public IExtensionData[] CalculateReferencedExtensionData(string content)
{
Hashtable datas = new Hashtable();
ContentSourceManager.SmartContentPredicate predicate = new ContentSourceManager.SmartContentPredicate();
SimpleHtmlParser p = new SimpleHtmlParser(content);
for (Element el; null != (el = p.Next());)
{
if (predicate.IsMatch(el))
{
BeginTag bt = el as BeginTag;
Attr idAttr = bt.GetAttribute("id");
if (idAttr != null) //Synchronized WP posts will strip ID attrs (bug 488143)
{
string smartContentSourceId;
string smartContentId;
string smartContentElementId = idAttr.Value;
ContentSourceManager.ParseContainingElementId(smartContentElementId, out smartContentSourceId, out smartContentId);
IExtensionData data = GetExtensionData(smartContentId);
if (data != null)
datas[smartContentId] = data;
}
}
}
return (IExtensionData[])ArrayHelper.CollectionToArray(datas.Values, typeof(IExtensionData));
}
protected internal override void HandleResult(string homepageHtml, ITestResults results)
{
Regex regex = new Regex(Regex.Escape(guid1) + "(.*?)" + Regex.Escape(guid2));
SimpleHtmlParser parser = new SimpleHtmlParser(homepageHtml);
for (Element e = parser.Next(); e != null; e = parser.Next())
{
if (e is Text)
{
Match m = regex.Match(e.ToString());
if (m.Success)
{
string str = m.Groups[1].Value;
if (str == HtmlUtils.EscapeEntities(TEST_STRING))
results.AddResult("requiresHtmlTitles", YES);
else if (str == HtmlUtils.EscapeEntities(HtmlUtils.EscapeEntities(TEST_STRING)))
results.AddResult("requiresHtmlTitles", NO);
else
results.AddResult("requiresHtmlTitles", "[ERROR] (value was: " + str + ")");
return;
}
}
}
throw new InvalidOperationException("Title encoding test failed--title was not detected");
}
public FormFactory(Stream s)
{
using (StreamReader reader = new StreamReader(s))
{
parser = new SimpleHtmlParser(reader.ReadToEnd());
}
}
public virtual string Execute(string html)
{
SimpleHtmlParser parser = new SimpleHtmlParser(html);
StringBuilder output = new StringBuilder(html.Length);
Element next;
while (null != (next = parser.Next()))
output.Append(Replace(next));
return output.ToString();
}
public virtual string Execute(string html)
{
SimpleHtmlParser parser = new SimpleHtmlParser(html);
StringBuilder output = new StringBuilder(html.Length);
Element next;
while (null != (next = parser.Next()))
{
output.Append(Replace(next));
}
return(output.ToString());
}
public string ScanAndPreserve(string html)
{
StringBuilder sb = new StringBuilder(html.Length);
SimpleHtmlParser p = new SimpleHtmlParser(html);
Element e;
while (null != (e = p.Next()))
{
if (!(e is BeginTag))
{
sb.Append(html, e.Offset, e.Length);
continue;
}
BeginTag bt = (BeginTag)e;
if (bt.NameEquals("div"))
{
switch (bt.GetAttributeValue("class"))
{
case ContentSourceManager.EDITABLE_SMART_CONTENT:
case ContentSourceManager.SMART_CONTENT:
sb.Append(html, e.Offset, e.Length);
sb.Append(p.CollectHtmlUntil("div"));
sb.Append("</div>");
continue;
}
}
if (!(bt.NameEquals("object")
|| bt.NameEquals("embed")
|| bt.NameEquals("noembed")
|| bt.NameEquals("script")))
{
sb.Append(html, e.Offset, e.Length);
continue;
}
else
{
string collected = p.CollectHtmlUntil(bt.Name);
string preserve = bt.RawText + collected + "</" + bt.Name + ">";
string preserveId = Guid.NewGuid().ToString("N");
preserved[preserveId] = preserve;
sb.AppendFormat("<span id=\"preserve{0}\" class=\"{1}\">", preserveId, PRESERVE_CLASS);
sb.Append(preserve);
sb.Append("</span>");
}
}
return sb.ToString();
}
