool>false</bool></edit>
 </match>
 <match target="font">
  <test name="family"><string>DotumChe</string></test>
  <edit name="globaladvance"><bool>false</bool></edit>
 </match>
 <match target="font">
  <test name="family"><string>BatangChe</string></test>
  <edit name="globaladvance"><bool>false</bool></edit>
 </match>
 <match target="font">
  <test name="family"><string>GungsuhChe</string></test>
  <edit name="globaladvance"><bool>false</bool></edit>
 </match>
<!--
 The Bitstream Vera fonts have GASP entries suggesting that hinting be
 disabled below 8 ppem, but FreeType ignores those, preferring to use
 the data found in the instructed hints.  The initial Vera release
 didn't include the right instructions in the 'prep' table. Fix this
 by disabling hinting manually at smaller sizes (< 8ppem)
 -->
 <match target="font">
  <test name="family">
   <string>Bitstream Vera Sans</string>
  </test>
  <test name="pixelsize" compare="less">
   <double>7.5</double>
  </test>
  <edit name="hinting">
   <bool>false</bool>
  </edit>
 </match>
 <match target="font">
  <test name="family">
   <string>Bitstream Vera Serif</string>
  </test>
  <test name="pixelsize" compare="less">
   <double>7.5</double>
  </test>
  <edit name="hinting">
   <bool>false</bool>
  </edit>
 </match>
 <match target="font">
  <test name="family">
   <string>Bitstream Vera Sans Mono</string>
  </test>
  <test name="pixelsize" compare="less">
   <double>7.5</double>
  </test>
  <edit name="hinting">
   <bool>false</bool>
  </edit>
 </match>
<!--
  Load per-user customization file
-->
 <include ignore_missing="yes">~/.fonts.conf</include>
<!--
  Load local system customization file
-->
 <include ignore_missing="yes">conf.d</include>
 <include ignore_missing="yes">local.conf</include>
<!--
  Provide required aliases for standard names
-->
 <alias>
  <family>serif</family>
  <prefer>
   <family>Bitstream Vera Serif</family>
   <family>Tahoma</family>
   <family>SimSun</family>
   <family>NSimSun</family>
   <family>Arial</family>
   <family>Times New Roman</family>
   <family>Thorndale AMT</family>
   <family>Luxi Serif</family>
   <family>Nimbus Roman No9 L</family>
   <family>Times</family>
   <family>Frank Ruehl</family>
   <family>Kochi Mincho</family>
   <family>AR PL SungtiL GB</family>
   <family>AR PL Mingti2L Big5</family>
   <family>ＭＳ 明朝</family>
   <family>Baekmuk Batang</family>
   <family>FreeSerif</family>
  </prefer>
 </alias>
 <alias>
  <family>sans-serif</family>
  <prefer>
  
   <family>Tahoma</family>
   <family>SimSun</family>
   <family>Bitstream Vera Sans</family>
   <family>NSimSun</family>
   <family>Arial</family>
   <family>Verdana</family>
   <family>Arial</family>
   <family>Albany AMT</family>
   <family>Luxi Sans</family>
   <family>Nimbus Sans L</family>
   <family>Helvetica</family>
   <family>Nachlieli</family>
   <family>Kochi Gothic</family>
   <family>AR PL KaitiM GB</family>
   <family>AR PL KaitiM Big5</family>
   <family>ＭＳ ゴシック</family>
   <family>Baekmuk Dotum</family>
   <family>SimSun</family>
   <family>FreeSans</family>
  </prefer>
 </alias>
 <alias>
  <family>monospace</family>
  <prefer>
   <family>Tahoma</family>
   <family>SimSun</family>
   <family>NSimSun</family>
   <family>Arial</family>
   <family>Bitstream Vera Sans Mono</family>
   <family>Andale Mono</family>
   <family>Courier New</family>
   <family>Cumberland AMT</family>
   <family>Luxi Mono</family>
   <family>Nimbus Mono L</family>
   <family>Courier</family>
   <family>Miriam Mono</family>
   <family>Kochi Gothic</family>
   <family>AR PL KaitiM GB</family>
   <family>Baekmuk Dotum</family>
   <family>FreeMono</family>
  </prefer>
 </alias>
<!--
 Artificial oblique for fonts without an italic or oblique version
 -->
 
