(1)Tapestry5学习小结:中文问题

(1)    Tapestry中文问题

在目前的Tapestry5.0.5-SNAPSHOT版本中，Tapestry还没有提供对中文的支持。查找了Tapestry的邮件列表，看到如下的解决办法。

首先，可以写一个 UTF8properties的java文件。该文件可以到w3c的网站上找到，代码如下： 

public class Utf8Properties extends Properties {

    /**
     * use serialVersionUID from JDK 1.1.X for interoperability
     */
    private static final long serialVersionUID = 5907218757225133892L;

    /**
     * Encoding used to read properties from a file
     */
    public static final String ENCODING                    = "UTF-8";

    /**
     * Characters used to write comment lines in a property file
     */
    private static final String COMMENT                    = "#!";

    /**
     * Possible Separator between key and value of a property in a property
     * file
     */
    private static final String keyValueSeparators       = "=: \t\r\n\f";

    /**
     * Creates an empty property list with no default values.
     *
     * @see java.util.Properties#Properties()
     */
    public Utf8Properties() {
      this(null);
    }

    /**
     * Creates an empty property list with the specified defaults.
     *
     * @param   defaults   the defaults.
     * @see java.util.Properties#Properties(java.util.Properties)
     */
    public Utf8Properties(Properties defaults) {
      this.defaults = defaults;
    }

    /**
     * Reads a property list (key and element pairs) from the input
     * stream. The stream is assumed to be using the UTF-8
     * character encoding or compatible.
     * Characters can be written with their unicode escape sequence.
     *
     * @param      inStream   the input stream.
     * @exception IOException if an error occurred when reading from the
     *               input stream.
     * @throws        IllegalArgumentException if the input stream contains a
     *          malformed Unicode escape sequence.
     * @see java.util.Properties#load(java.io.InputStream)
     */
    public synchronized void load(InputStream inStream) throws IOException {
      BufferedReader in = new BufferedReader(new InputStreamReader(inStream, ENCODING));
      String line = in.readLine();

      while(line != null) {
          line = removeWhiteSpaces(line);
          if(!line.equals("") && COMMENT.indexOf(line.charAt(0)) == -1) {
            // Removes the beginning separators
            String property = line;
            // Reads the whole property if it is on multiple lines
            while(continueLine(line)) {
                property = property.substring(0, property.length() - 1);
                line = in.readLine();
                property += line;
            }
            //property = new String(property.getBytes(ENCODING), ENCODING);

            if(!property.equals("")) {
                int endOfKey = 0;
                // calculates the ending index of the key
                while(endOfKey < property.length() &&
                   (keyValueSeparators.indexOf(property.charAt(endOfKey)) == -1)) {
                  endOfKey++;
                }
                String key   = property.substring(0, endOfKey);
                String value = property.substring(endOfKey + 1, property.length());

                key   = loadConversion(key);
                value = loadConversion(removeWhiteSpaces(value));

                put(key, value);
                 For debugging only
                //System.out.println("key: " + key);
                //System.out.println("value: " + value);
                //System.out.println("-----------");
            }
          }
          line = in.readLine();
      }
    }

    /**
     * A simple method to remove white spaces
     * at the beginning of a String
     * @param       line  the String to treat
     * @return      the same String without white spaces at the beginning
     */
    public static String removeWhiteSpaces(String line) {
      int index = 0;
      while(index < line.length() && keyValueSeparators.indexOf(line.charAt(index)) != -1) {
          index++;
      }
      return line.substring(index, line.length());
    }

    /**
     * Replaces all characters preceded by a '\' with the corresponding special
     * character and converts unicode escape sequences to their value
     * @param       line  the String to treat
     * @return      the converted line
     */
    private String loadConversion(String line) {
      StringBuffer val = new StringBuffer(line.length());

      int index = 0;

      // Replace all the "\." substrings with their corresponding escaped characters
      for(; index < line.length(); index++) {
          char currentChar = line.charAt(index);
          if(currentChar == '\\') {
            index++;
            currentChar = line.charAt(index);
            switch(currentChar) {
            case 't':
                currentChar = '\t';
                break;
            case 'r':
                currentChar = '\r';
                break;
            case 'n':
                currentChar = '\n';
                break;
            case 'f':
                currentChar = '\f';
                break;
            case 'u':
                index++;
                // Read the xxxx
                int value=0;
                for (int i=0; i<4; i++) {
                  currentChar = line.charAt(index++);
                  //System.out.println(currentChar);
                  switch (currentChar) {
                  case '0': case '1': case '2': case '3': case '4':
                  case '5': case '6': case '7': case '8': case '9':
                      value = (value << 4) + currentChar - '0';
                      break;
                  case 'a': case 'b': case 'c':
                  case 'd': case 'e': case 'f':
                      value = (value << 4) + 10 + currentChar - 'a';
                      break;
                  case 'A': case 'B': case 'C':
                  case 'D': case 'E': case 'F':
                      value = (value << 4) + 10 + currentChar - 'A';
                      break;
                  default:
                      throw new IllegalArgumentException(
                      "Malformed \\uxxxx encoding.");
                  }
                }
                // index must point on the last character of the escaped
                // sequence to avoid missing the next character
                index--;
                currentChar = (char) value;
            default:
                break;
            }
          }
          val.append(currentChar);
      }

