// Once collected, you get only one visit from the root tag block (hopefully HTML)
public void visit(TagBlock tb) {
// We are not going to parse attribute information in pages that don't have an HTML tag block
if (tb.startTag.tagName.equalsIgnoreCase("HTML")) {
HTMLElement he;
TagBlock tbe;
// Let's go looking for the head tag
for (Enumeration be = tb.body.elements(); be.hasMoreElements(); ) {
he = (HTMLElement) be.nextElement();
if (he.getClass() == TagBlock.class) {
tbe = (TagBlock) he;
if (tbe.startTag.tagName.equalsIgnoreCase("HEAD")) {
// Found the head tag block, now let's look for the page title and any meta tags
for (Enumeration hbe = tbe.body.elements(); hbe.hasMoreElements(); ) {
he = (HTMLElement) hbe.nextElement();
if (he.getClass() == TagBlock.class) {
tbe = (TagBlock) he;
if (tbe.startTag.tagName.equalsIgnoreCase("TITLE")) {
// Should only be one item in title body, hopefully it's a text element
for (Enumeration e = tbe.body.elements(); e.hasMoreElements(); ) {
he = (HTMLElement) e.nextElement();
if (he.getClass() == Text.class) {
pageTitle = HttpUtility.HtmlDecode(((Text) he).text).Trim();
break;
}
}
}
} else if (he.getClass() == Tag.class) {
Tag tg = (Tag) he;
// See if this is a META tag
if (tg.tagName.equalsIgnoreCase("META")) {
// Convert this tag and it's attributes into a more .NET
// friendly collection
MetaTag metaTag = new MetaTag(tg);
// Copy attributes in Java vector into .NET ArrayList...
for (Enumeration ae = tg.attributeList.attributes.elements();
ae.hasMoreElements(); )
metaTag.get_Attributes().Add((Attribute) ae.nextElement());
tagList.Add(metaTag);
}
}
}
// Once we've found the head element, we can leave...
break;
}
}
}
}
}
private void GetMetaTags(MetaTags tags, String Name, boolean stopAfterFirst) {
MetaTag tag;
for (int x = 0; x < tagList.get_Count(); x++) {
tag = get_Item(x);
if (tag != null) {
if (tag.get_Name().equalsIgnoreCase(Name)) {
tags.Add(tag);
if (stopAfterFirst) break;
}
}
}
}