 <match target="font">
  <!-- check to see if the font is roman -->
  <test name="slant">
   <const>roman</const>
  </test>
  <!-- check to see if the pattern requested non-roman -->
  <test target="pattern" name="slant" compare="not_eq">
   <const>roman</const>
  </test>
  <!-- multiply the matrix to slant the font -->
  <edit name="matrix" mode="assign">
   <times>
    <name>matrix</name>
    <matrix><double>1</double><double>0.2</double>
     <double>0</double><double>1</double>
    </matrix>
   </times>
  </edit>
  <!-- pretend the font is oblique now -->
  <edit name="slant" mode="assign">
   <const>oblique</const>
  </edit>
 </match>
<!--
 Synthetic emboldening for fonts that do not have bold face available
 -->
 <match target="font">
  <!-- check to see if the font is just regular -->
  <test name="weight" compare="less_eq">
   <int>100</int>
  </test>
  <!-- check to see if the pattern requests bold -->
  <test target="pattern" name="weight" compare="more_eq">
   <int>200</int>
  </test>
  <!-- set the embolden flag -->
  <edit name="embolden" mode="assign">
   <bool>true</bool>
  </edit>
 </match>

  <match target="font" >
                <test compare="more" name="pixelsize" qual="any">
                         <double>6</double>
                </test>
                <test compare="less" name="pixelsize" qual="any">
                         <double>16</double>
                </test>
                <edit mode="assign" name="antialias" >
                         <bool>false</bool>
                </edit>
        </match>

 <config>
<!--
  These are the default Unicode chars that are expected to be blank
  in fonts.  All other blank chars are assumed to be broken and
  won't appear in the resulting charsets
 -->
  <blank>
   <int>0x0020</int> <!-- SPACE -->
   <int>0x00A0</int> <!-- NO-BREAK SPACE -->
   <int>0x00AD</int> <!-- SOFT HYPHEN -->
   <int>0x034F</int> <!-- COMBINING GRAPHEME JOINER -->
   <int>0x0600</int> <!-- ARABIC NUMBER SIGN -->
   <int>0x0601</int> <!-- ARABIC SIGN SANAH -->
   <int>0x0602</int> <!-- ARABIC FOOTNOTE MARKER -->
   <int>0x0603</int> <!-- ARABIC SIGN SAFHA -->
   <int>0x06DD</int> <!-- ARABIC END OF AYAH -->
   <int>0x070F</int> <!-- SYRIAC ABBREVIATION MARK -->
   <int>0x115F</int> <!-- HANGUL CHOSEONG FILLER -->
   <int>0x1160</int> <!-- HANGUL JUNGSEONG FILLER -->
   <int>0x1680</int> <!-- OGHAM SPACE MARK -->
   <int>0x17B4</int> <!-- KHMER VOWEL INHERENT AQ -->
   <int>0x17B5</int> <!-- KHMER VOWEL INHERENT AA -->
   <int>0x180E</int> <!-- MONGOLIAN VOWEL SEPARATOR -->
   <int>0x2000</int> <!-- EN QUAD -->
   <int>0x2001</int> <!-- EM QUAD -->
   <int>0x2002</int> <!-- EN SPACE -->
   <int>0x2003</int> <!-- EM SPACE -->
   <int>0x2004</int> <!-- THREE-PER-EM SPACE -->
   <int>0x2005</int> <!-- FOUR-PER-EM SPACE -->
   <int>0x2006</int> <!-- SIX-PER-EM SPACE -->
   <int>0x2007</int> <!-- FIGURE SPACE -->
   <int>0x2008</int> <!-- PUNCTUATION SPACE -->
   <int>0x2009</int> <!-- THIN SPACE -->
   <int>0x200A</int> <!-- HAIR SPACE -->
   <int>0x200B</int> <!-- ZERO WIDTH SPACE -->
   <int>0x200C</int> <!-- ZERO WIDTH NON-JOINER -->
   <int>0x200D</int> <!-- ZERO WIDTH JOINER -->
   <int>0x200E</int> <!-- LEFT-TO-RIGHT MARK -->
   <int>0x200F</int> <!-- RIGHT-TO-LEFT MARK -->
   <int>0x2028</int> <!-- LINE SEPARATOR -->
   <int>0x2029</int> <!-- PARAGRAPH SEPARATOR -->
   <int>0x202A</int> <!-- LEFT-TO-RIGHT EMBEDDING -->
   <int>0x202B</int> <!-- RIGHT-TO-LEFT EMBEDDING -->
   <int>0x202C</int> <!-- POP DIRECTIONAL FORMATTING -->
   <int>0x202D</int> <!-- LEFT-TO-RIGHT OVERRIDE -->
   <int>0x202E</int> <!-- RIGHT-TO-LEFT OVERRIDE -->
   <int>0x202F</int> <!-- NARROW NO-BREAK SPACE -->
   <int>0x205F</int> <!-- MEDIUM MATHEMATICAL SPACE -->
   <int>0x2060</int> <!-- WORD JOINER -->
   <int>0x2061</int> <!-- function APPLICATION -->
   <int>0x2062</int> <!-- INVISIBLE TIMES -->
   <int>0x2063</int> <!-- INVISIBLE SEPARATOR -->
   <int>0x206A</int> <!-- INHIBIT SYMMETRIC SWAPPING -->
   <int>0x206B</int> <!-- ACTIVATE SYMMETRIC SWAPPING -->
   <int>0x206C</int> <!-- INHIBIT ARABIC FORM SHAPING -->
   <int>0x206D</int> <!-- ACTIVATE ARABIC FORM SHAPING -->
   <int>0x206E</int> <!-- NATIONAL DIGIT SHAPES -->
   <int>0x206F</int> <!-- NOMINAL DIGIT SHAPES -->
   <int>0x3000</int> <!-- IDEOGRAPHIC SPACE -->
   <int>0x3164</int> <!-- HANGUL FILLER -->
   <int>0xFEFF</int> <!-- ZERO WIDTH NO-BREAK SPACE -->
   <int>0xFFA0</int> <!-- HALFWIDTH HANGUL FILLER -->
   <int>0xFFF9</int> <!-- INTERLINEAR ANNOTATION ANCHOR -->
   <int>0xFFFA</int> <!-- INTERLINEAR ANNOTATION SEPARATOR -->
   <int>0xFFFB</int> <!-- INTERLINEAR ANNOTATION TERMINATOR -->
  </blank>
<!--
  Rescan configuration every 30 seconds when FcFontSetList is called
 -->
  <rescan>
   <int>30</int>
  </rescan>
 </config>
</fontconfig>

