src/Gargantext/RCT.hs

   1 {-|
   2 Module      : Gargantext.
   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 -}
  11
  12 {-# LANGUAGE NoImplicitPrelude #-}
  13
  14 module Gargantext.RCT where
  15
  16 import Gargantext.Prelude
  17
  18 foo :: Int
  19 foo = undefined
  20 --import Data.Text (Text, words)
  21 --import Data.Attoparsec.Text (anyChar, isEndOfLine, Parser, takeTill, many1, endOfLine, space, manyTill)
  22 --import Control.Applicative (many)
  23
  24 -- RCT is the acronym for Referential ConText (of Text)
  25 -- at the begin there was a byte
  26 -- then a char
  27 -- Char -> RCT [Char]
  28
  29 -- then a list of chars called a string, we call it a Form
  30 -- (removing all weird charachters which are not alphanumeric)
  31
  32 -- Form -> RCT Sentence
  33
  34 -- These forms compose the RCT Sentence
  35 -- an ngrams is composed with multiple forms
  36
  37 -- Paragraph = [Sentence]
  38
  39 -- type Title = Paragraph
  40 -- data Block = [Paragraph]
  41 -- Block is taken form Pandoc
  42
  43 -- data Document = [Block]
  44
  45 -- Set of databases
  46 -- Database
  47 -- Set of Articles
  48 --      Article
  49 --      Paragraph (abstract + title)
  50 -- Sentence - Ngrams - Forms
  51
  52
  53
  54 --separateurs :: Parser Text
  55 --separateurs = dropWhile isEndOfLine
  56
  57 --paragraphs :: Parser [Text]
  58 --paragraphs = many paragraph
  59 --
  60 --paragraph :: Parser Text
  61 --paragraph = takeTill isEndOfLine <* many1 endOfLine
  62 --
  63 -- forms :: Text -> [Text]
  64 -- forms = words
  65
  66