internal static IElementPredicate Parse(string criterion)
{
SimpleHtmlParser parser = new SimpleHtmlParser(criterion);
Element el = parser.Next();
if (el == null)
{
Debug.Fail("Criterion was null");
throw new ArgumentException("Criterion was null");
}
if (parser.Next() != null)
{
Debug.Fail("Too many criteria");
throw new ArgumentException("Too many criteria");
}
if (el is BeginTag)
{
BeginTag tag = (BeginTag)el;
if (tag.HasResidue || tag.Unterminated)
{
Debug.Fail("Malformed criterion");
throw new ArgumentException("Malformed criterion");
}
RequiredAttribute[] attributes = new RequiredAttribute[tag.Attributes.Length];
for (int i = 0; i < attributes.Length; i++)
{
attributes[i] = new RequiredAttribute(tag.Attributes[i].Name, tag.Attributes[i].Value);
}
return(new BeginTagPredicate(tag.Name, attributes));
}
else if (el is EndTag)
{
return(new EndTagPredicate(((EndTag)el).Name));
}
else if (el is Text)
{
return(new TextPredicate(el.RawText));
}
else if (el is Comment)
{
return(new CommentPredicate(el.RawText));
}
else
{
Debug.Fail("Invalid criterion \"" + criterion + "\"");
throw new ArgumentException("Invalid criterion \"" + criterion + "\"");
}
}
/// <summary>
/// Clones active smart content contained in the provided HTML, and disables unknown smart content.
/// </summary>
public static string PrepareSmartContentHtmlForEditorInsertion(string html, IContentSourceSidebarContext sourceContext)
{
StringBuilder output = new StringBuilder();
ContentSourceManager.SmartContentPredicate predicate = new ContentSourceManager.SmartContentPredicate();
SimpleHtmlParser p = new SimpleHtmlParser(html);
for (Element el; null != (el = p.Next());)
{
if (predicate.IsMatch(el))
{
BeginTag bt = el as BeginTag;
Attr idAttr = bt.GetAttribute("id");
String contentSourceId, contentItemId;
ContentSourceManager.ParseContainingElementId(idAttr.Value, out contentSourceId, out contentItemId);
ISmartContent smartContent = sourceContext.FindSmartContent(contentItemId);
if (smartContent != null)
{
String newId = Guid.NewGuid().ToString();
sourceContext.CloneSmartContent(contentItemId, newId);
if (RefreshableContentManager.ContentSourcesWithRefreshableContent.Contains(contentSourceId))
{
IExtensionData extensionData = sourceContext.FindExtentsionData(newId);
Debug.Assert(extensionData != null);
// Since we just made a new id for the smart content just about to be inserted
// we want to give it a chance to get a callback because its callback might have happened while
// it was on the clipboard(in the event of cut). This means the refreshable content manager doesnt know
// to watch out for this smart content on paste, it only knows to look out for who created it. Thus
// we just force the callback, and if it didnt need it, nothing will happen.
if (extensionData.RefreshCallBack == null)
{
extensionData.RefreshCallBack = DateTime.UtcNow;
}
}
idAttr.Value = ContentSourceManager.MakeContainingElementId(contentSourceId, newId);
}
else
{
ContentSourceManager.RemoveSmartContentAttributes(bt);
}
}
output.Append(el.ToString());
}
return output.ToString();
}
private static Element[] Elements(string html)
{
ArrayList elements = new ArrayList();
SimpleHtmlParser parser = new SimpleHtmlParser(html);
Element el;
while (null != (el = parser.Next()))
elements.Add(el);
return (Element[])elements.ToArray(typeof(Element));
}
/// <summary>
/// Balances the HTML and safely truncates it, using a custom algorithm
/// to determine how much each character/string counts against maxCost.
