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)));
}
private void EmitTagAndAttributes(string tagName, Tag tag)
{
BeginTag beginTag = tag as BeginTag;
if (beginTag != null)
{
Emit(string.Format(CultureInfo.InvariantCulture, "<{0}", tagName));
foreach (Attr attr in beginTag.Attributes)
{
if (PreserveAttributes.Contains(attr.Name.ToUpper(CultureInfo.InvariantCulture)))
{
Emit(" " + attr.ToString());
}
}
if (beginTag.Complete)
{
Emit("/");
}
Emit(">");
}
else
{
if (tagName.ToUpper(CultureInfo.InvariantCulture) != HTMLTokens.Br && tagName.ToUpper(CultureInfo.InvariantCulture) != HTMLTokens.Img)
{
Emit(string.Format(CultureInfo.InvariantCulture, "</{0}>", tagName));
}
}
}
private void HandleTextarea(HtmlForm parentForm, BeginTag textareaTag)
{
string name = textareaTag.GetAttributeValue("name");
string value = parser.CollectTextUntil("textarea");
new Textarea(parentForm, name, value);
}
private static bool FindAndRemoveEmptyTag(int startIndex, Element[] els)
{
for (int j = startIndex; j < els.Length - 1; j++)
{
BeginTag begin = els[j] as BeginTag;
if (begin != null)
{
for (int k = j + 1; k < els.Length; k++)
{
if (els[k] == null)
{
continue; // keep looking
}
if (!(els[k] is EndTag) || !((EndTag)els[k]).NameEquals(begin.Name))
{
break; // no match
}
// It's a match, delete start and end tags and restart
els[j] = null;
els[k] = null;
return(true);
}
}
}
return(false);
}
public bool IsMatch(Element e)
{
BeginTag tag = e as BeginTag;
if (tag == null)
{
return(false);
}
if (!tag.NameEquals("meta"))
{
return(false);
}
if (tag.GetAttributeValue("name") != "generator")
{
return(false);
}
string generator = tag.GetAttributeValue("content");
if (generator == null || CaseInsensitiveComparer.DefaultInvariant.Compare("blogger", generator) != 0)
{
return(false);
}
return(true);
}
Size FindSizeAttribute(string input)
{
Size size = new Size(_width, _height);
if (string.IsNullOrEmpty(input))
{
return(size);
}
try
{
RequiredAttribute[] attrWidth = new RequiredAttribute[] { new RequiredAttribute("width"), new RequiredAttribute("height") };
IElementPredicate predicate = new OrPredicate(new BeginTagPredicate("embed", attrWidth), new BeginTagPredicate("object", attrWidth));
HtmlExtractor ex = new HtmlExtractor(input);
if (ex.Seek(predicate).Success)
{
BeginTag tag = (BeginTag)ex.Element;
size = new Size(Convert.ToInt32(tag.GetAttributeValue("width"), CultureInfo.InvariantCulture), Convert.ToInt32(tag.GetAttributeValue("height"), CultureInfo.InvariantCulture));
}
}
catch (Exception ex)
{
Trace.Fail("Exception thrown while trying to find video size: " + ex);
}
return(size);
}
internal string Transform(BeginTag tag, string reference)
{
if (UrlHelper.IsUrl(reference))
{
Uri localReferenceUri = new Uri(reference);
/*
* If we need to drop a hint to the photo uploader about
* whether Lightbox-like preview is enabled, so that we know to link to
* the image itself rather than the photo "self" page on photos.live.com;
* this is where we would figure that out (by looking at the tag) and
* pass that info through to the DoUploadWork call.
