public string Capture(int timeoutMs)
{
// flag indicating whether we should continue with the capture
bool continueCapture = true;
// request the page
HttpWebResponse response = RequestPage(TargetUrl, timeoutMs);
OnHeadersReceived(response.Headers, ref continueCapture);
if (!continueCapture)
{
throw new OperationCancelledException();
}
// transfer it to a stream
MemoryStream pageStream = new MemoryStream();
using (Stream responseStream = response.GetResponseStream())
StreamHelper.Transfer(responseStream, pageStream);
pageStream.Seek(0, SeekOrigin.Begin);
// allow filter on content
OnContentReceived(new StreamReader(pageStream).ReadToEnd(), ref continueCapture);
if (!continueCapture)
{
throw new OperationCancelledException();
}
pageStream.Seek(0, SeekOrigin.Begin);
// Read the stream into a lightweight HTML doc. We use from LightWeightHTMLDocument.FromIHTMLDocument2
// instead of LightWeightHTMLDocument.FromStream because from stream improperly shoves a saveFrom declaration
// above the docType (bug 289357)
IHTMLDocument2 doc = HTMLDocumentHelper.StreamToHTMLDoc(pageStream, TargetUrl, false);
LightWeightHTMLDocument ldoc = LightWeightHTMLDocument.FromIHTMLDocument2(doc, TargetUrl, true);
// download references
FileBasedSiteStorage siteStorage = new FileBasedSiteStorage(DestinationPath, "index.htm");
PageToDownload page = new PageToDownload(ldoc, TargetUrl, siteStorage.RootFile);
PageAndReferenceDownloader downloader = new PageAndReferenceDownloader(new PageToDownload[] { page }, siteStorage);
downloader.Download(new TimeoutProgressHost(timeoutMs));
// return path to captured page
return(Path.Combine(DestinationPath, siteStorage.RootFile));
}
/// <summary>
/// Used to actually commit the HTML to disk
/// </summary>
/// <param name="pageInfo">The PageToDownload to write</param>
/// <param name="downloadedReferences">A hashtable of download references</param>
/// <param name="storage">The storage to write the file into</param>
private void WriteHtmlToDisk(PageToDownload pageInfo, FileBasedSiteStorage storage)
{
// Set the character set for this document
pageInfo.LightWeightHTMLDocument.MetaData.Charset = Encoding.UTF8.WebName;
string html = string.Empty;
// Replace references to any URL that we downloaded!
foreach (PageToDownload pageToDownload in _pagesToDownload)
{
if (!pageToDownload.IsRootPage)
{
pageInfo.LightWeightHTMLDocument.AddUrlToEscape(new UrlToReplace(pageToDownload.UrlToReplace, pageInfo.GetRelativeUrlForReferencedPage(pageToDownload)));
}
}
foreach (ReferenceToDownload referenceToDownload in _referencesToDownload.Values)
{
ReferenceToDownload downloadedReference = ((ReferenceToDownload)_referencesToDownload[referenceToDownload.AbsoluteUrl]);
// Since we consolidated references, replace the UrToReplace from the original reference
// with the relativePath to the reference that actually got downloaded
string path = downloadedReference.GetRelativeUrlForReference(pageInfo);
pageInfo.LightWeightHTMLDocument.AddUrlToEscape(new UrlToReplace(referenceToDownload.AbsoluteUrl, path));
}
html = pageInfo.LightWeightHTMLDocument.GenerateHtml();
// finally, write the html out to disk
string destination = Path.Combine(_siteStorage.BasePath, pageInfo.RelativePath);
Stream htmlStream = _siteStorage.Open(destination, AccessMode.Write);
using (StreamWriter writer = new StreamWriter(htmlStream, Encoding.UTF8))
{
writer.Write(html);
}
// if this is the entry page, write the path token and root file name
if (pageInfo.IsRootPage)
{
this._pathToken = pageInfo.ReferencedFileRelativePath;
_siteStorage.RootFile = pageInfo.FileName;
}
}
public string Capture(int timeoutMs)
{
