{-# LANGUAGE FlexibleContexts #-} {-# LANGUAGE OverloadedStrings #-} {-# LANGUAGE RecordWildCards #-} {-# LANGUAGE TupleSections #-} module Internal.Statement ( readImport ) where import Data.Bifunctor import qualified Data.ByteString.Lazy as BL import Data.Csv import Data.Either import qualified Data.List as L import qualified Data.Map as M import Data.Maybe import Data.Ord import qualified Data.Text as T import qualified Data.Text.Encoding as TE import Data.Time import qualified Data.Vector as V import Internal.Types import Internal.Utils import Numeric.Natural -- TODO this probably won't scale well (pipes?) readImport :: Import -> IO [BalTx] readImport Import { impPaths = ps , impMatches = ms , impTxOpts = ns , impDelim = d , impSkipLines = n -- , impTx = f } = do rs <- L.sort . concat <$> mapM (readImport_ n d ns) ps let (ts, es, notfound) = matchRecords ms rs mapM_ putStrLn $ reverse es mapM_ print notfound return ts readImport_ :: Natural -> Word -> TxOpts -> FilePath -> IO [TxRecord] readImport_ n delim tns p = do bs <- BL.readFile p case decodeByNameWithP (parseTxRecord tns) opts $ skip bs of Left m -> putStrLn m >> return [] Right (_, v) -> return $ catMaybes $ V.toList v where opts = defaultDecodeOptions { decDelimiter = fromIntegral delim } skip = BL.intercalate "\n" . L.drop (fromIntegral n) . BL.split 10 -- TODO handle this better, this maybe thing is a hack to skip lines with -- blank dates but will likely want to make this more flexible parseTxRecord :: TxOpts -> NamedRecord -> Parser (Maybe TxRecord) parseTxRecord TxOpts {..} r = do d <- r .: TE.encodeUtf8 toDate if d == "" then return Nothing else do a <- parseRational toAmountFmt =<< r .: TE.encodeUtf8 toAmount e <- r .: TE.encodeUtf8 toDesc os <- M.fromList <$> mapM (\n -> (n, ) <$> r .: TE.encodeUtf8 n) toOther d' <- parseTimeM True defaultTimeLocale toDateFmt d return $ Just $ TxRecord d' a e os matchRecords :: [Match] -> [TxRecord] -> ([BalTx], [String], [Match]) matchRecords ms rs = ( catMaybes ts , T.unpack <$> (es ++ bu) -- TODO record number of times each match hits for debugging , notfound ) where (matched, unmatched, notfound) = matchAll (matchPriorities ms) rs (es, ts) = partitionEithers $ fmap Just . balanceTx <$> catMaybes matched bu = fmap (\x -> T.pack $ "unmatched: " ++ show x) unmatched matchPriorities :: [Match] -> [MatchGroup] matchPriorities = fmap matchToGroup . L.groupBy (\a b -> mPriority a == mPriority b) . L.sortOn (Down . mPriority) matchToGroup :: [Match] -> MatchGroup matchToGroup ms = uncurry MatchGroup $ first (L.sortOn mDate) $ L.partition (isJust . mDate) ms -- TDOO could use a better struct to flatten the maybe date subtype data MatchGroup = MatchGroup { mgDate :: [Match] , mgNoDate :: [Match] } deriving (Show) data Zipped a = Zipped ![a] ![a] data Unzipped a = Unzipped ![a] ![a] ![a] initZipper :: [a] -> Zipped a initZipper = Zipped [] resetZipper :: Zipped a -> Zipped a resetZipper = initZipper . recoverZipper recoverZipper :: Zipped a -> [a] recoverZipper (Zipped as bs) = reverse as ++ bs zipperSlice :: (a -> b -> Ordering) -> b -> Zipped a -> Either (Zipped a) (Unzipped a) zipperSlice f x = go where go z@(Zipped _ []) = Left z go z@(Zipped bs (a:as)) = case f a x of GT -> go $ Zipped (a:bs) as EQ -> Right $ goEq (Unzipped