*/
bool isLightboxCloneEnabled = false;
_referenceFixer._fileUploadWorker.DoUploadWork(reference, _uploader, isLightboxCloneEnabled);
ISupportingFile supportingFile = _fileService.GetFileByUri(localReferenceUri);
if (supportingFile != null)
{
Uri uploadUri = supportingFile.GetUploadInfo(_uploader.DestinationContext).UploadUri;
if (uploadUri != null)
{
return(UrlHelper.SafeToAbsoluteUri(uploadUri));
}
}
}
return(reference);
}
protected override void OnBeginTag(BeginTag tag)
{
if (FlagIsSet(Flag.RemovePartialTags) && tag.Unterminated)
{
return;
}
//remove all illegal attributes from the tag
foreach (Attr attr in tag.Attributes)
{
if (IsIllegalAttribute(attr))
{
attr.Value = string.Empty;
}
}
if (tag.NameEquals("script"))
{
Debug.WriteLine("Script tag");
}
if (IsRegexMatch(IllegalTagTreeName, tag.Name))
{
suspendTagDepth++;
}
else if (!IsIllegalTag(tag) && suspendTagDepth == 0)
{
PushStartTag(tag.Name);
base.OnBeginTag(tag);
}
}
public string RestorePreserved(string html)
{
StringBuilder sb = new StringBuilder();
HtmlExtractor ex = new HtmlExtractor(html);
int pos = 0;
while (ex.Seek("<span class='" + PRESERVE_CLASS + "'>").Success)
{
sb.Append(html, pos, ex.Element.Offset - pos);
pos = ex.Element.Offset;
BeginTag bt = (BeginTag)ex.Element;
string elementId = bt.GetAttributeValue("id");
Match m = Regex.Match(elementId ?? "", @"^preserve([a-zA-Z0-9]+)$");
if (m.Success)
{
string preserveId = m.Groups[1].Value;
string preservedValue;
if (preserved.TryGetValue(preserveId, out preservedValue))
{
sb.Append(preservedValue);
ex.CollectTextUntil("span");
if (ex.Element == null)
{
pos = html.Length;
}
else
{
pos = ex.Parser.Position;
}
}
}
}
sb.Append(html, pos, html.Length - pos);
return(sb.ToString());
}
protected override string Replace(Element el)
{
if (el is BeginTag)
{
BeginTag beginTag = (BeginTag)el;
if (beginTag.NameEquals("a"))
{
Attr href = beginTag.GetAttribute("href");
if (href != null && href.Value != null)
{
href.Value = ConvertUrl(href.Value);
return(beginTag.ToString());
}
}
else if (beginTag.NameEquals("img"))
{
Attr src = beginTag.GetAttribute("src");
if (src != null && src.Value != null)
{
src.Value = ConvertUrl(src.Value);
return(beginTag.ToString());
}
}
}
return(base.Replace(el));
}
public bool IsBlogger()
{
if (Regex.IsMatch(homepageUrl, @"^http://.+\.blogspot\.com($|/)", RegexOptions.IgnoreCase) ||
Regex.IsMatch(homepageUrl, @"^http(s)?://(www\.)?blogger\.com($|/)", RegexOptions.IgnoreCase) ||
new HtmlExtractor(html).Seek(new BloggerGeneratorCriterion()).Success)
{
return(true);
}
HtmlExtractor ex = new HtmlExtractor(html);
while (ex.Seek("<link href rel='service.post' type='application/atom+xml'>").Success)
{
BeginTag bt = (BeginTag)ex.Element;
string atomHref = bt.GetAttributeValue("href");
// these obsolete Blogger atom links can't be used, but are
// still a good indication that it's Blogger
if (atomHref.StartsWith("https://www.blogger.com/atom/", StringComparison.OrdinalIgnoreCase))
{
return(true);
}
// any other blogger or blogspot atom link will be considered a match
if (Regex.IsMatch(atomHref, @"^https?\:\/\/.+\.blog(ger|spot)\.com\/.*", RegexOptions.IgnoreCase))
{
return(true);
}
}
return(false);
}
protected override void OnBeginTag(BeginTag tag)
{
if (IsPreserveWhitespaceTag(tag.Name))
{
preserveLinebreaksDepth++;
}
base.OnBeginTag(tag);
}
private void ModifyMetaDataAttribute(BeginTag tag, string attributeName, string valueToChangeTo)
{
Attr valueAttr = tag.GetAttribute(attributeName);
if (valueAttr != null && valueToChangeTo != null)
{
valueAttr.Value = valueToChangeTo;
}
}
private void ModifyMetaDataAsNecessary(BeginTag tag)
{
if (_metaData == null)
{
return;
}
Attr nameAttr = tag.GetAttribute(HTMLTokens.Name);
if (nameAttr == null)
{
nameAttr = tag.GetAttribute(HTMLTokens.HttpEquiv);
}
if (nameAttr != null)
{
switch (nameAttr.Value.ToUpper(CultureInfo.InvariantCulture))
{
case (HTMLTokens.Author):
ModifyMetaDataAttribute(tag, HTMLTokens.Content, _metaData.Author);
break;
case (HTMLTokens.ContentType):
ModifyMetaDataAttribute(tag, HTMLTokens.Content, string.Format(CultureInfo.InvariantCulture, "text/html; {0}", _metaData.Charset));
break;
case (HTMLTokens.Charset):
ModifyMetaDataAttribute(tag, HTMLTokens.Content, _metaData.Charset);