// flag indicating whether we should continue with the capture
bool continueCapture = true ;
// request the page
HttpWebResponse response = RequestPage(TargetUrl, timeoutMs);
OnHeadersReceived(response.Headers, ref continueCapture) ;
if ( !continueCapture )
throw new OperationCancelledException() ;
// transfer it to a stream
MemoryStream pageStream = new MemoryStream();
using ( Stream responseStream = response.GetResponseStream() )
StreamHelper.Transfer(responseStream, pageStream);
pageStream.Seek(0, SeekOrigin.Begin) ;
// allow filter on content
OnContentReceived( new StreamReader(pageStream).ReadToEnd(), ref continueCapture ) ;
if ( !continueCapture )
throw new OperationCancelledException() ;
pageStream.Seek(0, SeekOrigin.Begin) ;
// Read the stream into a lightweight HTML doc. We use from LightWeightHTMLDocument.FromIHTMLDocument2
// instead of LightWeightHTMLDocument.FromStream because from stream improperly shoves a saveFrom declaration
// above the docType (bug 289357)
IHTMLDocument2 doc = HTMLDocumentHelper.StreamToHTMLDoc(pageStream, TargetUrl, false);
LightWeightHTMLDocument ldoc = LightWeightHTMLDocument.FromIHTMLDocument2(doc, TargetUrl, true);
// download references
FileBasedSiteStorage siteStorage = new FileBasedSiteStorage(DestinationPath, "index.htm");
PageToDownload page = new PageToDownload(ldoc, TargetUrl, siteStorage.RootFile);
PageAndReferenceDownloader downloader = new PageAndReferenceDownloader(new PageToDownload[]{page}, siteStorage) ;
downloader.Download(new TimeoutProgressHost(timeoutMs)) ;
// return path to captured page
return Path.Combine(DestinationPath, siteStorage.RootFile) ;
}
/// <summary>
/// Constructs a new asynchrous page download.
/// </summary>
/// <param name="pagesToDownload">The array of pagesToDownload</param>
/// <param name="siteStorage">The File based site storage into which to place the page and references</param>
/// <param name="target">The ISynchronizeInvoke that will be used to call back</param>
/// <param name="throwOnFailure">Indicates whether downloader should throw on failure, or just
/// log the failure and continue</param>
public PageAndReferenceDownloader(PageToDownload[] pagesToDownload, FileBasedSiteStorage siteStorage, bool throwOnFailure)
{
_siteStorage = siteStorage;
_pagesToDownload = pagesToDownload;
_throwOnFailure = throwOnFailure;
}
/// <summary>
/// Constructs a new asynchrous page download.
/// </summary>
/// <param name="pagesToDownload">The array of pagesToDownload</param>
/// <param name="siteStorage">The File based site storage into which to place the page and references</param>
/// <param name="target">The ISynchronizeInvoke that will be used to call back</param>
public PageAndReferenceDownloader(PageToDownload[] pagesToDownload, FileBasedSiteStorage siteStorage)
: this(pagesToDownload, siteStorage, true)
{
}
public DownloadWorkItem(ReferenceToDownload reference, FileBasedSiteStorage siteStorage, IProgressHost progressHost)
{
this.reference = reference;
this.siteStorage = siteStorage;
this.progressHost = progressHost;
}
/// <summary>
/// Used to actually commit the HTML to disk
/// </summary>
/// <param name="pageInfo">The PageToDownload to write</param>
/// <param name="downloadedReferences">A hashtable of download references</param>
/// <param name="storage">The storage to write the file into</param>
private void WriteHtmlToDisk(PageToDownload pageInfo, FileBasedSiteStorage storage)
{
// Set the character set for this document
pageInfo.LightWeightHTMLDocument.MetaData.Charset = Encoding.UTF8.WebName;
string html = string.Empty;
// Replace references to any URL that we downloaded!
