applied HTML improvements from Nick Kew, allowing to do more checking to

* HTMLparser.c include/libxml/HTMLparser.h: applied HTML
  improvements from Nick Kew, allowing to do more checking
  to HTML elements and attributes.
Daniel

1 files changed, 43 insertions, 1 deletions

diff --git a/include/libxml/HTMLparser.h b/include/libxml/HTMLparser.h
index f0921371..c6d8899c 100644
--- a/include/libxml/HTMLparser.h
+++ b/include/libxml/HTMLparser.h
@@ -28,7 +28,8 @@ typedef xmlDocPtr htmlDocPtr;
 typedef xmlNodePtr htmlNodePtr;
 
 /*
- * Internal description of an HTML element.
+ * Internal description of an HTML element, representing HTML 4.01
+ * and XHTML 1.0 (which share the same structure).
  */
 typedef struct _htmlElemDesc htmlElemDesc;
 typedef htmlElemDesc *htmlElemDescPtr;
@@ -42,6 +43,23 @@ struct _htmlElemDesc {
     char dtd;           /* 1: only in Loose DTD, 2: only Frameset one */
     char isinline;      /* is this a block 0 or inline 1 element */
     const char *desc;   /* the description */
+
+/* NRK Jan.2003
+ * New fields encapsulating HTML structure
+ *
+ * Bugs:
+ *	This is a very limited representation.  It fails to tell us when
+ *	an element *requires* subelements (we only have whether they're
+ *	allowed or not), and it doesn't tell us where CDATA and PCDATA
+ *	are allowed.  Some element relationships are not fully represented:
+ *	these are flagged with the word MODIFIER
+ */
+    const char** subelts;		/* allowed sub-elements of this element */
+    const char* defaultsubelt;	/* subelement for suggested auto-repair
+					   if necessary or NULL */
+    const char** attrs_opt;		/* Optional Attributes */
+    const char** attrs_depr;		/* Additional deprecated attributes */
+    const char** attrs_req;		/* Required attributes */
 };
 
 /*
@@ -110,6 +128,30 @@ int			htmlParseChunk		(htmlParserCtxtPtr ctxt,
 						 const char *chunk,
 						 int size,
 						 int terminate);
+
+/* NRK/Jan2003: further knowledge of HTML structure
+ */
+typedef enum {
+  HTML_NA = 0 ,		/* something we don't check at all */
+  HTML_INVALID = 0x1 ,
+  HTML_DEPRECATED = 0x2 ,
+  HTML_VALID = 0x4 ,
+  HTML_REQUIRED = 0xc /* VALID bit set so ( & HTML_VALID ) is TRUE */
+} htmlStatus ;
+
+/* Using htmlElemDesc rather than name here, to emphasise the fact
+   that otherwise there's a lookup overhead
+*/
+htmlStatus htmlAttrAllowed(const htmlElemDesc*, const xmlChar*, int) ;
+int htmlElementAllowedHere(const htmlElemDesc*, const xmlChar*) ;
+htmlStatus htmlElementStatusHere(const htmlElemDesc*, const htmlElemDesc*) ;
+htmlStatus htmlNodeStatus(const htmlNodePtr, int) ;
+#define htmlDefaultSubelement(elt) elt->defaultsubelt
+#define htmlElementAllowedHereDesc(parent,elt) \
+	htmlElementAllowedHere((parent), (elt)->name)
+#define htmlRequiredAttrs(elt) (elt)->attrs_req
+
+
 #ifdef __cplusplus
 }
 #endif