占座先

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

 走在地铁里，走在大街上，走在每个封闭或者不封闭的角落，没有窒息的感觉，也没有新鲜的空气
，众生随着太阳的时间而活动，傍晚的万家灯火，深夜，一个一个的熄灭，直到熄灭，众生皆苦，
每个人，每天都在挣扎，穷人为了生存，富人为了包富，不穷不富的人则担心自己随时会变穷，
真实，其实是很荒芜的，懂得真实，其实是一种痛苦，对于眼前的真实，没有任何一个人可以改变
人会逝去，世界也有消忘得这一天，其实在昏昏噩噩中的状态是最幸福的，因为从来不会意识到 那一天，终结的那一天，在昏昏噩噩中，直到生命的 最后一刻
 
我看到的是什么，分明是 matrix。人能感知这个世界，不是因为他感知这个世界，而是这个世界告诉了他什么，一切都只是一个完整的充满过程的，不能说是实体但却是 是 实体的东西。
 
我们仍然生活在matrix中 ，苏醒的时候就是最痛苦的时候
真实，虚幻，对与matrix来说，亦真亦假，世界每天运转着，不会停息，直到他停止的那一刻
人每天忙碌着，直到他死亡的那一刻，在这个Matrix里面，你又看到了什么？

此方法，适用于 Utf编码的网页
 
 
在Utf8编码的网页上，双字节文字其实是使用UTF16编码 来表现的，当通过模拟表单的方式提交到后台的时候，因为UTF16不是标准的编码方式，所以一些程序无法识别！
 