bs [a] as) LT -> Left z goEq z@(Unzipped _ _ []) = z goEq z@(Unzipped bs cs (a:as)) = case f a x of GT -> goEq $ Unzipped (a:bs) cs as EQ -> goEq $ Unzipped bs (a:cs) as LT -> z zipperMatch :: Unzipped Match -> TxRecord -> (Zipped Match, Maybe (Maybe RawTx)) zipperMatch (Unzipped bs cs as) x = go [] cs where go _ [] = (Zipped bs $ cs ++ as, Nothing) go prev (m:ms) = case matches m x of Nothing -> go (m:prev) ms res@(Just _) -> let ps = reverse prev ms' = maybe ms (:ms) (matchDec m) in (Zipped bs $ ps ++ ms' ++ as, res) zipperMatch' :: Zipped Match -> TxRecord -> (Zipped Match, Maybe (Maybe RawTx)) zipperMatch' z x = go z where go (Zipped bs (a:as)) = case matches a x of Nothing -> go (Zipped (a:bs) as) res -> (Zipped (maybe bs (:bs) $ matchDec a) as, res) go z' = (z', Nothing) matchDec :: Match -> Maybe Match matchDec m@Match { mTimes = t } = if t' == Just 0 then Nothing else Just $ m { mTimes = t' } where t' = fmap pred t matchAll :: [MatchGroup] -> [TxRecord] -> ([Maybe RawTx], [TxRecord], [Match]) matchAll = go ([], []) where go (matched, unused) gs rs = case (gs, rs) of (_, []) -> (matched, [], unused) ([], _) -> (matched, rs, unused) (g:gs', _) -> let (ts, unmatched, us) = matchGroup g rs in go (ts ++ matched, us ++ unused) gs' unmatched matchGroup :: MatchGroup -> [TxRecord] -> ([Maybe RawTx], [TxRecord], [Match]) matchGroup MatchGroup { mgDate = ds, mgNoDate = ns } rs = (md ++ mn, unmatched, filter ((/= Nothing) . mTimes) $ ud ++ un) where (md, rest, ud) = matchDates ds rs (mn, unmatched, un) = matchNonDates ns rest matchDates :: [Match] -> [TxRecord] -> ([Maybe RawTx], [TxRecord], [Match]) matchDates ms = go ([], [], initZipper ms) where go (matched, unmatched, z) [] = (matched, reverse unmatched, recoverZipper z) go (matched, unmatched, z) (r:rs) = case zipperSlice findDate r z of Left res -> go (matched, r:unmatched, res) rs Right res -> let (z', p) = zipperMatch res r (m, u) = case p of Just p' -> (p':matched, unmatched) Nothing -> (matched, r:unmatched) in go (m, u, z') rs findDate m r = maybe EQ (`compareDate` trDate r) $ mDate m matchNonDates :: [Match] -> [TxRecord] -> ([Maybe RawTx], [TxRecord], [Match]) matchNonDates ms = go ([], [], initZipper ms) where go (matched, unmatched, z) [] = (matched, reverse unmatched, recoverZipper z) go (matched, unmatched, z) (r:rs) = let (z', res) = zipperMatch' z r (m, u) = case res of Just x -> (x:matched, unmatched) Nothing -> (matched, r:unmatched) in go (m, u, resetZipper z') rs balanceTx :: RawTx -> Either T.Text BalTx balanceTx t@Tx { txSplits = ss } = do bs <- balanceSplits ss return $ t { txSplits = bs } balanceSplits :: [RawSplit] -> Either T.Text [BalSplit] balanceSplits ss = fmap concat <$> mapM (uncurry bal) $ groupByKey $ fmap (\s -> (sCurrency s, s)) ss where hasValue s@(Split { sValue = Just v }) = Right s { sValue = v } hasValue s = Left s bal cur rss | length rss < 2 = Left $ T.append "Need at least two splits to balance: " cur | otherwise = case partitionEithers $ fmap hasValue rss of ([noVal], val) -> Right $ noVal { sValue = foldr (\s x -> x - sValue s) 0 val } : val ([], val) -> Right val _ -> Left $ T.append "Exactly one split must be blank: " cur groupByKey :: Ord k => [(k, v)] -> [(k, [v])] groupByKey = M.toList . M.fromListWith (++) . fmap (second (:[]))