Temporarily remove blobbing from plagiarism to prepare for integration with Submitty (#39)

* Restrict reports to top matches

* Compare in the correct order

* Avoid blobify altogether at first to determine course of action

* Increase winnowing bounds for C

* Fix tilde/exclamation mark lexer bug for C

* Add depth counter

* Update version

Author: chameco

lichen.cabal

@@ -1,5 +1,5 @@
 name:                lichen
-version:             0.2.3
+version:             0.2.4
 synopsis:            Plagiarism Detection and Other Static Analysis
 description:         Please see README.md
 homepage:            https://github.com/Submitty/AnalysisTools

src/Lichen/Config/Languages.hs

@@ -55,7 +55,7 @@ langDummy :: Language
 langDummy = Language [] dummy (WinnowConfig 0 0) (const $ pure ()) dummy
 
 langC :: Language
-langC = Language [".c", ".h", ".cpp", ".hpp", ".C", ".H", ".cc"] C.lex (WinnowConfig 9 5) (smartRead :: String -> Erring C.Tok) dummy
+langC = Language [".c", ".h", ".cpp", ".hpp", ".C", ".H", ".cc"] C.lex (WinnowConfig 16 9) (smartRead :: String -> Erring C.Tok) dummy
 
 langPython :: Language
 langPython = Language [".py"] Python.lex (WinnowConfig 16 9) (smartRead :: String -> Erring Python.Tok) Python.parse

src/Lichen/Plagiarism/Render/Compare.hs

@@ -66,7 +66,7 @@ renderBoth dir (fp, t) (fp', t') = do
         s' <- T.IO.readFile (dir </> sq t')
         let es = T.replace "\t" "        " s
             es' = T.replace "\t" "        " s'
-            (p, p') = blobify (toPosList es fp) (toPosList es' fp')
+            (p, p') = (toPosList es fp, toPosList es' fp')
         return (mconcat . fmap colorize $ splitInto es p, mconcat . fmap colorize $ splitInto es' p')
 
 renderCompare :: (Show a, Eq a) => FilePath -> (Double, (Fingerprints, a), (Fingerprints, a)) -> IO H.Html