private void ParseLineGroup(
ParseInput input
)
{
ParseResult result;
if (MarkdownHeading.CanParseFrom(input))
{
result = MarkdownHeading.ParseFrom(input);
}
else if (MarkdownHorizontalRule.CanParseFrom(input))
{
result = MarkdownHorizontalRule.ParseFrom(input);
}
else if (MarkdownQuote.CanParseFrom(input))
{
result = MarkdownQuote.ParseFrom(input);
}
else if (MarkdownList.CanParseFrom(input))
{
result = MarkdownList.ParseFrom(input);
}
else
{
result = MarkdownParagraph.ParseFrom(input);
}
foreach (IHtmlable entry in result.GetContent())
{
Content.AddLast(entry);
}
}
private static ParseResult ParseDoubleLineHeading(
ParseInput input
)
{
ArraySegment <string> lines = input.Lines();
ParseResult result = new ParseResult();
int level;
if (lines[1].StartsWith("="))
{
level = 1;
}
else
{
level = 2;
}
MarkdownHeading element = new MarkdownHeading(
level,
MarkdownParser.ParseInnerText(
new ParseInput(
input,
lines[0]
)
)
);
lines[0] = "";
lines[1] = "";
result.Success = true;
result.AddContent(
element
);
return(result);
}
public static bool CanParseFrom(
ParseInput input
)
{
string line = input.FirstLine;
ReferencedUrl[] urls = input.Urls;
if (
regexImageImmediateNoTitle.Match(line).Success ||
regexImageImmediateWithTitle.Match(line).Success
)
{
return(true);
}
Match linkMatch = regexImageReference.Match(line);
if (linkMatch.Success)
{
string reference = linkMatch.Groups[2].Value;
foreach (ReferencedUrl url in urls)
{
if (url.Reference == reference)
{
return(true);
}
}
}
return(false);
}
public static ParseResult ParseFrom(
ParseInput input
)
{
string line = input.FirstLine;
if (!CanParseFrom(input))
{
// Return a failed result if this cannot be parsed
ParseResult result = new ParseResult();
result.Line = line;
return(result);
}
// Otherwise, parse and return result
if (line.StartsWith("**"))
{
return(ParseStrongSection(
input,
"**"
));
}
else
{
return(ParseStrongSection(
input,
"__"
));
}
}
public static bool CanParseFrom(
ParseInput input
)
{
return(
CanParseSingleLineHeading(input) ||
CanParseDoubleLineHeading(input)
);
}
public static bool CanParseFrom(
ParseInput input
)
{
return(
regexOrderedListLine.Match(input.FirstLine).Success ||
regexUnorderedListLine.Match(input.FirstLine).Success
);
}
private static bool CanParseDoubleLineHeading(
ParseInput input
)
{
bool isDoubleLineHeading = false;
if (input.Lines().Count > 1)
{
isDoubleLineHeading = regexDoubleLineHeading.Match(input.Lines()[1]).Success;
}
return(isDoubleLineHeading);
}
public ParseInput(
ParseInput toCopy,
string line
)
{
Urls = toCopy.Urls;
lines = new string[]
{
line
};
startIndex = 0;
elements = 1;
}
private static void RemoveListIndicators(
ParseInput input
)
{
ArraySegment <string> lines = input.Lines();
for (int i = 0; i < lines.Count; i++)
{
string truncated = lines[i];
Match lineOrderedContentMatch = regexOrderedListLine.Match(lines[i]);
Match lineUnorderedContentMatch = regexUnorderedListLine.Match(lines[i]);
if (
lineOrderedContentMatch.Success ||
lineUnorderedContentMatch.Success
)
{
if (lineOrderedContentMatch.Success)
{
truncated = lineOrderedContentMatch.Groups[1].Value;
}
else if (lineUnorderedContentMatch.Success)
{
truncated = lineUnorderedContentMatch.Groups[1].Value;
}
int spaces = 0;
// Count spaces
while (
(spaces < truncated.Length) &&
(truncated[spaces] == ' ')
)
{
spaces++;
}
// If there are fewer than 5 spaces, remove all
if (spaces < 5)
{
lines[i] = truncated.Substring(spaces);
}
else
{
// More than five, just remove one space
lines[i] = truncated.Substring(1);
}
}
else
{
lines[i] = lines[i];
}
}
}
public static bool CanParseFrom(
ParseInput input
)
{
string line = input.FirstLine;
// Check if the format is correct (* or - plus whitespace only)
bool correctFormat = regexHorizontalRule.Match(line).Success;
// Check there are enough - or * characters (3++)
bool enoughNonWhitespace = (
(line.Length - line.Replace("*", "").Replace("-", "").Length)
> 2
);
return(correctFormat && enoughNonWhitespace);
}
// Given a text snippet, parse a plain text section from its start
