/// <summary> Creates a new instance of a Lexer.</summary>
/// <param name="connection">The url to parse.
/// </param>
/// <exception cref="ParserException">If an error occurs opening the connection.
/// </exception>
public Lexer(HttpProtocol connection):this(new Page(connection))
{
}
/// <summary> Construct a parser using the provided HttpProtocol.
/// A feedback object printing to System.Console is used.
/// </summary>
/// <param name="connection">HttpProtocol object.
/// </param>
/// <throws> ParserException If the creation of the underlying Lexer </throws>
/// <summary> cannot be performed.
/// </summary>
public Parser(HttpProtocol connection):this(connection, STDOUT)
{
}
/// <summary> Construct a page reading from a URL connection.</summary>
/// <param name="connection">A fully conditioned connection. The connect()
/// method will be called so it need not be connected yet.
/// </param>
/// <exception cref="ParserException">An exception object wrapping a number of
/// possible error conditions, some of which are outlined below.
/// <li>IOException If an i/o exception occurs creating the
/// source.</li>
/// <li>UnsupportedEncodingException if the character set specified in the
/// HTTP header is not supported.</li>
/// </exception>
public Page(HttpProtocol connection)
{
if (null == connection)
{
throw new System.ArgumentException("HttpProtocol cannot be null");
}
Connection = connection;
mBaseUrl = null;
}
/// <summary> Constructor for custom HTTP access.
/// This would be used to create a parser for a URLConnection that needs
/// a special setup or negotiation conditioning beyond what is available
/// from the {@link #getConnectionManager ConnectionManager}.
/// </summary>
/// <param name="connection">A fully conditioned connection. The connect()
/// method will be called so it need not be connected yet.
/// </param>
/// <param name="fb">The object to use for message communication.
/// </param>
/// <throws> ParserException If the creation of the underlying Lexer </throws>
/// <summary> cannot be performed.
/// </summary>
public Parser(HttpProtocol connection, IParserFeedBack fb):this(new Lexer(connection), fb)
{
this.m_obHttpProtocol = connection;
}
private void GetPageContent(HttpProtocol obProtocol, bool bIsRefresh)
{
if(m_bHasContent && !bIsRefresh)
{
return;
}
if(obProtocol == null)
{
throw new ArgumentNullException("obProtocol", "Null HttpProtocol object specified");
}
lock(this)
{
ParserStream stream = null;
System.String type = String.Empty;
System.String charset = String.Empty;
try
{
m_obProtocolOutput = obProtocol.GetProtocolOutput();
if (m_obProtocolOutput.Status.Code == HttpProtocolStatus.SUCCESS)
{
m_bHasContent = true;
this.m_HttpContentProperties = m_obProtocolOutput.Content.ContentProperties;
type = this.ContentType;
charset = GetCharset(type);
stream = new ParserStream(new System.IO.MemoryStream(m_obProtocolOutput.Content.ContentData));
}
if (null != stream)
{
mSource = new InputStreamSource(stream,charset,m_obProtocolOutput.Content.ContentData.Length);
}
}
catch (System.Exception e)
{
throw new ParserException("Failed to get page content", e);
}
mUrl = obProtocol.URL.ToString();
mIndex = new PageIndex(this);
}
}