foreach (PageToDownload pageToDownload in _pagesToDownload)
if (!pageToDownload.IsRootPage)
pageInfo.LightWeightHTMLDocument.AddUrlToEscape(new UrlToReplace(pageToDownload.UrlToReplace, pageInfo.GetRelativeUrlForReferencedPage(pageToDownload)));
foreach (ReferenceToDownload referenceToDownload in _referencesToDownload.Values)
{
ReferenceToDownload downloadedReference = ((ReferenceToDownload)_referencesToDownload[referenceToDownload.AbsoluteUrl]);
// Since we consolidated references, replace the UrToReplace from the original reference
// with the relativePath to the reference that actually got downloaded
string path = downloadedReference.GetRelativeUrlForReference(pageInfo);
pageInfo.LightWeightHTMLDocument.AddUrlToEscape(new UrlToReplace(referenceToDownload.AbsoluteUrl, path));
}
html = pageInfo.LightWeightHTMLDocument.GenerateHtml();
// finally, write the html out to disk
string destination = Path.Combine(_siteStorage.BasePath, pageInfo.RelativePath);
Stream htmlStream = _siteStorage.Open(destination, AccessMode.Write);
using (StreamWriter writer = new StreamWriter(htmlStream, Encoding.UTF8))
{
writer.Write(html);
}
// if this is the entry page, write the path token and root file name
if (pageInfo.IsRootPage)
{
this._pathToken = pageInfo.ReferencedFileRelativePath;
_siteStorage.RootFile = pageInfo.FileName;
}
}
/// <summary>
/// Download a reference, providing progress
/// </summary>
/// <param name="reference">The reference to download</param>
/// <param name="fileStorage">The storage to download the refernce into</param>
/// <param name="progressHost">The progressHost to provide feedback to</param>
private void DownloadReference(ReferenceToDownload reference, FileBasedSiteStorage fileStorage, IProgressHost progressHost)
{
UrlDownloadToFile downloader;
string fullPath;
downloader = new UrlDownloadToFile();
downloader.TimeoutMs = 30000;
if (progressHost.CancelRequested)
throw new OperationCancelledException();
// make sure that the directory exists
fullPath = Path.Combine(fileStorage.BasePath, reference.RelativePath);
string directory = Path.GetDirectoryName(fullPath);
if (!Directory.Exists(directory))
Directory.CreateDirectory(directory);
// Make sure there aren't any conflicts
lock (this)
{
string newFileName = Path.GetFileName(fullPath);
do
{
fullPath = PathHelper.GetNonConflictingPath(fullPath);
newFileName = Path.GetFileName(fullPath);
reference.SetFileName(ref newFileName);
} while (newFileName != Path.GetFileName(fullPath));
using (File.Open(fullPath, FileMode.Create, FileAccess.ReadWrite, FileShare.None))
{
}
}
string downloadUrl = reference.AbsoluteUrl;
if (UrlHelper.IsFileUrl(downloadUrl))
downloadUrl = HttpUtility.UrlDecode(downloadUrl);
downloader.Url = downloadUrl.Trim();
downloader.FilePath = fullPath;
downloader.ShowSecurityUI = true;
if (CredentialsContext != null)
downloader.CredentialsContext = CredentialsContext;
try
{
downloader.Download(progressHost);
}
catch (COMException e)
{
// If the file couldn't be downloaded, this doesn't matter. But log it
Trace.WriteLine("Didn't download file: " + downloader.Url + " " + e.ToString());
}
// Fix the filename of the downloaded entity to have the correct extension
string contentType = downloader.ContentType;
if (contentType != null && File.Exists(fullPath))
{
