commons-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From bur...@apache.org
Subject cvs commit: jakarta-commons-sandbox/feedparser/src/java/org/apache/commons/feedparser/locate DiscoveryLocator.java FeedLocator.java ProbeLocator.java
Date Fri, 20 Aug 2004 21:44:08 GMT
burton      2004/08/20 14:44:07

  Modified:    feedparser/src/java/org/apache/commons/feedparser
                        FeedList.java
               feedparser/src/java/org/apache/commons/feedparser/locate
                        DiscoveryLocator.java FeedLocator.java
                        ProbeLocator.java
  Log:
  Fixed bugs WRT the Atom feedparser RFC
  
  Revision  Changes    Path
  1.2       +25 -3     jakarta-commons-sandbox/feedparser/src/java/org/apache/commons/feedparser/FeedList.java
  
  Index: FeedList.java
  ===================================================================
  RCS file: /home/cvs/jakarta-commons-sandbox/feedparser/src/java/org/apache/commons/feedparser/FeedList.java,v
  retrieving revision 1.1
  retrieving revision 1.2
  diff -u -r1.1 -r1.2
  --- FeedList.java	23 Apr 2004 17:45:43 -0000	1.1
  +++ FeedList.java	20 Aug 2004 21:44:06 -0000	1.2
  @@ -77,10 +77,32 @@
        *
        * @author <a href="mailto:burton@peerfear.org">Kevin Burton</a>
        */
  -    public void setAdAtomFeed( FeedReference adAtomFeed ) { 
  +    public void setAdAtomFeed( FeedReference ref ) { 
           
  -        this.adAtomFeed = adAtomFeed;
  +        this.adAtomFeed = ref;
           
  +    }
  +
  +    public void setFirstAdAtomFeed( FeedReference ref ) {
  +
  +        // > The order of the autodiscovery elements is significant.
  +        // > The first element SHOULD point to the publisher's
  +        // > preferred feed for the document.
  +
  +        if ( getAdAtomFeed() == null )
  +            setAdAtomFeed( ref );
  +
  +    }
  +
  +    public void setFirstAdRSSFeed( FeedReference ref ) {
  +
  +        // > The order of the autodiscovery elements is significant.
  +        // > The first element SHOULD point to the publisher's
  +        // > preferred feed for the document.
  +
  +        if ( getAdRSSFeed() == null )
  +            setAdRSSFeed( ref );
  +
       }
   
   }
  
  
  
  1.13      +32 -8     jakarta-commons-sandbox/feedparser/src/java/org/apache/commons/feedparser/locate/DiscoveryLocator.java
  
  Index: DiscoveryLocator.java
  ===================================================================
  RCS file: /home/cvs/jakarta-commons-sandbox/feedparser/src/java/org/apache/commons/feedparser/locate/DiscoveryLocator.java,v
  retrieving revision 1.12
  retrieving revision 1.13
  diff -u -r1.12 -r1.13
  --- DiscoveryLocator.java	13 Aug 2004 23:19:24 -0000	1.12
  +++ DiscoveryLocator.java	20 Aug 2004 21:44:06 -0000	1.13
  @@ -24,6 +24,8 @@
   
   /**
    *
  + * http://xml.coverpages.org/draft-ietf-atompub-autodiscovery-00.txt
  + * 
    * @author <a href="mailto:burton@apache.org">Kevin A. Burton</a>
    */
   public class DiscoveryLocator {
  @@ -33,13 +35,21 @@
        * this.
        */
       static Pattern element_pattern =
  -        Pattern.compile( "<link[^>]+" );
  +        Pattern.compile( "<link[^>]+",
  +                         Pattern.CASE_INSENSITIVE );
   
       /**
  -     * Regex to match on 
  +     * Regex to match on attributes.
        */
  +
  +    // > Attribute values MUST be one of the following: enclosed in double
  +    // > quotes, enclosed in single quotes, or not enclosed in quotes at all.
  +    //
  +    // FIXME: we don't YET support this.
  +    
       static Pattern attr_pattern =
  -        Pattern.compile( "([a-zA-Z]+)=[\"']([^\"']+)[\"']" );
  +        Pattern.compile( "([a-zA-Z]+)=[\"']([^\"']+)[\"']",
  +                         Pattern.CASE_INSENSITIVE );
   
       static HashSet mediatypes = new HashSet();
   