public static ParseResult ParseFrom(
ParseInput input,
bool force
)
{
string line = input.FirstLine;
ParseResult result = new ParseResult();
int indexFirstSpecialCharacter = FindUnescapedSpecial(
line
);
// If force, then ensure at least one character is consumed
if (
force &&
(indexFirstSpecialCharacter == 0)
)
{
indexFirstSpecialCharacter++;
}
// If an empty string is `parsed` then fail
if (indexFirstSpecialCharacter == 0)
{
result.Line = line;
return(result);
}
else if (indexFirstSpecialCharacter != line.Length)
{
// There is some special character in the string
// Add as plain text only the content up to where it starts
MarkdownText element = new MarkdownText(
line.Substring(0, indexFirstSpecialCharacter)
);
result.AddContent(element);
result.Line = line.Substring(indexFirstSpecialCharacter);
result.Success = true;
}
else
{
// If there are no special sections, everything is plain text
MarkdownText element = new MarkdownText(
line
);
result.AddContent(element);
result.Line = "";
result.Success = true;
}
return(result);
}
public static ParseResult ParseFrom(
ParseInput input
)
{
string line = input.FirstLine;
ParseResult result = new ParseResult();
if (!CanParseFrom(input))
{
// Fail immediately if we cannot parse this text as strikethrough
result.Line = line;
return(result);
}
int j = 2;
// Find closing tildes
while (
(j < line.Length) &&
!(
(line.Substring(j - 1, 2) == "~~") &&
(line[j - 2] != '\\')
)
)
{
j++;
}
if (j >= line.Length)
{
// Fail if we cannot find the closing squiggles
result.Line = line;
return(result);
}
// Parse everything inside the stars
MarkdownStrikethrough element = new MarkdownStrikethrough(
MarkdownParser.ParseInnerText(
new ParseInput(
input,
line.Substring(2, j - 3)
)
)
);
result.AddContent(element);
result.Line = line.Substring(j + 1);
result.Success = true;
return(result);
}
// Shared code for parsing emphasis sections
public static ParseResult ParseFrom(
ParseInput input
)
{
string line = input.FirstLine;
ParseResult result = new ParseResult();
if (!CanParseFrom(input))
{
// Fail immediately if this string cannot be parsed
result.Line = line;
return(result);
}
int j = 1;
// Find closing `
while (
(j < line.Length) &&
!(
(line[j] == '`') &&
(line[j - 1] != '\\')
)
)
{
j++;
}
if (j >= line.Length)
{
// If we cannot parse, then return line as is
result.Line = line;
return(result);
}
// Parse everything inside the backticks
MarkdownCodeInline element = new MarkdownCodeInline(
MarkdownParser.ParseInnerText(
new ParseInput(
input,
line.Substring(1, j - 1)
)
)
);
result.AddContent(element);
result.Line = line.Substring(j + 1);
result.Success = true;
return(result);
}
public static ParseResult ParseFrom(
ParseInput input
)
{
ParseResult result = new ParseResult();
if (!CanParseFrom(input))
{
return(result);
}
input.FirstLine = "";
result.Success = true;
result.AddContent(
new MarkdownHorizontalRule()
);
return(result);
}
public static ParseResult ParseFrom(
ParseInput input
)
{
if (CanParseSingleLineHeading(input))
{
return(ParseSingleLineHeading(input));
}
else if (CanParseDoubleLineHeading(input))
{
return(ParseDoubleLineHeading(input));
}
else
{
return(new ParseResult());
}
}
public static ParseResult ParseFrom(
ParseInput input
)
{
string line = input.FirstLine;
ParseResult result = new ParseResult();
if (!CanParseFrom(input))
{
// Return a failed result if cannot parse from this line
result.Line = line;
return(result);
}
// Otherwise parse and return result
Match contentMatch = regexParseable.Match(line);
string content;
if (contentMatch.Groups[1].Value.Length != 0)
{
content = contentMatch.Groups[1].Value;
}
else
{
content = contentMatch.Groups[2].Value;
}
// Parse everything inside the stars
MarkdownEmphasis element = new MarkdownEmphasis(
MarkdownParser.ParseInnerText(
new ParseInput(
input,
content
)
)
);
result.AddContent(element);
result.Line = line.Substring(
content.Length + 2
);
result.Success = true;
return(result);
}
// Shared code for parsing strong sections
