//
// methods
//
#region public IPDFComponent GetComponent(IContentParser parser, string name)
/// <summary>
/// Returns a new component for the parser based on the specified tag name.
/// </summary>
/// <param name="parser"></param>
/// <param name="name"></param>
/// <returns>The instaniated component or null if the name is not recognised</returns>
public IPDFComponent GetComponent(IHtmlContentParser parser, string name, out HtmlComponentType type)
{
IPDFComponent proxy = null;
IParserComponentFactory innerfact;
if (null != _last && _lastName == name)
{
proxy = _last.GetComponent(parser, name, out type);
}
else if (_knowntags.TryGetValue(name, out innerfact))
{
_last = innerfact;
_lastName = name;
proxy = innerfact.GetComponent(parser, name, out type);
}
else
{
_last = null;
_lastName = null;
type = HtmlComponentType.Unknown;
proxy = GetUnknownComponent(parser, name);
}
if (proxy is Component)
{
((Component)proxy).Tag = name;
}
return(proxy);
}
/// <summary>
/// Creates a new instance of of the ParserResult struct
/// </summary>
/// <param name="parsed"></param>
/// <param name="offsetStart"></param>
/// <param name="offsetEnd"></param>
/// <param name="start"></param>
public HTMLParserResult(IPDFComponent parsed, HtmlComponentType type, string value, int offsetStart, int offsetEnd, bool start)
{
_value = value;
_parsed = parsed;
_type = type;
_start = offsetStart;
_end = offsetEnd;
_isStart = start;
_valid = true;
}
/// <summary>
/// Returns true if the current cursor is over a CDATA, Processing Instruction, DocType or Comment section.
/// </summary>
/// <param name="isCdata"></param>
/// <returns></returns>
protected bool IsNonHtmlTag(out HtmlComponentType type, out string terminator)
{
//CData is <![CDATA[ ... content ... ]]>
//Comment is <!-- ... content ... -->
//Processing Instruction is <? .... instruction ... ?>
//Doc Type is <!DOCTYPE ... type ... >
//Quick check on second char first
if (this.Source.Offset + 1 < this.Source.Length)
{
if (this.Source.Peek(1) == HTMLCommentSecondChar) //second char for comment, doctype and CData are the same.)
{
//Full check on comment string
if (this.Source.Offset + HTMLCommentStart.Length < this.Source.Length &&
this.Source.Substring(HTMLCommentStart.Length) == HTMLCommentStart)
{
this.Source.Offset += HTMLCommentStart.Length;
type = HtmlComponentType.Comment;
terminator = HTMLCommentEnd;
return(true);
}
//Full check on CDATA string
if (this.Source.Offset + HTMLCDATAStart.Length < this.Source.Length && this.Source.Substring(HTMLCDATAStart.Length) == HTMLCDATAStart)
{
this.Source.Offset += HTMLCDATAStart.Length;
type = HtmlComponentType.CData;
terminator = HTMLCDATAEnd;
return(true);
}
if (this.Source.Offset + HTMLDocTypeStart.Length < this.Source.Length && this.Source.Substring(HTMLDocTypeStart.Length) == HTMLDocTypeStart)
{
this.Source.Offset += HTMLDocTypeStart.Length;
type = HtmlComponentType.DocType;
terminator = HTMLDocTypeEnd;
return(true);
}
}
else if (this.Source.Offset + HTMLProcessingInstructionStart.Length < this.Source.Length && this.Source.Substring(HTMLProcessingInstructionStart.Length) == HTMLProcessingInstructionStart)
{
this.Source.Offset += HTMLProcessingInstructionStart.Length;
type = HtmlComponentType.ProcessingInstruction;
terminator = HTMLProcessingInstructionEnd;
return(true);
}
}
type = HtmlComponentType.None;
terminator = string.Empty;
return(false);
}
protected virtual bool ShouldSkipHtmlType(HtmlComponentType type)
{
switch (type)
{
case HtmlComponentType.DocType:
return(this.Parser.Settings.SkipDocType);
case HtmlComponentType.Comment:
return(this.Parser.Settings.SkipComments);
case HtmlComponentType.ProcessingInstruction:
return(this.Parser.Settings.SkipProcessingInstructions);
case HtmlComponentType.CData:
return(this.Parser.Settings.SkipCData);
case HtmlComponentType.Unknown:
return(this.Parser.Settings.SkipUnknownTags);
default:
return(false);
}
}
/// <summary>
/// Based on the current source reads in a single HTML tag and all it's attributes.
/// If it is a known component (based on the parsers component factory) then this will be returned, otherwise null.
