/// <summary>
/// Handles rendering of Tip/Warning/Important/Caution/Note
/// blocks.
///
/// </summary>
/// <returns></returns>
/// <remarks>
/// Note: Markdown is expected to be in LineFeed only mode for line breaks (StringUtils.NormalizeLinefeeds()).
/// If you have CR/LF the value needs to be fixed up. RenderExtensions automatically fix up inbound Markdown
/// to normalized linefeeds for rendering, but if you test locally make sure to pre-format the args.Markdown
/// </remarks>
public void ParseNoteTipWarningImportant(ModifyMarkdownArguments args)
{
var matches = TipNoteWarningImportantFileRegEx.Matches(args.Markdown);
foreach (Match match in matches)
{
string value = match.Value.Trim();
var lines = StringUtils.GetLines(value);
var sb = new StringBuilder();
for (int i = 0; i < lines.Length; i++)
{
string line = lines[i];
if (i == 0)
{
// note header
if (line.TrimStart().StartsWith("> [!"))
{
string icon = "fa-info-circle";
var word = StringUtils.ExtractString(line, "> [!", "]");
switch (word)
{
case "NOTE":
case "TIP":
icon = "fa-info-circle";
break;
case "WARNING":
case "CAUTION":
case "IMPORTANT":
icon = "fa-warning";
break;
}
icon = $"<i class='fa {icon}'></i> ";
sb.AppendLine("<div class=\"" + word + "\">");
sb.AppendLine($"<h5>{icon}{word}</h5>");
sb.AppendLine();
}
else
{
// content line - could be empty
sb.AppendLine(line.TrimStart(' ', '>'));
}
}
else
{
// content line - could be empty
sb.AppendLine(line.TrimStart(' ', '>'));
}
}
sb.AppendLine();
sb.AppendLine("</div>");
sb.AppendLine();
args.Markdown = args.Markdown.Replace(value, sb.ToString());
}
}
/// <summary>
/// Parses DocFx include files in the format of:
///
/// [!include[title](relativePathToFileToInclude>)]
///
/// Should run **prior** to Markdown parsing of the main document
/// as it will embed the file content as is.
/// </summary>
/// <param name="markdown"></param>
/// <returns></returns>
protected void ParseDocFxIncludeFiles(ModifyMarkdownArguments args)
{
var matches = includeFileRegEx.Matches(args.Markdown);
foreach (Match match in matches)
{
string value = match.Value;
//string title = StringUtils.ExtractString(value, "[!include[", "]");
string file = StringUtils.ExtractString(value, "](", ")]");
if (string.IsNullOrEmpty(file))
{
continue;
}
if (file.StartsWith("~"))
{
file = mmApp.Model.Window.FolderBrowser.FolderPath + file.Substring(1);
}
string filePath;
if (file.Contains(@":\"))
{
filePath = file;
}
else
{
filePath = mmApp.Model.ActiveDocument?.Filename;
if (string.IsNullOrEmpty(filePath))
{
continue;
}
filePath = Path.GetDirectoryName(filePath);
if (filePath == null)
{
continue;
}
}
string includeFile = Path.Combine(filePath, file);
includeFile = FileUtils.NormalizePath(includeFile);
if (!File.Exists(includeFile))
{
continue;
}
var markdownDocument = new MarkdownDocument();
markdownDocument.Load(includeFile);
string includeContent = markdownDocument.RenderHtml();
args.Markdown = args.Markdown.Replace(value, includeContent);
}
}
public void BeforeMarkdownRendered(ModifyMarkdownArguments args)
{
if (mmApp.Configuration.MarkdownOptions.MarkdownParserName.Contains("DocFx"))
{
return;
}
ParseDocFxIncludeFiles(args);
ParseNoteTipWarningImportant(args);
ParseXrefTags(args);
}
/// <summary>
/// Process all BeforeRenderMarkdown Extensions
/// </summary>
/// <param name="markdown"></param>
/// <param name="document"></param>
public void ProcessAllBeforeMarkdownRenderedHooks(ModifyMarkdownArguments args)
