Author: ltheussl
Date: Mon Apr 13 19:55:04 2009
New Revision: 764589
URL: http://svn.apache.org/viewvc?rev=764589&view=rev
Log:
[DOXIA-239] Handle non-ASCII characters in anchors and id's
Modified:
maven/doxia/doxia/trunk/doxia-core/src/main/java/org/apache/maven/doxia/util/HtmlTools.java
maven/doxia/doxia/trunk/doxia-core/src/test/java/org/apache/maven/doxia/util/DoxiaUtilsTest.java
maven/doxia/doxia/trunk/doxia-core/src/test/java/org/apache/maven/doxia/util/HtmlToolsTest.java
Modified:
maven/doxia/doxia/trunk/doxia-core/src/main/java/org/apache/maven/doxia/util/HtmlTools.java
URL:
http://svn.apache.org/viewvc/maven/doxia/doxia/trunk/doxia-core/src/main/java/org/apache/maven/doxia/util/HtmlTools.java?rev=764589&r1=764588&r2=764589&view=diff
==============================================================================
---
maven/doxia/doxia/trunk/doxia-core/src/main/java/org/apache/maven/doxia/util/HtmlTools.java
(original)
+++
maven/doxia/doxia/trunk/doxia-core/src/main/java/org/apache/maven/doxia/util/HtmlTools.java
Mon Apr 13 19:55:04 2009
@@ -257,92 +257,19 @@
/**
* Construct a valid id.
- * <p>
- * According to the <a
href="http://www.w3.org/TR/html4/types.html#type-name";>
- * HTML 4.01 specification section 6.2 SGML basic types</a>:
- * </p>
- * <p>
- * <i>ID and NAME tokens must begin with a letter ([A-Za-z]) and may be
- * followed by any number of letters, digits ([0-9]), hyphens ("-"),
- * underscores ("_"), colons (":"), and periods (".").</i>
- * </p>
*
* <p>
- * According to <a href="http://www.w3.org/TR/xhtml1/#C_8";>XHTML 1.0
- * section C.8. Fragment Identifiers</a>:
- * </p>
- * <p>
- * <i>When defining fragment identifiers to be backward-compatible, only
- * strings matching the pattern [A-Za-z][A-Za-z0-9:_.-]* should be
used.</i>
+ * <b>Note</b>: this method is identical to {...@link
DoxiaUtils#encodeId(String)},
+ * the rules to encode an id are laid out there.
* </p>
*
- * <p>
- * To achieve this we need to convert the <i>id</i> String. Two conversions
- * are necessary and one is done to get prettier ids:
- * </p>
- * <ol>
- * <li>If the first character is not a letter, prepend the id with the
- * letter 'a'</li>
- * <li>A space is replaced with an underscore '_'</li>
- * <li>Remove whitespace at the start and end before starting to
process</li>
- * </ol>
- *
- * <p>
- * For letters, the case is preserved in the conversion.
- * </p>
- *
- * <p>
- * Here are some examples:
- * </p>
- * <pre>
- * HtmlTools.encodeId( null ) = null
- * HtmlTools.encodeId( "" ) = ""
- * HtmlTools.encodeId( " _ " ) = "a_"
- * HtmlTools.encodeId( "1" ) = "a1"
- * HtmlTools.encodeId( "1anchor" ) = "a1anchor"
- * HtmlTools.encodeId( "_anchor" ) = "a_anchor"
- * HtmlTools.encodeId( "a b-c123 " ) = "a_b-c123"
- * HtmlTools.encodeId( " anchor" ) = "anchor"
- * HtmlTools.encodeId( "myAnchor" ) = "myAnchor"
- * </pre>
- *
- * <b>Note</b>: this method is intentionally similar to {...@link
DoxiaUtils#encodeId(String)}.
- *
* @param id The id to be encoded.
* @return The trimmed and encoded id, or null if id is null.
+ * @see {...@link DoxiaUtils#encodeId(java.lang.String)}.