break;
case (HTMLTokens.Description):
ModifyMetaDataAttribute(tag, HTMLTokens.Content, _metaData.Description);
break;
case (HTMLTokens.Generator):
ModifyMetaDataAttribute(tag, HTMLTokens.Content, _metaData.Generator);
break;
case (HTMLTokens.CopyRight):
ModifyMetaDataAttribute(tag, HTMLTokens.Content, _metaData.CopyRight);
break;
case (HTMLTokens.Keywords):
ModifyMetaDataAttribute(tag, HTMLTokens.Content, _metaData.KeywordString);
break;
case (HTMLTokens.Pragma):
ModifyMetaDataAttribute(tag, HTMLTokens.Content, _metaData.Pragma);
break;
case (HTMLTokens.Robots):
ModifyMetaDataAttribute(tag, HTMLTokens.Content, _metaData.Robots);
break;
}
_emittedMetaData.Add(nameAttr.Value.ToUpper(CultureInfo.InvariantCulture));
}
}
public static ImageViewer DetectImageViewer(string html, string sourceUrl)
{
List <ImageViewer> viewers = imageViewers;
LazyLoader <List <Regex> > regexes = new LazyLoader <List <Regex> >(delegate
{
List <Regex> regexList = new List <Regex>(viewers.Count);
foreach (ImageViewer v in viewers)
{
regexList.Add(new Regex(v.Pattern, RegexOptions.CultureInvariant));
}
return(regexList);
});
HtmlExtractor ex = new HtmlExtractor(html);
while (ex.Seek("<script src>").Success)
{
BeginTag tag = (BeginTag)ex.Element;
string src = tag.GetAttributeValue("src");
if (String.IsNullOrEmpty(src))
{
continue;
}
try
{
if (!UrlHelper.IsUrl(src))
{
// We need absolute URLs.
src = UrlHelper.EscapeRelativeURL(sourceUrl, src);
}
Uri srcUri = new Uri(src);
if (srcUri.IsAbsoluteUri)
{
// WinLive 248276: We want just the path portion since there could be an additional query or
// fragment on the URL that our regexs can't handle.
src = srcUri.GetLeftPart(UriPartial.Path);
}
}
catch (UriFormatException)
{
// We'll just use the regex on the raw attribute value.
}
List <Regex> regexList = regexes.Value;
for (int i = 0; i < regexList.Count; i++)
{
if (regexList[i].IsMatch(src))
{
return(viewers[i]);
}
}
}
return(null);
}
protected override void OnBeginTag(BeginTag tag)
{
if (tag.NameEquals(HTMLTokens.Body))
{
bodyBeginTag = tag;
}
base.OnBeginTag(tag);
}
private bool TagPermittedAboveBody(BeginTag tag)
{
foreach (string permittedAboveBody in _permittedBeforeBody)
{
if (tag.NameEquals(permittedAboveBody))
{
return(true);
}
}
return(false);
}
private string AddHtmlReference(BeginTag tag, string reference)
{
if (UrlHelper.IsUrl(reference))
{
ISupportingFile supportingFile = _editingContext.SupportingFileService.GetFileByUri(new Uri(reference));
if (supportingFile != null && supportingFile.Embedded)
{
AddReference(supportingFile);
}
}
return(reference);
}
private void HandleInput(HtmlForm parentForm, BeginTag inputTag)
{
string type = inputTag.GetAttributeValue("type");
if (type != null)
{
type = type.Trim().ToLowerInvariant();
}
string name = inputTag.GetAttributeValue("name");
string value = inputTag.GetAttributeValue("value");
switch (type)
{
case "password":
new Textbox(parentForm, name, value);
break;
case "checkbox":
{
int dummy;
bool isChecked = inputTag.GetAttribute("checked", true, 0, out dummy) != null;
new Checkbox(parentForm, name, value, isChecked);
break;
}
case "radio":
{
int dummy;
bool isChecked = inputTag.GetAttribute("checked", true, 0, out dummy) != null;
new Radio(parentForm, name, value, isChecked);
break;
}
case "submit":
new SubmitButton(parentForm, name, value);
break;
case "image":
new ImageButton(parentForm, name, value, inputTag.GetAttributeValue("src"));
break;
case "hidden":
new Hidden(parentForm, name, value);
break;
case "text":
default:
new Textbox(parentForm, name, value);
break;
}
}
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());
}
private string EnumerateLocalFileReference(BeginTag tag, string reference)
{
// check if the local file is contained in the supporting file directory
string localFilePath = new Uri(reference).LocalPath;
if (this.ContainsFile(localFilePath))
{
_supportingFileScratchList.Add(localFilePath);
}
// don't transform the url
return(reference);
}
public static string Trim(string html, bool onlyTrimParagraphs)
{
Element[] els = Elements(html);
int pos;
// First, go backwards over the list, deleting
// all <br> and whitespace. Stop as soon as
// significant content is encountered.
