/// <summary> Internal routine that actually does the parsing. The caller
/// can pass either an InputStream or file name. If both are passed,
/// the file name is preferred.
/// </summary>
internal Node ParseInternal(Stream input, string file, Stream Output, TidyMessageCollection messages)
{
Lexer lexer;
Node document = null;
Node doctype;
Out o = new OutImpl(); /* normal output stream */
PPrint pprint;
/* ensure config is self-consistent */
_options.Adjust();
if (file != null)
{
input = new FileStream(file, FileMode.Open, FileAccess.Read);
}
else if (input == null)
{
input = Console.OpenStandardInput();
}
if (input != null)
{
lexer = new Lexer(new ClsStreamInImpl(input, _options.CharEncoding, _options.TabSize), _options);
lexer.messages = messages;
/*
store pointer to lexer in input stream
to allow character encoding errors to be
reported
*/
lexer.input.Lexer = lexer;
/* Tidy doesn't alter the doctype for generic XML docs */
if (_options.XmlTags)
{
document = ParserImpl.parseXMLDocument(lexer);
}
else
{
document = ParserImpl.parseDocument(lexer);
if (!document.CheckNodeIntegrity())
{
Report.BadTree(lexer);
return null;
}
Clean cleaner = new Clean(_options.tt);
/* simplifies <b><b> ... </b> ...</b> etc. */
cleaner.NestedEmphasis(document);
/* cleans up <dir>indented text</dir> etc. */
cleaner.List2BQ(document);
cleaner.BQ2Div(document);
/* replaces i by em and b by strong */
if (_options.LogicalEmphasis)
{
cleaner.EmFromI(document);
}
if (_options.Word2000 && cleaner.IsWord2000(document, _options.tt))
{
/* prune Word2000's <![if ...]> ... <![endif]> */
cleaner.DropSections(lexer, document);
/* drop style & class attributes and empty p, span elements */
cleaner.CleanWord2000(lexer, document);
}
/* replaces presentational markup by style rules */
if (_options.MakeClean || _options.DropFontTags)
{
cleaner.CleanTree(lexer, document);
}
if (!document.CheckNodeIntegrity())
{
Report.BadTree(lexer);
return null;
}
doctype = document.FindDocType();
if (document.Content != null)
{
if (_options.Xhtml)
{
lexer.SetXhtmlDocType(document);
}
else
{
lexer.FixDocType(document);
}
if (_options.TidyMark)
{
lexer.AddGenerator(document);
}
}
/* ensure presence of initial <?XML version="1.0"?> */
if (_options.XmlOut && _options.XmlPi)
{
lexer.FixXmlPI(document);
}
if (document.Content != null)
{
Report.ReportVersion(lexer, doctype);
Report.ReportNumWarnings(lexer);
}
}
// Try to close the InputStream but only if if we created it.
if ((file != null) && (input != Console.OpenStandardOutput()))
{
try
{
input.Close();
}
catch (IOException)
{
}
}
if (lexer.messages.Errors > 0)
{
Report.NeedsAuthorIntervention(lexer);
}
o.State = StreamIn.FSM_ASCII;
o.Encoding = _options.CharEncoding;
if (lexer.messages.Errors == 0)
{
if (_options.BurstSlides)
{
Node body;
body = null;
/*
remove doctype to avoid potential clash with
markup introduced when bursting into slides
*/
/* discard the document type */
doctype = document.FindDocType();
if (doctype != null)
{
Node.DiscardElement(doctype);
}
/* slides use transitional features */
lexer.versions |= HtmlVersion.Html40Loose;
/* and patch up doctype to match */
if (_options.Xhtml)
{
lexer.SetXhtmlDocType(document);
}
else
{
lexer.FixDocType(document);
}
/* find the body element which may be implicit */
body = document.FindBody(_options.tt);
if (body != null)
{
pprint = new PPrint(_options);
Report.ReportNumberOfSlides(lexer, pprint.CountSlides(body));
pprint.CreateSlides(lexer, document);
}
else
{
Report.MissingBody(lexer);
}
}
else if (Output != null)
{
pprint = new PPrint(_options);
o.Output = Output;
if (_options.XmlTags)
{
pprint.PrintXmlTree(o, (short) 0, 0, lexer, document);
}
else
{
pprint.PrintTree(o, (short) 0, 0, lexer, document);
}
pprint.FlushLine(o, 0);
}
}
Report.ErrorSummary(lexer);
}
return document;
}
public static Node parseXMLDocument(Lexer lexer)
{
Node node, document, doctype;
document = lexer.NewNode();
document.Type = Node.RootNode;
doctype = null;
lexer.Options.XmlTags = true;
while (true)
{
node = lexer.GetToken(Lexer.IgnoreWhitespace);
if (node == null)
{
break;
}
/* discard unexpected end tags */
if (node.Type == Node.EndTag)
{
Report.Warning(lexer, null, node, Report.UNEXPECTED_ENDTAG);
continue;
}
/* deal with comments etc. */
if (Node.InsertMisc(document, node))
{
continue;
}
if (node.Type == Node.DocTypeTag)
{
if (doctype == null)
{
Node.InsertNodeAtEnd(document, node);
doctype = node;
}
else
{
Report.Warning(lexer, document, node, Report.DISCARDING_UNEXPECTED);
}
// TODO
continue;
}
/* if start tag then parse element's content */
if (node.Type == Node.StartTag)
{
Node.InsertNodeAtEnd(document, node);
parseXMLElement(lexer, node, Lexer.IgnoreWhitespace);
}
}
if (doctype != null && !lexer.CheckDocTypeKeyWords(doctype))
{
Report.Warning(lexer, doctype, null, Report.DTYPE_NOT_UPPER_CASE);
}
/* ensure presence of initial <?XML version="1.0"?> */
if (lexer.Options.XmlPi)
{
lexer.FixXmlPI(document);
}
return document;
}