/// </summary>
/// <param name="name">The actual characters read as the tag name</param>
/// <param name="autoend">Set to true if this tag / component should always end after it has begun (does not contain other components / tags)</param>
/// <param name="isEndMarker">Set to true if this is the end marker of another tag (</span> etc)</param>
/// <returns>The parsed component or null</returns>
private IPDFComponent GetCurrentTag(out string name, out HtmlComponentType type, out bool autoend, out bool isEndMarker)
{
autoend = false;
isEndMarker = false;
name = string.Empty;
this.Source.MoveNext();
char cur = this.Source.Current;
if (cur == HTMLEndMarker) //We are parsing a closing tag and want the inner text
{
if (this.Source.MoveNext())
{
cur = this.Source.Current;
isEndMarker = true;
}
else
{
type = HtmlComponentType.None;
return(null);
}
}
this.Buffer.Clear();
while (!this.Source.EOS)
{
cur = this.Source.Current;
if (cur == HTMLEndMarker || cur == HTMLCloseTag || cur == HTMLWhiteSpace)
{
break;
}
else
{
this.Buffer.Append(cur);
}
this.Source.MoveNext();
}
name = this.Buffer.ToString();
if (isEndMarker)
{
if (this.ParsedPath.Count > 0)
{
int index = this.ParsedPath.IndexOf(name);
if (index == this.ParsedPath.Count - 1)
{
//Last one so just pop it
HTMLParserResult prev = this.ParsedPath.Pop();
type = prev.Type;
return(prev.Parsed);
}
else if (index > -1)
{
HTMLParserResult[] tags = this.ParsedPath.PopToTag(name);
this.TagsToClose.PushRange(tags);
HTMLParserResult prev = this.TagsToClose.Pop();
type = prev.Type;
return(prev.Parsed);
}
}
//Orphan end marker
type = HtmlComponentType.None;
return(null);
}
else if (IsSkippedTag(name))
{
string endTag = HTMLStartTag.ToString() + HTMLEndMarker.ToString() + name + HTMLCloseTag.ToString();
MoveToEnd(endTag);
type = HtmlComponentType.None;
return(null);
}
IPDFComponent parsed = this._owner.ComponentFactory.GetComponent(this.Parser, name, out type);
if (!this.Source.EOS)
{
if (cur == HTMLWhiteSpace)
{
ReadAttributes(parsed, name);
}
else
{
while (cur == HTMLEndMarker || cur == HTMLCloseTag)
{
if (!this.Source.MoveNext())
{
break;
}
cur = this.Source.Current;
}
if (!this.Source.EOS)
{
this.Source.MovePrev(); //extra character for looking past marker removed
}
}
}
if (null != parsed)
{
autoend = !this._owner.ComponentFactory.IsContainerComponent(this.Parser, parsed, name);
}
else
{
autoend = false;
}
return(parsed);
}
public override IPDFComponent GetComponent(IHtmlContentParser parser, string name, out HtmlComponentType type)
{
type = HtmlComponentType.Body;
HTMLBody body = new HTMLBody();
body.ID = "html-body";
return(body);
}
public override IPDFComponent GetComponent(IHtmlContentParser parser, string name, out HtmlComponentType type)
{
type = HtmlComponentType.Link;
return(new HTMLAnchor());
}
public override IPDFComponent GetComponent(IHtmlContentParser parser, string name, out HtmlComponentType type)
{
type = HtmlComponentType.HorizontalLine;
return(new HTMLHorizontalRule());
}
public override IPDFComponent GetComponent(IHtmlContentParser parser, string name, out HtmlComponentType type)
{
SpanBase span;
type = HtmlComponentType.Span;
switch (name.ToLower())
{
case "b":
case "strong":
span = new HTMLBoldSpan();
break;
case "i":
case "em":
span = new HTMLItalicSpan();
break;
case "u":
span = new HTMLUnderlinedSpan();
break;
case ("font"):
span = new HTMLFontSpan();
break;
default:
span = new HTMLSpan();
break;
}
return(span);
}
/// <summary> /// Abstract method that inheritors implement to create and return specific types of components. /// </summary> /// <param name="parser"></param> /// <param name="name"></param> /// <returns></returns> public abstract IPDFComponent GetComponent(IHtmlContentParser parser, string name, out HtmlComponentType type);
public override IPDFComponent GetComponent(IHtmlContentParser parser, string name, out HtmlComponentType type)
{
type = HtmlComponentType.Preformatted;
return(new HTMLPreformatted());
}
public override IPDFComponent GetComponent(IHtmlContentParser parser, string name, out HtmlComponentType type)
{
type = HtmlComponentType.TableCell;
return(new HTMLTableHeaderCell());
}
public override IPDFComponent GetComponent(IHtmlContentParser parser, string name, out HtmlComponentType type)
{
type = HtmlComponentType.Panel;
return(new HTMLParagraph());
}
public override IPDFComponent GetComponent(IHtmlContentParser parser, string name, out HtmlComponentType type)
{
type = HtmlComponentType.Heading;
HeadingBase head;
switch (name.ToLower())
{
case "h1":
head = new HTMLHead1();
break;
case "h2":
head = new HTMLHead2();
break;
case "h3":
head = new HTMLHead3();
break;
case "h4":
head = new HTMLHead4();
break;
case "h5":
head = new HTMLHead5();
break;
case "h6":
head = new HTMLHead6();
break;
default:
head = null;
break;
}
return(head);
}
public override IPDFComponent GetComponent(IHtmlContentParser parser, string name, out HtmlComponentType type)
{
type = HtmlComponentType.ListItem;
return(new HTMLListDefinitionItem());
}
public override IPDFComponent GetComponent(IHtmlContentParser parser, string name, out HtmlComponentType type)
{
type = HtmlComponentType.List;
return(new HTMLListOrdered());
}
public override IPDFComponent GetComponent(IHtmlContentParser parser, string name, out HtmlComponentType type)
{
type = HtmlComponentType.Image;
return(new HTMLImage());
}
public override IPDFComponent GetComponent(IHtmlContentParser parser, string name, out HtmlComponentType type)
{
type = HtmlComponentType.Panel;
if (String.Equals("blockquote", name, StringComparison.OrdinalIgnoreCase))
{
return(new HTMLBlockQuote());
}
else if (String.Equals("fieldset", name, StringComparison.OrdinalIgnoreCase))
{
return(new HTMLFieldSet());
}
else if (String.Equals("legend", name, StringComparison.OrdinalIgnoreCase))
{
return(new HTMLLegend());
}
else
{
return(new HTMLDiv());
}
}