private static ParseResult ParseStrongSection(
ParseInput input,
string delimiter
)
{
string line = input.FirstLine;
ParseResult result = new ParseResult();
int j = 2;
// Find closing two characters
while (
(j < line.Length) &&
!(
(line.Substring(j - 1, 2) == delimiter) &&
(line[j - 2] != '\\')
)
)
{
j++;
}
if (j >= line.Length)
{
// Fail if closing characters cannot be found
result.Line = line;
return(result);
}
// Parse everything inside the strong section delimiters
MarkdownStrong element = new MarkdownStrong(
MarkdownParser.ParseInnerText(
new ParseInput(
input,
line.Substring(2, j - 3)
)
)
);
result.AddContent(element);
result.Line = line.Substring(j + 1);
result.Success = true;
// Return the line string minus the content we parsed
return(result);
}
public static ParseResult ParseFrom(
ParseInput lines,
bool innerParagraph
)
{
ParseResult result = new ParseResult();
ParseResult innerResult;
IHtmlable returnedElement;
// If the list items content contains another list
if (MarkdownList.CanParseFrom(lines))
{
innerResult = MarkdownList.ParseFrom(lines);
returnedElement = new MarkdownListItem(
innerResult.GetContent()
);
}
else
{
// Otherwise, if the item content should go in a paragraph
if (innerParagraph)
{
innerResult = MarkdownParagraph.ParseFrom(lines);
returnedElement = new MarkdownListItem(
innerResult.GetContent()
);
}
else
{
// line item content should not go in a paragraph
returnedElement = new MarkdownListItem(
MarkdownParser.ParseInnerText(lines)
);
}
}
result.Success = true;
result.AddContent(
returnedElement
);
return(result);
}
public MarkdownParser(
string[] lines
)
{
// Assume success
Success = true;
// Store parsed content as we go
Content = new LinkedList <IHtmlable>();
// Urls potentially referenced by anchors/images
Urls = new LinkedList <ReferencedUrl>();
// Extract all 'footnote' style urls
ParseReferencedUrls(lines);
// Create new parse input to pass to parsers
ParseInput input = new ParseInput(
Utils.LinkedListToArray(Urls),
lines,
0,
lines.Length
);
// Parsing printed content
int currentIndex = 0;
while (currentIndex < lines.Length)
{
// Parse some lines
ParseLineGroup(
input.Slice(
currentIndex,
lines.Length - currentIndex
)
);
// Move to the next non-empty line
while (
(currentIndex < lines.Length) &&
ContainsOnlyWhitespace(lines[currentIndex])
)
{
currentIndex++;
}
}
}
private static ParseResult ParseSingleLineHeading(
ParseInput input
)
{
ArraySegment <string> lines = input.Lines();
ParseResult result = new ParseResult();
// Calculate heading level, (maximum 6)
int level = 0;
while (
(level < 6) &&
(lines[0][level] == '#')
)
{
level++;
}
Match contentMatch = regexSingleLineHeading.Match(lines[0]);
string content = StripLeadingCharacter(
StripTrailingCharacter(
contentMatch.Groups[1].Value,
'#'
),
' '
);
lines[0] = "";
result.Success = true;
result.AddContent(
new MarkdownHeading(
level,
MarkdownParser.ParseInnerText(
new ParseInput(
input,
content
)
)
)
);
return(result);
}
public static bool CanParseFrom(
ParseInput input
)
{
ArraySegment <string> lines = input.Lines();
if (!regexBacktickSectionOpen.Match(lines[0]).Success)
{
return(false);
}
else
{
for (int i = 1; i < lines.Count; i++)
{
if (regexBacktickSectionClose.Match(lines[i]).Success)
{
return(true);
}
}
}
return(false);
}
public static ParseResult ParseFrom(
ParseInput input
)
{
ParseResult result = new ParseResult();
if (!CanParseFrom(input))
{
return(result);
}
ArraySegment <string> lines = input.Lines();
lines[0] = "";
LinkedList <IHtmlable> innerContent = new LinkedList <IHtmlable>();
int i = 1;
while (!regexBacktickSectionClose.Match(lines[i]).Success)
{
innerContent.AddLast(
new MarkdownText(
lines[i]
)
);
lines[i] = "";
i++;
}
// Remember to clear final line (closing backticks)
lines[i] = "";
MarkdownCodeBlock blockCodeElement = new MarkdownCodeBlock(
Utils.LinkedListToArray(innerContent)
);
result.Success = true;
result.AddContent(blockCodeElement);
return(result);
}
public static ParseResult ParseFrom(
ParseInput input
)
{
