// ------------------------------------------------------------------ #endregion #region Private methods. // ------------------------------------------------------------------ /// <summary> /// Checks the type of the verify link. /// </summary> /// <param name="absoluteUri">The absolute URI.</param> /// <param name="linkType">Type of the link.</param> /// <returns></returns> private UriType CheckVerifyLinkType( Uri absoluteUri, UriType linkType) { if (linkType == UriType.Resource) { // The original. return(linkType); } else { if (DoIsProcessableUri(absoluteUri, linkType)) { // Ensure PDFs don't get parsed. string head = ResourceDownloader.DownloadHead( absoluteUri, _options); if (string.IsNullOrEmpty(head)) { return(UriType.Resource); } else { head = head.ToLowerInvariant(); if (head.Contains(@"pdf") || head.Contains(@"application") || head.Contains(@"image")) { return(UriType.Resource); } else { Debug.Assert( head.Contains(@"text"), @"no text document type but marked as content."); // The original. return(linkType); } } } else { return(UriType.Resource); } } }
// ------------------------------------------------------------------ #endregion #region Private methods. // ------------------------------------------------------------------ /// <summary> /// Process one single URI with a document behind (i.e. no /// resource URI). /// </summary> /// <param name="uriInfo">The URI info.</param> /// <param name="depth">The depth.</param> private void ProcessUrl( DownloadedResourceInformation uriInfo, int depth) { Trace.WriteLine( string.Format( @"Processing URI '{0}', with depth {1}.", uriInfo.AbsoluteUri.AbsoluteUri, depth)); string ext = DownloadedResourceInformation.CorrectFileExtension(DownloadedResourceInformation.TryExtractFileExtension(uriInfo.AbsoluteUri)); if (ext == ".html" && _settings.Options.MaximumLinkDepth >= 0 && depth > _settings.Options.MaximumLinkDepth) { Trace.WriteLine( string.Format( @"Depth {1} exceeds maximum configured depth. Ending recursion " + @"at URI '{0}'.", uriInfo.AbsoluteUri.AbsoluteUri, depth)); } else if (depth > _maxDepth) { Trace.WriteLine( string.Format( @"Depth {1} exceeds maximum allowed recursion depth. " + @"Ending recursion at URI '{0}' to possible continue later.", uriInfo.AbsoluteUri.AbsoluteUri, depth)); // Add myself to start there later. // But only if not yet process, otherwise we would never finish. if (_settings.HasDownloadedUri(uriInfo)) { Trace.WriteLine( string.Format( @"URI '{0}' was already downloaded. NOT continuing later.", uriInfo.AbsoluteUri.AbsoluteUri)); } else { _settings.AddDownloadedResourceInfo(uriInfo); // Finished the function. Trace.WriteLine( string.Format( @"Added URI '{0}' to continue later.", uriInfo.AbsoluteUri.AbsoluteUri)); } } else { // If we are in asynchron mode, periodically check for stopps. if (processAsyncBackgroundWorker != null) { if (processAsyncBackgroundWorker.CancellationPending) { throw new StopProcessingException(); } } // -- // Notify event sinks about this URL. if (ProcessingUrl != null) { ProcessingUrlEventArgs e = new ProcessingUrlEventArgs( uriInfo, depth); ProcessingUrl(this, e); } // -- if (uriInfo.IsProcessableUri) { if (_settings.HasDownloadedUri(uriInfo)) { Trace.WriteLine( string.Format( @"URI '{0}' was already downloaded. Skipping.", uriInfo.AbsoluteUri.AbsoluteUri)); } else { Trace.WriteLine( string.Format( @"URI '{0}' was not already downloaded. Processing.", uriInfo.AbsoluteUri.AbsoluteUri)); if (uriInfo.LinkType == UriType.Resource) { Trace.WriteLine( string.Format( @"Processing resource URI '{0}', with depth {1}.", uriInfo.AbsoluteUri.AbsoluteUri, depth)); byte[] binaryContent; ResourceDownloader.DownloadBinary( uriInfo.AbsoluteUri, out binaryContent, _settings.Options); ResourceStorer storer = new ResourceStorer(_settings); storer.StoreBinary( binaryContent, uriInfo); _settings.AddDownloadedResourceInfo(uriInfo); _settings.PersistDownloadedResourceInfo(uriInfo); } else { Trace.WriteLine( string.Format( @"Processing content URI '{0}', with depth {1}.", uriInfo.AbsoluteUri.AbsoluteUri, depth)); string textContent; string encodingName; Encoding encoding; byte[] binaryContent; ResourceDownloader.DownloadHtml( uriInfo.AbsoluteUri, out textContent, out encodingName, out encoding, out binaryContent, _settings.Options); ResourceParser parser = new ResourceParser( _settings, uriInfo, textContent); List <UriResourceInformation> linkInfos = parser.ExtractLinks(); ResourceRewriter rewriter = new ResourceRewriter(_settings); textContent = rewriter.ReplaceLinks( textContent, uriInfo); ResourceStorer storer = new ResourceStorer(_settings); storer.StoreHtml( textContent, encoding, uriInfo); // Add before parsing childs. _settings.AddDownloadedResourceInfo(uriInfo); foreach (UriResourceInformation linkInfo in linkInfos) { DownloadedResourceInformation dlInfo = new DownloadedResourceInformation( linkInfo, uriInfo.LocalFolderPath, uriInfo.LocalBaseFolderPath); // Recurse. ProcessUrl(dlInfo, depth + 1); // Do not return or break immediately if too deep, // because this would omit certain pages at this // recursion level. } // Persist after completely parsed childs. _settings.PersistDownloadedResourceInfo(uriInfo); } Trace.WriteLine( string.Format( @"Finished processing URI '{0}'.", uriInfo.AbsoluteUri.AbsoluteUri)); } } else { Trace.WriteLine( string.Format( @"URI '{0}' is not processable. Skipping.", uriInfo.AbsoluteUri.AbsoluteUri)); } } }