bin/gargantext-cli/Main.hs

   1 {-|
   2 Module      : Main.hs
   3 Description : Gargantext starter
   4 Copyright   : (c) CNRS, 2017-Present
   5 License     : AGPL + CECILL v3
   6 Maintainer  : team@gargantext.org
   7 Stability   : experimental
   8 Portability : POSIX
   9
  10 Main specifications to index a corpus with a term list
  11
  12  -}
  13
  14 {-# LANGUAGE DataKinds         #-}
  15 {-# LANGUAGE DeriveGeneric     #-}
  16 {-# LANGUAGE FlexibleInstances #-}
  17 {-# LANGUAGE NoImplicitPrelude #-}
  18 {-# LANGUAGE OverloadedStrings #-}
  19 {-# LANGUAGE StandaloneDeriving #-}
  20 {-# LANGUAGE TypeOperators     #-}
  21 {-# LANGUAGE Strict             #-}
  22
  23 module Main where
  24
  25 import qualified Data.Vector as DV
  26
  27 import Gargantext.Prelude
  28 import Data.Text (Text)
  29
  30 import Gargantext.Text.Parsers.CSV (readCsv)
  31 import Gargantext.Text.List.CSV (fromCsvListFile)
  32
  33 main :: IO ()
  34 main = do
  35   [corpusfile, termListFile, outputFile] <- readParams
  36
  37   -- corpus :: [Text]
  38   corpus   <- DV.toList <$> map DV.csv_abstract <$> readCsv corpusFile
  39
  40   -- termListMap :: [Text]
  41   termList <- termListMap <$> fromCsvListFile termListFile
  42
  43   let corpusIndexed = indexCorpusWith corpus termList
  44   let cooc = cooccurrences corpusIndexed
  45
  46   writeFile outputFile cooc