JSPWiki文件名转换问题

2022-07-15 80酷酷网 80kuku.com

　　js|文件名|问题|转换这几天研究了一下JSPWiki把网页存到硬盘上的代码,想出了一个解决的办法. JSPWiki把文件名转换成%xx的格式是为了能在硬盘上存储各个国家语言的文件,如果我们只在中文环境下使用,则没有必要进行这样的转换. JSPWiki里的pageProvider和attachmentProvider负责了上面的转换,而这两个provider可以在jspwiki.properties里面设置,所以我们继承一下JSPWiki里提供的类,再在配置文件里设置一下,很容易就让JSPWiki不进行文件名转换了. 我是通过继承NativeFileNameVersioningFileProvider和BasicAttachmentProvider类实现子类的.可惜BasicAttachmentProvider中负责文件名转换的方法是static的,我在子类里没办法把它重载掉,这个问题已经向JSPWiki的开发者提出修改意见了.现在page名可以存为中文了,附件名还要等到JSPWiki的BasicAttachmentProvider修改了以后才能实现. 附件里是编译完的代码,把它拷贝到jspwiki的lib目录里,再在jspwiki.properties里设置"jspwiki.pageProvider =net.sf.startemplate.NativeFileNameVersioningFileProvider"就可以把page存为中文名的文件了.这个provider和已有的%xx形式的page不能正常工作,需要把这些page的名称改成未编码的中文格式. 至于utf8编码的文件用notepad修改后会在最前面添加3个字符,那都是微软做的孽,不用微软的或者基于微软的文本编辑器就行了.(微软为了识别unicode的编码,在文件最开始加了标识符,而其他公司并不承认这个标准,所以才导致这些标识符会被显示出来) 浏览器对网址会转换成%xx形式后再向服务器,服务器收到以后也都有相应的逆向转换,这个应该没有什么问题. 很高兴和你交流!! 以下是我的代码:两个provider和一个工具类

