src/Gargantext/RCT.hs

   1 module Gargantext.RCT where
   2
   3 import Gargantext.Prelude
   4
   5 foo :: Int
   6 foo = undefined
   7 --import Data.Text (Text, words)
   8 --import Data.Attoparsec.Text (anyChar, isEndOfLine, Parser, takeTill, many1, endOfLine, space, manyTill)
   9 --import Control.Applicative (many)
  10
  11 -- RCT is the acronym for Referential ConText (of Text)
  12 -- at the begin there was a byte
  13 -- then a char
  14 -- Char -> RCT [Char]
  15
  16 -- then a list of chars called a string, we call it a Form
  17 -- (removing all weird charachters which are not alphanumeric)
  18
  19 -- Form -> RCT Sentence
  20
  21 -- These forms compose the RCT Sentence
  22 -- an ngrams is composed with multiple forms
  23
  24 -- Paragraph = [Sentence]
  25
  26 -- type Title = Paragraph
  27 -- data Block = [Paragraph]
  28 -- Block is taken form Pandoc
  29
  30 -- data Document = [Block]
  31
  32 -- Set of databases
  33 -- Database
  34 -- Set of Articles
  35 --      Article
  36 --      Paragraph (abstract + title)
  37 -- Sentence - Ngrams - Forms
  38
  39
  40
  41 --separateurs :: Parser Text
  42 --separateurs = dropWhile isEndOfLine
  43
  44 --paragraphs :: Parser [Text]
  45 --paragraphs = many paragraph
  46 --
  47 --paragraph :: Parser Text
  48 --paragraph = takeTill isEndOfLine <* many1 endOfLine
  49 --
  50 -- forms :: Text -> [Text]
  51 -- forms = words
  52
  53
  54