  @@ -88,17 +98,28 @@
   
                   //expand the href
                   String href = (String)attributes.get( "href" );
  +
  +                // http://xml.coverpages.org/draft-ietf-atompub-autodiscovery-00.txt
  +                
  +                // > The href attribute MUST be present in an Atom autodiscovery element,
  +                // > and its value MUST be the URI [RFC2396] of an Atom feed.  The value
  +                // > MAY be a relative URI, and if so, clients MUST resolve it to a
full
  +                // > URI (section 5 of [RFC2396]) using the document's base URI (section
  +                // > 12.4 of HTML 4 [W3C.REC-html401-19991224]).
  +
                   href = ResourceExpander.expand( resource, href );
   
                   FeedReference feedReference = new FeedReference( href, type );
                   
  +                feedReference.title = (String)attributes.get( "title" );
  +                
                   list.add( feedReference );
   
                   if ( type.equals( FeedReference.ATOM_MEDIA_TYPE ) )
  -                    list.setAdAtomFeed( feedReference );
  -
  +                    list.setFirstAdAtomFeed( feedReference );
  +                    
                   if ( type.equals( FeedReference.RSS_MEDIA_TYPE ) )
  -                    list.setAdRSSFeed( feedReference );
  +                    list.setFirstAdRSSFeed( feedReference );
   
               }
               
  @@ -123,8 +144,11 @@
   
           while ( m.find( index ) ) {
   
  -            String name = m.group( 1 );
  -            String value = m.group( 2 );
  +            String name = m.group( 1 ).toLowerCase().trim();
  +            String value = m.group( 2 ).toLowerCase().trim();
  +
  +            if ( "".equals( value ) ) 
  +                value = null; 
   
               map.put( name, value );
               
  
  
  
  1.17      +7 -1      jakarta-commons-sandbox/feedparser/src/java/org/apache/commons/feedparser/locate/FeedLocator.java
  
  Index: FeedLocator.java
  ===================================================================
  RCS file: /home/cvs/jakarta-commons-sandbox/feedparser/src/java/org/apache/commons/feedparser/locate/FeedLocator.java,v
  retrieving revision 1.16
  retrieving revision 1.17
  diff -u -r1.16 -r1.17
  --- FeedLocator.java	18 Aug 2004 18:23:01 -0000	1.16
  +++ FeedLocator.java	20 Aug 2004 21:44:06 -0000	1.17
  @@ -85,13 +85,19 @@
           //String resource = "file:///projects/feedparser/tests/locate5.html";
           //String resource = "file:///projects/feedparser/tests/locate6.html";
   
  +        String resource = "file:///projects/feedparser/tests/locate8.html";
  +
           //String resource = "http://blogs.sun.com/roller/page/gonzo";
   
           //String resource = "http://gonze.com/weblog/";
   
           //String resource = "http://codinginparadise.org/";
   
  -        String resource= "http://www.thealarmclock.com/mt/";
  +
  +        //        String resource = "http://bucsfishingreport.com/pMachine/weblog.php";
  +        
  +        //String resource = "http://www.livejournal.com/community/indiexiankids/";
  +//String resource= "http://www.thealarmclock.com/mt/";
           
           //String resource = "http://guinness.joeuser.com";
           
  
  
  
  1.8       +9 -0      jakarta-commons-sandbox/feedparser/src/java/org/apache/commons/feedparser/locate/ProbeLocator.java
  
  Index: ProbeLocator.java
  ===================================================================
  RCS file: /home/cvs/jakarta-commons-sandbox/feedparser/src/java/org/apache/commons/feedparser/locate/ProbeLocator.java,v
  retrieving revision 1.7
  retrieving revision 1.8
  diff -u -r1.7 -r1.8
  --- ProbeLocator.java	11 Aug 2004 20:48:09 -0000	1.7
  +++ ProbeLocator.java	20 Aug 2004 21:44:06 -0000	1.8
  @@ -31,6 +31,15 @@
   
       static HashMap probeMapping = new HashMap();
   
  +    //FIXME: also just try common path names.  Do this in ORDER so that I find
  +    //beter metadata feeds sooner
  +    //
  +    //      /atom.xml
  +    //      /atom.xml
  +    //      /rss.xml
  +    //      /index.xml
  +    //      /index.rdf
  +    
       static {
   
           // We can use:
  
  
  

---------------------------------------------------------------------
To unsubscribe, e-mail: commons-dev-unsubscribe@jakarta.apache.org
For additional commands, e-mail: commons-dev-help@jakarta.apache.org


Mime
View raw message