ArraySegment <string> lines = input.Lines();
ParseResult result = new ParseResult();
if (!CanParseFrom(input))
{
return(result);
}
int endQuoteSection = FindEndOfQuoteSection(
lines
);
string[] truncatedLines = new string[endQuoteSection];
// Remove quote arrows and spaces, if needed
for (int i = 0; i < endQuoteSection; i++)
{
string truncated = lines[i];
if (lines[i].StartsWith(">"))
{
truncated = truncated.Substring(1);
int spaces = 0;
// Count spaces
while (
(spaces < truncated.Length) &&
(truncated[spaces] == ' ')
)
{
spaces++;
}
// If there are fewer than 5 spaces, remove all
if (spaces < 5)
{
truncatedLines[i] = truncated.Substring(spaces);
}
else
{
// More than five, just remove one space
truncatedLines[i] = truncated.Substring(1);
}
}
else
{
truncatedLines[i] = lines[i];
}
// Remove original line
lines[i] = "";
}
/*
* The truncated lines should be parsed as any other line group
* and wrapped in a blockquote element
*/
MarkdownParser parser = new MarkdownParser(
truncatedLines
);
MarkdownQuote quoteElement = new MarkdownQuote(
parser.ContentAsArray()
);
result.Success = true;
result.AddContent(
quoteElement
);
return(result);
}
public static bool CanParseFrom(
ParseInput input
)
{
return(input.FirstLine.StartsWith(">"));
}
public static ParseResult ParseFrom(
ParseInput input
)
{
ArraySegment <string> lines = input.Lines();
ParseResult result = new ParseResult();
if (!CanParseFrom(input))
{
return(result);
}
/*
* Work out the list type from the first line
* before we start to mangle the line contents
*/
MarkdownElementType type;
if (regexOrderedListLine.Match(input.FirstLine).Success)
{
type = MarkdownElementType.OrderedList;
}
else
{
type = MarkdownElementType.UnorderedList;
}
int endListSection = FindEndOfListSection(
lines
);
// New array segment with only parsed content
ArraySegment <string> listLines = new ArraySegment <string>(
lines.Array,
lines.Offset,
endListSection
);
// Need to split into groups per list item
int currentIndex = 0;
// Hold the parsed list items as we go
LinkedList <IHtmlable> listItems = new LinkedList <IHtmlable>();
// Track whether list item contents should be in a paragraph
bool whitespaceLineBefore = false;
bool whitespaceLineAfter = false;
while (currentIndex < endListSection)
{
int endIndex = FindEndOfListItem(
listLines,
currentIndex
);
/*
* There is a whitespace line between
* this list item and the following one
* and this list item isn't the final one
*/
whitespaceLineAfter = (
(endIndex < listLines.Count) &&
(
ContainsOnlyWhitespace(
listLines[endIndex - 1]
)
)
);
// Create new parse input for this list item
ParseInput listItemLines = new ParseInput(
input.Urls,
listLines.Array,
listLines.Offset + currentIndex,
endIndex - currentIndex
);
RemoveListIndicators(
listItemLines
);
ParseResult nextListItem = MarkdownListItem.ParseFrom(
listItemLines,
whitespaceLineBefore || whitespaceLineAfter
);
foreach (
IHtmlable entry
in nextListItem.GetContent()
)
{
listItems.AddLast(
entry
);
}
// Jump over lines just parsed
currentIndex += (endIndex - currentIndex);
// Whitespace after previous entry becomes before next entry
whitespaceLineBefore = whitespaceLineAfter;
/*
* If there's further whitespace after parsed section
* (for some reason) then jump over this too
*/
while (
(currentIndex < listLines.Count) &&
(
ContainsOnlyWhitespace(
lines[currentIndex]
)
)
)
{
currentIndex++;
}
}
result.Success = true;
result.AddContent(
new MarkdownList(
Utils.LinkedListToArray(
listItems
),
type
)
);
return(result);
}
public static ParseResult ParseFrom(
ParseInput input
)
{
string line = input.FirstLine;
ReferencedUrl[] urls = input.Urls;
ParseResult result = new ParseResult();
if (
!CanParseFrom(input)
)
{
return(result);
}
Match linkMatch;
// Format: [text](url)
linkMatch = regexLinkImmediate.Match(line);
if (linkMatch.Success)
{
string text = linkMatch.Groups[1].Value;
string url = linkMatch.Groups[2].Value;
result.Success = true;
result.Line = regexLinkImmediate.Replace(
line,
""
);
result.AddContent(
