blob: 384687a6ae3a7a29737440eb43b209a52b490619 (
plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
|
{-# LANGUAGE FlexibleContexts #-}
{-# LANGUAGE ScopedTypeVariables #-}
{- |
Module : Text.Pandoc.Readers.RST
Copyright : Copyright (C) 2006-2020 John MacFarlane
License : GNU GPL, version 2 or above
Maintainer : John MacFarlane <jgm@berkeley.edu>
Stability : alpha
Portability : portable
Conversion from CSV to a 'Pandoc' table.
-}
module Text.Pandoc.Readers.CSV ( readCSV ) where
import Data.Text (Text)
import qualified Data.Text as T
import Text.Pandoc.CSV (parseCSV, defaultCSVOptions)
import Text.Pandoc.Definition
import qualified Text.Pandoc.Builder as B
import Text.Pandoc.Class (PandocMonad)
import Text.Pandoc.Shared (crFilter)
import Text.Pandoc.Error
import Text.Pandoc.Options (ReaderOptions)
import Control.Monad.Except (throwError)
readCSV :: PandocMonad m
=> ReaderOptions -- ^ Reader options
-> Text -- ^ Text to parse (assuming @'\n'@ line endings)
-> m Pandoc
readCSV _opts s =
case parseCSV defaultCSVOptions (crFilter s) of
Right (r:rs) -> return $ B.doc $ B.table capt
(zip aligns widths)
(TableHead nullAttr hdrs)
[TableBody nullAttr 0 [] rows]
(TableFoot nullAttr [])
where capt = B.emptyCaption
numcols = length r
toplain = B.simpleCell . B.plain . B.text . T.strip
toRow = Row nullAttr . map toplain
toHeaderRow l = if null l then [] else [toRow l]
hdrs = toHeaderRow r
rows = map toRow rs
aligns = replicate numcols AlignDefault
widths = replicate numcols ColWidthDefault
Right [] -> return $ B.doc mempty
Left e -> throwError $ PandocParsecError s e
|