You are viewing a plain text version of this content. The canonical link for it is here.
Posted to commits@xalan.apache.org by jk...@apache.org on 2002/10/17 22:31:15 UTC

cvs commit: xml-xalan/java/src/org/apache/xml/dtm DTMFilter.java

jkesselm    2002/10/17 13:31:15

  Modified:    java/src/org/apache/xml/dtm Tag: xslt20 DTMFilter.java
  Log:
  SHOW_ITEM changes
  
  Revision  Changes    Path
  No                   revision
  
  
  No                   revision
  
  
  1.2.16.1.2.1 +31 -13    xml-xalan/java/src/org/apache/xml/dtm/DTMFilter.java
  
  Index: DTMFilter.java
  ===================================================================
  RCS file: /home/cvs/xml-xalan/java/src/org/apache/xml/dtm/DTMFilter.java,v
  retrieving revision 1.2.16.1
  retrieving revision 1.2.16.1.2.1
  diff -u -r1.2.16.1 -r1.2.16.1.2.1
  --- DTMFilter.java	14 Aug 2002 19:45:33 -0000	1.2.16.1
  +++ DTMFilter.java	17 Oct 2002 20:31:15 -0000	1.2.16.1.2.1
  @@ -170,27 +170,45 @@
      * a function.  Make sure this does not conflict with
      * {@link org.w3c.dom.traversal.NodeFilter}.
      * <p>
  -   * %REVIEW% Might be safer to start from higher bits and work down,
  -   * to leave room for the DOM to expand its set of constants... Or,
  -   * possibly, to create a DTM-specific field for these additional bits.
  +   * NOTE the need to avoid colliding with SHOW_ITEM.
  +   * <p>
  +   * NOTE that if we introduce any more node types, we may have to
  +   * push this up and/or individually pick out the appropriate bits in
  +   * SHOW_ITEM.
      */
     public static final int SHOW_BYFUNCTION = 0x00010000;
  -  
  +
     /**
  -   * Special bit for supporting "item" in ItemType..
  +   * Special bit for supporting "item" in ItemType. "An Item is either an
  +   * atomic value or a node." 
  +   * <p>
  +   * As a mask, then, SHOW_ITEM ought to accept
  +   * just about anything except a Sequence -- and with an occurances count,
  +   * it might accept sequences too.
      * <p>
  -   * %REVIEW% Might be safer to start from higher bits and work down,
  -   * to leave room for the DOM to expand its set of constants... Or,
  -   * possibly, to create a DTM-specific field for these additional bits.
  +   * %REVIEW%
  +   * That suggests that SHOW_ITEM should be pretty much the same as SHOW_ALL.
  +   * For now, I'm going to maintain a slight difference by leaving some of the
  +   * high bits unset, since we need space for SHOW_UNTYPED and SHOW_BYFUNCTION
  +   * ... but I suspect that we should reconsider the design of those flags,
  +   * to avoid muddying the waters.
  +   * <p>
  +   * NOTE that if we introduce any more node types, we may have to
  +   * push this up and/or individually pick out the appropriate bits in
  +   * SHOW_ITEM.
      */
  -  public static final int SHOW_ITEM = 0x00020000;
  +  public static final int SHOW_ITEM = 0x0000FFFF;
  +
   
     /**
  -   * Special bit for supporting "untyped" in ItemType..
  +   * Special bit for supporting "untyped" in ItemType. This should match
  +   * only atomic values (*not* nodes!) which are bare instances of xs:anySimpleType.
  +   * <p>
  +   * NOTE the need to avoid colliding with SHOW_ITEM.
      * <p>
  -   * %REVIEW% Might be safer to start from higher bits and work down,
  -   * to leave room for the DOM to expand its set of constants... Or,
  -   * possibly, to create a DTM-specific field for these additional bits.
  +   * NOTE that if we introduce any more node types, we may have to
  +   * push this up and/or individually pick out the appropriate bits in
  +   * SHOW_ITEM.
      */
     public static final int SHOW_UNTYPED = 0x00040000;
   
  
  
  

---------------------------------------------------------------------
To unsubscribe, e-mail: xalan-cvs-unsubscribe@xml.apache.org
For additional commands, e-mail: xalan-cvs-help@xml.apache.org