Changeset 5246

Timestamp:

2003-08-22T09:43:11+12:00 (21 years ago)

Author:

jmt12

Message:

Removed obsolete encoding/decoding code (but not the greenstone ones)

File:

• trunk/gli/src/org/greenstone/gatherer/util/Utility.java (modified) (6 diffs)

trunk/gli/src/org/greenstone/gatherer/util/Utility.java

@@ -99 +99 @@
     static final public String DICTIONARY = "dictionary";
     static final public String DLS_MDS = "dls.mds";
+    static final public String ENCODING = "UTF-8";
     static final public String ENGLISH_VALUE = "en";
     /** Definition of an important directory name, in this case the etc (or extra information) directory for the collection. */
@@ -197 +198 @@
     return new TreePath(temp);
     }
+
+    /** Decodes a string of text so its safe to use in a Greenstone configuration file. Esentially replaces "\n" with a newline.
+     * @param raw The <strong>String</strong> before decoding, read from the configuration file..
+     * @return A <strong>String</strong> ready to be placed in a component.
+     */
+    static public String decodeGreenstone(String raw) {
+    raw = raw.replaceAll("&apos;", "\'");
+    raw = raw.replaceAll("&gt;", ">");
+    raw = raw.replaceAll("&lt;", "<");
+    raw = raw.replaceAll("&quot;", "\"");
+    raw = raw.replaceAll("&#39;", "\'");
+    raw = raw.replaceAll("\\\\n", "\n");
+    return raw;
+    }
+
     /** Takes a rfc2616 'safe' String and translates it back into its 'unsafe' form. Basically the native c wget decode_string() function, but without pointer stuff. If searches through the String looking for the pattern %xy where x and y are hexidecimal digits and where xy maps to a character.<BR> If x or y are not hexidecimal or % is followed by a \0 then the pattern is left as is.
      * @param encoded The url-safe <strong>String</strong> to be decoded.
@@ -247 +263 @@
     return result;
     }
-    /** Encodes a string of text so its safe to use in a Greenstone configuration file. Esentially replaces newlines with their escaped form.
-     * @param raw The <strong>String</strong> before encoding.
-     * @return A <strong>String</strong> which is safe to write to the configuration file.
-     */
-    static final private char AMPERSTAMP_CHAR = '&';
-    static final private char ESCAPE_CHAR = '\\';
-    static final private char GREATER_THAN_CHAR = '>';
-    static final private char LESS_THAN_CHAR = '<';
-    static final private char NEWLINE_CHAR = '\n';
-    static final private char QUOTE_CHAR = '\'';
-    static final private char SPEECH_CHAR = '\"';
-    static final private String ENCODED_AMPERSTAMP_STR = "&amp;";
-    static final private String ENCODED_GREATER_THAN_STR = "&gt;";
-    static final private String ENCODED_LESS_THAN_STR = "&lt;";
-    static final private String ENCODED_SPEECH_STR = "&quot;";
-    static final private String ESCAPED_NEWLINE_STR = "\\n";
-
-    /** Decodes a string of text so its safe to use in a Greenstone configuration file. Esentially replaces "\n" with a newline.
-     * @param raw The <strong>String</strong> before decoding, read from the configuration file..
-     * @return A <strong>String</strong> ready to be placed in a component.
-     */
-    static public String decodeGreenstone(String raw) {
-    raw = raw.replaceAll("&apos;", "\'");
-    raw = raw.replaceAll("&gt;", ">");
-    raw = raw.replaceAll("&lt;", "<");
-    raw = raw.replaceAll("&quot;", "\"");
-    raw = raw.replaceAll("&#39;", "\'");
-    raw = raw.replaceAll("\\\\n", "\n");
+
+    static public String encodeGreenstone(String raw) {
+    raw = raw.replaceAll("<", "&lt;");
+    raw = raw.replaceAll(">", "&gt;");
+    raw = raw.replaceAll("\n", "\\\\n");
     return raw;
-    }
-
-    static public String encodeGreenstone(String raw) {
-    // Once again regex fails to provide the power necessary for me to change strings. What I need to do is replace "<" and ">" with "&lt;" and "&gt;", and replace "\<" and "\>" with "<" and ">".
-    StringBuffer processed = new StringBuffer();
-    int index = 0;
-    while(index < raw.length()) {
-        char c = raw.charAt(index);
-        switch(c) {
-        // Replace a normal new line character with "\n"
-            case NEWLINE_CHAR:
-            processed.append(ESCAPED_NEWLINE_STR);
-            break;
-        // Replace "\<" with "<", or with "\&lt;" if this is for XML. Similar requirements for "\>".
-            case ESCAPE_CHAR:
-            if(index + 1 < raw.length()) {
-            char d = raw.charAt(index + 1);
-            if(d == LESS_THAN_CHAR) {
-                processed.append(LESS_THAN_CHAR);
-                index++;
-                break;
-            }
-            else if(d == GREATER_THAN_CHAR) {
-                processed.append(GREATER_THAN_CHAR);
-                index++;
-                break;
-            }
-            }
-            // I have no idea how this would happen, but I better watch for it anyway
-            processed.append(c);
-            break;
-        // Replace "<" with "&lt;"
-            case LESS_THAN_CHAR:
-            processed.append(ENCODED_LESS_THAN_STR);
-            break;
-        // Replace ">" with "&gt;"
-            case GREATER_THAN_CHAR:
-            processed.append(ENCODED_GREATER_THAN_STR);
-            break;
-            default:
-            processed.append(c);
-        }
-        index++;
-    }
-    return processed.toString();
-    }
-    /** When retrieve text for, or from the collect.cfg file it may contain characters that can't go into a DOM such as "<" and ">". We also might already have encoded versions "&lt;" and "&gt;". Thus we must encode the former, and double encode the latter. */
-    static public String encodeXML(String raw) {
-    StringBuffer processed = new StringBuffer();
-    int index = 0;
-    while(index < raw.length()) {
-        char c = raw.charAt(index);
-        switch(c) {
-        case GREATER_THAN_CHAR:
-        processed.append(ENCODED_GREATER_THAN_STR);
-        break;
-        case LESS_THAN_CHAR:
-        processed.append(ENCODED_LESS_THAN_STR);
-        break;
-        case AMPERSTAMP_CHAR:
-        processed.append(ENCODED_AMPERSTAMP_STR);
-        break;
-        default:
-        processed.append(c);
-        }
-        index++;
-    }
-    return processed.toString();
     }
 
@@ -356 +281 @@
         // Create an output format for our document.
         OutputFormat f = new OutputFormat(document);
+        f.setEncoding(ENCODING);
         f.setIndenting(true);
         f.setLineWidth(0);
         f.setPreserveSpace(false);
         // Create the necessary writer stream for serialization.
-        OutputStreamWriter osw = new OutputStreamWriter(os);
+        OutputStreamWriter osw = new OutputStreamWriter(os, ENCODING);
         Writer w               = new BufferedWriter(osw);
         // Generate a new serializer from the above.
@@ -824 +750 @@
         try {
         URL url = ClassLoader.getSystemResource(filename);
-        file = new File(URLDecoder.decode(url.getFile(), "UTF-8"));
+        file = new File(URLDecoder.decode(url.getFile(), ENCODING));
         url = null;
         }
@@ -847 +773 @@
     try {
         FileInputStream fis   = new FileInputStream(file);
-        InputStreamReader isr = new InputStreamReader(fis);
+        InputStreamReader isr = new InputStreamReader(fis, ENCODING);
         Reader r              = new BufferedReader(isr);
         InputSource isc       = new InputSource(r);