src/Gargantext/Text/Context.hs

   1 {-|
   2 Module      : Gargantext.Text.Context
   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 Context of text management tool
  11
  12 -}
  13
  14 {-# LANGUAGE NoImplicitPrelude #-}
  15 {-# LANGUAGE OverloadedStrings #-}
  16
  17 module Gargantext.Text.Context where
  18
  19 import Data.Text (Text, pack, unpack, length)
  20 import Data.String (IsString)
  21
  22 import Text.HTML.TagSoup
  23 import Gargantext.Text
  24 import Gargantext.Prelude hiding (length)
  25
  26 data SplitBy = Paragraph | Sentences | Chars
  27
  28
  29 splitBy :: SplitBy -> Int -> Text -> [Text]
  30 splitBy Chars     n = map pack . chunkAlong n n . unpack
  31 splitBy Sentences n = map unsentences . chunkAlong n n  . sentences
  32 splitBy Paragraph _ = map removeTag   . filter isTagText . parseTags
  33   where
  34     removeTag :: IsString p => Tag p -> p
  35     removeTag (TagText x) = x
  36     removeTag (TagComment x) = x
  37     removeTag _          = ""
  38
  39