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