public static string GetContentDisplayHtml(this string content, bool isSecureConnection)
{
content = HtmlUtil.AposToHtml(HtmlUtil.LineBreaksToHtml(HtmlUtil.StripHtmlComments(content)));
content = isSecureConnection
? ReplaceScheme(content, "http:", "https:")
: ReplaceScheme(content, "https:", "http:");
return(content);
}
public void Summarize(string summaryText, IList <string> bulletPointsText, string contentHtml,
bool includeContent, out string digestText, out string bodyText)
{
var digestBuilder = new StringBuilder(DigestMaxSize);
// Add summary.
digestBuilder.Append(summaryText);
// Add bullet points.
if (bulletPointsText != null)
{
foreach (var bulletPoint in bulletPointsText)
{
if (!string.IsNullOrEmpty(bulletPoint))
{
digestBuilder.AppendFormat(" \x2022\x00a0{0}", bulletPoint.Trim()); // • text
}
}
}
// Check whether we need to go through the content.
// TODO: consider using MS Search HTML IFilter instead
var contentText = HttpUtility.HtmlDecode(
HtmlUtil.StripHtmlTags(
HtmlUtil.AposToHtml(contentHtml)));
contentText = contentText.Replace('\x0095', '\x2022'); // replace Windows Western bullet with Unicode bullet
if (!includeContent && digestBuilder.Length > 0)
{
digestText = digestBuilder.ToString();
bodyText = contentText;
return;
}
// Add the content extract (up to DigestMaxSize in total).
Analyzer analyzer = new SimpleAnalyzer();
var contentStream = analyzer.tokenStream(Field, new java.io.StringReader(contentText));
var offset = (OffsetAttribute)contentStream.addAttribute(typeof(OffsetAttribute));
Fragmenter fragmenter = new SimpleFragmenter(DigestMaxSize - digestBuilder.Length);
fragmenter.start(contentText, contentStream);
var endOffset = 0;
while (contentStream.incrementToken())
{
endOffset = offset.endOffset();
if (fragmenter.isNewFragment())
{
break;
}
}
if (digestBuilder.Length > 0)
{
digestBuilder.Append(' ');
}
digestBuilder.Append(contentText, 0, endOffset);
digestBuilder.Append(" ...");
digestText = digestBuilder.ToString();
bodyText = contentText.Substring(endOffset);
}
