static void HtmlParser06()
{
HtmlParser p;
HtmlElement *root = p.Parse("<ul><p>ignore<li><br><meta><li><ol><li></ul><dropme>");
assert(9 == p.ElementsCount());
assert(0 == p.TotalAttrCount());
assert(str::Eq("ul", root->name));
assert(!root->next);
HtmlElement *el = root->GetChildByName("li");
assert(el);
assert(str::Eq(el->down->name, "br"));
assert(str::Eq(el->down->next->name, "meta"));
assert(!el->down->next->next);
el = root->GetChildByName("li", 1);
assert(el);
assert(!el->next);
el = el->GetChildByName("ol");
assert(!el->next);
assert(str::Eq(el->down->name, "li"));
assert(!el->down->down);
}
static void WalkChmTocOrIndex(EbookTocVisitor *visitor, HtmlElement *list, UINT cp, bool isIndex, int level=1)
{
CrashIf(!list->NameIs("ul"));
// some broken ToCs wrap every <li> into its own <ul>
for (; list && list->NameIs("ul"); list = list->next) {
for (HtmlElement *el = list->down; el; el = el->next) {
if (!el->NameIs("li"))
continue; // ignore unexpected elements
bool valid = (isIndex ? VisitChmIndexItem : VisitChmTocItem)(visitor, el, cp, level);
if (!valid)
continue; // skip incomplete elements and all their children
HtmlElement *nested = el->GetChildByName("ul");
// some broken ToCs have the <ul> follow right *after* a <li>
if (!nested && el->next && el->next->NameIs("ul"))
nested = el->next;
if (nested)
WalkChmTocOrIndex(visitor, nested, cp, isIndex, level + 1);
}
}
}