string suggestedExtension = MimeHelper.GetExtensionFromContentType(contentType);
if (suggestedExtension == null)
suggestedExtension = UrlHelper.GetExtensionForUrl(downloader.FinalUrl);
if (Path.GetExtension(fullPath) != suggestedExtension)
{
string newFilePath = Path.ChangeExtension(fullPath, suggestedExtension);
string newFileName = Path.GetFileName(newFilePath);
// Try to reset the name until we can both agree
while (true)
{
newFilePath = PathHelper.GetNonConflictingPath(newFilePath);
newFileName = Path.GetFileName(newFilePath);
FileHelper.Rename(fullPath, newFilePath);
reference.SetFileName(ref newFileName);
if (newFileName != Path.GetFileName(newFilePath))
{
try
{
File.Delete(newFilePath);
}
catch (Exception e) { Debug.Fail("Unable to delete failed temp file: " + e.ToString()); }
}
else
{
break;
}
}
fullPath = newFilePath;
}
}
// Handle anything special we need to do for stylesheet and js file references
if (Path.GetExtension(fullPath) == ".css" && File.Exists(fullPath))
{
string fileContents = string.Empty;
using (StreamReader reader = new StreamReader(fullPath))
fileContents = reader.ReadToEnd();
if (fileContents != string.Empty)
{
LightWeightCSSReplacer cssReplacer = new LightWeightCSSReplacer(fileContents);
// fix up references
foreach (ReferenceToDownload referenceInfo in _referencesToDownload.Values)
cssReplacer.AddUrlToReplace(new UrlToReplace(referenceInfo.UrlToReplace, referenceInfo.FileName));
string newCss = cssReplacer.DoReplace();
using (StreamWriter writer = new StreamWriter(fullPath, false))
writer.Write(newCss);
}
}
}
/// <summary>
/// Constructs a new asynchronous page download.
/// </summary>
/// <param name="pagesToDownload">The array of pagesToDownload</param>
/// <param name="siteStorage">The File based site storage into which to place the page and references</param>
/// <param name="target">The ISynchronizeInvoke that will be used to call back</param>
/// <param name="throwOnFailure">Indicates whether downloader should throw on failure, or just
/// log the failure and continue</param>
public PageAndReferenceDownloader(PageToDownload[] pagesToDownload, FileBasedSiteStorage siteStorage, bool throwOnFailure)
{
_siteStorage = siteStorage;
_pagesToDownload = pagesToDownload;
_throwOnFailure = throwOnFailure;
}
/// <summary>
/// Constructs a new asynchronous page download.
/// </summary>
/// <param name="pagesToDownload">The array of pagesToDownload</param>
/// <param name="siteStorage">The File based site storage into which to place the page and references</param>
/// <param name="target">The ISynchronizeInvoke that will be used to call back</param>
public PageAndReferenceDownloader(PageToDownload[] pagesToDownload, FileBasedSiteStorage siteStorage) : this(pagesToDownload, siteStorage, true)
{
}
/// <summary>
/// Download a reference, providing progress
/// </summary>
/// <param name="reference">The reference to download</param>
/// <param name="fileStorage">The storage to download the reference into</param>
/// <param name="progressHost">The progressHost to provide feedback to</param>
private void DownloadReference(ReferenceToDownload reference, FileBasedSiteStorage fileStorage, IProgressHost progressHost)
{
if (IsBase64EmbededImage(reference.AbsoluteUrl))
{
return;
}
UrlDownloadToFile downloader;
string fullPath;
downloader = new UrlDownloadToFile();