/// </summary>
public static string Balance(string html, int maxCost, HTMLBalancerCostFilter costFilter, bool ellipsis)
{
bool appendEllipsis = false;
SimpleHtmlParser parser = new SimpleHtmlParser(html);
ArrayList openTags = new ArrayList();
StringBuilder output = new StringBuilder();
long balance = 0; // long to make sure that int32.MaxValue does not cause overflow
if (costFilter == null)
costFilter = new DefaultCostFilter();
Element el;
while (null != (el = parser.Next()))
{
if (el is StyleElement ||
el is ScriptElement ||
el is Comment ||
el is MarkupDirective)
{
continue;
}
long lenLeft = Math.Max(0, maxCost - balance - LengthToClose(costFilter, openTags));
if (el is Tag)
{
if (el is BeginTag && ((BeginTag)el).Unterminated)
continue; // skip corrupted tags
if (TagCost(costFilter, openTags, (Tag)el) > lenLeft)
break; // don't use this tag; we're done
else
{
RegisterTag(openTags, (Tag)el);
output.Append(el.ToString());
balance += costFilter.ElementCost(el);
}
}
else if (el is Text)
{
if (costFilter.ElementCost(el) > lenLeft)
{
// shrink down the text to fit
output.Append(costFilter.TruncateText((Text)el, (int)lenLeft));
appendEllipsis = true;
break;
}
else
{
// plenty of room
output.Append(el.ToString());
balance += costFilter.ElementCost(el);
}
//update the text end index
}
else
{
if (costFilter.ElementCost(el) > lenLeft)
break;
else
{
output.Append(el.ToString());
balance += costFilter.ElementCost(el);
}
}
}
// Append an ellipsis if we truncated text
// We use "..." here rather than TextHelper.Ellipsis, because some mail clients don't understand "\u2026".
if (ellipsis && appendEllipsis)
output.Append("...");
for (int i = openTags.Count - 1; i >= 0; i--)
{
output.Append(MakeEndTag((string)openTags[i]));
}
return output.ToString();
}
private static void EditorContext_PerformTemporaryFixupsToEditedHtml(TemporaryFixupArgs args)
{
string html = args.Html;
if (html.Contains("table"))
{
StringBuilder output = new StringBuilder(html.Length);
SimpleHtmlParser parser = new SimpleHtmlParser(html);
for (Element el; null != (el = parser.Next());)
{
output.Append(html, el.Offset, el.Length);
if (el is BeginTag &&
((BeginTag)el).NameEquals("td"))
{
Element e = parser.Peek(0);
if (e is EndTag && ((EndTag)e).NameEquals("td"))
output.Append(" ");
}
}
args.Html = output.ToString();
}
}
private static void DetachExtendedEntryBehavior(TemporaryFixupArgs args)
{
string html = args.Html;
if (html.Contains(EXTENDED_ENTRY_ID))
{
//replace the EXTENDED_ENTRY_ID behavior div with the <!--more--> comment
StringBuilder output = new StringBuilder(html.Length);
SimpleHtmlParser parser = new SimpleHtmlParser(html);
SmartPredicate splitDiv = new SmartPredicate(String.Format(CultureInfo.InvariantCulture, "<div id='{0}'>", EXTENDED_ENTRY_ID));
for (Element el; null != (el = parser.Next());)
{
if (splitDiv.IsMatch(el))
{
Element e = parser.Peek(0);
if (e is EndTag && ((EndTag)e).NameEquals("div"))
{
output.Append(BlogPost.ExtendedEntryBreak);
parser.Next();
}
}
else
output.Append(html, el.Offset, el.Length);
}
args.Html = output.ToString();
}
}
public HtmlTextSource(SimpleHtmlParser parser)
{
this._parser = parser;
}
/// <summary>
/// Namespaced tags come with Office 2007 clipboard data and result in weird
/// namespace declarations being inserted as text into the DOM.
