class HTML::Document
:nodoc:
text into a tree of nodes.
A top-level HTML document. You give it a body of text, and it will parse that
:nodoc:
def find(conditions)
conditions. The conditions are interpreted differently for different node
Search the tree for (and return) the first node that matches the given
def find(conditions) @root.find(conditions) end
def find_all(conditions)
conditions. The conditions are interpreted differently for different node
Search the tree for (and return) all nodes that match the given
def find_all(conditions) @root.find_all(conditions) end
def initialize(text, strict=false, xml=false)
def initialize(text, strict=false, xml=false) tokenizer = Tokenizer.new(text) @root = Node.new(nil) node_stack = [ @root ] while token = tokenizer.next node = Node.parse(node_stack.last, tokenizer.line, tokenizer.position, token, strict) node_stack.last.children << node unless node.tag? && node.closing == :close if node.tag? if node_stack.length > 1 && node.closing == :close if node_stack.last.name == node.name if node_stack.last.children.empty? node_stack.last.children << Text.new(node_stack.last, node.line, node.position, "") end node_stack.pop else open_start = node_stack.last.position - 20 open_start = 0 if open_start < 0 close_start = node.position - 20 close_start = 0 if close_start < 0 msg = <<EOF.strip ring attempt to close #{node_stack.last.name} with #{node.name} ened at byte #{node_stack.last.position}, line #{node_stack.last.line} osed at byte #{node.position}, line #{node.line} tributes at open: #{node_stack.last.attributes.inspect} xt around open: #{text[open_start,40].inspect} xt around close: #{text[close_start,40].inspect} strict ? raise(msg) : warn(msg) end elsif !node.childless?(xml) && node.closing != :close node_stack.push node end end end end