解决方法：将utf16转换为utf8，然后使用escape编码，就 不会有这个问题了
 
 
解决方案代码如下：
 
 
function utf16to8(str) {
    var out, i, len, c;
    out = "";
    len = str.length;
    for(i = 0; i < len; i++) {
 c = str.charCodeAt(i);
 if ((c >= 0x0001) && (c <= 0x007F)) {
     out += str.charAt(i);
 } else if (c > 0x07FF) {
     out += String.fromCharCode(0xE0 | ((c >> 12) & 0x0F));
     out += String.fromCharCode(0x80 | ((c >>  6) & 0x3F));
     out += String.fromCharCode(0x80 | ((c >>  0) & 0x3F));
 } else {
     out += String.fromCharCode(0xC0 | ((c >>  6) & 0x1F));
     out += String.fromCharCode(0x80 | ((c >>  0) & 0x3F));
 }
    }
    return out;
}
function utf8to16(str) {
    var out, i, len, c;
    var char2, char3;
    out = "";
    len = str.length;
    i = 0;
    while(i < len) {
 c = str.charCodeAt(i++);
 switch(c >> 4)
 {
   case 0: case 1: case 2: case 3: case 4: case 5: case 6: case 7:
     // 0xxxxxxx
     out += str.charAt(i-1);
     break;
   case 12: case 13:
     // 110x xxxx   10xx xxxx
     char2 = str.charCodeAt(i++);
     out += String.fromCharCode(((c & 0x1F) << 6) | (char2 & 0x3F));
     break;
   case 14:
     // 1110 xxxx  10xx xxxx  10xx xxxx
     char2 = str.charCodeAt(i++);
     char3 = str.charCodeAt(i++);
     out += String.fromCharCode(((c & 0x0F) << 12) |
        ((char2 & 0x3F) << 6) |
        ((char3 & 0x3F) << 0));
     break;
 }
    }
    return out;
}
 

1.解决连接问题：
参照我的个人空间中的上面两篇文章，同时要注意，Apache一定要在/etc/profile中的环境变量或者你自己设置在自己环境中的相关的环境变量生效后，启动，你的设置才会有效 ,下面的问题也是一样
 
2.Oralce10g 服务端和客户端的乱码问题:
 
基本原理：服务器端和客户端的编码设置要一致，避免服务器自动转换，可解决乱码问题
 
我是在中文环境下装的oracle
 
我在服务端 select * from prop$
发现 NLS_LANGUAGE是AMERICAN
NLS_CHARACTERSET 是 ZHS16GBK

此时客户端上设置 如下 设置环境变量，注意，设置完之后参看我第一条说的问题！，一定要看
 
For windows set NLS_LANG = AMERICAN_AMERICA.ZHS16GBK
For Linux export NLS_LANG = AMERICAN_AMERICA.ZHS16GBK
设置好以后，最好重启，释放系统内存，重新启动
 
都ok后，启动Apache,此时php和oracle连接的乱码一并可以解决！

接上一篇！
 

1.在Debian上安装Oracle客户端，如果不能安装，可使用强制方法安装
./runInstaller -ignoreSysPrereqs
 
2.安装Instant Client 详细内容见上篇
 
 
3.设置LD_LIBRARY_PATH,追加oracle client所在的目录下的lib
 
4.设置TNS_ADMIN，方法同上篇
 
 
5.用客户端配置网络服务
 
 
6.重启你的系统后，重启apache
 
此时你就可以用 oci:dbname=yoursercive的方式连接了

好久 没有去逛过 书城，昨天为了求一个书城才有的资料，不得不千里迢迢来到 书城。
不是为了别的，就为了找一个关于 页面布局配色方面的资料，你要说别处还真不是没有，但是找起来不方便呢。

于是3步并作2步的跑到了销售电脑技术类书籍的地方，可真是不看不知道一看吓一跳，几乎所有的讲技术的书籍，都是在讲软件的 用法，天啊，这也叫技术。

翻了半天，找网页布局和色彩的相关资料，找了半天，都是每天书上都是photoshop，天啊

我又不是来 研究photoshop的。

最终结论，现在的人越来越浮躁，都想一步登天。所以能获利的，都是能沉得住气的人

     星期五，我离开了，那是我在万商的最后一天。真奇怪，每当到了这最后一天的时候人总有点点懒散，但是仍然挣扎的，要把没有了结的事情了解到，晚上的散伙反还算是吃的有滋有味，猛然发现自己的生活多么的有趣。今天，呆在了家里用自己存下的钱买了台电脑，专门用来研究技术的电脑。我在家附近的地方四处走动，猛然发现四处都有新鲜感，今天下午的时候，走在城中，清河路上，发现竟然是如此的冷清，人稀稀拉拉的，不成群，也许是真的不热闹，也许是上海真的台拥挤了，拥挤的，连一个人过马路的时候几乎分配不到马路的面积！

      喘了两天，想到自己明天不要上班，还真是感到欣慰，记得刚开始工作的时候，在外面没有房子住，最后不得不天天奔波于家里和公司之间，还没有睡醒，就要起床，出发，还有四处走动，就要回家。时间和自己打起了游击战。似乎那个时候，我就是时间的俘虏。想象现在，其实后面仍然要很多的事情要做，上周疲于奔命，竟然连号称可以连续工作7天7夜，不吃不喝的人也病倒了，天啊！我生活在这个时代，究竟是幸运还是无可奈何呢？

      忙碌的日子又快到了，接下来是什么呢？