/// </summary>
public static string StripNamespacedTagsAndCommentsAndMarkupDirectives(string html)
{
StringBuilder output = new StringBuilder(html.Length);
SimpleHtmlParser parser = new SimpleHtmlParser(html);
for (Element el; null != (el = parser.Next());)
{
if (el is Tag && ((Tag)el).Name.IndexOf(':') >= 0)
continue;
if (el is Comment)
continue;
if (el is MarkupDirective)
continue;
if (el is BeginTag)
{
foreach (Attr attr in ((BeginTag)el).Attributes)
{
if (ILLEGAL_ATTR_REGEX.IsMatch(attr.Name))
((BeginTag)el).RemoveAttribute(attr.Name);
}
}
output.Append(el.ToString());
}
html = output.ToString();
return html;
}
public static string ConvertNewLinesToBr(string html)
{
SimpleHtmlParser parser = new SimpleHtmlParser(html);
StringBuilder sb = new StringBuilder();
Element ele = parser.Next();
while (ele != null)
{
if (ele is Text)
sb.Append(ele.RawText.Replace("\r\n", "<br/>"));
else
sb.Append(ele.RawText);
ele = parser.Next();
}
return sb.ToString();
}
protected override bool ShouldAllowNewLineInsert(string html)
{
SimpleHtmlParser p = new SimpleHtmlParser(html);
for (Element el; null != (el = p.Next());)
{
BeginTag tag = el as BeginTag;
if (tag != null && tag.NameEquals("img"))
{
// Don't allow new lines after emoticons.
string classNames = tag.GetAttributeValue("class");
if (!String.IsNullOrEmpty(classNames) && classNames.Contains(Emoticon.CLASS_NAME))
return false;
}
}
return base.ShouldAllowNewLineInsert(html);
}
/// <summary>
/// Reposition the extractor back to the beginning of the
/// HTML.
/// </summary>
/// <returns>Returns this. This allows chaining together of calls,
/// like this:
///
/// if (ex.Seek(...).Success || ex.Reset().Seek(...).Success) { ... }
/// </returns>
public HtmlExtractor Reset()
{
lastMatch = null;
parser = new SimpleHtmlParser(html);
return this;
}
public HtmlExtractor(string html)
{
this.html = html;
this.parser = new SimpleHtmlParser(html);
}
public HtmlExtractor(Stream data, Encoding encoding)
{
using (StreamReader reader = new StreamReader(data, encoding))
html = reader.ReadToEnd();
this.parser = new SimpleHtmlParser(html);
}
internal static IElementPredicate Parse(string criterion)
{
SimpleHtmlParser parser = new SimpleHtmlParser(criterion);
Element el = parser.Next();
if (el == null)
{
Trace.Fail("Criterion was null");
throw new ArgumentException("Criterion was null");
}
if (parser.Next() != null)
{
Trace.Fail("Too many criteria");
throw new ArgumentException("Too many criteria");
}
if (el is BeginTag)
{
BeginTag tag = (BeginTag)el;
if (tag.HasResidue || tag.Unterminated)
{
Trace.Fail("Malformed criterion");
throw new ArgumentException("Malformed criterion");
}
RequiredAttribute[] attributes = new RequiredAttribute[tag.Attributes.Length];
for (int i = 0; i < attributes.Length; i++)
attributes[i] = new RequiredAttribute(tag.Attributes[i].Name, tag.Attributes[i].Value);
return new BeginTagPredicate(tag.Name, attributes);
}
else if (el is EndTag)
{
return new EndTagPredicate(((EndTag)el).Name);
}
else if (el is Text)
{
return new TextPredicate(el.RawText);
}
else if (el is Comment)
{
return new CommentPredicate(el.RawText);
}
else
{
Trace.Fail("Invalid criterion \"" + criterion + "\"");
throw new ArgumentException("Invalid criterion \"" + criterion + "\"");
}
}
/// <summary>
/// Walks the current contents to find smart content areas. When one is found, it calls the operation on the smart content. The operation has a chance
/// to return new content. If the content is non-null it will replace the current content.