if (onlyTrimParagraphs)
{
pos = 1 + FindCleanupIndexForParagraphTrim(els);
}
else
{
pos = 1 + FindLastVisibleElementAndRemoveWhitespace(els);
}
// pos now points to the index where whitespace cleanup should begin
// Remove empty pairs of invisible tags, e.g. <b></b>. Each time
// a pair is removed, start over, because the removal
// of an empty pair may create another empty pair, e.g. <p><i></i></p>
while (FindAndRemoveEmptyTag(pos, els))
{
}
// Remove extra unmatched <p> begin tags.
for (int i = pos; i < els.Length; i++)
{
BeginTag bt = els[i] as BeginTag;
if (bt != null && bt.NameEquals("p"))
{
els[i] = null;
}
}
// Concatenate all the elements that are left.
StringBuilder output = new StringBuilder(html.Length);
foreach (Element el in els)
{
if (el != null)
{
output.Append(el.RawText);
}
}
return(output.ToString());
}
/// <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());
}
protected override void OnBeginTag(BeginTag tag)
{
if (tag != null && LightWeightHTMLDocument.AllUrlElements.ContainsKey(tag.Name.ToUpper(CultureInfo.InvariantCulture)))
{
Attr attr = tag.GetAttribute((string)LightWeightHTMLDocument.AllUrlElements[tag.Name.ToUpper(CultureInfo.InvariantCulture)]);
if (attr != null)
{
string url = attr.Value;
if (!UrlHelper.IsUrl(url))
{
attr.Value = UrlHelper.EscapeRelativeURL(_baseUrl, url);
}
}
}
base.OnBeginTag(tag);
}
protected override void OnBeginTag(BeginTag tag)
{
if (tag.NameEquals(HTMLTokens.Title) && !tag.Complete)
{
_inTitle = true;
}
if (TagsToPreserve.Contains(tag.Name.ToUpper(CultureInfo.InvariantCulture)))
{
EmitTagAndAttributes(tag.Name, tag);
}
else if (ReplaceTags.ContainsKey(tag.Name.ToUpper(CultureInfo.InvariantCulture)))
{
EmitTagAndAttributes((string)ReplaceTags[tag.Name.ToUpper(CultureInfo.InvariantCulture)], tag);
}
}
private bool IsIllegalTag(BeginTag tag)
{
if (IsRegexMatch(IllegalTagName, tag.Name))
{
return(true);
}
else if (FlagIsSet(Flag.RemoveStyles) && tag.NameEquals("link"))
{
//if this link element is a stylesheet, it is illegal
Attr relAttr = tag.GetAttribute("rel");
if (relAttr != null && relAttr.Value != null && relAttr.Value.ToUpperInvariant().Trim() == "STYLESHEET")
{
return(true);
}
}
return(false);
}
/// <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);
}
protected override void OnBeginTag(BeginTag tag)
{
if (tag != null && LightWeightHTMLDocument.AllUrlElements.ContainsKey(tag.Name.ToUpper(CultureInfo.InvariantCulture)))
{
Attr attr = tag.GetAttribute((string)LightWeightHTMLDocument.AllUrlElements[tag.Name.ToUpper(CultureInfo.InvariantCulture)]);
if (attr != null)
{
string oldRef = attr.Value;
string newRef = _referenceFixer(tag, attr.Value);
attr.Value = newRef;
if (oldRef != newRef && _referenceFixed != null)
{
//notify the reference fixed callback that a reference was fixed.
_referenceFixed(oldRef, newRef);
}
}
}
base.OnBeginTag(tag);
}
public string FixReferences(BeginTag tag, string reference)
{
// protect against unexpected/empty input
if (!UrlHelper.IsUrl(reference))
{
return(reference);
}
Uri referenceUri = new Uri(reference);
if (referenceUri.IsFile)
{
return(_fixer(tag, reference));
}
else
{
return(reference);
}
}
protected override void OnBeginTag(BeginTag tag)
{
if (tag.NameEquals(HTMLTokens.Ul))
{
unorderedListLevel++;
}
else if (tag.NameEquals(HTMLTokens.Ol))
{
orderedListLevel++;
}
else if ((unorderedListLevel < 1) &&
(orderedListLevel < 1) &&
(tag.NameEquals(HTMLTokens.Li)))
{
hasIncompleteList = true;
}
base.OnBeginTag(tag);
}