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

   1 {-|
   2 Module      : Gargantext.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.Text.Corpus.Parsers.GrandDebat
  27   where
  28
  29 import Data.Aeson (ToJSON, FromJSON)
  30 import qualified Data.ByteString.Lazy as DBL
  31 import qualified Data.JsonStream.Parser as P
  32 import Data.Maybe (Maybe())
  33 import Data.Text (Text)
  34 import qualified Data.Text as Text
  35 import GHC.Generics (Generic)
  36
  37 import Gargantext.Core (Lang(..))
  38 import Gargantext.Database.Admin.Types.Hyperdata (HyperdataDocument(..), ToHyperdataDocument, toHyperdataDocument)
  39 import Gargantext.Prelude
  40 import Gargantext.Prelude.Utils
  41
  42
  43 data GrandDebatReference = GrandDebatReference
  44   { id        :: !(Maybe Text)
  45   , reference :: !(Maybe Text)
  46   , title     :: !(Maybe Text)
  47
  48   , createdAt :: !(Maybe Text)
  49   , publishedAt   :: !(Maybe Text)
  50   , updatedAt     :: !(Maybe Text)
  51
  52   , trashed       :: !(Maybe Bool)
  53   , trashedStatus :: !(Maybe Text)
  54
  55   , authorId      :: !(Maybe Text)
  56   , authorType    :: !(Maybe Text)
  57   , authorZipCode :: !(Maybe Text)
  58
  59   , responses     :: !(Maybe [GrandDebatResponse])
  60   }
  61   deriving (Show, Generic)
  62
  63
  64 data GrandDebatResponse = GrandDebatResponse
  65   { questionId     :: !(Maybe Text)
  66   , questionTitle  :: !(Maybe Text)
  67   , value          :: !(Maybe Text)
  68   , formattedValue :: !(Maybe Text)
  69   }
  70   deriving (Show, Generic)
  71
  72 instance FromJSON GrandDebatResponse
  73 instance FromJSON GrandDebatReference
  74
  75 instance ToJSON GrandDebatResponse
  76 instance ToJSON GrandDebatReference
  77
  78
  79 instance ToHyperdataDocument GrandDebatReference
  80   where
  81     toHyperdataDocument (GrandDebatReference id' _ref title'
  82                                    _createdAt' publishedAt' _updatedAt
  83                                    _trashed _trashedStatus
  84                                    _authorId authorType' authorZipCode'
  85                                    responses') =
  86       HyperdataDocument (Just "GrandDebat") id'
  87                          Nothing Nothing Nothing Nothing
  88                          title' authorType' authorType' authorZipCode'
  89                         (toAbstract <$> responses')
  90                          publishedAt'
  91                          Nothing Nothing Nothing Nothing Nothing Nothing
  92                         (Just $ Text.pack $ show FR)
  93         where
  94           toAbstract = (Text.intercalate " . ") . ((filter (/= "")) . (map toSentence))
  95           toSentence (GrandDebatResponse _id _qtitle _qvalue r) = case r of
  96             Nothing -> ""
  97             Just r' -> case Text.length r' > 10 of
  98                 True  -> r'
  99                 False -> ""
 100
 101 instance ReadFile [GrandDebatReference]
 102   where
 103     -- | read json: 3 version below are working but with increased optimization
 104     --readFile fp = maybe [] identity <$> decode <$> DBL.readFile fp
 105     --readFile fp = either (panic . Text.pack) identity <$> P.eitherDecode <$> DBL.readFile fp
 106     readFile' fp = P.parseLazyByteString (P.arrayOf P.value) <$> DBL.readFile fp
 107
 108
 109