*/
public static String encodeId( String id )
{
- if ( id == null )
- {
- return null;
- }
-
- id = id.trim();
- int length = id.length();
- StringBuffer buffer = new StringBuffer( length );
-
- for ( int i = 0; i < length; ++i )
- {
- char c = id.charAt( i );
-
- if ( ( i == 0 ) && ( !Character.isLetter( c ) ) )
- {
- buffer.append( "a" );
- }
-
- if ( c == ' ' )
- {
- buffer.append( "_" );
- }
- else if ( ( Character.isLetterOrDigit( c ) ) || ( c == '-' ) || (
c == '_' ) || ( c == ':' )
- || ( c == '.' ) )
- {
- buffer.append( c );
- }
- }
-
- return buffer.toString();
+ return DoxiaUtils.encodeId( id );
}
/**
@@ -355,31 +282,7 @@
*/
public static boolean isId( String text )
{
- if ( text == null || text.length() == 0 )
- {
- return false;
- }
-
- for ( int i = 0; i < text.length(); ++i )
- {
- char c = text.charAt( i );
-
- if ( i == 0 && !Character.isLetter( c ) )
- {
- return false;
- }
-
- if ( c == ' ' )
- {
- return false;
- }
- else if ( !Character.isLetterOrDigit( c ) && c != '-' && c != '_'
&& c != ':' && c != '.' )
- {
- return false;
- }
- }
-
- return true;
+ return DoxiaUtils.isValidId( text );
}
private HtmlTools()
Modified:
maven/doxia/doxia/trunk/doxia-core/src/test/java/org/apache/maven/doxia/util/DoxiaUtilsTest.java
URL:
http://svn.apache.org/viewvc/maven/doxia/doxia/trunk/doxia-core/src/test/java/org/apache/maven/doxia/util/DoxiaUtilsTest.java?rev=764589&r1=764588&r2=764589&view=diff
==============================================================================
---
maven/doxia/doxia/trunk/doxia-core/src/test/java/org/apache/maven/doxia/util/DoxiaUtilsTest.java
(original)
+++
maven/doxia/doxia/trunk/doxia-core/src/test/java/org/apache/maven/doxia/util/DoxiaUtilsTest.java
Mon Apr 13 19:55:04 2009
@@ -155,6 +155,8 @@
assertEquals( DoxiaUtils.encodeId( " anchor" ), "anchor" );
assertEquals( DoxiaUtils.encodeId( "myAnchor" ), "myAnchor" );
assertEquals( DoxiaUtils.encodeId( "my&Anchor" ), "my%26Anchor" );
+ assertEquals( DoxiaUtils.encodeId( "HÃ¥kon" ), "H%c3%a5kon" );
+ assertEquals( DoxiaUtils.encodeId( "TheuÃl" ), "Theu%c3%9fl" );
}
/**
@@ -178,5 +180,6 @@
assertTrue( DoxiaUtils.isValidId( "a:" ) );
assertTrue( DoxiaUtils.isValidId( "a." ) );
assertTrue( DoxiaUtils.isValidId( "index.html" ) );
+ assertFalse( DoxiaUtils.isValidId( "TheuÃl" ) );
}
}
Modified:
maven/doxia/doxia/trunk/doxia-core/src/test/java/org/apache/maven/doxia/util/HtmlToolsTest.java
URL:
http://svn.apache.org/viewvc/maven/doxia/doxia/trunk/doxia-core/src/test/java/org/apache/maven/doxia/util/HtmlToolsTest.java?rev=764589&r1=764588&r2=764589&view=diff
==============================================================================
---
maven/doxia/doxia/trunk/doxia-core/src/test/java/org/apache/maven/doxia/util/HtmlToolsTest.java
(original)
+++
maven/doxia/doxia/trunk/doxia-core/src/test/java/org/apache/maven/doxia/util/HtmlToolsTest.java
Mon Apr 13 19:55:04 2009
@@ -63,6 +63,8 @@
assertEquals( HtmlTools.encodeId( "a b-c123 " ), "a_b-c123" );
assertEquals( HtmlTools.encodeId( " anchor" ), "anchor" );
assertEquals( HtmlTools.encodeId( "myAnchor" ), "myAnchor" );
+ assertEquals( HtmlTools.encodeId( "HÃ¥kon" ), "H%c3%a5kon" );
+ assertEquals( HtmlTools.encodeId( "TheuÃl" ), "Theu%c3%9fl" );
}
/**
@@ -76,10 +78,9 @@
"http://www.example.com/?This is a simple test." ),
"http://www.example.com/?This%20is%20a%20simple%20test."; );
- // TODO: the & is not encoded?
- //assertEquals( HtmlTools.encodeURL(
- // "http://www.example.com/?This is a simple & short test." ),
- //
"http://www.example.com/?This%20is%20a%20simple%20%26%20short%20test."; );
+ assertEquals( HtmlTools.encodeURL(
+ "http://www.example.com/?This is a simple & short test." ),
+
"http://www.example.com/?This%20is%20a%20simple%20&%20short%20test."; );
}
/**
@@ -101,6 +102,7 @@
assertTrue( HtmlTools.isId( "a-" ) );
assertTrue( HtmlTools.isId( "a:" ) );
assertTrue( HtmlTools.isId( "a." ) );
+ assertFalse( HtmlTools.isId( "TheuÃl" ) );
}
/**