{
foreach (var extension in RenderExtensions)
{
try
{
extension.BeforeMarkdownRendered(args);
}
catch (Exception ex)
{
mmApp.Log($"BeforeMarkdownRendered RenderExtension failed: {extension.GetType().Name}", ex);
}
}
}
/// <summary>
/// Handle Right To Left ACE Editor markdown removal so rendering looks correct
/// </summary>
/// <param name="args"></param>
public void BeforeMarkdownRendered(ModifyMarkdownArguments args)
{
// Right to Left Rendering fix up from ACE Editor
if (mmApp.Configuration.Editor.EnableRightToLeft)
{
// HACK: Strip ACE embed RTL/LTR Transition character
var bytes = args.MarkdownDocument.Encoding.GetBytes(args.Markdown);
var bytesToRemove = new byte[] { 0xe2, 0x80, 0xAb };
if (DataUtils.IndexOfByteArray(bytes, bytesToRemove) > -1)
{
var newbytes = DataUtils.RemoveBytes(bytes, bytesToRemove);
args.Markdown = args.MarkdownDocument.Encoding.GetString(newbytes);
}
}
}
/// <summary>
/// Check for ```markdown blocks and replace them with DIV blocks
/// </summary>
/// <param name="args"></param>
public void BeforeMarkdownRendered(ModifyMarkdownArguments args)
{
while (true)
{
string extract = StringUtils.ExtractString(args.Markdown, "\n```mermaid", "```", returnDelimiters: true);
if (string.IsNullOrEmpty(extract))
{
break;
}
string newExtract = extract.Replace("```mermaid", "<div class=\"mermaid\">")
.Replace("```", "</div>");
args.Markdown = args.Markdown.Replace(extract, newExtract);
}
}
public void ParseXrefTags(ModifyMarkdownArguments args)
{
var matches = XRefRegEx.Matches(args.Markdown);
foreach (Match match in matches)
{
string value = match.Value.Trim();
var link = StringUtils.ExtractString(value, "<xref:", ">")?.TrimEnd('/');
if (link == null)
{
return;
}
string title = null;
var filePath = FixUpRootPath(args, link);
if (File.Exists(filePath))
{
string fileContent = File.ReadAllText(filePath);
title = StringUtils.GetLines(fileContent)
.FirstOrDefault(l => l.StartsWith("# ") || l.StartsWith("## ") || l.StartsWith("### "));
}
else if (File.Exists(Path.ChangeExtension(filePath, "md")))
{
string fileContent = File.ReadAllText(Path.ChangeExtension(filePath, "md"));
title = StringUtils.GetLines(fileContent)
.FirstOrDefault(l => l.StartsWith("# ") || l.StartsWith("## ") || l.StartsWith("### "));
}
if (string.IsNullOrEmpty(title))
{
title = link;
}
else
{
title = title.TrimStart('#', ' ', '\t');
}
var html = $"<a href=\"{link}\">{WebUtility.HtmlEncode(title)}</a>";
args.Markdown = args.Markdown.Replace(value, html);
}
}
/// <summary>
/// Handles rendering of Tip/Warning/Important/Caution/Note
/// blocks.
///
/// </summary>
/// <returns></returns>
/// <remarks>
/// Note: Markdown is expected to be in LineFeed only mode for line breaks (StringUtils.NormalizeLinefeeds()).
/// If you have CR/LF the value needs to be fixed up. RenderExtensions automatically fix up inbound Markdown
/// to normalized linefeeds for rendering, but if you test locally make sure to pre-format the args.Markdown
/// </remarks>
public void ParseNoteTipWarningImportant(ModifyMarkdownArguments args)
{
var matches = TipNoteWarningImportantFileRegEx.Matches(args.Markdown);
foreach (Match match in matches)
{
string value = match.Value.Trim();
var lines = StringUtils.GetLines(value);
var sb = new StringBuilder();
for (int i = 0; i < lines.Length; i++)
{
string line = lines[i];
if (i == 0)
{
// note header
if (line.TrimStart().StartsWith("> [!"))