downloader.TimeoutMs = 30000;
if (progressHost.CancelRequested)
{
throw new OperationCancelledException();
}
// make sure that the directory exists
fullPath = Path.Combine(fileStorage.BasePath, reference.RelativePath);
string directory = Path.GetDirectoryName(fullPath);
if (!Directory.Exists(directory))
{
Directory.CreateDirectory(directory);
}
// Make sure there aren't any conflicts
lock (this)
{
string newFileName = Path.GetFileName(fullPath);
do
{
fullPath = PathHelper.GetNonConflictingPath(fullPath);
newFileName = Path.GetFileName(fullPath);
reference.SetFileName(ref newFileName);
} while (newFileName != Path.GetFileName(fullPath));
using (File.Open(fullPath, FileMode.Create, FileAccess.ReadWrite, FileShare.None))
{
}
}
string downloadUrl = reference.AbsoluteUrl;
if (UrlHelper.IsFileUrl(downloadUrl))
{
downloadUrl = HttpUtility.UrlDecode(downloadUrl);
}
downloader.Url = downloadUrl.Trim();
downloader.FilePath = fullPath;
downloader.ShowSecurityUI = true;
if (CredentialsContext != null)
{
downloader.CredentialsContext = CredentialsContext;
}
try
{
downloader.Download(progressHost);
}
catch (COMException e)
{
// If the file couldn't be downloaded, this doesn't matter. But log it
Trace.WriteLine("Didn't download file: " + downloader.Url + " " + e.ToString());
}
// Fix the filename of the downloaded entity to have the correct extension
string contentType = downloader.ContentType;
if (contentType != null && File.Exists(fullPath))
{
string suggestedExtension = MimeHelper.GetExtensionFromContentType(contentType);
if (suggestedExtension == null)
{
suggestedExtension = UrlHelper.GetExtensionForUrl(downloader.FinalUrl);
}
if (Path.GetExtension(fullPath) != suggestedExtension)
{
string newFilePath = Path.ChangeExtension(fullPath, suggestedExtension);
string newFileName = Path.GetFileName(newFilePath);
// Try to reset the name until we can both agree
while (true)
{
newFilePath = PathHelper.GetNonConflictingPath(newFilePath);
newFileName = Path.GetFileName(newFilePath);
FileHelper.Rename(fullPath, newFilePath);
reference.SetFileName(ref newFileName);
if (newFileName != Path.GetFileName(newFilePath))
{
try
{
File.Delete(newFilePath);
}
catch (Exception e) { Debug.Fail("Unable to delete failed temp file: " + e.ToString()); }
}
else
{
break;
}
}
fullPath = newFilePath;
}
}
// Handle anything special we need to do for stylesheet and js file references
if (Path.GetExtension(fullPath) == ".css" && File.Exists(fullPath))
{
string fileContents = string.Empty;
using (StreamReader reader = new StreamReader(fullPath))
fileContents = reader.ReadToEnd();
if (fileContents != string.Empty)
{
LightWeightCSSReplacer cssReplacer = new LightWeightCSSReplacer(fileContents);
// fix up references
foreach (ReferenceToDownload referenceInfo in _referencesToDownload.Values)
{
cssReplacer.AddUrlToReplace(new UrlToReplace(referenceInfo.UrlToReplace, referenceInfo.FileName));
}
string newCss = cssReplacer.DoReplace();
using (StreamWriter writer = new StreamWriter(fullPath, false))
writer.Write(newCss);
}
}
}
public DownloadWorkItem(ReferenceToDownload reference, FileBasedSiteStorage siteStorage, IProgressHost progressHost)
{
this.reference = reference;
this.siteStorage = siteStorage;
this.progressHost = progressHost;
}
/// <summary>
/// Fixes up the downloaded supporting files that were in the template.
