[PARSERS] Iramuteq file format parser

gargantext.cabal

@@ -5,7 +5,7 @@ cabal-version: 1.12
 -- see: https://github.com/sol/hpack
 
 name:           gargantext
-version: 0.0.6.9.9.1
+version:        0.0.6.9.9.1
 synopsis:       Search, map, share
 description:    Please see README.md
 category:       Data
@@ -192,6 +192,7 @@ library
       Gargantext.Core.Text.Corpus.Parsers.Date.Attoparsec
       Gargantext.Core.Text.Corpus.Parsers.FrameWrite
       Gargantext.Core.Text.Corpus.Parsers.GrandDebat
+      Gargantext.Core.Text.Corpus.Parsers.Iramuteq
       Gargantext.Core.Text.Corpus.Parsers.Isidore
       Gargantext.Core.Text.Corpus.Parsers.Json2Csv
       Gargantext.Core.Text.Corpus.Parsers.RIS

src/Gargantext/Core/Text/Corpus/Parsers/Iramuteq.hs

+{-|
+Module      : Gargantext.Core.Text.Corpus.Parsers.WOS
+Description : 
+Copyright   : (c) CNRS, 2017-Present
+License     : AGPL + CECILL v3
+Maintainer  : team@gargantext.org
+Stability   : experimental
+Portability : POSIX
+
+Here is a longer description of this module, containing some
+commentary with @some markup@.
+-}
+
+
+module Gargantext.Core.Text.Corpus.Parsers.Iramuteq (parseIramuteqFile, notices) where
+
+import Control.Applicative
+import Data.Attoparsec.ByteString (Parser, takeTill, parseOnly)
+import Data.Attoparsec.ByteString.Char8 (isEndOfLine, takeWhile, endOfLine)
+import Data.ByteString (ByteString)
+import Prelude hiding (takeWhile, take, concat, readFile, lines, concat)
+import qualified Data.ByteString as DB
+
+parseIramuteqFile :: String -> IO (Either String [[(ByteString, ByteString)]])
+parseIramuteqFile fp = do
+  txts <- DB.readFile fp
+  pure $ parseOnly notices txts
+
+-------------------------------------------------------------
+notices :: Parser [[(ByteString, ByteString)]]
+notices = do
+  ns <- (many notice)
+  pure ns
+
+notice :: Parser [(ByteString, ByteString)]
+notice = do
+    hs <- headers
+    ns <- takeWhile (/= '*')
+    pure $ hs <> [("text", ns)]
+
+-----------------------------------------------------------------
+headers :: Parser [(ByteString, ByteString)]
+headers = parseOf header fields
+
+header :: Parser ByteString
+header = "**** " *> takeTill isEndOfLine <* endOfLine
+
+-----------------------------------------------------------------
+fields :: Parser [(ByteString, ByteString)]
+fields = many (parseOf field fieldTuple)
+
+field :: Parser ByteString
+field =   "*" *> takeWhile (/= ' ') <* " "
+      <|> "*" *> takeWhile (/= '\n')
+
+fieldTuple :: Parser (ByteString, ByteString)
+fieldTuple = do
+  name <- takeWhile (/= '_') <* "_"
+  rest <- takeWhile (/= '\n')
+  pure (name,rest)
+
+-----------------------------------------------------------------
+constP :: Parser a -> ByteString -> Parser a
+constP p t = case parseOnly p t of
+  Left _ -> empty
+  Right a -> return a
+
+parseOf :: Parser ByteString -> Parser a -> Parser a
+parseOf ptxt pa = bothParse <|> empty
+  where
+    bothParse = ptxt >>= constP pa
+
+

src/Gargantext/Core/Text/Corpus/Parsers/RIS.hs

@@ -16,7 +16,6 @@ citation programs to exchange data.
 
 -}
 
-
 module Gargantext.Core.Text.Corpus.Parsers.RIS (parser, onField, fieldWith, lines) where
 
 import Data.List (lookup)