src/Gargantext/Core/Text/Corpus/Parsers/GrandDebat.hs

   1 {-|
   2 Module      : Gargantext.Core.Text.Corpus.Parsers.GrandDebat
   3 Description : Grand Debat Types
   4 Copyright   : (c) CNRS, 2017-Present
   5 License     : AGPL + CECILL v3
   6 Maintainer  : team@gargantext.org
   7 Stability   : experimental
   8 Portability : POSIX
   9
  10 _flowCorpusDebat :: FlowCmdM env err m
  11                  => User -> Either CorpusName [CorpusId]
  12                  -> Limit -> FilePath
  13                  -> m CorpusId
  14 _flowCorpusDebat u n l fp = do
  15   docs <- liftBase ( splitEvery 500
  16                  <$> take l
  17                  <$> readFile' fp
  18                  :: IO [[GD.GrandDebatReference ]]
  19                  )
  20   flowCorpus u n (Multi FR) (map (map toHyperdataDocument) docs)
  21
  22
  23 -}
  24
  25
  26 module Gargantext.Core.Text.Corpus.Parsers.GrandDebat
  27   where
  28
  29 import Data.Aeson (ToJSON, FromJSON)
  30 import Data.Text (Text)
  31 import GHC.Generics (Generic)
  32 import Gargantext.Core (Lang(..))
  33 import Gargantext.Database.Admin.Types.Hyperdata (HyperdataDocument(..), ToHyperdataDocument, toHyperdataDocument)
  34 import Gargantext.Prelude
  35 import Gargantext.Database.GargDB
  36 import qualified Data.ByteString.Lazy as DBL
  37 import qualified Data.JsonStream.Parser as P
  38 import qualified Data.Text as Text
  39
  40 data GrandDebatReference = GrandDebatReference
  41   { id        :: !(Maybe Text)
  42   , reference :: !(Maybe Text)
  43   , title     :: !(Maybe Text)
  44
  45   , createdAt :: !(Maybe Text)
  46   , publishedAt   :: !(Maybe Text)
  47   , updatedAt     :: !(Maybe Text)
  48
  49   , trashed       :: !(Maybe Bool)
  50   , trashedStatus :: !(Maybe Text)
  51
  52   , authorId      :: !(Maybe Text)
  53   , authorType    :: !(Maybe Text)
  54   , authorZipCode :: !(Maybe Text)
  55
  56   , responses     :: !(Maybe [GrandDebatResponse])
  57   }
  58   deriving (Show, Generic)
  59
  60
  61 data GrandDebatResponse = GrandDebatResponse
  62   { questionId     :: !(Maybe Text)
  63   , questionTitle  :: !(Maybe Text)
  64   , value          :: !(Maybe Text)
  65   , formattedValue :: !(Maybe Text)
  66   }
  67   deriving (Show, Generic)
  68
  69 instance FromJSON GrandDebatResponse
  70 instance FromJSON GrandDebatReference
  71
  72 instance ToJSON GrandDebatResponse
  73 instance ToJSON GrandDebatReference
  74
  75
  76 instance ToHyperdataDocument GrandDebatReference
  77   where
  78     toHyperdataDocument (GrandDebatReference id' _ref title'
  79                                    _createdAt' publishedAt' _updatedAt
  80                                    _trashed _trashedStatus
  81                                    _authorId authorType' authorZipCode'
  82                                    responses') =
  83       HyperdataDocument (Just "GrandDebat") id'
  84                          Nothing Nothing Nothing Nothing
  85                          title' authorType' authorType' authorZipCode'
  86                         (toAbstract <$> responses')
  87                          publishedAt'
  88                          Nothing Nothing Nothing Nothing Nothing Nothing
  89                         (Just $ Text.pack $ show FR)
  90         where
  91           toAbstract = (Text.intercalate " . ") . ((filter (/= "")) . (map toSentence))
  92           toSentence (GrandDebatResponse _id _qtitle _qvalue r) = case r of
  93             Nothing -> ""
  94             Just r' -> case Text.length r' > 10 of
  95                 True  -> r'
  96                 False -> ""
  97
  98 instance ReadFile [GrandDebatReference]
  99   where
 100     -- | read json: 3 version below are working but with increased optimization
 101     --readFile fp = maybe [] identity <$> decode <$> DBL.readFile fp
 102     --readFile fp = either (panic . Text.pack) identity <$> P.eitherDecode <$> DBL.readFile fp
 103     readFile' fp = P.parseLazyByteString (P.arrayOf P.value) <$> DBL.readFile fp
 104
 105
 106