/// <summary>
/// Determines if the input document is valid, against the whitelist.
/// </summary>
/// <remarks>
/// It is considered valid if all the tags and attributes
/// in the input HTML are allowed by the whitelist.
/// <p/>
/// This method can be used as a validator for user input forms.
/// An invalid document will still be cleaned successfully
/// using the
/// <see cref="Clean(Supremes.Nodes.Document)">Clean(Supremes.Nodes.Document)</see>
/// document. If using as a validator, it is recommended to still clean the document
/// to ensure enforced attributes are set correctly, and that the output is tidied.
/// </remarks>
/// <param name="dirtyDocument">document to test</param>
/// <returns>true if no tags or attributes need to be removed; false if they do</returns>
public bool IsValid(Document dirtyDocument)
{
Validate.NotNull(dirtyDocument);
Document clean = Document.CreateShell(dirtyDocument.BaseUri);
int numDiscarded = CopySafeNodes(dirtyDocument.Body, clean.Body);
return numDiscarded == 0;
}
// current doc we are building into
// the stack of open elements
// current base uri, for creating new elements
// currentToken is used only for error tracking.
// null when not tracking errors
internal virtual void InitialiseParse(string input, string baseUri, ParseErrorList errors)
{
Validate.NotNull(input, "String input must not be null");
Validate.NotNull(baseUri, "BaseURI must not be null");
doc = new Document(baseUri);
reader = new CharacterReader(input);
this.errors = errors;
tokeniser = new Tokeniser(reader, errors);
stack = new DescendableLinkedList<Element>();
this.baseUri = baseUri;
}
/// <summary>
/// Creates a new, clean document, from the original dirty document,
/// containing only elements allowed by the whitelist.
/// </summary>
/// <remarks>
/// The original document is not modified.
/// Only elements from the dirt document's <c>body</c> are used.
/// </remarks>
/// <param name="dirtyDocument">Untrusted base document to clean.</param>
/// <returns>cleaned document.</returns>
public Document Clean(Document dirtyDocument)
{
Validate.NotNull(dirtyDocument);
Document clean = Document.CreateShell(dirtyDocument.BaseUri);
if (dirtyDocument.Body != null)
{
// frameset documents won't have a body. the clean doc will have empty body.
CopySafeNodes(dirtyDocument.Body, clean.Body);
}
return clean;
}
public void Init()
{
html = Dcsoup.Parse(htmlString);
}