{
var word = StringUtils.ExtractString(line, "> [!", "]");
sb.AppendLine("<div class=\"" + word + "\">");
sb.AppendLine($"<h5>{word}</h5>");
sb.AppendLine();
}
else
{
// content line - could be empty
sb.AppendLine(line.TrimStart(' ', '>'));
}
}
else
{
// content line - could be empty
sb.AppendLine(line.TrimStart(' ', '>'));
}
}
sb.AppendLine();
sb.AppendLine("</div>");
sb.AppendLine();
args.Markdown = args.Markdown.Replace(value, sb.ToString());
}
}
public string FixUpRootPath(ModifyMarkdownArguments args, string link)
{
if (string.IsNullOrEmpty(link))
{
return(link);
}
var filePath = link;
if (filePath.StartsWith("~/") || filePath.StartsWith("/"))
{
filePath = StringUtils.ReplaceStringInstance(filePath, "~/", "", 1, false);
if (filePath.StartsWith("/"))
{
filePath = filePath.Substring(1);
}
filePath = args.MarkdownDocument.PreviewWebRootPath + "\\" + filePath;
}
else
{
var path = Path.GetDirectoryName(args.MarkdownDocument.Filename);
if (!string.IsNullOrEmpty(path))
{
filePath = Path.Combine(path, filePath);
}
else
{
filePath = args.MarkdownDocument.PreviewWebRootPath + "\\" + filePath;
}
}
filePath = FileUtils.NormalizePath(filePath);
return(filePath);
}
public void BeforeMarkdownRendered(ModifyMarkdownArguments args)
{
ParseDocFxIncludeFiles(args);
ParseNoteTipWarningImportant(args);
}
/// <summary>
/// Parses DocFx include files in the format of:
///
/// [!include[title](relativePathToFileToInclude>)]
///
/// Should run **prior** to Markdown parsing of the main document
/// as it will embed the file content as is.
/// </summary>
/// <returns></returns>
public void ParseDocFxIncludeFiles(ModifyMarkdownArguments args)
{
var matches = includeFileRegEx.Matches(args.Markdown);
foreach (Match match in matches)
{
string value = match.Value;
//string title = StringUtils.ExtractString(value, "[!include[", "]");
string file = StringUtils.ExtractString(value, "](", ")]");
if (string.IsNullOrEmpty(file))
{
continue;
}
bool isCode = value.StartsWith("[!code-");
string syntax = isCode ? StringUtils.ExtractString(value, "[!code-", "[]") : null;
string filePath;
bool hasPreviewWebPath = !string.IsNullOrEmpty(args.MarkdownDocument.PreviewWebRootPath);
// Fix up paths
if (hasPreviewWebPath && file.StartsWith("~/"))
{
file = Path.Combine(args.MarkdownDocument.PreviewWebRootPath, file.Substring(2));
}
else if (hasPreviewWebPath && file.StartsWith("~") || file.StartsWith("/"))
{
file = Path.Combine(args.MarkdownDocument.PreviewWebRootPath, file.Substring(1));
}
if (file.Contains(@":\") || file.Contains(":/"))
{
filePath = file;
}
else
{
var lastDoc = nestedDocs.LastOrDefault();
if (string.IsNullOrEmpty(lastDoc))
{
filePath = mmApp.Model.ActiveDocument?.Filename;
filePath = Path.GetDirectoryName(filePath);
}
else
{
filePath = Path.GetDirectoryName(lastDoc);
}
if (string.IsNullOrEmpty(filePath))
{
continue;
}
}
string includeFile = Path.Combine(filePath, file);
includeFile = FileUtils.NormalizePath(includeFile);
if (!File.Exists(includeFile))
{
// escape the embedded link ([] so it doesn't expand
args.Markdown = args.Markdown.Replace("[]", "[]");
continue;
}
var includedContent = File.ReadAllText(includeFile);
if (isCode)
{
includedContent = $"```{syntax}\n{includedContent.Trim()}\n```";
}
nestedDocs.Push(includeFile);
// We need to process nested content
var markdownDocument = new MarkdownDocument();
markdownDocument.PreviewWebRootPath = args.MarkdownDocument.PreviewWebRootPath;
markdownDocument.CurrentText = includedContent;
markdownDocument.OnBeforeDocumentRendered(ref includedContent);
args.Markdown = args.Markdown.Replace(value, includedContent);
nestedDocs.Pop();
}
}