/// </summary>
/// <param name="blogTemplateFile"></param>
/// <param name="storage"></param>
/// <param name="supportingFilesDir"></param>
protected internal virtual void FixupDownloadedFiles(string blogTemplateFile, FileBasedSiteStorage storage, string supportingFilesDir)
{
templateStrategy.FixupDownloadedFiles(blogTemplateFile, storage, supportingFilesDir);
}
private string DownloadTemplateFiles(string templateContents, string templateUrl, IProgressHost progress)
{
progress.UpdateProgress(Res.Get(StringId.ProgressDownloadingSupportingFiles));
FileBasedSiteStorage files = new FileBasedSiteStorage(_blogTemplateDir);
// convert the string to a stream
MemoryStream templateStream = new MemoryStream();
StreamWriter writer = new StreamWriter(templateStream, Encoding.UTF8);
writer.Write(templateContents);
writer.Flush();
templateStream.Seek(0, SeekOrigin.Begin);
//read the stream into a lightweight HTML. Note that we use from LightWeightHTMLDocument.FromIHTMLDocument2
//instead of LightWeightHTMLDocument.FromStream because from stream improperly shoves a saveFrom declaration
//above the docType (bug 289357)
IHTMLDocument2 doc = HTMLDocumentHelper.StreamToHTMLDoc(templateStream, templateUrl, true);
LightWeightHTMLDocument ldoc = LightWeightHTMLDocument.FromIHTMLDocument2(doc, templateUrl, true, false);
PageDownloadContext downloadContext = new PageDownloadContext(0);
ApplyCredentials(downloadContext, templateUrl);
using (PageToDownloadFactory downloadFactory = new PageToDownloadFactory(ldoc, downloadContext, _parentControl))
{
//calculate the dependent styles and resources
ProgressTick tick = new ProgressTick(progress, 50, 100);
downloadFactory.CreatePagesToDownload(tick);
tick.UpdateProgress(100, 100);
//download the dependent styles and resources
tick = new ProgressTick(progress, 50, 100);
PageAndReferenceDownloader downloader = new PageAndReferenceDownloader(downloadFactory.PagesToDownload, files);
this.ApplyCredentials(downloader, templateUrl);
downloader.Download(tick);
tick.UpdateProgress(100, 100);
//Expand out the relative paths in the downloaded HTML file with absolute paths.
//Note: this is necessary so that template resources are not improperly resolved relative
// to the location of the file the editor is editing.
string blogTemplateFile = Path.Combine(_blogTemplateDir, files.RootFile);
string origFile = blogTemplateFile + ".token";
File.Move(blogTemplateFile, origFile);
string absPath = String.Format(CultureInfo.InvariantCulture, "file:///{0}/{1}", _blogTemplateDir.Replace('\\', '/'), downloader.PathToken);
TextHelper.ReplaceInFile(origFile, downloader.PathToken, blogTemplateFile, absPath);
File.Delete(origFile);
//fix up the files
FixupDownloadedFiles(blogTemplateFile, files, downloader.PathToken);
//complete the progress.
progress.UpdateProgress(100, 100);
File.WriteAllText(blogTemplateFile + ".path", absPath);
return blogTemplateFile;
}
}
/// <summary>
/// Fixes up the CSS styles of in the template to make them editor safe.
/// </summary>
/// <param name="blogTemplateFile"></param>
/// <param name="storage"></param>
/// <param name="supportingFilesDir"></param>
protected internal override void FixupDownloadedFiles(string blogTemplateFile, FileBasedSiteStorage storage, string supportingFilesDir)
{
//Fix up the downloaded files to remove position:relative CSS settings (fixes bugs 244951, 287589, 287556)
string supportingFilesPath = Path.Combine(storage.BasePath, supportingFilesDir);
if (Directory.Exists(supportingFilesPath))
{
string[] files = Directory.GetFiles(supportingFilesPath);
foreach (string file in files)
{
if (file.EndsWith(".css", StringComparison.OrdinalIgnoreCase))
{
string cssPath = Path.Combine(supportingFilesPath, file);
string newCssContent = "";
using (StreamReader sr = new StreamReader(cssPath))
{
string cssContent = sr.ReadToEnd();
newCssContent = new LightWeightCSSReplacer(cssContent).DoReplace();
}
using (StreamWriter sw = new StreamWriter(cssPath, false, Encoding.UTF8))
{
sw.Write(newCssContent);
}
}
}
}
//fix up the styles in the HTML document
string templateFilePath = Path.Combine(supportingFilesPath, blogTemplateFile);
string newTemplateContent = "";
using (StreamReader sr = new StreamReader(templateFilePath))
{
string docContent = sr.ReadToEnd();
newTemplateContent = new LightWeightDocumentCssReplacer(docContent).DoReplace();
}
using (StreamWriter sw = new StreamWriter(templateFilePath, false, Encoding.UTF8))
{
sw.Write(newTemplateContent);
}
}
