Fixed blockquote parser; implemented unordered list parser and relevant

sub-parsers

src/MdToHTML.hs

@@ -272,8 +272,62 @@ parseQuotedLines =
 -- Parse a blockquote, which is a greater-than sign followed by a paragraph.
 parseBlockquote :: ReadP MdToken
 parseBlockquote = do
-  char '>'
+  quotedLines <- parseQuotedLines
-  Blockquote <$> (parseBlockquote <++ parsePara) -- Parse another blockquote or a regular paragraph, wrap it in a blockquote.
+  -- remaining <- look
+  -- let quotedLines = fst $ leftmostLongestParse parseQuotedLines remaining
+  -- string (init $ unlines quotedLines)
+  let parsedQuotedLines = fst $ leftmostLongestParse (many1 (parseBlockquote <++ parsePara)) (init $ unlines quotedLines) -- unlines joins the lines together with a newline, and adds a trailing newline. init removes the trailing newline.
+  return (Blockquote parsedQuotedLines)
+
+-- Parse a nested list item.
+parseUListNested :: ReadP MdToken
+parseUListNested = do
+  --  firstChar <- string "    " <++ string "\t"
+  --  skipSpaces
+  --  restOfLine <- manyTill get (void (char '\n') <++ eof)
+  --  let restOfLineParsed = fst $ leftmostLongestParse parseLine restOfLine
+  --  return restOfLineParsed
+  let firstCharParser = string "    " <++ string "\t"
+  let restOfLineParser = manyTill get (void (char '\n') <++ eof)
+  lines <- greedyParse1 (firstCharParser *> restOfLineParser)
+  let linesParsed = fst $ leftmostLongestParse parseUnorderedList (init $ unlines lines)
+  return linesParsed
+
+-- Parse an unordered list line item.
+parseUListLineItem :: ReadP MdToken
+parseUListLineItem = do
+  firstChar <- choice (map char ['*', '+', '-'])
+  skipSpaces
+  restOfLine <- manyTill get (void (char '\n') <++ eof)
+  let restOfLineParsed = fst $ leftmostLongestParse parseLine restOfLine
+  nestedList <- parseUListNested <++ return (Unit "")
+  return $ Line [restOfLineParsed, nestedList]
+
+-- Parse an unordered list paragraph item.
+-- This is defined as a line item, followed by an empty line, followed by one or more
+-- lines indented by a space or tab.
+parseUListParaItem :: ReadP MdToken
+parseUListParaItem = do
+  firstLine <- parseUListLineItem
+  char '\n'
+  lines <- greedyParse1 ((string "    " <|> string "\t") *> parseTillEol)
+  let res = fst $ leftmostLongestParse (greedyParse1 parsePara) (init $ unlines lines)
+  char '\n'
+  return $ Document (Para firstLine : res) -- I only wrap this in a document because I want some way of converting [MdToken] to MdToken, without any overhead. There is no other reason to wrap it in a Document.
+
+-- This is hacky as hell
+-- parsedParas <- manyTillLazy parsePara (string "\n\n" *> choice (map char "*-+"))
+-- return $ Document parsedParas -- I wrap this in a document because I want some way of converting [MdToken] to MdToken, without any overhead. There is no other reason to wrap it in a Document.
+
+-- Parse an unordered list item, which can be a line item or another list.
+parseUListItem :: ReadP MdToken
+parseUListItem = parseUListParaItem <++ parseUListLineItem <++ parseUListNested
+
+-- Parse an unordered list.
+parseUnorderedList :: ReadP MdToken
+parseUnorderedList = do
+  lineItems <- greedyParse1 parseUListItem
+  return $ UnordList lineItems
 
 -- Parse a document, which is multiple paragraphs.
 parseDocument :: ReadP MdToken