/// </summary>
/// <param name="contents">the raw HTML string whose structured blocks will be replaced.</param>
/// <param name="operation">Delegate for generating replacement content.</param>
/// <param name="editMode">If true, then the element's stylename will be activated for editing</param>
/// <param name="continueOnError">
/// true - if the plugin throws an exception, it keeps crawling the DOM
/// false - if a plugin throws an exception, it stops processing the DOM and return empty string
/// null - if a plugin throws an exception, this function will rethrow it
/// </param
/// <returns>the contents with structured blocks replaced.</returns>
internal static string PerformOperation(string contents, SmartContentOperation operation, bool editMode, IContentSourceSidebarContext sourceContext, bool? continueOnError)
{
//replace all structured content blocks with their editor HTML
//string html = PostBodyPreprocessor.Preprocess(contents);
StringBuilder sb = new StringBuilder();
SimpleHtmlParser parser = new SimpleHtmlParser(contents);
for (Element e = parser.Next(); e != null; e = parser.Next())
{
if (e is BeginTag)
{
BeginTag beginTag = (BeginTag)e;
string elementClassName = beginTag.GetAttributeValue("class");
if (ContentSourceManager.IsSmartContentClass(elementClassName))
{
ISmartContent sContent = null;
try
{
string contentSourceId, contentItemId;
string blockId = beginTag.GetAttributeValue("id");
if (blockId != null)
{
ContentSourceManager.ParseContainingElementId(blockId, out contentSourceId, out contentItemId);
ContentSourceInfo contentSource = sourceContext.FindContentSource(contentSourceId);
if (contentSource != null && contentSource.Instance is SmartContentSource)
{
SmartContentSource sSource = (SmartContentSource)contentSource.Instance;
sContent = sourceContext.FindSmartContent(contentItemId);
if (sContent != null)
{
//write the div with the appropriate className
string newClassName = editMode ? ContentSourceManager.EDITABLE_SMART_CONTENT : ContentSourceManager.SMART_CONTENT;
beginTag.GetAttribute("class").Value = newClassName;
//replace the inner HTML of the div with the source's editor HTML
string content = parser.CollectHtmlUntil("div");
sb.Append(e.ToString());
operation(sourceContext, sSource, sContent, ref content);
sb.Append(content);
sb.Append("</div>");
continue;
}
}
}
}
catch (Exception ex)
{
Trace.WriteLine(String.Format(CultureInfo.InvariantCulture, "Error loading smart content item\r\n{0}", ex));
sContent = null;
if (continueOnError == null)
throw;
if (!continueOnError.Value)
return String.Empty;
}
if (sContent == null)
{
//this element references an unknown smart content, so it should not be editable
Attr classAttr = beginTag.GetAttribute("class");
classAttr.Value = ContentSourceManager.SMART_CONTENT;
}
}
}
sb.Append(e.ToString());
}
return sb.ToString();
}
/// <summary>
/// Decides if, after the given HTML is inserted, we should insert a new line.
