/// <summary>
/// Loads the HTML document from the specified TextReader.
/// </summary>
/// <param name="reader">The TextReader used to feed the HTML data into the document. May not be null.</param>
public void Load(TextReader reader)
{
// all Load methods pass down to this one
if (reader == null)
{
throw new ArgumentNullException("reader");
}
_onlyDetectEncoding = false;
if (OptionCheckSyntax)
{
_openednodes = new Hashtable();
}
else
{
_openednodes = null;
}
if (OptionUseIdAttribute)
{
_nodesid = new Hashtable();
}
else
{
_nodesid = null;
}
StreamReader sr = reader as StreamReader;
if (sr != null)
{
try
{
// trigger bom read if needed
sr.Peek();
}
catch
{
// void on purpose
}
_streamencoding = sr.CurrentEncoding;
_text = new ImplStreamAsArray (sr);
}
else
{
_streamencoding = null;
// Expensive, but cannot avoid since TextReader doesnt have any length of the underlying data
_text = new DummyStreamAsArray (reader.ReadToEnd());
}
_declaredencoding = null;
// SLIM: _text = reader.ReadToEnd();
_documentnode = CreateNode(HtmlNodeType.Document, 0);
Parse();
if (OptionCheckSyntax)
{
foreach(HtmlNode node in _openednodes.Values)
{
if (!node._starttag) // already reported
{
continue;
}
string html;
if (OptionExtractErrorSourceText)
{
html = node.OuterHtml;
if (html.Length > OptionExtractErrorSourceTextMaxLength)
{
html = html.Substring(0, OptionExtractErrorSourceTextMaxLength);
}
}
else
{
html = string.Empty;
}
AddError(
HtmlParseErrorCode.TagNotClosed,
node._line, node._lineposition,
node._streamposition, html,
"End tag </" + node.Name + "> was not found");
}
// we don't need this anymore
_openednodes.Clear();
}
}
/// <summary>
/// Detects the encoding of an HTML text provided on a TextReader.
/// </summary>
/// <param name="reader">The TextReader used to feed the HTML. May not be null.</param>
/// <returns>The detected encoding.</returns>
public Encoding DetectEncoding(TextReader reader)
{
if (reader == null)
{
throw new ArgumentNullException("reader");
}
_onlyDetectEncoding = true;
if (OptionCheckSyntax)
{
_openednodes = new Hashtable();
}
else
{
_openednodes = null;
}
if (OptionUseIdAttribute)
{
_nodesid = new Hashtable();
}
else
{
_nodesid = null;
}
StreamReader sr = reader as StreamReader;
if (sr != null)
{
_streamencoding = sr.CurrentEncoding;
_text = new ImplStreamAsArray (sr);
}
else
{
_streamencoding = null;
// Expensive, but cannot avoid since TextReader doesnt have any length of the underlying data
_text = new DummyStreamAsArray (reader.ReadToEnd());
}
_declaredencoding = null;
// SLIM: _text = reader.ReadToEnd();
_documentnode = CreateNode(HtmlNodeType.Document, 0);
// this is a hack, but it allows us not to muck with the original parsing code
try
{
Parse();
}
catch(EncodingFoundException ex)
{
_lastnodes.Clear();
return ex.Encoding;
}
return null;
}
