public LightWeightHTMLReplacer(string html, string url, LightWeightHTMLMetaData metaData, bool fragmentMode) : base(html)
{
_url = url;
_metaData = metaData;
if (fragmentMode)
{
_firstTag = false;
_seenHead = true;
_seenBody = true;
}
}
public LightWeightHTMLReplacer(string html, string url, LightWeightHTMLMetaData metaData, bool fragmentMode) : base(html)
{
_url = url;
_metaData = metaData;
if (fragmentMode)
{
_firstTag = false;
_seenHead = true;
_seenBody = true;
}
}
public static LightWeightHTMLDocument FromStream(Stream stream, string url, string name)
{
if (!stream.CanSeek)
{
string filePath = TempFileManager.Instance.CreateTempFile();
using (FileStream file = new FileStream(filePath, FileMode.Open))
StreamHelper.Transfer(stream, file);
return(LightWeightHTMLDocument.FromFile(filePath, url, name));
}
else
{
Encoding currentEncoding = Encoding.Default;
LightWeightHTMLDocument lwDoc = null;
using (StreamReader reader = new StreamReader(stream, currentEncoding))
{
lwDoc = LightWeightHTMLDocument.FromString(reader.ReadToEnd(), url, name, true);
// If there is no metadata that disagrees with our encoding, just return the DOM read with default decoding
LightWeightHTMLMetaData metaData = new LightWeightHTMLMetaData(lwDoc);
if (metaData != null && metaData.Charset != null)
{
try
{
// The decoding is different than the encoding used to read this document, reread it with correct encoding
Encoding encoding = Encoding.GetEncoding(metaData.Charset);
if (encoding != currentEncoding)
{
reader.DiscardBufferedData();
stream.Seek(0, SeekOrigin.Begin);
using (StreamReader reader2 = new StreamReader(stream, encoding))
{
lwDoc = LightWeightHTMLDocument.FromString(reader2.ReadToEnd(), url, name, true);
}
}
}
catch (NotSupportedException)
{
// The encoding isn't supported on this system
}
catch (ArgumentException)
{
// The encoding isn't an encoding that the OS even knows about (its probably
// not well formatted or misspelled or something)
}
}
}
return(lwDoc);
}
}
/// <summary>
/// Any setting that is derivaed from the homepage html needs to be in this function. This function is turned
/// on and off when detecting blog seetings through the IncludeHomePageSettings. None of these checks will be run
/// if the internet is not active. As each check is made, it does not need to be applied back the _content until the end
/// at which time it will write the settings back to the registry.
/// </summary>
private void DetectHomePageSettings()
{
if (_homepageAccessor.HtmlDocument == null) return;
IDictionary homepageSettings = new Hashtable();
Debug.Assert(!UseManifestCache, "This code will not run correctly under the manifest cache, due to option overrides not being set");
LightWeightHTMLMetaData metaData = new LightWeightHTMLMetaData(_homepageAccessor.HtmlDocument);
if (metaData.Charset != null)
{
try
{
homepageSettings.Add(BlogClientOptions.CHARACTER_SET, metaData.Charset);
}
catch (NotSupportedException)
{
//not an actual encoding
}
}
string docType = new LightWeightHTMLMetaData(_homepageAccessor.HtmlDocument).DocType;
if (docType != null)
{
bool xhtml = docType.IndexOf("xhtml", StringComparison.OrdinalIgnoreCase) >= 0;
if (xhtml)
{
homepageSettings.Add(BlogClientOptions.REQUIRES_XHTML, true.ToString(CultureInfo.InvariantCulture));
}
}
//checking whether blog is rtl
HtmlExtractor extractor = new HtmlExtractor(_homepageAccessor.HtmlDocument.RawHtml);
if (extractor.Seek(new OrPredicate(
new SmartPredicate("<html dir>"),
new SmartPredicate("<body dir>"))).Success)
{
BeginTag tag = (BeginTag)extractor.Element;
string dir = tag.GetAttributeValue("dir");
if (String.Compare(dir, "rtl", StringComparison.OrdinalIgnoreCase) == 0)
{
homepageSettings.Add(BlogClientOptions.TEMPLATE_IS_RTL, true.ToString(CultureInfo.InvariantCulture));