/// </summary>
protected virtual bool ShouldAllowNewLineInsert(string html)
{
SimpleHtmlParser p = new SimpleHtmlParser(html);
for (Element el; null != (el = p.Next());)
{
if (el is BeginTag && (((BeginTag)el).NameEquals("div") || ((BeginTag)el).NameEquals("img")))
{
return true;
}
}
return false;
}
public HtmlExtractor(string html)
{
this.html = html;
this.parser = new SimpleHtmlParser(html);
}
/// <summary>
/// Namespaced tags come with Office 2007 clipboard data and result in weird
/// namespace declarations being inserted as text into the DOM. (Bug 303784)
/// </summary>
private static string StripNamespacedTags(string html)
{
StringBuilder output = new StringBuilder(html.Length);
SimpleHtmlParser parser = new SimpleHtmlParser(html);
for (Element el; null != (el = parser.Next());)
{
if (el is Tag && ((Tag)el).Name.IndexOf(':') >= 0)
continue;
output.Append(el.RawText);
}
html = output.ToString();
return html;
}
public FormFactory(string html)
{
parser = new SimpleHtmlParser(html);
}
private static string BalanceHtml(string html)
{
StringBuilder sb = new StringBuilder(html.Length + 10);
SimpleHtmlParser parser = new SimpleHtmlParser(html);
Element el;
while (null != (el = parser.Next()))
{
if (el is BeginTag)
{
BeginTag bt = (BeginTag)el;
if (!ElementFilters.RequiresEndTag(bt.Name))
bt.Complete = true;
}
sb.Append(el.ToString());
}
return sb.ToString();
}
private string ThinInternal(string html, bool preserveImages, bool strict, params ModifyReplacement[] modifyReplacements)
{
Hashtable replacements = _tagSpecs;
if (strict)
{
replacements = _tagSpecsStrict;
}
if (modifyReplacements != null)
{
replacements = (Hashtable)replacements.Clone();
foreach (ModifyReplacement modifyReplacement in modifyReplacements)
modifyReplacement(replacements);
}
// Will hold the results of the leading whitespace buffer.
// This buffer may or may not make it into the final result,
// depending on whether any block-level tags are present.
StringBuilder leadingOutput = new StringBuilder(10);
// Will hold the results of everything else.
StringBuilder mainOutput = new StringBuilder(html.Length);
// references whichever output buffer is current.
StringBuilder output = leadingOutput;
SimpleHtmlParser parser = new SimpleHtmlParser(html);
Element el;
bool preserveWhitespace = false; // <pre> blocks should preserve whitespace
WhitespaceBuffer whitespaceBuffer = new WhitespaceBuffer();
whitespaceBuffer.Promote(WhitespaceClass.Paragraph); // Insert an implicit <p> unless the first non-whitespace element is a block
bool hasBlock = false;
while (null != (el = parser.Next()))
{
if (el is Tag)
{
Tag t = (Tag)el;
string lowerName = t.Name.ToLower(CultureInfo.InvariantCulture);
TagDesc desc = (TagDesc)replacements[lowerName];
// if this tag is not in the table, drop it
if (desc == null)
continue;
// Replace tag with substitute tag if necessary (e.g. <DIV> becomes <P>)
string tagName = desc.Substitute;
if (tagName == null)
tagName = lowerName;
// special case for images
if (!preserveImages && tagName == TAG_IMG)
continue;
bool beginTag = el is BeginTag;
ElementClass elClass = WhitespaceBuffer.ClassifyTag(tagName, desc.TagType);
hasBlock |= (elClass == ElementClass.Block || elClass == ElementClass.Paragraph || elClass == ElementClass.Break);
if (!preserveWhitespace && WhitespaceBuffer.ProcessElementClass(ref whitespaceBuffer, output, elClass, true))
continue;
output = mainOutput;
if (beginTag)
{
WriteBeginTag(desc, tagName, ((BeginTag)el).Attributes, output);
if (tagName == TAG_PRE)
preserveWhitespace = true;
}
else if (el is EndTag)
{
if (!((EndTag)el).Implicit && desc.TagType != TagType.Empty)
{
output.Append(string.Format(CultureInfo.InvariantCulture, "</{0}>", tagName));
}
if (tagName == TAG_PRE)
preserveWhitespace = false;
}
}
else if (el is Text)
{
string text = el.RawText;
text = HtmlUtils.EscapeEntities(HtmlUtils.UnEscapeEntities(text, HtmlUtils.UnEscapeMode.NonMarkupText));
if (!preserveWhitespace && WhitespaceBuffer.ProcessElementClass(ref whitespaceBuffer, output, WhitespaceBuffer.ClassifyText(text), false))
continue;
output = mainOutput;
output.Append(text);
}
}
if (hasBlock && ReferenceEquals(mainOutput, output))
output.Insert(0, leadingOutput.ToString());
// The whitespace buffer may not be empty at this point. That's OK--we want to drop trailing whitespace
return output.ToString();
}
/// <summary>
/// Converts tag names, attribute names, and style text to lowercase.
