src/Data/Gargantext/RCT.hs

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