}
}
if (_homepageAccessor.HtmlDocument != null)
{
string html = _homepageAccessor.OriginalHtml;
ImageViewer viewer = DhtmlImageViewers.DetectImageViewer(html, _context.HomepageUrl);
if (viewer != null)
{
homepageSettings.Add(BlogClientOptions.DHTML_IMAGE_VIEWER, viewer.Name);
}
}
_context.HomePageOverrides = homepageSettings;
}
public static IHTMLDocument2 StreamToHTMLDoc(Stream stream, string baseUrl, bool escapePaths)
{
if (!stream.CanSeek)
{
MemoryStream mStream = new MemoryStream();
StreamHelper.Transfer(stream, mStream);
mStream.Seek(0, SeekOrigin.Begin);
stream = mStream;
}
string htmlContent = null;
Encoding currentEncoding = Encoding.Default;
LightWeightHTMLDocument lwDoc = null;
using (StreamReader reader = new StreamReader(stream, currentEncoding))
{
htmlContent = reader.ReadToEnd();
lwDoc = LightWeightHTMLDocument.FromString(htmlContent, baseUrl, true);
// If there is no metadata that disagrees with our encoding, just return the DOM read with default decoding
LightWeightHTMLMetaData metaData = new LightWeightHTMLMetaData(lwDoc);
if (metaData != null && metaData.Charset != null)
{
try
{
// The decoding is different than the encoding used to read this document, reread it with correct encoding
Encoding encoding = Encoding.GetEncoding(metaData.Charset);
if (encoding != currentEncoding)
{
reader.DiscardBufferedData();
stream.Seek(0, SeekOrigin.Begin);
using (StreamReader reader2 = new StreamReader(stream, encoding))
{
htmlContent = reader2.ReadToEnd();
}
}
}
catch (NotSupportedException)
{
// The encoding isn't supported on this system
}
catch (ArgumentException)
{
// The encoding isn't an encoding that the OS even knows about (its probably
// not well formatted or misspelled or something)
}
}
}
//now that the html content is in loaded in the right encoding, convert it into a document.
IHTMLDocument2 doc2 = StringToHTMLDoc(htmlContent, baseUrl, escapePaths, escapePaths);
return doc2;
}
public static LightWeightHTMLDocument FromStream(Stream stream, string url, string name)
{
if (!stream.CanSeek)
{
string filePath = TempFileManager.Instance.CreateTempFile();
using (FileStream file = new FileStream(filePath, FileMode.Open))
StreamHelper.Transfer(stream, file);
return LightWeightHTMLDocument.FromFile(filePath, url, name);
}
else
{
Encoding currentEncoding = Encoding.Default;
LightWeightHTMLDocument lwDoc = null;
using (StreamReader reader = new StreamReader(stream, currentEncoding))
{
lwDoc = LightWeightHTMLDocument.FromString(reader.ReadToEnd(), url, name, true);
// If there is no metadata that disagrees with our encoding, just return the DOM read with default decoding
LightWeightHTMLMetaData metaData = new LightWeightHTMLMetaData(lwDoc);
if (metaData != null && metaData.Charset != null)
{
try
{
// The decoding is different than the encoding used to read this document, reread it with correct encoding
Encoding encoding = Encoding.GetEncoding(metaData.Charset);
if (encoding != currentEncoding)
{
reader.DiscardBufferedData();
stream.Seek(0, SeekOrigin.Begin);
using (StreamReader reader2 = new StreamReader(stream, encoding))
{
lwDoc = LightWeightHTMLDocument.FromString(reader2.ReadToEnd(), url, name, true);
}
}
}
catch (NotSupportedException)
{
// The encoding isn't supported on this system
}
catch (ArgumentException)
{
// The encoding isn't an encoding that the OS even knows about (its probably
// not well formatted or misspelled or something)
}
}
}
return lwDoc;
}
}
public LightWeightHTMLReplacer(string html, string url, LightWeightHTMLMetaData metaData) : this(html, url, metaData, false)
{
}
public LightWeightHTMLReplacer(string html, string url, LightWeightHTMLMetaData metaData) : this(html, url, metaData, false)
{
}