/// </summary>
private string CleanupHtml(string html, bool xml)
{
bool needsCleanup;
do
{
needsCleanup = false;
StringBuilder output = new StringBuilder(html.Length);
SimpleHtmlParser htmlParser = new SimpleHtmlParser(html);
for (Element el; null != (el = htmlParser.Next());)
{
if (el is BeginTag)
{
BeginTag bt = (BeginTag)el;
if (RemoveMeaninglessTags(htmlParser, bt))
{
// Since we are removing a tag, we will want to clean up again, since that might mean
// there will be another tag to remove
needsCleanup = true;
continue;
}
output.Append("<");
output.Append(bt.Name.ToLower(CultureInfo.InvariantCulture));
foreach (Attr attr in bt.Attributes)
{
if (attr.NameEquals("contenteditable") || attr.NameEquals("atomicselection") ||
attr.NameEquals("unselectable"))
continue;
output.Append(" ");
output.Append(attr.Name.ToLower(CultureInfo.InvariantCulture));
if (attr.Value != null)
{
string attrVal = attr.Value;
if (attr.NameEquals("style"))
attrVal = LowerCaseCss(attrVal);
else if (attr.Name == attr.Value)
attrVal = attrVal.ToLower(CultureInfo.InvariantCulture);
output.AppendFormat("=\"{0}\"",
xml
? HtmlUtils.EscapeEntitiesForXml(attrVal, true)
: HtmlUtils.EscapeEntities(attrVal));
}
}
if (bt.HasResidue)
{
if (bt.Attributes.Length == 0)
output.Append(" ");
output.Append(bt.Residue);
}
if (bt.Complete)
output.Append(" /");
output.Append(">");
}
else if (el is EndTag)
{
output.AppendFormat("</{0}>", ((EndTag)el).Name.ToLower(CultureInfo.InvariantCulture));
}
else if (el is Text)
{
string textHtml = HtmlUtils.TidyNbsps(el.RawText);
if (xml)
textHtml =
HtmlUtils.EscapeEntitiesForXml(
HtmlUtils.UnEscapeEntities(textHtml, HtmlUtils.UnEscapeMode.NonMarkupText), false);
output.Append(textHtml);
}
else if (el is StyleText)
output.Append(el.RawText.ToLower(CultureInfo.InvariantCulture));
else
output.Append(el.RawText);
}
html = output.ToString();
} while (needsCleanup);
return html;
}
public static bool ContainsUnbalancedDivs(string html)
{
int tags = 0;
SimpleHtmlParser p = new SimpleHtmlParser(html);
for (Element e; (e = p.Next()) != null;)
{
if (e is Tag && ((Tag)e).NameEquals("div"))
{
if (e is BeginTag)
++tags;
else
--tags;
}
}
return tags != 0;
}
public HtmlExtractor(Stream data, Encoding encoding)
{
using (StreamReader reader = new StreamReader(data, encoding))
html = reader.ReadToEnd();
this.parser = new SimpleHtmlParser(html);
}
/// <summary>
/// Is the tag a meaningless tag such as <p></p> or <a href="..."></a> or <a href="..."> </a>
/// </summary>
/// <param name="htmlParser"></param>
/// <param name="bt"></param>
/// <returns></returns>
private static bool RemoveMeaninglessTags(SimpleHtmlParser htmlParser, BeginTag bt)
{
// Look to see if the tag is a <p> without any attributes
if ((bt.NameEquals("p") && bt.Attributes.Length == 0 && !bt.HasResidue))
{
Element e = htmlParser.Peek(0);
// Look to see if thereis a matching end tag to the element we are looking at
if (e != null && e is EndTag && ((EndTag)e).NameEquals("p"))
{
// eat up the end tag
htmlParser.Next();
return true;
}
}