   1:/**
   2: * Created on 2005-9-8 by Dengber.
   3: */
   4:package net.sf.startemplate;
   5:
   6:import org.apache.log4j.Logger;
   7:
   8:import com.ecyrd.jspwiki.providers.VersioningFileProvider;
   9:
  10:/**
  11: * A privider for jspwiki.
  12: * This class behaves the same with <code>VersioningFileProvider</code> provided
  13: * by jspwiki,excepting it saves wiki pages into harddisk with the name inputted
  14: * by the user. This is different with <code>VersioningFileProvider</code>, it
  15: * will save wiki pages with a encoded name, such as %E6%B5%8B%E8%AF%95.
  16: * Though it is a good solution to save files whos name may contains any 
  17: * charachers in the world, it is not READABLE.
  18: * <br /><br />
  19: * For many people, they use jspwiki only in a intranet, the character problem 
  20: * above is nothing to them. Saving wiki pages with the native charachter names 
  21: * may working well to them. So this class is a choice to that people: the wiki
  22: * pages in harddisk have a readable name.
  23: * <br /><br />
  24: * DANGER: If the jspwiki user can input a character that can not be proccessed
  25: * by your server, not use this class.
  26: * <br /><br />
  27: * How to use:Copy the jar file that contains this class to <b>lib</b> 
  28: * directory of your jspwiki webapp, and set the following value to 
  29: * jspwiki.properties:
  30: * jspwiki.pageProvider=net.sf.startemplate.NativeFileNameVersioningFileProvider 
  31: * 
  32: * author Dengber dengbergmail.com
  33: * 
  34: */
  35:public class NativeFileNameVersioningFileProvider extends
  36:        VersioningFileProvider {
  37:    private static final Logger log = Logger
  38:            .getLogger(NativeFileNameVersioningFileProvider.class);
  39:
  40:    /**
  41:     * Convert string to the formatting that can be as file name.
  42:     * 
  43:     * param pageName
  44:     * return
  45:     */
  46:    protected String mangleName(String pageName) {
  47:        final String[] WINDOWS_DEVICE_NAMES = { "con", "prn", "nul",
  48:            "aux", "lpt1", "lpt2", "lpt3", "lpt4", "lpt5", "lpt6", "lpt7",
  49:            "lpt8", "lpt9", "com1", "com2", "com3", "com4", "com5", "com6",
  50:            "com7", "com8", "com9" };
  51:
  52:        log.debug("mangleName(" + pageName + ")");
  53:        pageName = TextUtil.encodeFileName(pageName);
  54:
  55:        boolean windowsHackNeeded = false;
  56:        String os = System.getProperty("os.name").toLowerCase();
  57:
  58:        if (os.startsWith("windows") || os.equals("nt")) {
  59:            windowsHackNeeded = true;
  60:        }
  61:
  62:        if (windowsHackNeeded) {
  63:            String pn = pageName.toLowerCase();
  64:            for (int i = 0; i < WINDOWS_DEVICE_NAMES.length; i++) {
  65:                if (WINDOWS_DEVICE_NAMES[i].equals(pn)) {
  66:                    pageName = "$$$" + pageName;
  67:                }
  68:            }
  69:        }
  70:
  71:        log.debug("mangleName()=" + pageName);
  72:        return pageName;
  73:    }
  74:
  75:    /**
  76:     * Convert string converted by <code>mangleName()</code> to its orginal 
  77:     * value.
  78:     * 
  79:     * param pageName
  80:     * return 
  81:     */
  82:    protected String unmangleName(String pageName) {
  83:        log.debug("unmangleName(" + pageName + ")");
  84:        boolean windowsHackNeeded = false;
  85:        String os = System.getProperty("os.name").toLowerCase();
  86:
  87:        if (os.startsWith("windows") || os.equals("nt")) {
  88:            windowsHackNeeded = true;
  89:        }
  90:
  91:        if (windowsHackNeeded && pageName.startsWith("$$$")
  92:                && pageName.length() > 3) {
  93:            pageName = pageName.substring(3);
  94:        }
  95:
  96:        log.debug("unmangleName()=" + pageName);
  97:        return TextUtil.decodeFileName(pageName);
  98:    }
  99:
 100:}
 101:

   1:/**
   2: * Created on 2005-9-11 by Dengber.
   3: */
   4:package net.sf.startemplate;
   5:
   6:import org.apache.log4j.Logger;
   7:
   8:import com.ecyrd.jspwiki.providers.BasicAttachmentProvider;
   9:
  10:/**
  11: * A privider for jspwiki.
  12: * This class behaves the same with <code>VersioningFileProvider</code> provided
  13: * by jspwiki,excepting it saves wiki pages into harddisk with the name inputted
  14: * by the user. This is different with <code>VersioningFileProvider</code>, it
  15: * will save wiki pages with a encoded name, such as %E6%B5%8B%E8%AF%95.
  16: * 
  17: * see net.sf.startemplate.NativeFileNameVersioningFileProvider
  18: * 
  19: * author Dengber dengbergmail.com
  20: *
  21: */
  22:public class NativeFileNameAttachmentProvider extends BasicAttachmentProvider {
  23:    private static final Logger log = Logger.getLogger(NativeFileNameAttachmentProvider.class);
  24:    
  25:    /**
  26:     * Convert string to the formatting that can be as file name.
  27:     * 
  28:     * param pageName
  29:     * return
  30:     */
  31:    protected String mangleName(String pageName) {
  32:        log.debug("mangleName(" + pageName + "");
  33:        return TextUtil.encodeFileName(pageName);
  34:    }
  35:
  36:    /**
  37:     * Convert string converted by <code>mangleName()</code> to its orginal 
  38:     * value.
  39:     * 
  40:     * param pageName
  41:     * return 
  42:     */
  43:    protected String unmangleName(String pageName) {
  44:        log.debug("unmangleName(" + pageName + "");
  45:        return TextUtil.decodeFileName(pageName);
  46:    }
  47:}
  48:

   1:/**
   2: * Created on 2005-9-11 by Dengber.
   3: */
   4:package net.sf.startemplate;
   5:
   6:import org.apache.log4j.Logger;
   7:
   8:/**
   9: * A utility class for string processing.
  10: * 
  11: * author Dengber dengbergmail.com
  12: * 
  13: */
  14:public class TextUtil {
  15:
  16:    /**
  17:     * Encode \/:*?<>|% to %nn%nn formatting.
  18:     * 
  19:     * param fileName
  20:     * return
  21:     */
  22:    public static String encodeFileName(String fileName) {
  23:        StringBuffer result = new StringBuffer();
  24:        final String HEX_DIGITS = "0123456789ABCDEF";
  25:
  26:        for (int i = 0; i < fileName.length(); i++) {
  27:            char c = fileName.charAt(i);
  28:            switch (c) {
  29:            case '\\':
  30:            case '/':
  31:            case ':':
  32:            case '*':
  33:            case '?':
  34:            case '\"':
  35:            case '<':
  36:            case '>':
  37:            case '|':
  38:            case '%':
  39:                result.append('%');
  40:                //the character above is all one-byte character,
  41:                //so we only need to encode the most right 8 bit.
  42:                result.append(HEX_DIGITS.charAt((c & 0xF0) >> 4));
  43:                result.append(HEX_DIGITS.charAt(c & 0x0F));
  44:                break;
  45:            default:
  46:                result.append(c);
  47:            }
  48:        }
  49:
  50:        return result.toString();
  51:    }
  52:
  53:    /**
  54:     * Decode %nn%nn formatting string to its orginal value .
  55:     * 
  56:     * param fileName
  57:     * return
  58:     */
  59:    public static String decodeFileName(String fileName) {
  60:        StringBuffer result = new StringBuffer();
  61:
  62:        for (int i = 0; i < fileName.length(); i++) {
  63:            char c = fileName.charAt(i);
  64:            switch (c) {
  65:            case '%':
  66:                char upper = fileName.charAt(++i);
  67:                char lower = fileName.charAt(++i);
  68:                char ch = (char) Integer.parseInt("" + upper + lower, 16);
  69:                result.append(ch);
  70:                break;
  71:
  72:            default:
  73:                result.append(c);
  74:                break;
  75:            }
  76:        }
  77:
  78:        return result.toString();
  79:    }
  80:
  81:}
  82:

======= 2005-09-07 22:26:40 您在来信中写道：======= > dengber，您好！> > 我从sf.net上面下载了你的startemplate，看看不错。> > 我最近正在将jspwiki部署到我公司上，作为一种工作日志、工作总结、技术文章的共享平台。有个问题想问你一下，不知道你是否了解，如何让jspwiki里面用中文写的超级链接，所形成的实际文件名，可以成为中文编码呢？因为目前都会变成%xx%xx的形式来表示中文。> > 我知道jspwiki是利用java写的，它的中文处理机制，完全就是java的机制，就是将所有编码，统一成UTF-8编码。而windows2000或以上的系统的默认编码，一般是unicode，这两者不太一样。我不知道你试过直接编辑jspwiki的数据文件没，我试过，发现只能利用UltraEdit之类的可以支持UTF-8的编辑器，才可以成功编写，而如果利用windows自带的notepad来编辑的话，其文件头部会自动被加上3个字节。> > 另外，当在浏览器里面，鼠标指向一个含中文字符的链接的时候，会被自动转换成可能试URL所要求的表示形式，就是上述的两个加了百分号的数字来表示一个汉字的方式。> > 所以我想，要想解决这个问题，是不是需要解决上述两个问题。> > 不知道你是否研究过这个，请不吝赐教。　　