src/Gargantext/Text/Parsers/CSV.hs

   1 {-|
   2 Module      : Gargantext.Text.Parsers.CSV
   3 Description :
   4 Copyright   : (c) CNRS, 2017-Present
   5 License     : AGPL + CECILL v3
   6 Maintainer  : team@gargantext.org
   7 Stability   : experimental
   8 Portability : POSIX
   9
  10 Here is a longer description of this module, containing some
  11 commentary with @some markup@.
  12 -}
  13
  14 {-# LANGUAGE NoImplicitPrelude #-}
  15 {-# LANGUAGE OverloadedStrings #-}
  16 {-# LANGUAGE DeriveGeneric      #-}
  17
  18 module Gargantext.Text.Parsers.CSV where
  19
  20 import GHC.Generics (Generic)
  21 import GHC.IO (FilePath)
  22 import Data.Either (Either(Left, Right))
  23 import Data.Text (Text)
  24 import Control.Applicative
  25 import qualified Data.ByteString.Lazy as BL
  26 import Data.Csv
  27 import qualified Data.Vector as V
  28 import Data.Text (pack)
  29
  30 import Data.Char (ord)
  31 import Gargantext.Prelude
  32
  33 data CsvDoc = CsvDoc
  34     { title  :: !Text
  35     , source :: !Text
  36     , publication_year  :: !Int
  37     , publication_month :: !Int
  38     , publication_day   :: !Int
  39     , abstract          :: !Text
  40     , authors           :: !Text
  41     }
  42     deriving (Show, Generic)
  43
  44 instance FromNamedRecord CsvDoc where
  45   parseNamedRecord r = CsvDoc <$> r .: "title"
  46                               <*> r .: "source"
  47                               <*> r .: "publication_year"
  48                               <*> r .: "publication_month"
  49                               <*> r .: "publication_day"
  50                               <*> r .: "abstract"
  51                               <*> r .: "authors"
  52
  53 instance ToNamedRecord CsvDoc where
  54   toNamedRecord (CsvDoc t s py pm pd abst aut) =
  55     namedRecord [ "title"  .= t
  56                 , "source" .= s
  57                 , "publication_year"  .= py
  58                 , "publication_month" .= pm
  59                 , "publication_day"   .= pd
  60                 , "abstract"          .= abst
  61                 , "authors"           .= aut
  62                 ]
  63
  64
  65 csvDecodeOptions :: DecodeOptions
  66 csvDecodeOptions = (defaultDecodeOptions {decDelimiter = fromIntegral $ ord '\t'} )
  67
  68 csvEncodeOptions :: EncodeOptions
  69 csvEncodeOptions = (defaultEncodeOptions {encDelimiter = fromIntegral $ ord '\t'} )
  70
  71
  72 readCsv :: FilePath -> IO (Header, V.Vector CsvDoc)
  73 readCsv fp = do
  74     csvData <- BL.readFile fp
  75     case decodeByNameWith csvDecodeOptions csvData of
  76       Left e    -> panic (pack e)
  77       Right csvDocs -> pure csvDocs
  78
  79 writeCsv :: FilePath -> (Header, V.Vector CsvDoc) -> IO ()
  80 writeCsv fp (h, vs) = BL.writeFile fp $
  81                       encodeByNameWith csvEncodeOptions h (V.toList vs)
  82