classify 2012.10.23 → 2013.11.6

raw patch · 4 files changed

+45/−20 lines, 4 files

Files

• classify.cabal +1/−1
• src/Data/Classify/Parser.hs +32/−0
• src/Data/Classify/Rank.hs +11/−1
• src/Data/Classify/Television.hs +1/−18

@@ -1,6 +1,6 @@ Cabal-version:  >=1.6 Name:           classify-Version:        2012.10.23+Version:        2013.11.6 Build-Type:     Simple License:        PublicDomain License-File:   LICENSE

@@ -280,3 +280,35 @@         rets -> fst (head (sortBy (comparing (length.snd)) rets))  ++++++++++++stripName :: String -> [String]+stripName = nub . map (filter isAlphaNum . filter (not . isSpace)) . expand . map toLower+  where+    alts = [ ("the", "")+           , ("newyork", "ny")+           , ("los angeles", "la")+           , ("and", "")+           , ("&", "and") ]+    doesMatch key [] = Nothing+    doesMatch key (x:xs)+      | fst x `isPrefixOf` key = Just x+      | otherwise              = doesMatch key xs+    expand lst = lst : concatMap expand [ before ++ new ++ drop (length old) after+                 | (before, after) <- split lst+                 , Just (old, new) <- [doesMatch after alts] ]+    worker lst = expand lst++split :: String -> [(String,String)]+split str = zip (inits str) (tails str)++

@@ -59,13 +59,17 @@ -- Tests -------------------------------------------------------------- -testSeries = Map.fromList [ (P.trunc t,t) | (_,t,_) <- testData]+testSeries = Map.fromList [ (src,t) | (_,t,_) <- testData, src <- P.stripName t]  v24 = "24" simpsons = "The Simpsons" colbert = "The Colbert Report" palms = "Hidden Palms" daily = "The Daily Show"+ncis = "NCIS"+ncis_la = "NCIS: Los Angeles"+law_svu = "Law & Order: SVU"+mike_molly = "Mike & Molly"  testData = [ ("24.S06E01.6AM.TO.7AM.PROPER.DVDRip.XviD-MEMETiC.avi", "24", Version 6 1)            , ("The.Simpsons.S18E09.PROPER.PDTV.XviD-2HD.avi", "The Simpsons", Version 18 9)@@ -111,6 +115,12 @@            , ("south.park.1109.dsr.xvid-sys.[VTV].avi", "South Park", Version 11 9)            , ("South Park - 12x09 - Breast Cancer Show Ever.avi", "South Park", Version 12 9)            , ("Merlin 2008 5x03 The Death Song Of Uther Pendragon REPACK HDTV x264-FoV", "Merlin", Version 5 3)+           , ("NCIS Los Angeles S05E07 HDTV x264-LOL", ncis_la, Version 5 7)+           , ("NCIS S11E07 HDTV x264-LOL", ncis, Version 11 7)+           , ("NCIS LA 4x14 (HDTV-x264-LOL) [VTV]", ncis_la, Version 4 14)+           , ("Law and Order SVU S14E14 HDTV x264-LOL", law_svu, Version 14 14)+           , ("Law & Order SVU 14x13 (HDTV-x264-LOL) [VTV]", law_svu, Version 14 13)+           , ("Mike and Molly S04E01 HDTV x264-LOL", mike_molly, Version 4 1)            ]  runTests = flip mapM_ testData $ \(str, series, expectedVersion) ->

@@ -26,7 +26,7 @@  import qualified Data.Map as Map import Data.Char                      ( isAlphaNum, isSpace, toLower )-import Data.List                      ( inits, tails, isPrefixOf )+import Data.List                      ( inits, tails, isPrefixOf, nub )  tryClassify :: [Name] -> String -> Maybe (Name, Version, Title) tryClassify knownNames inp =@@ -45,20 +45,3 @@     Nothing     -> error $ "Data.Classify.Television.classify: Failed to parse: " ++ show inp     Just result -> result -stripName :: String -> [String]-stripName = worker . filter isAlphaNum . filter (not . isSpace) . map toLower-  where-    alts = [ ("the", "")-           , ("newyork", "ny")-           , ("los angeles", "la") ]-    doesMatch key [] = Nothing-    doesMatch key (x:xs)-      | fst x `isPrefixOf` key = Just x-      | otherwise              = doesMatch key xs-    expand lst = lst : concatMap expand [ before ++ new ++ drop (length old) after-                 | (before, after) <- split lst-                 , Just (old, new) <- [doesMatch after alts] ]-    worker lst = expand lst--split :: String -> [(String,String)]-split str = zip (inits str) (tails str)