      return val.toString();
    }

    /**
     * Replaces special characters with their '2-chars' representation.
     * For example, '\n' becomes '\\' followed by 'n'
     * @param       line  the String to treat
     * @return      the resulting String
     */
    private String storeConversion(String line) {
      int length = line.length();
      StringBuffer outBuffer = new StringBuffer(length*2);

      for(int i = 0; i < length; i++) {
          char currentChar = line.charAt(i);
          switch(currentChar) {
          case '\\':
            outBuffer.append('\\');
            outBuffer.append('\\');
            break;
          case '\t':
            outBuffer.append('\\');
            outBuffer.append('t');
            break;
          case '\n':
            outBuffer.append('\\');
            outBuffer.append('n');
            break;
          case '\r':
            outBuffer.append('\\');
            outBuffer.append('r');
            break;
          case '\f':
            outBuffer.append('\\');
            outBuffer.append('f');
            break;
          default:
            outBuffer.append(currentChar);
          break;
          }
      }
      return outBuffer.toString();
    }

    /**
     * Indicates wether the property continues on the next line or not
     * @param       line  the beginning of the property that might be continued on the next line
     * @return      true if the propertiy continues on the following line, false otherwise
     */
    private boolean continueLine(String line) {
      if(line != null && !line.equals("")) {
          return line.charAt(line.length() - 1) == '\\';
      }
      return false;
    }

    /**
     * The same method as java.util.Properties.store(...)
     *
     * @param out  an output stream
     * @param header      a description of the property list
     * @see java.util.Properties#store(java.io.OutputStream, java.lang.String)
     */
    public void store(OutputStream out, String header) throws IOException {
      BufferedWriter output;
      output = new BufferedWriter(new OutputStreamWriter(out, ENCODING));
      if (header != null) {
          output.write("#" + header);
          output.newLine();
      }
      output.write("#" + new Date());
      output.newLine();
      // we do not want that a Thread could modify this Utf8Properties
      // while storing it
      synchronized (this) {
          Enumeration e = keys();
          while(e.hasMoreElements()) {
            String key = storeConversion((String)e.nextElement());
            String val = storeConversion((String)get(key));

            output.write(key + "=" + val);
            output.newLine();
          }
      }
      output.flush();
    }
}

 

然后将该文件编译后放到org\apache\tapestry\util包内。

接下来就可以修改其原文件了。要修改的原文件是org.apache.tapestry.internal.services包下的MessagesSourceImpl类。将该类的readProperties(Resource resource)方法中的java.util.Properties类改为我们编写的Utf8properties类。然后再修改org.apache.tapestry.internal.services包下的RequestImpl和ResponseImpl类文件。

将RequestImpl类文件中的方法改为如下内容：

public String getParameter(String name)
    {
        String param = _request.getParameter(name);
        if(param != null)
            try
            {
                param = new String(param.getBytes("ISO-8859-1"),"utf-8");
            }
            catch (UnsupportedEncodingException ex)
            {
                //won't error
            }
        return param;
    }

    public String[] getParameters(String name)
    {
        String[] params = _request.getParameterValues(name);
        try
        {
            for(int i=params.length-1;i>=0;i--){
                params[i] = new String(params[i].getBytes("ISO-8859-1"),"utf-8");
            }
        }
        catch (UnsupportedEncodingException ex)
        {
            //won't error
        }
        return params;
}

将ResponseImpl类文件中的方法改为如下内容：

public PrintWriter getPrintWriter(String contentType) throws IOException
    {
        notBlank(contentType, "contentType");
        if(contentType.matches("text/html.*"))
            _response.setCharacterEncoding("utf-8");
        _response.setContentType(contentType);

        return _response.getWriter();
    }
public OutputStream getOutputStream(String contentType) throws IOException
    {
        notBlank(contentType, "contentType");
        if(contentType.matches("text/html.*"))
            _response.setCharacterEncoding("utf-8");
        _response.setContentType(contentType);

        return _response.getOutputStream();
    }

最后重新编译打包就可以了。最后要注意的是此时我们所有应用用到的文件都要以utf-8格式进行编码.

 

该问题由LinuxBoy在邮件列表中提出，并得到了解决。

https://issues.apache.org/jira/browse/TAPESTRY-1294