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, here are logic of main types.
  11
  12 -}
  13
  14 {-# LANGUAGE NoImplicitPrelude #-}
  15 {-# LANGUAGE OverloadedStrings #-}
  16
  17 module Gargantext.Text.Context
  18   where
  19
  20 import Data.Text (Text, pack, unpack)
  21 import Data.String (IsString)
  22
  23 import Text.HTML.TagSoup (parseTags, isTagText, Tag(..))
  24 import Gargantext.Text
  25 import Gargantext.Prelude hiding (length)
  26
  27
  28 ------------------------------------------------------------------------
  29
  30 type Term = Text
  31 type MultiTerm = [Term]
  32 type Label = MultiTerm
  33
  34 type TermList = [(Label, [MultiTerm])]
  35
  36 type Sentence  a = [a] -- or a nominal group
  37 type Corpus    a = [Sentence a] -- a list of sentences
  38
  39 -- type ConText a = [Sentence a]
  40 -- type Corpus a = [ConText a]
  41
  42
  43 ------------------------------------------------------------------------
  44
  45 data SplitContext = Chars Int | Sentences Int | Paragraphs Int
  46
  47 tag :: Text -> [Tag Text]
  48 tag = parseTags
  49
  50 -- | splitBy contexts of Chars or Sentences or Paragraphs
  51 -- >> splitBy (Chars 0) "abcde"
  52 -- ["a","b","c","d","e"]
  53 -- >> splitBy (Chars 1) "abcde"
  54 -- ["ab","bc","cd","de"]
  55 -- >> splitBy (Chars 2) "abcde"
  56 -- ["abc","bcd","cde"]
  57 splitBy :: SplitContext -> Text -> [Text]
  58 splitBy (Chars     n)  = map pack        . chunkAlong (n+1) 1 . unpack
  59 splitBy (Sentences n)  = map unsentences . chunkAlong (n+1) 1 . sentences
  60 splitBy (Paragraphs _) = map unTag       . filter isTagText   . tag
  61   where
  62     unTag :: IsString p => Tag p -> p
  63     unTag (TagText x) = x
  64     unTag _           = ""
  65
  66