public void ParseBrokenHtmlNoStartTag()
{
var html = "Some text with broken html</p>";
StringBuilder builder = new StringBuilder();
HtmlTagTable tagTable = new HtmlTagTable(html);
Build(builder, html, tagTable);
Assert.Equal(html, builder.ToString());
}
public void ParseBrokenHtmlEndTagOnly()
{
var html = "</body>";
StringBuilder builder = new StringBuilder();
HtmlTagTable tagTable = new HtmlTagTable(html);
Build(builder, html, tagTable);
Assert.Equal(html, builder.ToString());
}
public void ParsedHtmlSameAsOriginal(string file)
{
var html = GetHtml(file);
StringBuilder builder = new StringBuilder();
HtmlTagTable tagTable = new HtmlTagTable(html);
Build(builder, html, tagTable);
output.WriteLine(html);
output.WriteLine(builder.ToString());
Assert.Equal(html, builder.ToString());
}
private static void Build(StringBuilder builder, string html, HtmlTagTable tagTable)
{
if (string.IsNullOrEmpty(html))
{
return;
}
var parser = new HtmlHelper(html, tagTable);
if (parser.ParseErrors.Count() > 0)
{
// Fall back on regex parsing
foreach (var chunk in Regex.Split(html, "(<.+?>)"))
{
if (Regex.IsMatch(chunk, "<.+?>"))
{
builder.Append(chunk);
}
else
{
builder.Append(chunk);
}
}
return;
}
foreach (var node in parser.Descendants())
{
if (node.NodeType == HtmlNodeType.Element)
{
if (!tagTable.Table[node.Name].HasEndTag)
{
builder.Append(parser.GetRawStartTag(node));
}
else if (tagTable.Table[node.Name].IsEndGhostTag)
{
builder.Append(parser.GetRawEndTag(node));
}
else
{
builder.Append(parser.GetRawStartTag(node));
Build(builder, node.InnerHtml, tagTable);
if (node.Closed)
{
builder.Append(parser.GetRawEndTag(node));
}
node.RemoveAllChildren();
}
}
else if (node.NodeType == HtmlNodeType.Text)
{
builder.Append(node.InnerText);
}
else
{
builder.Append(node.InnerHtml);
}
}
}
private List <IAbstractMarkupData> CreateMarkupData(string input, HtmlTagTable tagTable, HtmlEntitizer entitizer, ConversionItem conversionItem)
{
var markupList = new List <IAbstractMarkupData>();
var parser = new HtmlHelper(entitizer.Entitize(input, conversionItem.Search.Text), tagTable);
if (parser.ParseErrors.Any())
{
return(ParseTagsFallback(input, markupList));
}
foreach (var node in parser.Descendants())
{
if (node.NodeType == HtmlNodeType.Element)
{
if (!tagTable.Table[node.OriginalName].HasEndTag)
{
var ph = parser.GetRawStartTag(node);
var phTag = CreatePlaceHolderTag(ph);
markupList.Add(phTag);
}
else if (tagTable.Table[node.OriginalName].IsEndGhostTag)
{
var eTag = parser.GetRawEndTag(node);
markupList.Add(CreatePlaceHolderTag(eTag));
}
else
{
var stTag = parser.GetRawStartTag(node);
if (node.Closed)
{
var eTag = parser.GetRawEndTag(node);
var startTag = CreateStartTag(stTag);
var endTag = CreateEndTag(eTag);
var tagPair = CreateTagPair(startTag, endTag);
if (!ContainsTags(node.InnerHtml))
{
if (!string.IsNullOrEmpty(node.InnerHtml))
{
var itext = CreateIText(entitizer.DeEntitize(node.InnerHtml));
tagPair.Add(itext);
}
// Experimental:
// Creation of new formatting
CreateNewFormatting(node, tagPair);
}
else
{
var list = CreateMarkupData(node.InnerHtml, tagTable, entitizer, conversionItem);
foreach (var item in list)
{
tagPair.Add(item);
}
}
markupList.Add(tagPair);
}
else
{
var phTag = CreatePlaceHolderTag(stTag);
markupList.Add(phTag);
var list = CreateMarkupData(node.InnerHtml, tagTable, entitizer, conversionItem);
foreach (var item in list)
{
markupList.Add(item);
}
}
node.RemoveAllChildren();
}
}
else if (node.NodeType == HtmlNodeType.Text)
{
//check for data like {M:>0},{M:<0},{M:≥0} which is treated like a text but tghe intention is to convert them in placeholder
if (Regex.IsMatch(input, "(?:>|≤|<|≥)\\d+"))
{
var originalTextSplited =
entitizer.GetOriginalTextSplited(entitizer.DeEntitize(node.InnerText), conversionItem.Search.Text);
ReplaceSpecialPlaceholders(originalTextSplited, markupList);
}
else
{
markupList.Add(CreateIText(entitizer.DeEntitize(node.InnerText)));
}
}
else
{
markupList.Add(CreateIText(entitizer.DeEntitize(node.InnerHtml)));
}
}
return(markupList);
}