Started working on parsing changes to allow incomplete elements to be parsed (eg. opening bold, followed by text, without closing bold)

src/MdToHTML.hs

@@ -33,7 +33,7 @@ data MdToken
   | Blockquote [MdToken]
   | UnordList [MdToken]
   | OrdList [MdToken]
-  | Code String
+  | Code MdToken
   | Codeblock String
   | Link MdToken URL
   | Image MdToken ImgPath
@@ -51,11 +51,11 @@ instance Show MdToken where
   show (Line tokens) = concatMap show tokens
   show Linebreak = "<br>"
   show SingleNewline = " "
-  show HorizontalRule = "---------"
+  show HorizontalRule = "<hr>"
   show (Blockquote tokens) = "<blockquote>" ++ concatMap show tokens ++ "</blockquote>"
   show (UnordList tokens) = "<ul>" ++ concatMap (prepend "<li>" . append "</li>" . show) tokens ++ "</ul>"
   show (OrdList tokens) = "<ol>" ++ concatMap (prepend "<li>" . append "</li>" . show) tokens ++ "</ol>"
-  show (Code code) = show code
+  show (Code code) = "<code>" ++ show code ++ "</code>"
   show (Codeblock code) = show code
   show (Link txt url) = "<a href=\"" ++ getUrl url ++ "\">" ++ show txt ++ "</a>"
   show (Image txt imgPath) = "<img src=" ++ getPath imgPath ++ ">" ++ show txt ++ "</img>"
@@ -88,7 +88,9 @@ leftmostLongestParse parser input =
         Nothing -> (mempty, mempty)
         Just x -> x
 
-specialChars = "\\#*_[\n"
+specialChars = "\\#*_[\n`"
+
+escapableChars = '~' : specialChars
 
 -- Makes a parser greedy. Instead of returning all possible parses, only the longest one is returned.
 greedyParse :: ReadP a -> ReadP [a]
@@ -141,7 +143,7 @@ parseBold = parseBoldWith "**" <|> parseBoldWith "__"
   where
     parseBoldWith delim = do
       string delim
-      inside <- greedyParse1 parseLineToken
+      inside <- many1 parseLineToken
       string delim
       return (Bold (Line inside))
 
@@ -151,7 +153,7 @@ parseItalic = parseItalicWith "*" <|> parseItalicWith "_"
   where
     parseItalicWith delim = do
       string delim
-      inside <- greedyParse1 parseLineToken
+      inside <- many1 parseLineToken
       string delim
       return (Italic (Line inside))
 
@@ -163,6 +165,14 @@ parseStrikethrough = do
   string "~~"
   return (Strikethrough (Line inside))
 
+-- Parse code
+parseCode :: ReadP MdToken
+parseCode = do
+  string "`"
+  inside <- many1 get
+  string "`"
+  return (Code (Unit inside))
+
 -- Parse a link
 parseLink :: ReadP MdToken
 parseLink = do
@@ -188,27 +198,22 @@ parseSingleNewline = do
 parseEscapedChar :: ReadP MdToken
 parseEscapedChar = do
   char '\\'
-  escapedChar <- choice (map char specialChars) -- Parse any of the special chars.
+  escapedChar <- choice (map char escapableChars) -- Parse any of the special chars.
   return (Unit [escapedChar])
 
 -- Parse a character as a Unit.
 parseUnit :: ReadP MdToken
 parseUnit = do
-  text <- satisfy (`notElem` specialChars)
+  -- text <- satisfy (`notElem` specialChars)
+  text <- get
   return (Unit [text])
 
--- Parse a regular string as a Unit.
-parseString :: ReadP MdToken
-parseString = do
-  firstChar <- satisfy (/= '\n') -- Must parse at least one non-newline character here
-  text <- munch (`notElem` specialChars)
-  return (Unit (firstChar : text))
-
 lineParsers :: [ReadP MdToken]
 lineParsers =
   [ parseLinebreak,
     parseSingleNewline,
     parseEscapedChar,
+    parseCode,
     parseBold,
     parseItalic,
     parseStrikethrough,

src/MdToHtmlTest.hs

@@ -93,6 +93,13 @@ orderedListTests =
       check_equal "Unordered list in ordered list" "<ol><li>Item 1</li><li>Item 2<ul><li>Item 1</li><li>Item 2</li></ul></li><li>Item 3</li></ol>" (convert "1. Item 1\n2. Item 2\n    - Item 1\n    * Item 2\n4. Item 3")
     ]
 
+codeTests =
+  TestList
+    [ check_equal "Code by itself" "<p><code>Hello world!</code></p>" (convert "`Hello world!`"),
+      check_equal "Code in a paragraph" "<p>The following <code>text</code> is code</p>" (convert "The following `text` is code"),
+      check_equal "Code across paragraphs (shouldn't work" "<p>`Incomplete</p><p>Code`</p>" (convert "`Incomplete\n\nCode`")
+    ]
+
 integrationTests =
   TestList
     [ check_equal "Integration 1" "<h1>Sample Markdown</h1><p>This is some basic, sample markdown.</p><h2><b>Second</b> <i>Heading</i></h2>" (convert "# Sample Markdown\n\n This is some basic, sample markdown.\n\n  ## __Second__ _Heading_"),
@@ -127,6 +134,7 @@ tests =
       blockquoteTests,
       unorderedListTests,
       orderedListTests,
+      codeTests,
       integrationTests
     ]