new MarkdownLink(
MarkdownParser.ParseInnerText(
new ParseInput(
input,
text
)
),
url
)
);
}
// Format: [text][id] [id]: url
linkMatch = regexLinkReference.Match(line);
if (linkMatch.Success)
{
string text = linkMatch.Groups[1].Value;
string reference = linkMatch.Groups[2].Value;
foreach (ReferencedUrl url in urls)
{
if (url.Reference == reference)
{
result.Success = true;
result.Line = regexLinkReference.Replace(
line,
""
);
result.AddContent(
new MarkdownLink(
MarkdownParser.ParseInnerText(
new ParseInput(
input,
text
)
),
url.Url
)
);
}
}
}
// Format: [text] [text]: url
linkMatch = regexLinkSelfReference.Match(line);
if (linkMatch.Success)
{
string text = linkMatch.Groups[1].Value;
foreach (ReferencedUrl url in urls)
{
if (url.Reference == text)
{
result.Success = true;
result.Line = regexLinkSelfReference.Replace(
line,
""
);
result.AddContent(
new MarkdownLink(
MarkdownParser.ParseInnerText(
new ParseInput(
input,
text
)
),
url.Url
)
);
}
}
}
return(result);
}
public static bool CanParseFrom(
ParseInput input
)
{
return(regexParseable.Match(input.FirstLine).Success);
}
public static ParseResult ParseFrom(
ParseInput input
)
{
string line = input.FirstLine;
ReferencedUrl[] urls = input.Urls;
ParseResult result = new ParseResult();
if (
!CanParseFrom(input)
)
{
return(result);
}
Match linkMatch;
// Format: 
linkMatch = regexImageImmediateNoTitle.Match(line);
if (linkMatch.Success)
{
string text = linkMatch.Groups[1].Value;
string url = linkMatch.Groups[2].Value;
string title = "";
result.Success = true;
result.Line = regexImageImmediateNoTitle.Replace(
line,
""
);
result.AddContent(
new MarkdownImage(
url,
text,
title
)
);
}
// Format: 
linkMatch = regexImageImmediateWithTitle.Match(line);
if (linkMatch.Success)
{
string text = linkMatch.Groups[1].Value;
string url = linkMatch.Groups[2].Value;
string title = linkMatch.Groups[3].Value;
result.Success = true;
result.Line = regexImageImmediateWithTitle.Replace(
line,
""
);
result.AddContent(
new MarkdownImage(
url,
text,
title
)
);
}
// Format: [text][id] [id]: url (title optional)
linkMatch = regexImageReference.Match(line);
if (linkMatch.Success)
{
string text = linkMatch.Groups[1].Value;
string reference = linkMatch.Groups[2].Value;
foreach (ReferencedUrl url in urls)
{
if (url.Reference == reference)
{
result.Success = true;
result.Line = regexImageReference.Replace(
line,
""
);
result.AddContent(
new MarkdownImage(
url.Url,
text,
url.Title
)
);
}
}
}
return(result);
}
// Given a single line of text, parse this, including special (emph, etc...) sections
public static IHtmlable[] ParseInnerText(
ParseInput input
)
{
// Store parsed content as we go
LinkedList <IHtmlable> content = new LinkedList <IHtmlable>();
// Until the whole string has been consumed
while (input.FirstLine.Length > 0)
{
ParseResult result;
if (MarkdownStrong.CanParseFrom(input))
{
result = MarkdownStrong.ParseFrom(input);
}
else if (MarkdownStrikethrough.CanParseFrom(input))
{
result = MarkdownStrikethrough.ParseFrom(input);
}
else if (MarkdownEmphasis.CanParseFrom(input))
{
result = MarkdownEmphasis.ParseFrom(input);
}
else if (MarkdownCodeInline.CanParseFrom(input))
{
result = MarkdownCodeInline.ParseFrom(input);
}
else if (MarkdownLink.CanParseFrom(input))
{
result = MarkdownLink.ParseFrom(input);
}
else if (MarkdownImage.CanParseFrom(input))
{
result = MarkdownImage.ParseFrom(input);
}
else
{
result = MarkdownText.ParseFrom(
input,
false
);
}
/*
* If no parsing method suceeded
* for once character to be parsed as text
*/
if (!result.Success)
{
result = MarkdownText.ParseFrom(
input,
true
);
}
// Extract parsed content
foreach (IHtmlable entry in result.GetContent())
{
content.AddLast(entry);
}
// Update text to be parsed
input.FirstLine = result.Line;
}
IHtmlable[] contentArray = new IHtmlable[content.Count];
content.CopyTo(
contentArray,
0
);
return(contentArray);
}
private static bool CanParseSingleLineHeading(
ParseInput input
)
{
return(regexSingleLineHeading.Match(input.FirstLine).Success);
}