// Look to see if the tag is an <a> without a style/id/name attribute, but has an href... meaning the link is not useful
if ((bt.NameEquals("a") && bt.GetAttribute("name") == null && bt.GetAttributeValue("style") == null && bt.GetAttributeValue("id") == null && bt.GetAttributeValue("href") != null))
{
bool hadWhiteSpaceText = false;
Element e = htmlParser.Peek(0);
// Look to see if the a just has whitespace inside of it
if (e is Text && HtmlUtils.UnEscapeEntities(e.RawText, HtmlUtils.UnEscapeMode.NonMarkupText).Trim().Length == 0)
{
e = htmlParser.Peek(1);
hadWhiteSpaceText = true;
}
// Look to see if thereis a matching end tag to the element we are looking at
if (e != null && e is EndTag && ((EndTag)e).NameEquals("a"))
{
// if this was an <a> with whitespace in the middle eat it up
if (hadWhiteSpaceText)
htmlParser.Next();
// eat up the end tag
htmlParser.Next();
return true;
}
}
return false;
}
/// <summary>
/// Reposition the extractor back to the beginning of the
/// HTML.
/// </summary>
/// <returns>Returns this. This allows chaining together of calls,
/// like this:
///
/// if (ex.Seek(...).Success || ex.Reset().Seek(...).Success) { ... }
/// </returns>
public HtmlExtractor Reset()
{
lastMatch = null;
parser = new SimpleHtmlParser(html);
return(this);
}
public void Parse()
{
SimpleHtmlParser parser = new SimpleHtmlParser(_html);
OnDocumentBegin();
while (true)
{
Element currentElement = parser.Next();
BeginTag beginTag = currentElement as BeginTag;
if (beginTag != null)
{
OnBeginTag(beginTag);
continue;
}
EndTag endTag = currentElement as EndTag;
if (endTag != null)
{
OnEndTag(endTag);
continue;
}
ScriptLiteral literal = currentElement as ScriptLiteral;
if (literal != null)
{
OnScriptLiteral(literal);
continue;
}
Comment comment = currentElement as Comment;
if (comment != null)
{
OnComment(comment);
continue;
}
MarkupDirective markupDirective = currentElement as MarkupDirective;
if (markupDirective != null)
{
OnMarkupDirective(markupDirective);
continue;
}
ScriptText scriptText = currentElement as ScriptText;
if (scriptText != null)
{
OnScriptText(scriptText);
continue;
}
ScriptComment scriptComment = currentElement as ScriptComment;
if (scriptComment != null)
{
OnScriptComment(scriptComment);
continue;
}
StyleText styleText = currentElement as StyleText;
if (styleText != null)
{
OnStyleText(styleText);
continue;
}
StyleUrl styleUrl = currentElement as StyleUrl;
if (styleUrl != null)
{
OnStyleUrl(styleUrl);
continue;
}
StyleImport styleImport = currentElement as StyleImport;
if (styleImport != null)
{
OnStyleImport(styleImport);
continue;
}
StyleComment styleComment = currentElement as StyleComment;
if (styleComment != null)
{
OnStyleComment(styleComment);
continue;
}
StyleLiteral styleLiteral = currentElement as StyleLiteral;
if (styleLiteral != null)
{
OnStyleLiteral(styleLiteral);
continue;
}
Text text = currentElement as Text;
if (text != null)
{
OnText(text);
continue;
}
if (currentElement == null)
{
OnDocumentEnd();
return;
}
Debug.Fail("Unrecognized element in LightWeightHTMLDocumentIterator");
}
}
